**A Narrative Review of the Molecular Epidemiology and Laboratory Surveillance of Vaccine Preventable Bacterial Meningitis Agents:** *Streptococcus pneumoniae***,** *Neisseria meningitidis***,** *Haemophilus influenzae* **and** *Streptococcus agalactiae*

**Raymond S. W. Tsang**

Laboratory for Vaccine Preventable Bacterial Diseases, National Microbiology Laboratory, Public Health Agency of Canada, 1015 Arlington Street, Winnipeg, MB R3E 3R2, Canada; raymond.tsang@canada.ca; Tel.: +1-204-789-6020

**Abstract:** This narrative review describes the public health importance of four most common bacterial meningitis agents, *Streptococcus pneumoniae*, *Neisseria meningitidis*, *Haemophilus influenzae*, and *S. agalactiae* (group B *Streptococcus*). Three of them are strict human pathogens that normally colonize the nasopharynx and may invade the blood stream to cause systemic infections and meningitis. *S. agalactiae* colonizes the genito-gastrointestinal tract and is an important meningitis agent in newborns, but also causes invasive infections in infants or adults. These four bacteria have polysaccharide capsules that protect them against the host complement defense. Currently licensed conjugate vaccines (against *S. pneumoniae*, *H. influenza*, and *N. meningitidis* only but not *S. agalactiae*) can induce protective serum antibodies in infants as young as two months old offering protection to the most vulnerable groups, and the ability to eliminate carriage of homologous serotype strains in vaccinated subjects lending further protection to those not vaccinated through herd immunity. However, the serotype-specific nature of these vaccines have driven the bacteria to adapt by mechanisms that affect the capsule antigens through either capsule switching or capsule replacement in addition to the possibility of unmasking of strains or serotypes not covered by the vaccines. The post-vaccine molecular epidemiology of vaccine-preventable bacterial meningitis is discussed based on findings obtained with newer genomic laboratory surveillance methods.

**Keywords:** bacterial meningitis; *S. pneumoniae*; *N. meningitidis*; *H. influenzae*; *S. agalactiae*; conjugate vaccines; post-vaccine surveillance

### **1. Introduction**

Pyogenic bacterial meningitis is a life threatening condition that can progress rapidly leading to death. When the disease happens in infants, children, and young adults, it may instill fear due to the contagious and potentially deadly nature of the disease especially in outbreak situation. The three most common causes of acute bacterial meningitis are *Streptococcus pneumoniae*, *Neisseria meningitidis*, and *Haemophilus influenzae* [1]. This group of bacterial meningitis agents can cause disease in all ages of life from newborn to the elderly. The global burden of meningitis disease in 2016 was estimated to be 2.82 million cases, and 318,400 deaths were attributed to meningitis. The three most common pathogens (*S. pneumoniae*, *N. meningitidis*, and *H. influenzae*) were responsible for 55.7% and 57.2% of the meningitis cases and deaths, respectively [2]. Besides meningitis, *S. pneumoniae*, *H. influenzae*, and *N. meningitidis* can cause other forms of invasive diseases such as bacteremic pneumonia, septicemia, septic arthritis, pericarditis, etc. The risk of developing a major (such as hearing loss, seizures, motor deficit, cognitive impairment, hydrocephalus, and visual disturbance) or a minor (learning difficulties, language impairment, developmental de-


**Citation:** Tsang, R.S.W. A Narrative Review of the Molecular Epidemiology and Laboratory Surveillance of Vaccine Preventable Bacterial Meningitis Agents: *Streptococcus pneumoniae*, *Neisseria meningitidis*, *Haemophilus influenzae* and *Streptococcus agalactiae*. *Microorganisms* **2021**, *9*, 449. https://doi.org/10.3390/microorganisms9 020449

Academic Editor: James Stuart

Received: 11 January 2021 Accepted: 16 February 2021 Published: 22 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

lay) sequela from bacterial meningitis was estimated to be 12.8% and 8.6%, respectively [3]. Meningitis caused by *S. pneumoniae* carried the highest risk with a major sequela (24.7%), followed by *H. influenzae* (9.5%) and *N. meningitidis* (7.2%) [3]. Using meningococcal disease (which carries the lowest risk of developing a major sequela) as an example, the cost to care for a case who developed a major sequela was estimated to be £160,000 (US\$214,096) to £200,000 (US\$267,620) for the first year alone; and the corresponding figure over the lifetime of a case may be as high as £590,000 (US\$789,479) to £1,090,000 (US\$1,458,529) [4]. Since the incidence of meningitis and the risk of developing sequela are much higher in low- and middle-income countries, and the resources to care for those meningitis patients who develop severe sequela are often lacking in these countries, vaccines are probably the most cost-effective strategy for the control and potentially elimination of this devastating and fearful disease.

Although a number of other bacterial agents can cause meningitis, such as *Listeria monocytogenes*, *Escherichia coli,* and other enteric bacteria, group B *Streptococcus* (*S. agalactiae*) is gaining attention as a frequent cause of either early or late onset of invasive diseases such as pneumonia, sepsis, or meningitis in the newborn [5,6] as well as various forms of invasive diseases in pregnant women and non-pregnant adults [5,7]. The World Health Organization (WHO) has also identified group B *Streptococcus* together with *S. pneumoniae*, *N. meningitidis*, and *H. influenzae* as the four major bacterial meningitis agents to be included in its work plan and global vision to defeat meningitis by 2030 [8].

Capsule-based protein-conjugate vaccines that target the major serogroups of *N. meningitidis* and serotypes of *H. influenzae* and *S. pneumoniae* causing invasive diseases are now available and implemented in vaccination programs in many countries [9–11]. As a result, the epidemiology of bacterial meningitis has changed with the number of cases caused by strains covered by the vaccine decreased dramatically but at the same time disease due to serogroups or serotypes of the pathogens not included in the vaccine has emerged [12]. Since disease surveillance has been described by the WHO as one of the five major pillars on the road map to defeat meningitis [8], the objectives of this report are to describe (i) features of *S. pneumoniae*, *N. meningitidis*, *H. influenzae,* and *S. agalactiae* that may have implications for vaccination and surveillance; (ii) currently licensed vaccines against *S. pneumoniae*, *N. meningitidis*, and *H. influenzae*; (iii) changes in the epidemiology of invasive diseases caused by these three pathogens; (iv) traditional and newer laboratory surveillance methods; and (v) how lessons learned from surveillance of the three most common bacterial meningitis agents can inform the pre- and post-vaccine licensure surveillance of invasive group B *Streptococcus* (GBS) disease when capsule polysaccharide conjugate vaccines against GBS have been developed and are in clinical trials [5,13].

#### **2. Characteristics of** *S. pneumoniae***,** *N. meningitidis***,** *H. influenzae***, and** *S. agalactiae* **Important for Vaccination and Surveillance**

*S. pneumoniae*, *N. meningitidis*, and *H. influenzae* are respiratory pathogens that normally colonize the human respiratory tract where they serve as a reservoir of infection [14–16]. Another common characteristic of these three invasive bacterial agents is the polysaccharide capsules on their cell surface, which serve as serotyping antigens. The serotypes are traditionally identified by anti-capsular antibodies using agglutination methods (or the Quellung reaction for *S. pneumoniae*). The capsules also serve as protective antigens shielding the bacteria from the human host defense like phagocytosis and complement activation [17,18]. As the protective antigen, vaccines based on the capsule have been developed to target the most common serotypes of *H. influenzae*, *N. meningitidis*, and *S. pneumoniae* causing invasive infections [9–11]. Another feature that makes these bacteria successful pathogens is the plasticity of their genome and their recombinant nature [19–21].

Unlike *S. pneumoniae*, *N. meningitidis*, and *H. influenzae*, *S. agalactiae* colonizes the human genito-gastrointestinal tract. Not only does it cause meningitis in the newborn and various forms of invasive diseases in infants and adults, *S. agalactiae* is also known to cause disease in cattle [22,23] and may have the potential to transmit to human as a zoonotic pathogen [24]. Similar to *S. pneumoniae*, *N. meningitidis*, and *H. influenzae*, *S. agalactiae* also has a surface polysaccharide capsule that acts as virulence factor and protective antigen [5,13]. Its genome is also prone to participate in recombination events [25].

#### **3. Currently Licensed Vaccines for Control of Bacterial Meningitis**

Currently there are 6 serotypes of *H. influenzae* recognized [26], 10 serotypes of *S. agalactiae* [5,13], 12 serogroups for *N. meningitidis* [27], and 100 serotypes for *S. pneumoniae* [28,29] (Table 1). Non-encapsulated strains also exist in all three species, and are termed nontypeable (for *H. influenzae* and *S. pneumoniae*) or non-groupable (for *N. meningitidis*). Currently licensed vaccines to control some strains of *S. pneumoniae*, *N. meningitidis*, and *H. influenzae* are listed in Table 2, while vaccines for protection against *S. agalactiae* are not licensed yet but are in advanced stages of clinical trials for maternal immunization [30,31].

**Table 1.** Capsular antigens of *Haemophilus influenzae*, *Neisseria meningitidis*, *Streptococcus agalactiae*, and *S. pneumoniae*.


**\*** 100 different serotypes identified, please see references for full list.

**Table 2.** Licensed vaccines \* against *Haemophilus influenzae*, *Neisseria meningitidis* and *Streptococcus pneumoniae.*


\* Except for the 4 component MenB and the factor H binding protein vaccines, all the other vaccines described in this Table are polysaccharide based vaccines. † protein carriers include TT (tetanus toxoid), DT (diphtheria toxoid), CRM<sup>197</sup> (mutant diphtheria toxoid), OMP (outer membrane protein of *N. meningitides*).

The first bacterial meningitis vaccine developed was solely polysaccharide-based vaccine against serotype b *H. influenzae* (Hib) but it was soon discovered that plain polysaccharide vaccines are T-cell-independent antigens and do not induce protective antibodies in infants less than two years old [32], the most vulnerable age group for developing meningitis and invasive disease [33,34]. Coupling of the capsular polysaccharide to a protein carrier converts the vaccine to a T-cell dependent antigen that induces protective antibodies in infants as young as two months of age [10]. Another characteristic of the capsular polysaccharide vaccines is they are serotype-specific and offer protection against infection by the homologous serotype and do not offer protection against heterologous serotypes. Besides preventing invasive infections, the conjugate vaccines also reduce or eliminate respiratory carriage of and hence offer herd immunity to the larger community for the serotypes of these pathogens included in the vaccines [35–37]. The fact that conjugate vaccines are serotype-specific and can eliminate nasopharyngeal carriage of the homologous serotypes means their protective coverage is limited to the serotypes included in the vaccine and they can also alter the bacterial flora in the nasopharynx of vaccinated subjects.

The choice of which serotypes or serogroups to be included in the vaccines are based on the fact that not all serotypes or serogroups are equally virulent nor have the same prevalence in causing invasive diseases. For example, animal infection with isogenic mutants of *H. influenzae* that expressed different capsule serotype antigens has shown that serotype b is the most virulent, followed by serotype a [38]. In addition, most *N. meningitidis* isolates recovered from normally sterile body sites of invasive meningococcal disease (IMD) patients belong to six of the 12 recognized serogroups (A, B, C, W, X, and Y) [39,40]. Before the introduction of pneumococcal conjugate vaccines (PCVs), 10 serotypes (1, 4, 5, 6A 6B, 14, 18C, 19A, 19F, and 23F) were responsible for at least 50% of all invasive pneumococcal disease isolates from six different parts of the world; and in one region, they were responsible for over 80% of their invasive pneumococci [41].

Even before vaccine introduction, temporal and geographical variations in the serogroups of *N. meningitidis* responsible for IMD is well documented [39,40]. Differences in the serotypes involved in invasive pneumococcal disease (IPD) have also been reported from different parts of the world [42,43]. Before Hib conjugate vaccines were introduced, most invasive *H. influenzae* diseases were caused by Hib [33,34].

#### **4. Effects of Vaccine Pressure, and Immune and/or Antibiotic Selection**

Since currently licensed conjugate vaccines against *S. pneumoniae*, *N. meningitidis*, and *H. influenzae* do not offer universal coverage for all the serotypes or serogroups, immune pressure and selection against these pathogens can be expected to happen in their natural habitat as well as in the serotypes and serogroups that will cause invasive disease in the post-vaccine period.

In the presence of vaccine pressure, these bacterial pathogens may evolve to adapt by mainly two mechanisms that affect their capsule antigens (the vaccine targets): Capsule or serotype switching, and capsule or serotype replacement. Capsule switching involves two strains of a species exchanging their capsule polysaccharide synthesis (*cps*) genes resulting in a swap of their capsule antigens. For example, as shown in Figure 1a, a strain of genetic linage 1 and with a vaccine type capsule (depicted in green) exchanges its *cps* genes with a strain of genetic lineage 2 and with a non-vaccine capsule type (depicted in red). The end result will be the genetic linage 1 strain now carries *cps* genes for non-vaccine capsule type and expresses the non-vaccine capsule (red); while conversely the genetic linage 2 strain now expresses vaccine type capsule (green). Both *S. pneumoniae* and *N. meningitidis* have been reported to have capsule switching occurring spontaneously in the absence of vaccine pressure or, i.e., such capsule switching events have been reported prior to conjugate vaccine introduction [44,45]. Capsule switched strains can also be selected for by vaccine induced immune pressure and/or by wide spread antibiotic use if the capsule switched strain carries antibiotic resistance genes [41]. After capsule switching, the recipient strain will retain its original genetic background (usually determined by multi-

locus sequence typing) [46] but expresses a different, e.g., non-vaccine type of capsule. Frequent capsule switch in *N. meningitidis* from serogroup C to serogroup B, if it happens in a hypervirulent clone like ST-11, may be problematic since there are no capsule-based serogroup B meningococcal vaccines and protein-based meningococcal vaccines against serogroup B may not provide universal coverage against all serogroup B strains [47]. sule. Frequent capsule switch in *N. meningitidis* from serogroup C to serogroup B, if it happens in a hypervirulent clone like ST-11, may be problematic since there are no capsule-based serogroup B meningococcal vaccines and protein-based meningococcal vaccines against serogroup B may not provide universal coverage against all serogroup B strains [47].

the capsule switched strain carries antibiotic resistance genes [41]. After capsule switching, the recipient strain will retain its original genetic background (usually determined by multi-locus sequence typing) [46] but expresses a different, e.g., non-vaccine type of cap-

*Microorganisms* **2021**, *9*, x FOR PEER REVIEW 5 of 19

**Figure 1.** (**a**) Depiction of capsule switching between a genetic lineage 1\* strain with a vaccine-type capsule (colored green) and a genetic lineage 2\* strain with a non-vaccine type capsule (colored red). Diagram based on information from Swartley et al. [48]. (**b**) Illustration of capsule replacement when the vaccine capsule type (colored green) strain of genetic lineage 1\* is removed by the vaccine, leaving the strain of genetic lineage 2\* with non-vaccine capsule (colored red) to remain and proliferate. Diagram based on information from Lipstich M [49]. (**c**) Another scenario of capsule replacement when strain of genetic lineage 1\* with both vaccine (colored green) and nonvaccine (colored red) capsule types are present before vaccine introduction and after vaccine use, only the non-vaccine capsule type of genetic lineage 1\* strain remains. Diagram based on information from Lipstich M [49]. **Figure 1.** (**a**) Depiction of capsule switching between a genetic lineage 1\* strain with a vaccine-type capsule (colored green) and a genetic lineage 2\* strain with a non-vaccine type capsule (colored red). Diagram based on information from Swartley et al. [48]. (**b**) Illustration of capsule replacement when the vaccine capsule type (colored green) strain of genetic lineage 1\* is removed by the vaccine, leaving the strain of genetic lineage 2\* with non-vaccine capsule (colored red) to remain and proliferate. Diagram based on information from Lipstich M [49]. (**c**) Another scenario of capsule replacement when strain of genetic lineage 1\* with both vaccine (colored green) and non-vaccine (colored red) capsule types are present before vaccine introduction and after vaccine use, only the non-vaccine capsule type of genetic lineage 1\* strain remains. Diagram based on information from Lipstich M [49].

Capsule replacement happens when strains with vaccine capsule types that used to inhabit the nasopharynx have been eliminated by the conjugate vaccines and are now being replaced by strains expressing the non-vaccine capsule types. As a result, strains of non-vaccine capsule types may increase in prevalence and eventually cause disease. Strains with vaccine capsule type (depicted as green) and non-vaccine capsule type (depicted as red) can co-exist prior to the use of conjugate vaccines and the two capsule types may be of the different (Figure 1b) or the same (Figure 1c) genetic lineage. After selection by vaccine pressure, only strains of the non-vaccine capsule type (red) remain and expand Capsule replacement happens when strains with vaccine capsule types that used to inhabit the nasopharynx have been eliminated by the conjugate vaccines and are now being replaced by strains expressing the non-vaccine capsule types. As a result, strains of nonvaccine capsule types may increase in prevalence and eventually cause disease. Strains with vaccine capsule type (depicted as green) and non-vaccine capsule type (depicted as red) can co-exist prior to the use of conjugate vaccines and the two capsule types may be of the different (Figure 1b) or the same (Figure 1c) genetic lineage. After selection by vaccine pressure, only strains of the non-vaccine capsule type (red) remain and expand to fill the void that used to be occupied by strains of the vaccine type (Figure 1b). If strains of the same genetic lineage that expressed both vaccine and non-vaccine capsule types exist prior to vaccine use, then strains of the same genetic lineage still persist after vaccine use but they only carry the non-vaccine capsule type (Figure 1c). The phenomenon illustrated in

Figure 1b,c is sometimes referred to as "unmasking" versus true replacement when only the vaccine capsule type exists in the nasopharynx before vaccine use, and after vaccine use, the vaccine capsule type strain is removed and the void in the nasopharynx is being replaced by a non-vaccine capsule type strain. Strains of non-vaccine capsule types can also be further selected by widespread antibiotic use if they carry the corresponding antibiotic resistance genes [41].

#### **5. Molecular Epidemiology of Invasive Pneumococcal Disease (IPD), Invasive Meningococcal Disease (IMD), and Invasive** *H. influenzae* **Disease in the Post Conjugate Vaccine Era**

#### *5.1. IPD in the Post PCV Era*

Although PCVs are very effective in reducing the burden of IPD caused by vaccine serotypes in many countries, IPD due to non-vaccine serotypes is still a concern. The emergence of IPD due to non-vaccine serotypes may be due to either unmasking effect when vaccine serotypes are removed from their natural habitat of the human nasopharynx, allowing non-vaccine serotypes (which already exist) to expand and occupy the nasopharynx; or by replacement due to the non-vaccine serotypes which do not exist in the nasopharynx prior to PCV introduction but emerge to fill the void in the nasopharynx left behind by the vaccine type [50]. For example, prior to PCV7 introduction, serotypes 19A (a non-PCV7 vaccine serotype) existed at a level of about 7.5% in 2001 and increased to 16% in 2007 after PCV7 was introduced in 2000, before declining to about 3% in 2014 after PCV13 was introduced in 2010 [51]. The other mechanism responsible for the emergence of non-vaccine serotypes is genetic recombination between strains leading to capsule switching. This was illustrated in the emergence of some serotype 19A strains after PCV7 introduction by a genetic recombination between a vaccine covered serotype 4, sequence type (ST)-595 recipient strain with a donor strain of non-vaccine serotype 19 ST-199, providing the recipient ST-595 strain with the non-vaccine capsule serotype 19A [52]. Another study also demonstrated that serotype 19A may again escape the PCV13 selection by a further switch to serotype 15B [53]. Although in this later study the genetic recombination occurred in strains from pre-PCV7 period; nevertheless the mechanism of genetic recombination with a non-vaccine capsule type is present in pneumococci. The pneumococcal capsule locus is a hotspot for mutation including exhibiting a higher rate of genetic recombination compared to the rest of the pneumococcal genome [54]. However, pneumococcal capsule locus recombination that leads to capsule serotype switch does not appear to be random. For example, capsule switch between strains within a serogroup occurred more often than serotype switch involving strains between different serogroups [55]. Since many factors may govern the pneumococcal population structure and the associated serotypes, some have suggested the existence of epistatic factor contributing to the dynamic of the pneumococcal capsule genetics [55,56].

Another mechanism may explain the persistence of some vaccine serotype in the post PCV period. For example, serotype 3 (included in the PCV13 vaccine) persisted in the nasopharyngeal samples as well as in specimens from IPD patients despite PCV13 usage [51,57,58]. Genome sequencing of serotype 3 isolates obtained prior to and after introduction of PCV13 showed a different clade of serotype 3 has emerged in the post PCV13 period despite the fact that both pre- and post-PCV13 isolates were typed by MLST to belong to the same ST-180 clonal complex (CC). However, the new clade has been shown to have sub-capsular protein antigen changes, which could explain strains of the new clade have adapted to exist despite the presence of immunity induced by PCV13 [57].

Regardless of the vaccine escape mechanism, various non-PCV serotypes have emerged in places where PCV immunization programs have been implemented reflecting geographical differences in serotype prevalence and distribution [59]. Non-PCV serotypes like serotype 2, 8, 10A, 11A, 12F, 15A, 15B/C. 16F, 22F, 24F, 33F, and 35B/D, have been described as causes of IPD [41,59–63]. To deal with this increase in non-PCV serotypes, 15-valent and 20-valent PCVs have been developed and are now in early clinical trials [64,65]. However, in the post PCV era, predominance by a single or a few serotypes as causes of IPD was not

observed. Instead, increase serotype diversity of invasive pneumococci recovered from IPD cases has been observed [66], which may challenge the usefulness of increasing the valency of PCVs. Two editorials in 2007, "Invasive pneumococcal disease, the target is moving" [67] and "Serotype replacement in invasive pneumococcal disease: where do we go from here?" [68] appear to be just as relevant today after two decades of PCV use. Indeed, expert comments in 2021 still wrestle with the changing epidemiology of IPD due to shifting serotypes, and identify continuous surveillance as an important function in the control of IPD [69,70]. Ideally, a pan-pneumococcal universal vaccine would solve the problem of chasing after the emergence of non-vaccine serotypes as causes of IPD.

#### *5.2. IMD in the Post Conjugate Vaccine Era*

In the US, quadrivalent (A, C, W, Y) meningococcal conjugate vaccine was licensed in 2005 and recommended for the 11 to 18 years age group [71]. In the post-quadrivalent conjugate vaccine period of 2006-2010, no capsule or serogroup replacement was detected [72]. In the Canadian province of Quebec, outbreak due to a serogroup B strain of ST-269 appeared in 2013 [73] after two rounds of province wide vaccination against serogroup C meningococci (MenC) (first with plain polysaccharide vaccine in 1992-1993 and then with the MenC-conjugate vaccine in 2011) for control of outbreaks due to the hyper-virulent strain of ET-15 (ST-11) [74,75].

In Europe, serogroup B *N. meningitidis* was responsible for most IMD (73.6% in 2011) while an increase in serogroup Y IMD has been reported in a number of European countries [40]. Beginning in 2013, an increase in IMD due to serogroup W meningococci (MenW) has been reported from across Europe with both incidence rates of disease and the proportion of IMD isolates due to serogroup W showing yearly increase [76]. This increase in MenW disease in Europe was due to the introduction of a new ST-11 strain (different from the Hajj strain, which emerged during the 2000 Hajj pilgrimage in Saudi Arabia) from South America into the UK with further diversification to the 2013 UK strain, which spread through Europe [76,77]. This new MenW strain has also been reported to cause an increase in IMD in both Australia and Canada [78,79]. Expansion of a penicillin-resistant MenW ST-11 clone has also been described [80].

Before introduction of the monovalent meningococcal serogroup A conjugate vaccine, MenAfriVac, in 2010 [81], most meningococcal epidemics in Africa were mostly caused by serogroup A *N. meningitidis* (MenA). However, a serogroup X meningococcus (MenX) epidemic was reported from southwest Niger in 2004-2006 [82], and subsequently MenX outbreaks had occurred in Burkina Faso, Niger, Togo, and Uganda [83]. The MenX outbreak strain has been characterized as ST-181 CC with high experimental animal pathogenicity [84]. After introduction of the MenAfriVac, epidemics due to serogroups C, W, and X meningococci have been reported in the African meningitis belt countries. [85,86]. The MenC strain appeared to be a new strain typed as ST-10217, which has been shown to have arisen from a nongroupable strain recovered from a healthy carrier in Burkina Faso in 2012 prior to the emergence of the MenC ST-10217 and the MenC outbreak in 2013 [87]. The MenW strain causing outbreaks in Africa has been studied, and it appeared to be related to, and to have diversified from the 2000 Hajj strain [88].

Longitudinal carriage studies have been carried out in Africa to understand the epidemic nature of meningococcal meningitis in recent years both before and after the introduction of MenAfriVac in 2010. In the study carried out in one district of northern Ghana over the period of 1998 to 2005 before MenAfriVac was introduced, it was found that the colonized meningococcal population changed with time and matched temporally with the strain causing epidemics in the region [89]. Three successive waves of colonized meningococci were observed with ST-5 MenA, followed by ST-751 MenX, and ST-7 MenA. In the study to assess the effect of immunization with MenAfriVac, the carriage study has shown that the vaccine was both effective in control of MenA disease and in elimination of MenA from the respiratory tract of healthy carriers up to six or seven years after vaccine introduction. Like the other longitudinal carriage study in Ghana, a small percentage of

oropharyngeal samples contained MenW of ST-11 CC (0.48%) and MenC of ST-10217 CC (0.10%) [90]. These studies certainly pointed to the importance of meningococci in healthy carriers as contributors of infection and potential sources of epidemics; and that conjugate vaccination may further change the population of meningococci in the normal habitat of the human upper respiratory tract.

#### *5.3. Invasive H. influenzae Disease in the Post Hib Conjugate Vaccine Era*

Following introduction of the Hib conjugate vaccine in the early 1990s, the epidemiology of invasive *H. influenzae* disease in those countries with Hib vaccination programs have changed substantially in the past three decades. Non-typeable or non-encapsulated *H. influenzae* (NT-Hi) is now the most frequent cause of invasive *H. influenzae* disease worldwide [91]. In Europe, during the period of 2007 to 2014, NT-Hi was the most common type identified but 74.1% of their invasive encapsulated *H. influenzae* were typed as serotype f, followed by serotype e (21.4%) [92]. In contrast, in the U.S., although NT-Hi is also the most common cause of invasive *H. influenzae* disease, the incidence of serotype a invasive *H. influenzae* disease has increased by 13% annually during the period of 2002 to 2015 [93].The incidence of invasive disease caused by NT-Hi has increased by 3% annually while incidence of invasive *H. influenzae* disease due to other serotypes was either stable or decreasing. The global presence of serotype a *H. influenzae* (Hia) has been documented [94,95], and the severity of invasive Hia disease has been described [96–98] which called for a Hia vaccine development [99].

Genetic analysis of Hia has revealed a population biology very similar to Hib, i.e., (a) with two phylogenetic populations similar to the clonal divisions I and II descried for Hib; and (b) with most invasive Hia isolates clustered together in a phylogenetic population (named clonal division I as for the majority of invasive Hib strains), represented by isolates typed by MLST as ST-23 and many STs related to ST-23 as single, double, or triple locus variants [100,101]. Another clone within this larger genetic population of clonal division I and identified by MLST as ST-4 has been reported in Brazil to be associated with more severe disease and higher case fatality rate [102]. In contrast to clonal division I Hia, clonal division II Hia is rarely isolated from invasive disease cases in Canada [101] and has not been found associated with invasive disease in Alaska [100]. However, clonal division II Hia identified by MLST as ST-62 has been found in 75% (21/28) of the Hia invasive disease case isolates from children < 18 years old in Utah, United States [103]. This may suggest unique geographical distribution of Hia genotypes.

To understand the emergence of NT-Hi as a cause of invasive disease in the post Hib conjugate vaccine era, comparative genome studies have revealed that NT-Hi showed much higher genetic diversity when compared to Hib or other serotypes that have been regarded as more genetically conserved or clonal [104,105]. Non-encapsulated *S. pneumoniaae* have also been reported to have higher genetic diversity probably as a result of higher rates of genetic recombination as the capsule may serve as a barrier for foreign DNA uptake [21,106]. The higher genetic diversity of non-encapsulated *H. influenzae* may offer better adaptation to the host, e.g., by evading host immunity.

#### **6. Laboratory Surveillance of** *S. pneumoniae***,** *N. meningitidis***, and** *H. influenzae*

It is with this background of a changing microbial ecology in the nasopharynx of the human host as the bacterial pathogens are adapting to the vaccine pressure that laboratory surveillance of invasive bacterial meningitis pathogens are becoming increasingly important as well as challenging. The molecular typing methods for outbreak detection and surveillance of IMD, IPD, and invasive *H. influenzae* disease have been reviewed a decade ago with a focus on DNA sequencing methods [107]. They can be briefly summarized below as:

(1) Serogrouping and serotyping by the conventional method of using antisera to detect the capsular antigens, through either bacterial agglutination or Quelling reactions, has the value of detecting expression of the capsule antigens albeit the method may

sometimes be inaccurate [108]. Molecular method like PCR has been introduced to improve the detection and identification of serotypes, including that for *S. agalactiae* [109–112].

(2) Clonal analysis by multilocus sequence typing (MLST) is available for *S. pneumoniae*, *N. meningitidis,* and *H. influenzae* [46,113,114]. Strains are typed as STs and related STs are grouped together to form a CC.

(3) Target gene sequencing for fine typing:

The following genes have been proposed for typing of meningococci: *fetA,* which encodes an iron-regulated outer membrane protein; *porA*, which encodes the class 1 outer membrane protein PorA; *porB*, which encodes the class 2/3 outer membrane protein PorB [107]; and the newer protein-based MenB vaccine target genes, *fHbp*, *nhba*, and *nadA*, which encode for factor H binding protein, Neisseria heparin binding antigen, and the Neisseria adhesion A, respectively [115,116].

For *S. pneumoniae*, the *ply* and *lytA* genes, which encode for the pneumolysin and the autolysin, respectively; as well as the *pspA* gene, which encodes for the pneumococcal surface protein A, have been proposed as targets for potential typing purposes [107]. PspA has been reported to be associated with virulence and invasiveness of pneumococci [117]. Other gene markers associated with virulence have been suggested, including *pspC*, which encodes for the pneumococcal choline-binding protein C (PspC), for association with invasiveness of strains [117]. The *slaA* gene, which encodes for phospholipase A2, and four contiguous genes, one of which predicted as *pblB* that encodes a prophage tail protein, were either associated with the clinical disease of meningitis, or 30-day mortality rate, respectively [118].

The following genes have been proposed for typing of *H. influenzae*: *ompP2*, *ompP5*, *hmw1*, and *hmw2* [107]. The *omp2* and *omp5* genes encode for two different outer membrane proteins, a porin and a OmpA family protein, respectively. The *hmw1* and *hmw2* encode for HMW1 and HMW2, which are surface adhesion proteins (HMW stands for high molecular weight). A number of potential vaccine candidates have also been identified and they may have potential as further typing targets [119].

(4) Antibiotic susceptibility profile: Antibiogram can serve as a typing tool but more usefully in direct patient care as well as for surveillance purpose. Testing can be done by the disk diffusion method or quantitatively by the dilution assays (broth or agar dilution methods). Guidelines for the testing methods including the culture media, classes of antibiotics to be tested, as well as the interpretation of results have been published by both The European Committee on Antimicrobial Susceptibility Testing (EUCAST) [120] and the Clinical Laboratory Standards Institute's (CLSI's) Subcommittee on Antimicrobial Susceptibility Testing (AST) [121].

Besides the phenotypic methods, genetic prediction of antibiotic susceptibility has also been described by Harrison et al. [107]. Of the genes associated with decreased susceptibility or resistance to different types of antibiotics, the penicillin binding protein genes of *S. pneumoniae* that determines susceptibility towards penicillin are of special interest. The *cps* locus of *S. pneumoniae* is flanked by two of the penicillin binding protein genes, *pbp1a* and *pbp2x*; their juxtaposition sometimes allow capsule switching and transfer of the penicillin resistant genes to occur in a single recombination event.

#### **7. Whole Genome Sequencing (WGS) for Molecular Epidemiology and Genomic Surveillance of Vaccine-Preventable Bacterial Meningitis Agents**

For quite some years, MLST has been proven useful to classify isolates into clonal types and it has been applied to identify hypervirulent clones [122,123] and capsule switching events between serogroups or serotypes [44,45]. However, intra-clonal variations have been described, which may have implications in our understanding of the changing epidemiology of these vaccine-preventable diseases [57,117,124].

With the first bacterial genome sequenced and published in 1995 [125], there has been a very rapid development over the last two decades in sequencing technologies that include cost reduction as well as web-based bioinformatics platforms and pipelines to assemble and analyze genome sequences. As such, genome sequencing has now become a standard laboratory tool to study microbes. Many of our current understanding of the molecular epidemiology of a number of infectious diseases, including the common bacterial meningitis agents of *S. pneumoniae*, *N. meningitidis*, *H. influenzae,* and *S, agalactiae* are based on data obtained by whole-genome sequencing projects.

However, to make WGS a routine laboratory tool for the global surveillance of vaccine-preventable bacterial meningitis, additional work may still be required on standardization and harmonization of methodology, data analysis and nomenclature on top of the issues of data ownership, and confidentiality. A global partnership to study the genomes of *S. pneumoniae* has published an international definition of pneumococcal lineage [126]. For *N. meninigitidis*, nomenclature is currently based on historical convention [127] and an international committee is responsible for naming clonal complexes ( https://pubmlst.org/organisms/neisseria-spp/further-info (accessed on 15 December 2020)). Similar development for *H. influenzae* appears to be lacking for now. Traditional analysis of the population biology of encapsulated *H. influenzae* divided them into two clonal divisions [128], and WGS analysis of the recently emerged serotype a *H. influenzae* also revealed two populations [101] like the two clonal divisions described using multilocus enzyme electrophoresis of Hib [128]. However, the definition of lineages of non-typeable *H. influenzae* may need further study and discussion because their genetic background appear to be much more diverse than the encapsulated or serotypeable strains of *H. influenzae* [129].

WGS data can be used to predict results obtained by the traditional surveillance methods. Use of WGS to predict serotype of *S. pneumoniae* [130] and *H. influenzae* [131] as well as serogroup of *N. meningitidis* [132] have been described. A platform that uses WGS data for determination of MLST ST and clonal analysis has also been developed [133]. Use of WGS data to identify genetic typing markers and virulence factors has also been published [105,118]. Pipelines to apply WGS to predict antibiotic susceptibility of bacterial pathogens have been developed [134,135]. Improved sequencing technology has allowed direct non-culture genome sequencing from clinical specimens to identify the cause of culture negative fulminant fever [136]. This metagenomics approach has been applied to investigate a meningococcal outbreak in Liberia and the genome data identified the outbreak strain as identical to the unique serogroup C meningococcal strain causing outbreaks in West Africa [137]. The experience gained from WGS studies of *S. pneumoniae*, *N. meningitidis*, and to a lesser extent *H. influenzae*, would help to inform and prepare for the pre- and post-vaccine introduction surveillance of *S. agalactiae*. The platforms built for *S. pneumoniae*, *N. meningitidis*, and *H. influenzae* will likely shorten the deployment of these technologies to study *S. agalactiae*.

To enable a whole genome nucleotide sequence-based surveillance tool to complement conjugate vaccines in the global effort to defeat meningitis, a WHO-led partnership called Global Meningitis Genome Partnership (GMGP) was formed to coordinate, assist, and develop guidelines for using WGS data to identify and track the global epidemiology of common bacterial meningitis agents of *S. pneumoniae*, *N. meningitidis*, *H. influenzae*, and *S. agalactiae* [138]. This collaborative approach has the potential of building synergy between the international partners to achieve the goal of defeating vaccine-preventable bacterial meningitis by 2030.

#### **8. Chemoprophylaxis, Corticosteroids, and Experimental Immune Modulating Approaches for Prevention and as Adjuvant Therapeutic Agents of Bacterial Meningitis**

Although vaccines remain the primary tool to offer active protection against infections, chemoprophylaxis can prevent secondary cases by offering protection to close contacts and household members of index cases. Chemoprophylaxis can also offer protection to those immunized subjects before adequate level of adaptive immunity can be developed. Guidelines that define household members and close contacts of index cases of Hib and IMD as well as the choice and dosage of prophylactic antibiotics have been published [139,140]. For those requiring chemoprophylaxis to prevent IMD, a single dose of ciprofloxacin is

recommended, or rifampicin given twice daily for two days as an alternative. Other prophylactic antibiotics may include ceftriaxone, cefixime, and azithromycin. IMD patients treated with benzylpenicillin (which may not eliminate pharyngeal meningococci) are recommended to receive chemoprophylaxis that can eliminate nasopharyngeal carriage of meningococci before hospital discharge to prevent potential transmission to household members. Rifampicin once a day for four days or ciprofloxacin twice a day for five days are recommended prophylactic antibiotics for contacts of index cases of Hib. Other effective antibiotics may include ceftriaxone and azithromycin. Chemoprophylaxis is generally not recommended for close contacts of IPD patients. However, children with increased risk of IPD such as those with asplenia or sickle cell disease should receive daily prophylaxis with oral penicillin [141]. Public Health England also has guidelines of infection control, vaccination, and chemoprophylaxis (with rifampicin, penicillin, or azithromycin) for high risk individuals living in closed settings when outbreak or cluster of severe pneumococcal disease occur [142]. To prevent early onset of GBS in neonates, pregnant women should be offered screening for GBS and intrapartum antibiotic prophylaxis in indicated situations [143]. Besides chemoprophylaxis, immunization with the recommended vaccines for IMD, Hib, and IPD should be the primary tool for prevention of these vaccine preventable diseases.

Early treatment with dexamethasone reduced mortality and improved the outcome of adult patients with acute meningitis [144]. However, in a Cochrane review to study corticosteroids as an adjuvant therapy of bacterial meningitis, the authors found that corticosteroids did not reduce the overall mortality in meningitis patients but can reduce hearing loss and neurological sequelae [145]. The effect of corticosteroids on meningitis mortality and sequelae varied according to the bacterial agent causing meningitis [145]. Benefits of corticosteroids in treatment of meningitis patients have led to hypothesis and experimental approaches to modulate the immune response in order to decrease the harmful effects of inflammation and to improve the outcome of bacterial meningitis [146]. In one study, the benefit of prophylactic palmitoylethanolamide (a natural fatty acid amide) was demonstrated in a mouse model of *E. coli* meningitis to prolong survival and reduce symptoms by reducing inflammation and slowing the progression of infection [147]. Despite success as immunomodulation therapy for a number of auto-immune diseases such as arthritis and psoriasis, this approach, other than the use of dexamethasone, as adjuvant therapy of bacterial meningitis remain elusive and at the pre-clinical stages of development.

#### **9. Looking Ahead and What to Expect in the Post-Genomic Era of Meningitis Control**

The conjugate vaccines currently in use to control *S. pneumoniae*, *N. meningitidis*, and *H. influenzae*invasive infections have undoubtedly saved tens of thousands of lives [148] but there is no room for complacency because these vaccines do not offer universal coverage against all serotypes or serogroups of these pathogens. Therefore, laboratory surveillance couples with good epidemiological work remain important to monitor the trends of vaccinepreventable bacterial meningitis. We need to stay vigilant for diseases due to strains arising from the phenomenon of capsule switching and capsule replacement. For *H. influenzae*, the most common invasive strains now are non-encapsulated [91]. An increase in the detection of non-encapsulated *S. pneumoniae* has recently been reported [149]. Although most non-encapsulated *S. pneumoniae* do not cause IPD, their increase in prevalence may be concerning since they may serve as reservoirs of gene pools, including antibiotic resistance genes, for transfer into encapsulated *S. pneumoniae*. Another concern is the finding of hybrid capsules [150,151] including new capsule types due to recombination with a different *Streptococcus* species, for example, *S. mitis* [29]. Transfer of a *S. pneumoniae* capsule into a normally non-pathogenic or non-invasive *S. mitis* strain has also been reported [152]. With a large repertoire of capsule genes in *S pneumoniae*, and related *Streptococcus* species, there may be endless combinations for the organisms to take advantage of to evade vaccine immunity.

The genomic era seems to have opened up new opportunities like "reverse vaccinology" to quickly identify potential vaccine candidates [153]. Machine-learning and artificial intelligence have also been proposed to mine genomes for useful data and genes for potential applications [154].

#### **10. Conclusions**

Nowadays, we have powerful conjugate vaccines that target the most common bacterial meningitis agents (at least the most common invasive serotypes or serogroups) to not only prevent infections in the vulnerable age group, but also by eliminating nasopharyngeal carriage, to provide herd immunity to the non-vaccinated individuals. Conjugate vaccines have prevented millions of deaths from bacterial meningitis over the last two decades [2]. We now also have genomic tools that can read the complete coding sequences of bacteria for a never-before-seen gene-by-gene comparison at the nucleotide sequence level to identify and track the movement of strains (including new strains) and infections globally [76,84,97,133] in order to either quickly deploy vaccines or to develop newer vaccines for control. Nevertheless, we cannot be complacent as we have witnessed changes in the three bacterial meningitis agents after vaccine introduction. The significant increase of invasive *H. influenzae* disease due to non-encapsulated or non-typeable strains or the increase in Hia in some population in recent years are of concern [94–98]. The epidemiology of IMD in Africa has changed with much success in the deployment of the monovalent MenAfriVac leading to dramatic decreases in incidences of serogroup A diseases [81]. However, other vaccine-preventable serorgroups like W and C still continue to cause significant amount of disease when vaccines against these serogroups have not been deployed yet. The most problematic may be related to IPD due to non-vaccine serotypes emerging to cause disease after the sequential introduction of PCV7, PCV10 and PCV13 [50,52,53,57,58]. Whether this is related to the large number of serotypes of *S. pneumoniae* in contrast to the much smaller number of serotypes of *H. influenzae* or serogroups of *N. meningitidis* is unknown, but mathematical modelling suggested the number of serotypes might have an effect on strain replacement in nasopharyngeal carriage after vaccination [49]. Even though only 10 serotypes of *S. agalactiae* have been identified, its different ecology (genito-gastrointestinal colonizer versus pharyngeal colonizer) may make the effect of conjugate vaccines on the subsequent epidemiology difficult to predict.

In summary, we are in a much better position to control bacterial meningitis than ever before and surveillance continues to have a key role to play [69,70].

**Funding:** The author's laboratory surveillance activities on vaccine preventable bacterial diseases are supported by the Public Health Agency of Canada.

**Acknowledgments:** The author thanks Mei Ling Lam for her assistance in preparation of the tables, figures and the reference section.

**Conflicts of Interest:** The author has no conflict of interest to declare. Opinions expressed in this manuscript are those of the author and they do not represent the view of the National Microbiology Laboratory nor the Public Health Agency of Canada.

#### **References**


### *Review* **Bacterial Meningitis in Children: Neurological Complications, Associated Risk Factors, and Prevention**

**Abdulwahed Zainel <sup>1</sup> , Hana Mitchell 1,2 and Manish Sadarangani 1,2,\***


**Abstract:** Bacterial meningitis is a devastating infection, with a case fatality rate of up to 30% and 50% of survivors developing neurological complications. These include short-term complications such as focal neurological deficit and subdural effusion, and long-term complications such as hearing loss, seizures, cognitive impairment and hydrocephalus. Complications develop due to bacterial toxin release and the host immune response, which lead to neuronal damage. Factors associated with increased risk of developing neurological complications include young age, delayed presentation and *Streptococcus pneumoniae* as an etiologic agent. Vaccination is the primary method of preventing bacterial meningitis and therefore its complications. There are three vaccine preventable causes: *Haemophilus influenzae* type b (Hib), *S. pneumoniae,* and *Neisseria meningitidis.* Starting antibiotics without delay is also critical to reduce the risk of neurological complications. Additionally, early adjuvant corticosteroid use in Hib meningitis reduces the risk of hearing loss and severe neurological complications.

**Keywords:** bacterial meningitis; neurological sequelae; hearing loss; seizure; epilepsy; hydrocephalus; focal neurological deficit; vaccine; corticosteroid; dexamethasone

### **1. Introduction**

Acute bacterial meningitis is the most common bacterial central nervous system (CNS) infection. It is a devastating illness, especially in neonates (age < one month) and infants (age < one year). Bacterial meningitis has a high case-fatality rate of up to 30% [1–5], and as many as 50% of survivors develop neurological complications [1,3,6–10]—with outcomes highly dependent on patient's age and the infecting organism. This article provides an overview of the short and long-term neurological complications of bacterial meningitis, the associated risk factors, and available vaccines and therapies which may reduce the risk of complications.

### **2. Epidemiology and Etiology**

The incidence of bacterial meningitis in children differs by age group and is highest in infants aged younger than two months [11,12]. In the United States, the incidence rate during 2006–2007 in children under two months was 81 cases per 100,000, compared with 0.4 cases per 100,000 in children aged 11–17 years. Bacterial meningitis is more common in low and middle income countries (LMICs) compared to high income countries (HICs) [13,14]. For example, the incidence rate of meningitis in 2016 in all ages in South Sudan was 270 per 100,000 whereas in Australia it was 0.5 per 100,000 [14].

The most common organisms causing bacterial meningitis vary by age group (Table 1). Introduction of vaccines against *Haemophilus influenzae type b* (Hib)*, Neisseria meningitidis* and *Streptococcus pneumoniae* over the last three decades has led to a drastic decrease in the incidence rate of bacterial meningitis beyond the neonatal period in countries with these

**Citation:** Zainel, A.; Mitchell, H.; Sadarangani, M. Bacterial Meningitis in Children: Neurological Complications, Associated Risk Factors, and Prevention. *Microorganisms* **2021**, *9*, 535. https://doi.org/10.3390/ microorganisms9030535

Academic Editor: James Stuart

Received: 27 January 2021 Accepted: 3 March 2021 Published: 5 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

vaccines included as part of their routine infant and children immunization programs [15]. However, the case fatality rate has not changed significantly [16–20].

**Table 1.** Most common organism for different Age Groups.


\* GBS: Group B Streptococcus.

Hib was the leading cause of bacterial meningitis in the 1990s, but became uncommon in countries that introduced Hib immunization. However, it is still a frequent cause of bacterial meningitis in the countries were Hib vaccine is not universal or in unvaccinated children due to vaccine hesitancy [27]. In countries where Hib is now rare, *S. pneumoniae* has become the leading cause of bacterial meningitis outside the neonatal period, except in some European and Sub-Saharan African countries, where it is the second most common cause in this age group after *N. meningitidis* [26]. Group B Streptococcus (GBS) and *Escherichia coli* remain the leading causes of bacterial meningitis in neonates with little change in incidence rates over time [4,12]. Intrapartum antibiotic prophylaxis has reduced the risk of early onset, but not late-onset GBS meningitis [28,29]. Although, *Listeria monocytogenes* is an uncommon cause, it should be considered in neonates.

#### **3. Pathophysiology of Bacterial Meningitis**

Meningitis develops after the pathogen invades the CNS either though hematogenous route (bacteremia) or by direct extension secondary to sinusitis or mastoiditis and multiplies in the subarachnoid space. The presence of bacteria in the subarachnoid space leads to activation of the immune response, resulting in bacterial lysis. The presence of bacterial particles triggers a further inflammatory response with on-going migration of neutrophils across the blood–brain barrier and continuous cytokine and chemokine release (including IL-1B or CXCL1,2,5) (Figure 1) [7,30–35]. A persistent inflammatory state subsequently leads to decreased cerebral perfusion, cerebral edema, raised intracranial pressure, metabolic disturbances, and vasculitis, all contributing to neuronal injury and ischemia [32].

**Figure 1.** Pathophysiology of Neuronal Damage due to Bacterial Meningitis. **Figure 1.** Pathophysiology of Neuronal Damage due to Bacterial Meningitis.

#### **4. Bacterial Meningitis Complications 4. Bacterial Meningitis Complications**

There are many complications that are associated with bacterial meningitis. These include short-term complications such as seizures, focal neurological deficits and subdural effusions, and long-term complications such as hearing loss, cognitive impairment, hydrocephalus, learning disability, and epilepsy [8,36–38]. Neurological sequelae are more likely to happen in LMICs compared with HICs [37] due to delayed presentation to medical services, lack of access to healthcare and limited resources. Additionally, complications are likely under reported in LMICs (Table 2) [7]. There are many complications that are associated with bacterial meningitis. These include short-term complications such as seizures, focal neurological deficits and subdural effusions, and long-term complications such as hearing loss, cognitive impairment, hydrocephalus, learning disability, and epilepsy [8,36–38]. Neurological sequelae are more likely to happen in LMICs compared with HICs [37] due to delayed presentation to medical services, lack of access to healthcare and limited resources. Additionally, complications are likely under reported in LMICs (Table 2) [7].

**Table 2.** Long and Short-term Neurological Complications following pneumococcal and meningococcal meningitis in Low and Middle Income Countries (LMICs) and High Income Countries **Table 2.** Long and Short-term Neurological Complications following pneumococcal and meningococcal meningitis in Low and Middle Income Countries (LMICs) and High Income Countries (HICs).


Cognitive impairment 4–41% N/A\* 4% 12–19% [10,13,43,48,49] \*: Not avilable.

#### \*: Not avilable. *4.1. Short-Term Complications*

#### *4.1. Short-Term Complications*  4.1.1. Subdural Effusion

4.1.1. Subdural Effusion Subdural effusions occur in 20–39% of children with bacterial meningitis [50–52]. Subdural effusion is most common in infants (age < one year) compared to older children [8,51]. There is no significant difference in subdural effusions complication due to Hib meningitis compared to S. pneumoniae and N. meningitidis [51]. Most subdural effusions in the context of bacterial meningitis are asymptomatic, resolve spontaneously and rarely require intervention. Indications for drainage include an infected effusion or empyema, Subdural effusions occur in 20–39% of children with bacterial meningitis [50–52]. Subdural effusion is most common in infants (age < one year) compared to older children [8,51]. There is no significant difference in subdural effusions complication due to Hib meningitis compared to *S. pneumoniae* and *N. meningitidis* [51]. Most subdural effusions in the context of bacterial meningitis are asymptomatic, resolve spontaneously and rarely require intervention. Indications for drainage include an infected effusion or empyema, focal neurological signs or symptoms, or increased intracranial pressure [51].

focal neurological signs or symptoms, or increased intracranial pressure [51].

#### 4.1.2. Focal Neurological Deficit

Focal neurological deficit refers to a set of signs and symptoms resulting from a lesion localized to a specific anatomical site in central nervous system [53]. Examples include isolated limb weakness or hemiparesis, visual deficit or a speech impediment. They are estimated to occur in 3–14% of bacterial meningitis cases [41,42].

Acute focal neurological deficits after bacterial meningitis are usually due to ischemic stroke, but can also occur due to subdural empyema, cerebral abscess or intracranial bleeding [7]. Focal neurological deficits generally improve over months to years after the initial insult. Abscess collections can be surgically drained which usually leads to complete symptom resolution while focal deficits resulting from an ischemic event take longer to resolve. In a recent study, children with stroke following bacterial meningitis compared to children without stroke were less likely to have a normal neurological exam at discharge (21% vs. 76%) and within a seven years follow-up period (31% vs. 74%) [54]. Focal neurological deficits that persist after acute infection are not a very common complication after bacterial meningitis in childhood (Table 2); however, the incidence is highest for infections occurring in the first year of life [55].

#### *4.2. Long-Term Complications*

#### 4.2.1. Hearing Loss

Sensorial hearing loss is the most widely reported neurological sequelae of bacterial meningitis [7,27,44,51,56–59]. Hearing loss may develop from both the direct spread of bacterial products and as a result of the host inflammatory response in the meninges and CSF. When bacteria reach the cochlea, a severe labyrinthitis results, which leads to blood-labyrinth barrier breakage, and ultimately meningitis-associated hearing loss [7]. Around 10% of children with bacterial meningitis develop unilateral or bilateral sensorineural hearing loss [37,60]; 5% of children develop bilateral severe or profound hearing loss [37]. Hearing loss is a more common complication in infections caused by *S. pneumoniae* (14–32%), compared with *N. meningitidis* (4–23%) and *H. influenzae* (20%) [44,60]. Children with hearing loss are at risk of further developing balance disturbances [61] and speech and language delay [62], and are therefore at higher risk of having long-term behavioral problems [63].

Reversible deafness (i.e., transient hearing loss) has been documented in long term follow up of children with pneumococcal meningitis [7,61]. For example, a study done among children with pneumococcal meningitis in Bangladesh reported that 33% of children in short-term follow-up (30–40 days) had hearing loss, but only 18% had persistent hearing loss in long-term follow-up (6–24 months) following hospital. The difference was attributed to recovery of transient hearing impairment [43].

#### 4.2.2. Cognitive Impairment

Due to the irreversible neuronal damage that occurs during bacterial meningitis, the risk of developing long-term cognitive deficits and learning difficulties are significant [3,48]. The rates of cognitive impairment world-wide are difficult to estimate because there is no standardized method of measuring it and long-term data on meningitis survivors are rarely available.

In a Dutch study, 680 children between the age of 4 and 13 years who survived bacterial meningitis were followed up for 6 years after their meningitis episode. The study followed their educational, behavioral and general health issues. The survivors were compared to a control group of healthy school-age siblings and peers, with similar socioeconomic background. It was found that 30% of children with meningitis had problems with school achievement or concentration. Additionally, these children repeated a school year twice as often as the control group (16% vs. 8%). Moreover, the post-meningitis group were referred to special-needs school four times more frequently compared to control group [48].

In a Danish nationwide population-based cohort study, adults who had a bacterial meningitis in childhood were compared to control group that included a general population

of the same age and sex, their siblings and the siblings of meningitis patients. Adults who had childhood bacterial meningitis had lower educational achievements and economic self-sufficiency compared to control group. By the age of 35 years, 11%, 10.2% and 5.5% fewer had completed high school in meningococcal pneumococcal, and Hib meningitis, respectively. Additionally, 7.9%, 8.9% and 6.5% fewer had obtained a higher education in meningococcal pneumococcal, and Hib meningitis, respectively. Additionally, 3.8%, 10.6% and 4.3% had lower economic self-sufficiency in meningococcal pneumococcal, and Hib meningitis, respectively [49]. In a study done in Bangladesh, short (30–40 days) and long-term (6–24 months) follow-up revealed that 41% in both groups had deficits in mental development and 49% and 35%, respectively, had psychomotor delay [43]. Another example, in a study that was done in Brazil, 5.88% children developed learning disabilities, and 7.35% children had developmental delay [40].

Finally, psychiatric disease including anxiety and depression is likely under-recognized and underreported in meningitis survivors and contributes to cognitive difficulties and overall quality of life [64].

#### 4.2.3. Seizures and Epilepsy

One of the clinical presentations of bacterial meningitis is seizures [65,66]. In cases of bacterial meningitis with seizures, if seizures develop early during the illness and are easily controlled, permanent neurological complications are rarely of concern. However, if seizures are prolonged, difficult to control or develop 72 h after admission, neurological sequelae are more likely to occur and are usually suggestive of a cerebrovascular event [44,67]. In HICs, 1–5% of epilepsy cases are presumed to be due to CNS infection; including bacterial meningitis [68]. In Sub-Saharan Africa 26% of patients have epilepsy attributed to CNS infection [69].

In a neonatal bacterial meningitis study seizures have been more commonly associated with GBS then *E. coli* (41% vs. 25%) [21]. 71% of children who late seizure after bacterial meningitis had permanent focal neurological deficit [70].

#### 4.2.4. Hydrocephalus

Hydrocephalus incidence is around 7% of bacterial meningitis in children [71] and it is more common in neonates and infants; 25% [72,73]. It is more common in neonatal Gram negative meningitis [57]. Hydrocephalus may develop at the beginning of the illness or weeks later after diagnosis with bacterial meningitis. The most common type of hydrocephalus after bacterial meningitis is communicating hydrocephalus; seen in up to 52% of cases with hydrocephalus [74]. In communicating hydrocephalus CSF flows freely between the ventricles but is not adequately reabsorbed back into the blood stream. Depending on the size of hydrocephalus and resulting neurologic impairment temporary or permanent ventricular shunt placement may be required [75].

#### **5. Risk Factors**

There are many risk factors associated with neurological complications in bacterial meningitis (Table 3).

**Table 3.** Risk factors for developing neurological complication in Bacterial Meningitis.


\* N/A: Not available.

In general, infants are at higher risk of developing neurological complications compared to older children [3,9,23]. 71% of infants (aged < one year) with bacterial meningitis develop neurological complications compared to 38% in children aged one to five years and 10% in those aged six to 16 years [9]. Children younger than 12 months at time of diagnosis with bacterial meningitis are at increased risk of developing hydrocephalus, subdural effusion, seizure disorder and hearing loss [76]. Altered level of consciousness is associated with poor prognosis [6,9]. 82% of children with bacterial meningitis who developed neurological complications had altered level of consciousness on presentation; where, 39% of children with bacterial meningitis who did not develop neurological complications had altered level of consciousness [9]. The longer the duration that the child was unconscious, the worst the outcome is.

In bacterial meningitis, delayed presentation to hospital increase the risk of subdural effusion, hydrocephalus, hearing impairment and seizure disorder [76]. Although delayed presentation is one of the known risk factors for developing neurological complications, there is no universal definition for the duration of the delay. In one study, children admitted with duration of illness <48 h had a lower incidence of neurological complication (40%) compared to children who were admitted after 48 h of illness [9]. Children with *S. pneumoniae* meningitis have a higher risk of developing neurological complication (75% of *S. pneumoniae* meningitis cases) compared to *N. meningitidis* (25%) and Hib (20%) [9]. *S. pneumoniae* compared to *N. meningitidis* and Hib is associated with higher risk of symptomatic seizures, hydrocephalus, hearing loss and mental retardation [76]. Delay in starting antibiotics beyond 24–72 h has a poor prognosis and leads to increased risk of severe neurological complication such as hydrocephalus, subdural effusion, hearing loss, and seizure disorder [76]. In summary, young age, delayed presentation and *S. pneumoniae* as an etiologic agent were associated with increased risk of neurological complications in both HIC and LMIC settings [11,15,54,71,76–78].

#### **6. Prevention of Neurological Complication**

#### *6.1. Primary Prevention*

The most effective prevention of neurological complications from bacterial meningitis is preventing the infection though infant and childhood vaccination programs. Despite the development of multiple vaccines against the organisms causing bacterial meningitis, there continue to be many meningitis outbreaks caused by vaccine-preventable organisms [79,80]. There are currently vaccines against 3 of the organisms that cause bacterial meningitis: Hib, *N. meningitidis* (capsular groups A, B, C, W and Y) and 23 of the >90 serotypes of *S. pneumoniae* [4,15]. Hib conjugate vaccine targets only type b *H. influenzae*, and is given as three or four doses before 18 months of age [81]. There are two types of vaccine against *N. meningitidis*: Conjugate vaccines against capsular groups A, C, W, and Y and protein vaccines against group B. There are two types of vaccines against *S. pneumoniae*: Pneumococcal conjugate vaccines (PCV 10 against 10 serotypes, PCV 13 against 13 serotypes) and polysaccharide vaccine against 23 serotypes which is not routinely used in healthy children [82].

Routine vaccination can lead to development of community protection by indirect effect prevention of transmission within a population [83]. Since the introduction of pneumococcal conjugate vaccines (PCVs), the overall incidence of invasive pneumococcal disease (IPD) has dropped significantly, including in unimmunized children, highlighting these indirect effects [84,85]. For example, in South Africa, Morocco, Gambia, Mozambique, Kenya and Burkina Faso, 32–81% reduction in IPD has been reported after PCV introduction, with highest reduction in children aged under 24 months (55–89%) [86]. However, infections caused by non-vaccine serotypes infections have increased in some countries. In the United States, the proportion of IPD caused by non-vaccine serotype increased from 6% to 38% after the introduction of PCV7 vaccine; sometimes referred to as serotype replacement [85,87,88]. Overall, IPD incidence dropped.

The highest rate of meningococcal disease worldwide is in the Sub-Saharan Africa, specifically in the "meningitis belt" region where major epidemics occur every 5–12 years [89]. After the introduction of MenA vaccine to the Sub-Saharan Africa in 2010, there has been a 99% reduction in group A meningitis in this region [90] and capsular group W is the currently the commonest [91]. Rate of meningococcal meningitis are much lower in other parts of Sub-Saharan Africa, although longitudinal surveillance outside of the meningitis belt is limited [92]. Following wide-spread introduction of Hib vaccine, *S. pneumoniae* accounted for 65% of acute bacterial meningitis cases in Malawi, while the rates of *N. meningitidis* have remained constant at <5% [90].

### *6.2. Secondary Prevention of Complications*

#### 6.2.1. Antibiotic Therapy

It is important to have a high clinical suspicion of bacterial meningitis and start appropriate treatment without delay [93,94]. The empiric antibiotic choice should be based on the most likely causative agent for patient's age [73,95,96]. In children, third-generation cephalosporins, such as cefotaxime or ceftriaxone, are the usual empirical choice to cover the most common organisms—*S. pneumoniae* and *N. meningitidis*. Ampicillin should be added to cover *L. monocytogenes* in very young children; some guidelines recommended for younger than 3 months, others recommended this for those younger than 1 month [95,96].

#### 6.2.2. Corticosteroids

Neuronal damage due to acute bacterial meningitis is not only due to bacterial invasion to the subarachnoid space, but also due to the host's inflammatory response to this invasion [36]. The only widely researched agent that can limit subarachnoid inflammation is dexamethasone. The recommendation for the use of dexamethasone in bacterial meningitis unfortunately cannot be generalized and depends on the causative organism and the ability to administer dexamethasone within 1–12 h of administration of antibiotics [27,40,44,97].

In infant and children with Hib meningitis, administration of dexamethasone with antibiotics has shown a significant reduction in neurological sequelae rate (17%) compared to antibiotic only (23.4%) [27,59]. Additionally, rates of hearing loss were lower with dexamethasone use (12.9%) compared to antibiotics only (17.4%) [59,96,98,99]. In a large study done in Malawi, children with *H. influenzae* meningitis who received dexamethasone were less likely to have neurological sequelae compared to a placebo group (27% vs. 40%) [13]. Therefore, it is generally recommended that dexamethasone be administered before or with the first dose of antibiotics [1,13,96] when *H. influenzae* is confirmed or strongly suspected.

On the other hand, the use of dexamethasone in infants and children with pneumococcal meningitis is controversial as it has not been clearly proven to change the outcome [1,13,36,98,99]. Additionally, the use of dexamethasone for meningococcal meningitis was not proven to be effective in reduce neurological sequelae [96].

Better understanding of specific microbial and host factors contributing to CNS infection and inflammatory response may help in identification of new therapeutic targets and specific immunomodulatory regimens.

#### **7. Long-Term Follow Up**

All children who are diagnosed with meningitis should have a hearing assessment done before discharge or 1 month within discharge; even if hearing loss is not clinically suspected [13,61]. It is critical to perform audiology assessment month after diagnosis or earlier if possible, as up to 90% of children's cochlea with hearing loss due to meningitis can ossify, preventing appropriate treatment with cochlear implants [100,101].

Children with seizure disorders require antiepileptic medication and should ideally have long-term follow up by a neurologist [7,8,67,70]. Children with hearing loss and/ or intellectual disability will need neurodevelopmental follow up and support for speech, language and social development. Mental health concerns and psychiatric problems are likely underreported in children who were diagnosed with meningitis and periodic mental

health assessments by an appropriate specialist should be included into their long-term care [64].

#### **8. Conclusions**

Children with bacterial meningitis are at risk of developing neurological complications that include focal neurological deficits, subdural effusion, hearing loss, cognitive impairment, seizure disorder, and hydrocephalus. There is a need to optimize utilization of available vaccines and to develop vaccines for pathogens implicated in neonatal meningitis (GBS and *E. coli*). For children diagnosed with bacterial meningitis, starting antibiotic therapy without delay is critical for a good prognosis and to reduce the risk of developing neurological complications. So far, steroids are the only drug that can control inflammatory response, but effectiveness is limited to specific situations.

**Author Contributions:** A.Z., H.M., and M.S. conceived and designed the structure of the manuscript. A.Z. drafted the original version of the manuscript. All authors edited and approved the final version. All authors have read and agreed to the published version of the manuscript.

#### **Funding:** None.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** MS is supported via salary awards from the BC Children's Hospital Foundation, the Canadian Child Health Clinician Scientist Program and the Michael Smith Foundation for Health Research.

**Conflicts of Interest:** MS has been an investigator on projects funded by GlaxoSmithKline, Merck, Pfizer, Sanofi-Pasteur, Seqirus, Symvivo and VBI Vaccines. All funds have been paid to his institute, and he has not received any personal payments.

#### **References**


## *Article* **Field Evaluation of the Performance of Two Rapid Diagnostic Tests for Meningitis in Niger and Burkina Faso**

**Marc Rondy <sup>1</sup> , Mamadou Tamboura <sup>2</sup> , Fati Sidikou <sup>3</sup> , Issaka Yameogo <sup>4</sup> , Kambire Dinanibe <sup>5</sup> , Guetwende Sawadogo <sup>6</sup> , Chantal Kambire <sup>7</sup> , Halima Mainassara <sup>3</sup> , Ali Elhaj Mahamane <sup>3</sup> , Baruani Bienvenu <sup>8</sup> , Haladou Moussa <sup>8</sup> , Rasmata Ouedraogo <sup>2</sup> , Katya Fernandez 9,\*, Muhamed-Kheir Taha <sup>10</sup> and Olivier Ronveaux <sup>11</sup>**

	- haladoum@who.int (H.M.)

**Abstract:** New lateral flow tests for the diagnosis of *Neisseria meningitidis* (Nm) (serogroups A, C, W, X, and Y), MeningoSpeed, and *Streptococcus pneumoniae* (Sp), PneumoSpeed, developed to support rapid outbreak detection in Africa, have shown good performance under laboratory conditions. We conducted an independent evaluation of both tests under field conditions in Burkina Faso and Niger, in 2018–2019. The tests were performed in the cerebrospinal fluid of suspected meningitis cases from health centers in alert districts and compared to reverse transcription polymerase chain reaction tests performed at national reference laboratories (NRLs). Health staff were interviewed about feasibility. A total of 327 cases were tested at the NRLs, with 26% confirmed Nm (NmC 63% and NmX 37%) and 8% Sp. Sensitivity and specificity were, respectively, 95% (95% CI: 89–99) and 90% (95% CI: 86–94) for Nm and 92% (95% CI: 75–99) and 99% (95% CI: 97–100) for Sp. Positive and negative predictive values were, respectively, 77% (95% CI: 68–85) and 98% (95% CI: 95–100) for Nm and 86% (95% CI: 67–96) and 99% (95% CI: 98–100) for Sp. Concordance showed 82% agreement for Nm and 97% for Sp. Interviewed staff evaluated the tests as easy to use and to interpret and were confident in their readings. Results suggest overall good performance of both tests and potential usefulness in meningitis outbreak detection.

**Keywords:** meningitis; *Neisseria meningitidis*; *Streptococcus pneumoniae*; rapid diagnostic test; national reference laboratory; cerebrospinal fluid; Niger; Burkina Faso

### **1. Introduction**

Because of its high case fatality (around 10% [1,2]) and epidemic potential, meningococcal meningitis is a major public health threat [3], especially in the "meningitis belt". This region, which stretches from Senegal to Ethiopia, is characterized by a high seasonal incidence of meningococcal meningitis between January and June [4].

**Citation:** Rondy, M.; Tamboura, M.; Sidikou, F.; Yameogo, I.; Dinanibe, K.; Sawadogo, G.; Kambire, C.; Mainassara, H.; Mahamane, A.E.; Bienvenu, B.; et al. Field Evaluation of the Performance of Two Rapid Diagnostic Tests for Meningitis in Niger and Burkina Faso. *Microorganisms* **2021**, *9*, 832. https://doi.org/10.3390/ microorganisms9040832

Academic Editor: James Stuart

Received: 15 March 2021 Accepted: 11 April 2021 Published: 14 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Meningococcal meningitis caused by *Neisseria meningitidis* (Nm) bacteria may cause large outbreaks and six of the 12 Nm serogroups (A, B, C, W, X, and Y) are responsible for invasive forms of meningitis [5]. The epidemiology of Nm meningitis is evolving rapidly, and new variants are continually emerging [6].

A wide choice of meningococcal vaccines covering a variable number of serogroups is currently available [7]. In a context of a diversification of serogroups with epidemic potential in the meningitis belt, it is important to strengthen microbiological surveillance to prompt the early implementation of vaccination campaigns appropriate to circulating serogroups.

Pneumococcal meningitis causes clusters of cases and, less frequently, outbreaks, in the meningitis belt, even following introduction of pneumococcal conjugate vaccines. In 2019, following massive the reduction of NmA, with MenA vaccine rollout, *Streptococcus pneumoniae* (Sp) was found to be responsible for 40% of meningitis cases in countries in the region [8]. Pneumococcal meningitis has very high case fatality rates (36–66% in the meningitis belt [9]) and, given the difficulty of treatment, requires longer treatment protocols than those for meningococcal meningitis [10].

The microbiological diagnosis of meningitis is currently based on culture or, more frequently, on polymerase chain reaction (PCR) tests carried out by national reference laboratories (NRLs) on a sample of cerebrospinal fluid (CSF) [11]. This last procedure, which requires maintaining the sample in a cold chain in primary health care centers (PHCs) and during transport to the NRL, makes the early identification of circulating strains difficult. To reduce the time needed to detect circulating strains, diagnostic tests have been developed that can be used at bedside and provide a diagnosis in a few minutes. However, currently available rapid diagnostic tests (RDTs) to detect meningococcal meningitis have a short shelf life, are sensitive to heat, and cannot detect all circulating serogroups, making them of little utility in the meningitis belt context [12].

The rapid diagnostic tests (RDT) were first developed at the Institut Pasteur, Paris, as immunochromatographic tests [13,14]. They were thereafter transferred the BioSpeedia company who then developed MeningoSpeed and PneumoSpeed that can be used between 2 and 30 ◦C to diagnose, respectively, the five main meningococcal serogroups (A, C, W, X and Y) [15] found in the African meningitis belt and Sp meningitis. The tests are based on the use of antibodies directed against capsular polysaccharide of Nm (serogroups A, C, W, Y, and X) and against pneumococcal cell wall polysaccharide C that is common to all Sp isolates. Laboratory based tests suggest good sensitivities and specificities of these two RDTs [12].

These RDTs require the transfer of two drops of CSF in each cassette (three for Nm groups in A/W serogroups, Y/C serogroups and X serogroup; one for Sp meningitis) and a waiting time of 15 min. The reading process is similar to any RDTs, based on control and case lines appearance and they can be performed by all staff after a short demonstration.

The objective of the study was to measure the performance of the MeningoSpeed and PneumoSpeed RDTs in diagnosing Nm and Sp meningitis in CSF at the bedside among patients in PHCs in Niger and Burkina Faso in 2018 and 2019.

#### **2. Materials and Methods**

#### *2.1. Overview*

To measure the performance of the RDTs, we compared the results of RDTs conducted at bedside in PHCs with the results of PCRs obtained at NRLs. PHCs in districts where incidence exceeded 3 suspected cases per week per 100,000 inhabitants were invited to participate, and staff were trained, supervised and supplied with RDTs.

To measure the performance of the RDTs after sample transportation, RDTs were also repeated at the NRLs and results compared with PCR tests.

The study population consisted of any consenting patient aged at least two months admitted to a participating PHC with suspected bacterial meningitis (according to the WHO case definition [16]) (sudden onset of fever and stiff neck or other meningeal signs (including bulging fontanelle for patients under 12 months)) between April 2018 and June 2019. Suspected cases with contraindications for lumbar puncture or who refused to participate were excluded from the study.

According to national guidelines for suspected bacterial meningitis case management, medical staff collected CSF through lumbar puncture on every suspected case. For each patient included in the study, the two RDTs (MeningoSpeed and PneumoSpeed) were performed on CSF at the PHC. A 1 mL CSF tube was then refrigerated and sent to the NRL where an additional RDT and a real-time PCR test was performed and used as a gold standard.

PHC staff collected information on patients' symptoms; antibiotic treatment before admission; dates of onset of symptoms, lumbar puncture, RDT reading, and dispatch of sample to the NRL; lot number; and RDT results. PHC information was merged with NRL data, including the results of the PCR tests and the RDTs performed at the NRLs, and data were analyzed using the STATA software package version 14 [17]). The producer of the tests (BioSpeedia Company, Saint Etienne, France) did not fund the trail, did not contribute to the design and did not participate in the analysis of the results. None of the authors is employed by that company.

#### *2.2. Performance of the RDTs*

MeningoSpeed test results were classified as positive or negative per serogroup A, C, W, Y, or X (positive if positive for that serogroup; negative if negative for that serogroup, regardless of other serogroup's results) and for any Nm (positive if positive for any serogroup A, C, W, Y, or X; negative if negative for all these 5 serogroups). PneumoSpeed test results were classified as positive or negative for Sp meningitis.

We compared the results of the RDTs performed in the PHCs with the results of PCR tests performed at the NRLs to calculate the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the tests. The analyses were stratified by year of lumbar puncture (2018 vs. 2019) and epidemic period (defined as between January and June) versus non-epidemic period.

#### *2.3. RDT Feasibility and Acceptability*

PHC staff photographed the RDTs as they were reading them. The photographs were then reviewed and the RDTs blindly re-interpreted by independent readers. We compared the results of the RDTs as interpreted by the PHC staff and independent experts and quantified the level of concordance between them using the kappa coefficient [18]. We considered the agreement between the tests moderate if κ < 0.4, average or good if κ = 0.4 to 0.75 and excellent if κ > 0.75. National study coordinators conducted semi-structured interviews with RDT users to assess the acceptability and feasibility of the RDTs.

#### *2.4. Sample Size*

To estimate an expected 95% sensitivity of MeningoSpeed Nm meningitis identification, with an absolute precision of 5% and a 30% PCR positivity [19], we estimated that at least 243 suspected cases should be included in the study.

#### *2.5. Ethical Considerations*

The study followed the principles governing biomedical research involving human participation and was carried out in line with principles of the Declaration of Helsinki to ensure that the rights, integrity, and confidentiality of participants were protected. The agreements of the ethics committees of WHO (WHO ERC.0002926), Niger (deliberation N◦ 35/2017/CNRES) and Burkina Faso (deliberation N◦ 2017-10-156) were obtained before the start of the study as well as for its extension into 2019.

Written consent was obtained from suspected meningitis cases or from their parents or legal guardians. Standardized information was read to the potential participant. The confidentiality of the data collected, and the anonymity of the participants were ensured during and after the survey (no patient name appears in the databases).

#### **3. Results**

#### *3.1. RDT Performance*

Between 8 April 2018 and 30 June 2019, 421 people with suspected meningitis were admitted to the participating PHCs and tested with the RDTs. Of those people, 327 were eligible for the study, with completed questionnaires and PCR results entered into the database (246 in Niger, and 81 in Burkina Faso) and 198 (61%) were recruited during the epidemic period. The distribution of symptoms among suspected cases was typical of that found in the meningitis belt. A total of 106 (32%) and 28 (9%) patients tested positive for Nm and Sp, respectively, with the RDTs in the PHCs (Table 1).


**Table 1.** Characteristics and distribution of symptoms among suspected cases (N = 327), Niger and Burkina Faso, 2018–2019.

Of the 327 cases with PCR results obtained at an NRL, 86 (26%) were confirmed with Nm (NmC 63% and NmX 37%) and 26 (8%) with Sp (Table 2). The median time between lumbar puncture and PCR result was 18 days outside the epidemic period, and 22 days during the epidemic period.


**Table 2.** Concordance of results obtained by RDT in PHCs and NRLs with results obtained by PCR in NRLs, and serogroupspecific performances, Burkina Faso and Niger, 2018–2019.

RDT: rapid diagnostic test; PHC: primary health care center; NRL: national reference laboratory; PCR: polymerase chain reaction, 95% CI: 95% confidence intervel; PPV: positive predictive value; NPV: negative predictive value; Nm: Neisseria meningitidis; Sp: Streptococcus pneumoniae.

> Sensitivity and specificity were, respectively, 95% (95% CI: 89–99) and 90% (95% CI: 86–94) for any Nm, 93% (95% CI: 82–98), and 98% (95% CI: 95–99) for NmC, and 91% (95% CI: 75–98) and 96% (95% CI: 93–98) for NmX.

> PPV and NPV were, respectively, 77% (95% CI: 68–85) and 98% (95% CI: 95–100) for any Nm. PPV was better in 2019 than in 2018: 90% (95% CI: 81–96) vs. 49% (95% CI: 31–67), *p*-value < 0.01 (Table 3); and was better during epidemic months: 89% (95% CI: 79–95) versus 53% (95% CI: 35–70), *p*-value < 0.01 (Tables 3 and 4).

**Table 3.** Concordance of results obtained by RDT in PHCs with results obtained by PCR in NRLs, and serogroup specific performances, by year of lumbar puncture, Burkina Faso and Niger, 2018–2019.


RDT: rapid diagnostic test; PHC: primary health care center; NRL: national reference laboratory; PCR: polymerase chain reaction; 95% CI: 95% confidence intervel; PPV: positive predictive value; NPV: negative predictive value; Nm: Neisseria meningitidis; Sp: Streptococcus pneumoniae.

> For Sp, sensitivity and specificity were, respectively, 92% (95% CI: 75–99) and 99% (95% CI: 97–100), and PPV and NPV values were 86% (95% CI: 67–96) and 99% (95% CI: 98–100) (Table 2).

> Owing to a lack of tests and a strike by staff, it was only possible to repeat the RDTs at the NRLs in 147 of the 327 patients (45%). Performances of the RDTs at the NRLs and the health centers were comparable (Table 2).


**Table 4.** Concordance of results obtained by RDT in PHCs with results obtained by PCR in NRLs, and serogroup specific performances, by epidemic vs. non epidemic period, Burkina Faso and Niger, 2018–2019.

RDT: rapid diagnostic test; PHC: primary health care center; NRL: national reference laboratory; PCR: polymerase chain reaction; 95% CI: 95% confidence intervel; PPV: positive predictive value; NPV: negative predictive value; Nm: Neisseria meningitidis; Sp: Streptococcus pneumoniae.

#### *3.2. RDT Feasibility and Acceptability*

There was an 82% agreement for Nm and 97% for Sp between RDT results registered at the PHCs and photographs reviewed by an independent expert. The kappa coefficient suggested an excellent concordance for Sp and NmX, a good concordance for all Nm and moderate concordance for NmA, NmC, and NmW (Table 5).

**Table 5.** Concordance of the results of RDTs as interpreted live in the PHCs and by independent reading of photographs, Burkina Faso and Niger, 2018–2019.


RDT: rapid diagnostic test; PHC: primary health care center; Nm: Neisseria meningitidis; Sp: Streptococcus pneumoniae.

Eleven NmX, nine NmA, six NmC, two NmW, and one NmY were detected positive with RDTs but were tested negative by real-time PCR.

All nine positive NmA results identified in health centers were from Burkina Faso, eight in 2018 and one in 2019. Of these, two were identified as positive for NmA on photographs by the external expert. The authors confirmed that a faint line, suggestive of a positive NmA result was indeed visible in these two photographs (Figure 1). Of the eleven false positive NmX results, nine were from Burkina Faso and two from Niger. Of those, four photographs were reviewed and classified as negative by the independent experts. One photograph of a false positive NmC result was available and suggested a misreading by the PHC medical staff.

A total of 31 staff were interviewed about their experiences in using the RDTs (20 in Niger and 11 in Burkina Faso). Overall, respondents found the RDTs easy to use (average difficulty score: 1.9/10). They found their interpretation very easy (average difficulty score: 1.2/10) and had great confidence in their result (average lack of confidence score: 1.3/10).

**Figure 1.** Photographs of six false positive NmA RDTs by independent expert classification, Burkina Faso and Niger, 2018– 2019. **Figure 1.** Photographs of six false positive NmA RDTs by independent expert classification, Burkina Faso and Niger, 2018–2019.

#### A total of 31 staff were interviewed about their experiences in using the RDTs (20 in **4. Discussion**

#### Niger and 11 in Burkina Faso). Overall, respondents found the RDTs easy to use (average *Performance of the RDTs*

by the PHC medical staff.

difficulty score: 1.9/10). They found their interpretation very easy (average difficulty score: 1.2/10) and had great confidence in their result (average lack of confidence score: 1.3/10). **4. Discussion**  *Performance of the RDTs*  We described in this article the performance results from ready-to-use bedside RDTs, usable by any medical staff, storable at 2–30 °C, allowing to detect Nm and Sp within 15 min. Our results suggest that the RDTs for diagnosing Nm and Sp performed well under field conditions at PHC level, with sensitivity and specificity above 90%. They performed We described in this article the performance results from ready-to-use bedside RDTs, usable by any medical staff, storable at 2–30 ◦C, allowing to detect Nm and Sp within 15 min. Our results suggest that the RDTs for diagnosing Nm and Sp performed well under field conditions at PHC level, with sensitivity and specificity above 90%. They performed similarly whether used by medical staff at a PHC or technical staff at an NRL. These RDTs were described as acceptable and easy to use in a bedside context. Serogroup-specific Nm RDT performance could be measured for NmC and NmX and were in the same range, above 90%. The two serogroups were the most prevalent in West Africa in 2018–2019 [20,21].Further studies would be needed to validate their use in detecting NmA, NmW, and NmY due to low prevalence during our study period.

positive NmA result was indeed visible in these two photographs (Figure 1). Of the eleven false positive NmX results, nine were from Burkina Faso and two from Niger. Of those, four photographs were reviewed and classified as negative by the independent experts. One photograph of a false positive NmC result was available and suggested a misreading

similarly whether used by medical staff at a PHC or technical staff at an NRL. These RDTs were described as acceptable and easy to use in a bedside context. Serogroup-specific Nm RDT performance could be measured for NmC and NmX and were in the same range, above 90%. The two serogroups were the most prevalent in West Africa in 2018–2019 [20,21]. Further studies would be needed to validate their use in detecting NmA, NmW, and NmY due to low prevalence during our study period. By extending the study over two years we were able to obtain a sample size large enough to measure the performances of the RDTs with acceptable precision. The study protocol was well followed, despite challenges in systematizing the capture of photographs and the use of the RDTs at the NRLs. In addition, recruitment was lower in Burkina By extending the study over two years we were able to obtain a sample size large enough to measure the performances of the RDTs with acceptable precision. The study protocol was well followed, despite challenges in systematizing the capture of photographs and the use of the RDTs at the NRLs. In addition, recruitment was lower in Burkina Faso due to security issues along with a strike by health workers in 2019. Owing to these limitations, the precision of some secondary outcomes was low. Logistical issues between the PHCs and NRLs, potentially leading to alteration of samples and affecting the measured performance of the RDTs, cannot be excluded. In order to account for such a risk, the protocol included repetition of RDTs at the NRLs. Similar results between RDTs carried out at the PHCs and NRLs suggest that sample alteration was low in this study.

Faso due to security issues along with a strike by health workers in 2019. Owing to these limitations, the precision of some secondary outcomes was low. Logistical issues between Our results suggest a higher sensitivity of the RDT in detecting NmC in this field context than previously reported in laboratory conditions (95% versus 65%) [15]. This could be due to the fact that the circulating clone in Niger and Burkina Faso (NmC, clonal complex ST-10217) [19,22] is particularly well detected by the antibodies used in the MeningoSpeed test.

PPV for any Nm and NmX were at 77% and 73%, respectively, corresponding to a quarter of false positive RDT readings for these outcomes. Moreover, recurrent false positives for NmA were registered. Further analyses suggested that most of these findings were attributed to specific PHCs and that constant supervision and training of RDT users led to a significant decrease in false positives between the first and the second study years. NmA RDT issues, documented with photographs, were fed back to the manufacturer.

The original study protocol included the use of PneumoSpeed tests on urine samples. However, only 31 tests were reported, precluding any interpretation related to this objective. Considering its potential impact on patients' comfort and practice safety, specific studies should be implemented to measure the performance of RDT tests on urine.

Performance measurements are the results of both the quality of the test and the interpretive capacity of the user. Measurement of the concordance between the interpretations of the nursing staff and the independent expert made it possible to discuss this additional information bias linked to the evaluator. Although the proportion of photographs that could be read was low, concordance was generally good, suggesting a good understanding of the use of the RDTs by the PHC staff.

These field performance results met the WHO target product profile acceptable values for sensitivity and specificity, of >90% [23] and were superior or comparable to field performance values for existing meningococcal rapid tests (69–80% sensitivity and 81–94% specificity for latex agglutination tests and 89–92% sensitivity and 85–99 specificity for lateral flow test) [12]. Our field evaluation confirms the good performance of the tests in laboratory conditions and suggests that the tests are suitable for use in field conditions and that they are acceptable to health personnel, but that they should be accompanied by clear instructions and effective training. Considering this performance, their longer shelf-life and improved thermostability (but still below the desired target product profile value of 40◦ ), easier test procedures, and inclusion of all main Nm serogroups (including NmX), MeningoSpeed and PneumoSpeed are good candidate tests for the early detection of meningitis epidemics in Africa. However, this field study is only one step towards ensuring access to safe appropriate diagnostic tests of good quality. At any rate, ensuring confirmation of test results with a more specific test such as PCR will continue to be key, given the decrease in incidence of Nm A and anticipated decrease of other Nm serogroups with future vaccination efforts.

One of the priority goals identified by the Defeating meningitis by 2030 Roadmap is the improvement of diagnosis at all levels of care, through the development and access to diagnostic assays [24]. An expert group gathered by WHO in 2018 [25] identified three essential objectives for the development of in vitro diagnostic tests (IVD) for meningitis diagnosis, including the rapid detection of epidemics in the African meningitis belt. The MeningoSpeed and PneumoSpeed could potentially meet this specific need. Issues that remain to be addressed, before procuring the test more widely, are costs, further thermostability improvements, and scale-up of production capacity with a reliable quality management system. Until elimination of meningitis epidemics in the region is achieved, one of three visionary goals of the roadmap, the use of rapid diagnostic tests will remain an important tool for meningitis control in Africa.

**Author Contributions:** Conceptualization, M.-K.T. and O.R.; Formal analysis, M.R.; Investigation, M.T., F.S., I.Y., K.D. G.S., H.M. (Halima Mainassara), H.M. (Haladou Moussa) and R.O.; Methodology, M.R. and R.O.; Validation, O.R.; Writing—original draft, M.R.; Writing—review & editing, M.T., F.S., I.Y., K.D., G.S., C.K., H.M. (Halima Mainassara), A.E.M., B.B., H.M. (Haladou Moussa), R.O., K.F., M.-K.T. and O.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study followed the principles governing biomedical research involving human participation and was carried out in line the principles of the Declaration of Helsinki to ensure that the rights, integrity, and confidentiality of participants were protected. The agreements of the ethics committees of WHO (WHO ERC.0002926), Niger (deliberation N◦ 35/2017/CNRES) and Burkina Faso (deliberation N◦ 2017-10-156) were obtained before the start of the study as well as for its extension into 2019.

**Informed Consent Statement:** Written consent was obtained from suspected meningitis cases or from their parents or legal guardians. Standardized information was read to the potential participant. The confidentiality of the data collected and the anonymity of the participants were ensured during and after the survey (no patient name appears in the databases).

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Review* **Defeating Paediatric Tuberculous Meningitis: Applying the WHO "Defeating Meningitis by 2030: Global Roadmap"**

**Robindra Basu Roy 1,2,\*, Sabrina Bakeera-Kitaka <sup>3</sup> , Chishala Chabala <sup>4</sup> , Diana M Gibb <sup>2</sup> , Julie Huynh 5,6 , Hilda Mujuru <sup>7</sup> , Naveen Sankhyan <sup>8</sup> , James A Seddon 9,10, Suvasini Sharma <sup>11</sup>, Varinder Singh <sup>11</sup>, Eric Wobudeya <sup>3</sup> and Suzanne T Anderson <sup>2</sup>**


**Abstract:** Children affected by tuberculous meningitis (TBM), as well as their families, have needs that lie at the intersections between the tuberculosis and meningitis clinical, research, and policy spheres. There is therefore a substantial risk that these needs are not fully met by either programme. In this narrative review article, we use the World Health Organization (WHO) "Defeating Meningitis by 2030: global roadmap" as a starting point to consider key goals and activities to specifically defeat TBM in children. We apply the five pillars outlined in the roadmap to describe how this approach can be adapted to serve children affected by TBM. The pillars are (i) prevention; (ii) diagnosis and treatment; (iii) surveillance; (iv) support and care for people affected by meningitis; and (v) advocacy and engagement. We conclude by calling for greater integration between meningitis and TB programmes at WHO and at national levels.

**Keywords:** tuberculosis; tuberculous meningitis; TBM; children

### **1. Introduction**

Tuberculous meningitis (TBM) is a devastating childhood disease, with one in five affected children dying and only one in three surviving without disability [1]. Depending upon tuberculosis prevalence, the age range being studied, and whether it is a population or hospital-based setting, TBM accounts from 1 to 10% of childhood TB cases, although there is considerable uncertainty around these estimates, given the challenges of making a microbiological diagnosis, the difficulties in discriminating TBM from other

**Citation:** Basu Roy, R.; Bakeera-Kitaka, S.; Chabala, C.; Gibb, D.M; Huynh, J.; Mujuru, H.; Sankhyan, N.; Seddon, J.A; Sharma, S.; Singh, V.; et al. Defeating Paediatric Tuberculous Meningitis: Applying the WHO "Defeating Meningitis by 2030: Global Roadmap". *Microorganisms* **2021**, *9*, 857. https://doi.org/10.3390/ microorganisms9040857

Academic Editor: James Stuart

Received: 29 March 2021 Accepted: 13 April 2021 Published: 16 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

meningitides on clinical grounds, and underreporting [2,3]. The forthcoming publication of the first global road map on defeating meningitis from the World Health Organization (WHO) provides an opportune moment to highlight the specific challenges and goals needed to defeat TBM [4]. Issues affecting children with TBM and their families lie at the intersection of the tuberculosis (TB) and meningitis clinical, research, and policy spheres and, hence, there is an accompanying risk that the specific needs of this vulnerable group are not fully addressed (Table 1).



\* Documents primarily related to meningitis were searched for "TB" or "tuberculosis" whilst documents primarily related to tuberculosis were searched for "meningitis." p denotes page.

In this article we therefore apply the framework outlined in the October 2020 draft "defeating meningitis" road map to TBM. As a team of paediatricians working on the SURE trial, the largest ever randomised controlled treatment trial in paediatric TBM, we focus specifically upon TBM in children [7]. The "defeating meningitis" roadmap explicitly places its emphasis on the four main causes of acute bacterial meningitis (meningococcus, pneumococcus, *Haemophilus influenzae,* and group B streptococcus), although it also

highlights that many of the goals are applicable to reducing the burden of disease from all causes of meningitis (Table 1) [4]. The "defeating meningitis" roadmap sits within a network of interconnected issues related to TB including universal health coverage, antimicrobial resistance and inclusion of people with disabilities [4]. Its vision is "Towards a world free of meningitis," although this is qualified, as meningitis is heterogenous and not amenable to elimination or eradication. Three overarching visionary goals are proposed: (i) elimination of bacterial meningitis epidemics; (ii) reduction of cases and deaths from vaccine-preventable bacterial meningitis; (iii) reduction of disability and improved quality of life after meningitis of any cause [4]. From the outset, the limitations of applying the roadmap to paediatric TBM are clear, as although it is, to some extent, vaccine-preventable and a significant cause of disability, it is not a disease that occurs in epidemics [1,8].

We consider the five pillars of (i) prevention; (ii) diagnosis and treatment; (iii) disease surveillance; (iv) support and care for people affected by meningitis; and (v) advocacy and engagement, to outline how this approach can be adapted to serve children affected by TBM. We conclude by calling for greater integration of the approaches to eliminate TB and defeat meningitis, especially in light of the challenges and opportunities presented to healthcare systems around the world tackling the COVID-19 pandemic.

#### **2. Pillar 1: Prevention**

Prevention of paediatric TBM is closely related to the prevention of TB infection and thereby TB disease, and severe forms of disease such as TBM. Strategies to control TBM should focus on TB prevention with a special focus on vulnerable groups. In regions with high TB prevalence, children under 5 years are most commonly affected with TBM [2,9]. Others at risk of TBM include those living with HIV and other immunocompromised children. The WHO lists the three primary strategies for TB prevention, including (i) TB preventive therapy (TPT); (ii) prevention of transmission through infection control; and (iii) BCG vaccination [5]. It is recommended to use TPT to prevent TB infection from progressing to TB disease. This is particularly important in immunocompromised hosts and young children in whom the risk of severe forms of TB, including TBM, is high. The WHO recommends TPT for three high-risk groups: household contacts (especially children under five years of age), people living with HIV, and other clinical risk groups [10]. This latter group, which includes those initiating anti-TNF treatment or transplantation, prisoners, health workers, homeless people, and people who use drugs, represents a relatively small proportion of children at risk in high TB prevalence areas.

In March 2020, WHO updated recommendations on TPT [10]. Alongside established regimes, such as 6 or 9 months of daily isoniazid, 3 months of weekly rifapentine plus isoniazid, and 3 months of daily isoniazid and rifampicin, new recommendations included:


The inclusion of rifampicin and rifapentine in regimens for TB prevention allows shorter treatment and higher completion rates. There is ongoing research to improve uptake and scale-up TPT, such as the Unitaid-funded: IMPAACT 4TB and CaP TB studies [5]. Despite clear guidance, the numbers of household contacts, including children, provided with TPT globally, remain dismally low at 20–30% [5,10,11]. It is therefore not surprising that, in a case series of children with TBM, TPT was not provided to any of the children who had known exposure to an adult TB case [12].

TB infection, prevention and control is part of the End TB strategy. Particular areas of concern for preventing TB transmission through infection control include:

• Newborn care settings. There are many documented outbreaks of TB among neonates with the source case usually being a mother or a member of staff with potential to infect many babies in the neonatal care unit. Neonates are particularly vulnerable for acute onset or development of disseminated severe disease [13,14].


In 2019, WHO released new guidance, and recommended administrative, environmental, and personal protection measures to limit the spread of TB [15]. The COVID-19 pandemic has threatened to derail the global progress in TB control. Alarmingly, it has been estimated that the COVID-19 pandemic could cause an additional 6.3 million TB cases globally between 2020 and 2025 [16,17]. However, there are some positives too. The expertise and experience in rapid testing and contact tracing for COVID-19 could be leveraged for TB case tracing and testing. Additionally, digital communication strategies developed during COVID-19 could be expanded for remote care, follow-up, and support of TB patients [18,19]. Lastly, the enhanced awareness generated during the COVID-19 pandemic could help emphasise basic infection prevention strategies in healthcare settings, such as the use of personal protective equipment (PPE), cough etiquette, and respiratory isolation [5].

The third key strategy is neonatal BCG vaccination. This has been shown to protect against TBM and disseminated TB in children with protection lasting up to the age of 10 years [8,20,21]. Data on protection beyond 15 years are limited; however, a small number of trials and observational studies suggest that BCG vaccination may protect for longer. Importantly, there are limitations to the potential for BCG to further decrease incidence of TBM in childhood, as the majority of children with TBM have been vaccinated with BCG. An investigational TB vaccine candidate (M72/AS01E) was evaluated in adults infected with *Mycobacterium tuberculosis* (*M. tuberculosis)*. The vaccine efficacy at month 36 was 49.7% (90% confidence interval [CI], 12.1 to 71.2; 95% CI, 2.1 to 74.2) in prevention of TB disease [22]. It remains to be seen if this vaccine will prove to have the effectiveness needed for more widespread use (Table 2).


**Table 2.** Selected strategic goals reproduced from Pillar 1 of the "defeating meningitis" roadmap and suggested paediatric tuberculous meningitis (TBM)-related activities [4].


**Table 2.** *Cont*.

LMICs = low and middle-income countries.

#### **3. Pillar 2: Diagnosis and Treatment**

Early and accurate diagnosis of TBM remains challenging. Diagnostic and treatment delay are the most important predictors of mortality and disability [23]. The challenges are numerous: (1) non-specific symptoms leading to misdiagnosis; (2) inadequately sensitive diagnostic tools to confirm TBM; (3) suboptimal sample collection for laboratory processing; and (4) lack of standardised region-specific training of healthcare staff to identify TBM and initiate appropriate anti-TB therapy.

Mycobacterial cultures are considered the "gold standard" diagnostic test for TBM. However they take 2–6 weeks to yield a result and require specialised laboratory facilities and experience not available at all healthcare levels. The rapid molecular test, Xpert MTB/RIF, has emerged as an important diagnostic test for all forms of TB, such that WHO have recommended it as the initial TB diagnostic test, replacing the acid-fast bacilli smear [24]. Next-generation Xpert MTB/RIFUltra (Xpert Ultra) has a sensitivity of 44–77% and specificity approaching 100% in the cerebrospinal fluid (CSF) of adults with TBM [25–27]. Xpert MTB/RIF offers value for clinical decision-making as it allows early initiation of anti-TB therapy and information on the presence/absence of rifampicin resistance and, hence, whether treatment for multidrug-resistant (MDR) TBM is indicated. However, the test requires an expensive platform, access to electricity, and costly consumables. Xpert cartridges costs USD \$10 in countries that are allowed concessional pricing; however, for high TB burden countries, this equates to a substantial expenditure [28]. Although the next generation Xpert Ultra has slightly higher sensitivity than Xpert MTB/RIF it does not have a sufficiently high negative predictive value to exclude TBM when the result is negative. Microbiological confirmation of MDR-TB in children is uncommon. In the absence of a microbiological diagnosis, MDR-TBM requires careful history-taking

including recent exposure to an infectious MDR-TB source case or someone who failed TB treatment or died from suspected MDR-TB.

Xpert MTB/RIF and Xpert Ultra should not be used as the sole diagnostic test in TBM and every attempt should be made to obtain samples from elsewhere such as sputum, gastric aspirates, lymph node aspirates, biopsies, to support the diagnosis. Recent rapid diagnostic tests, such as TB-LAMP (loop-mediated isothermal amplification), which amplifies MTB DNA and is implementable at peripheral health centre level, and urinary TB FujiLAM, which detects *M. tuberculosis* lipoarabinomannan antigen, remove the need for advanced laboratory expertise [29–31]. However, information on their role in TBM diagnosis, diagnostic performance in the paediatric population, and impact on mortality have yet to be clearly elucidated. In the absence of an optimal and accessible diagnostic test, diagnosis of TBM still relies on clinical symptoms and signs, CSF parameters, and radiology (chest radiographs at the minimum and neuroimaging where available), and, where available, bacterial, viral and fungal microbiological tests to distinguish TBM from other infective causes of meningitis. A clinical decision tool (i.e., CHILD TB LP- altered Consciousness, caregiver HIV-infected, Illness length, Lethargy, focal neurological Deficit, failure to Thrive, Blood/serum sodium, CSF Lymphocytes, CSF Protein) developed to facilitate early diagnosis of childhood TBM has been shown to accurately classify microbiologically confirmed TBM (sensitivity 100%, specificity 90%) [32]. However, this algorithm requires prospective evaluation as a rapid diagnostic tool in children with CNS infections in numerous geographical settings with varying TB burden and risk factors.

Bacteriologic confirmation of meningitis caused by *M. tuberculosis* relies heavily on obtaining adequate volumes of CSF (at least 6 mL for mycobacterial testing and additional volume for tests to exclude other causes of meningitis, e.g., pyogenic bacterial meningitis, *Cryptococcus* sp., and viral meningoencephalitis) [33]. As TBM is paucibacillary in nature, maximising the number of mycobacteria in a sample will increase the probability of a positive result. This is achieved by obtaining large volumes of CSF with subsequent centrifugation (3000× *g* for 15 min) to concentrate *M. tuberculosis* into pellet form [34]. Increased focus on the importance of CSF collection and processing should be incorporated into TB guidelines, and in training of healthcare and laboratory staff (Table 3). Even in the absence of positive TB microbiology, CSF taken for microscopy and biochemistry is important for TBM diagnosis. This highlights the pivotal role of frontline clinicians in obtaining informed consent for the lumbar puncture as part of the routine diagnostic approach. Parental refusal to perform lumbar punctures in children with suspected central nervous system infection is common in low-middle income (LMIC) settings; a cross-sectional study of 215 families of Pakistani children who had indications for lumbar puncture showed that 33% refused [35]. Common reasons for refusal were lack of knowledge about the risks of the procedure (30%) and fear of paralysis of lower limbs (49%). High levels of illiteracy, stigma associated with the procedure and potential differences of opinion amongst the extended family impact on LP consent in many LMIC settings. Training of healthcare staff on how to counsel families in these settings, and guidance on the process of informed consent with real-world scenarios may reduce the frequency of parental refusal for LP, given its importance to securing a diagnosis of TBM. Current research is focussed on detecting novel blood, CSF, and urine biomarkers (transcriptome, proteome, and metabolome) to diagnose TBM in children [36–39]. Such approaches and new biomarkers of cerebral injury in CSF offer much promise for future diagnosis of TBM in children [40].


**Table 3.** Selected relevant strategic goals reproduced from Pillar 2 the "defeating meningitis" roadmap and suggested paediatric TBM-related activities [4].

**Table 3.** *Cont*.


CSF = cerebrospinal fluid, WHO = World Health Organisation, MTB = mycobacterium tuberculosis, Rif = rifampicin resistance, TB = tuberculosis, TBM = TB meningitis, HIV = human immunodeficiency virus, LAM = lipoarabinomannan, LAMP = loop-mediated isothermal amplification, TPP = target product profile, LMIC = low and middle income countries.

The treatment of TBM is urgent to limit disability and death, especially in a condition with delayed presentation and diagnosis. The WHO regimen for drug susceptible TBM is two months of HRZE ((INH: 10 mg/kg), rifampicin (RMP: 15 mg/kg), pyrazinamide (PZA: 35 mg/kg), and ethambutol (EMB: 20 mg/kg) (HRZE)) intensive phase followed by a continuation phase of ten months of HR treatment, which is 6 months longer than the treatment for pulmonary TB. The 2 HRZE used in the crucial initial 2 months of TB treatment is recognised to be suboptimal in penetrating the blood–brain barrier and new regimens are being put into clinical trials [7,41,42]. The pathophysiology of TBM includes severe inflammation in the brain with potential for raised intracranial pressure, hydrocephalus, ischaemia, and infarction [43]. Adjuvant treatment with steroids in the initial 6–8 weeks has been shown in a Cochrane systematic review to improve mortality, but to have little effect on disabling sequelae [44]. However, the benefits of immunomodulatory treatment may depend on genetically regulated levels of inflammation [45]. More recently, data from an adult treatment trial of TBM suggests that addition of adjunctive aspirin may improve clinical outcomes in certain patient subgroups at high dose and this is under evaluation in an ongoing paediatric clinical trial [7,46].

The optimal dosing, choice, route and duration of antitubercular agents in TBM and the context of an inflamed blood brain barrier remains the subject of debate. In Cape Town, South Africa, children with TBM are treated with a shorter, higher dose 6 month regimen: INH: 20 mg/kg, RMP: 20 mg/kg, PZA: 40 mg/kg and ethionamide: 20 mg/kg for 6 months [47]. Although excellent outcomes have been reported with this "Cape Town" regimen it has never been subjected to a randomised controlled trial and it is not clear whether the apparent improvements are due to the regimen or high quality supportive care [47]. Given the potential to inform global TB policy with improved CSF penetration, halving the duration of treatment and hence improving concordance, the recently commenced SURE trial will compare the 12-month WHO regimen with a modified 6-month intensified regimen (RMP 30 mg/kg; INH 20 mg/kg; ethionamide replaced with levofloxacin, which penetrates CSF well) in children under 15 years [7]. The impact of adjunctive high dose aspirin vs placebo on severe disability will also be evaluated.

Successful treatment of MDR-TBM in children presents further challenges. MDR-TB meningitis is associated with high mortality owing to delayed diagnosis of drug resistance, the absence of a standardised approach to the management and poor CSF penetration of many MDR-TB drugs [48,49]. Whilst it is recommended that treatment regimens for MDR-TB should include at least 4–5 effective drugs, and careful consideration of drugs that

penetrate the CSF well such as fluoroquinolones and linezolid, there is a lack of evidence to inform best antibiotic combination, duration and doses [50]. In light of the increasing threat of MDR-TB, pre-clinical studies evaluating CSF-brain penetrating properties of new TB drugs and clinical trials assessing optimal drug regimens to improve outcome are urgently needed (Table 3).

#### **4. Pillar 3: Disease Surveillance**

Systems for TB surveillance are some of the oldest in the world, having begun more than two centuries ago with the recording of TB mortality in England and Wales [51]. As defined by the WHO, the primary aim of global TB surveillance is to assess the progress of TB control activities in the context of the End TB strategy [5]. WHO collates all the annual surveillance data provided by countries and then generates global estimates. These are adjusted by correction factors to account for underreporting, over- and under-diagnosis [5]. Globally, an estimated 10 million (range, 8.9–11.0 million) people fell ill with TB in 2019 [5]. Extrapulmonary TB represented 16% of the 7.1 million incident cases that were notified in 2019 [5].

The backbone of TB surveillance is case notification, which is statutory in many countries. The most accepted international case definitions for notification are the 2013 WHO definitions, last updated in January 2020, which incorporate: bacteriological status, classification of disease as pulmonary and extra-pulmonary, history of previous TB treatment, HIV status, and drug resistance [52]. Surveillance is incorporated within the national TB control programs in most LMIC as a monitoring and evaluation tool. The focus of surveillance at present are the infectious pulmonary cases for which more data are provided, while all extra-pulmonary forms are reported collectively. For example, the Global TB report provides no further details about extra-pulmonary TB including TBM [5].

Use of electronic registers for TB cases allows the capture of more detailed information such as a breakdown of the extra-pulmonary sites of disease but these are not part of formal reporting at national and global level. The National TB Elimination Program of India kindly shared data on CNS TB from their national e-Register for this publication. They reported that CNS TB among under 18 years olds contributed to 1.06%, 1.27%, and 1.66% of the total extra-pulmonary TB cases in the years 2018, 2019 and 2020, respectively (personal communication to Varinder Singh).

Amongst the other challenges that compromise detailed surveillance for TBM is the lack of a clear case definition. As paediatric TBM is a paucibacillary disease, many infectious and non-infectious neurological diseases may be misdiagnosed as TBM, such as partially treated bacterial meningitis, viral encephalitis, autoimmune encephalitis, and subacute onset neurodegenerative diseases. This is compounded by a paucity of neuroimaging facilities or facilities for molecular TB diagnostics in low resource settings. TBM, despite its own specific diagnostic and therapeutic challenges, is neglected in most TB programmatic guidelines, which typically only address the duration of treatment and adjunctive corticosteroids [53,54]. Both under- and over-reporting of TBM, as a result of misdiagnosis, is also likely in low-resource settings.

As TBM diagnosis requires some key laboratory facilities and clinical expertise, it is often centralised to large hospitals. Thus, TBM surveillance requires data from hospitals, both in the public and private sector, and also from laboratories where microbiological testing is performed. Even in public hospitals, reporting has many challenges. In a study from China, 25% of cases that were documented in the hospital records were not reported to the public health authorities [55,56]. Factors cited that also apply in other settings were unqualified and overworked health personnel, poor supervision and accountability at local and national levels, and a complicated incohesive health information management system [55,56]. The situation is complex in the case of private practitioners as the system requires them to provide patient details from outside the state health system. Limited practitioner time is an important barrier to TB notification and therefore user-friendly interventions, such as mobile based notification, have potential to improve notification,

although are constrained by technology and internet access [57]. In a pilot study of the use of a mobile interface voice-based TB notification system, only 6% of private practitioners were found to use it [57].

Although TB surveillance is one of the most well-established and prevalent surveillance systems across the globe, there is a need for a well-defined strategy for TBM surveillance, which is poorly quantified at present. The data thus collected will be helpful to understand the disease, its trends and inform appropriate action (Table 4).

**Table 4.** Selected relevant strategic goals reproduced from Pillar 3 the "defeating meningitis" roadmap and suggested paediatric TBM-related activities [4].


#### **5. Pillar 4: Support and Care for People Affected by Meningitis**

For many childhood survivors of TBM, the majority of whom live in LMICs, severe illness results in significant neurodevelopmental sequelae. A meta-analysis on treatment outcomes in childhood TBM demonstrated neurological sequelae in 54% of survivors [1]. However, data on the physical, cognitive, and behavioural sequelae of TBM, which have lasting socioeconomic implications for patients and their families, are limited and rarely include long-term follow-up. Children living with disabilities in LMICs are likely to experience poorer health and quality of life, reduced school participation and high rates of poverty when compared with their non-disabled peers [58,59], yet this has not been evaluated in TBM survivors.

Common impairments documented post TBM are in cognition, learning, emotion, and behaviour, all potentially affecting educational attainment and future employment. Poor neurodevelopmental outcome is associated with younger age, delayed presentation and treatment initiation, clinical severity and hydrocephalus, highlighting the need for increasing awareness of TBM and better clinical and diagnostic tools for timely initiation of treatment and management of sequelae [9,60,61].

The United Nations Sustainable Development Goal (SDG) 4, Pillar 4 of the Roadmap to Defeating Meningitis by 2030, and United Nations Convention on the Rights of the Child together highlight the need for timely identification and management of sequelae, together with reliable, valid measures to evaluate preventive and interventional efforts as well as improved access to appropriate support and care services [4,62,63]. Achieving these goals, considered standard of care in most high-income countries, will be challenging for several reasons.

First, there is a paucity of robust and standardised neurodevelopmental assessment tools (NDATs) developed for, and with normal reference populations, across different geographical and cultural settings. To fully understand the burden of impairment caused by TBM will, therefore, require appropriately adapted, as well as new, locally developed NDATs to detect both early developmental and later, emergent speech, behavioural and cognitive difficulties together with adaptive function [64]. A number of NDATs that have

either been adapted for use in or designed for different geographical and cultural settings are now being used [65]. However, these often require a high level of skill and training for healthcare professionals and are time-consuming. Screening tools to detect childhood disability and suitable for use by a broader range of healthcare workers are evolving. For example, the WHO Disability Assessment Schedule (DAS) 2.0, a generic instrument developed for adults to assess health and disability, has been adapted for use in children in a number of settings [66–68]. The culturally neutral, WHO Indicators of Infant and Young Child Development caregiver report tool, which monitors pre-school children across multiple LMIC settings, has been developed with feasibility testing and piloting across a number of LMIC planned [69].

Second, for children in whom development disability is identified, access to health and rehabilitation interventions to improve functioning and quality of life are limited. Even when available, the uptake is low with numerous barriers to access including cost, transport, physical inaccessibility, lack of appropriately trained healthcare workers, as well as cultural beliefs and stigma around disability [70,71]. Moreover, caregivers are likely to experience high levels of depression and anxiety with limited family and community support [72].

Third, children with developmental disabilities are often excluded from programmes and clinical trials of early child development interventions as it may be difficult to quantify improvement when developmental progress is the primary outcome, further compounding delay and reducing learning opportunities [73].

Fourth, there is a dearth of good quality information from large studies on the spectrum of evolving disabilities in children of all ages post TBM and limited information from LMICs on what interventions are effective for children with disabilities in general, their cost effectiveness and scalability. Collaborative efforts between stakeholders, funders and researchers, will help to improve outcome for children and families dealing with the sequelae of TBM. As TB clinicians and researchers, we must strongly advocate for these initiatives (Table 5).

**Table 5.** Selected relevant strategic goals reproduced from Pillar 4 the "defeating meningitis" roadmap and suggested paediatric TBM-related activities [4].



**Table 5.** *Cont*.

#### **6. Pillar 5: Advocacy and Engagement**

Advocacy is key to bringing about changes to policies and practices at institutional, community and individual level [74]. Advocacy for TBM fits into the scope of the meningitis roadmap by working with partners to raise public and political awareness of TBM and its devastating effects in order to improve diagnosis, treatment, prevention and support for affected families [4].

There are existing frameworks for TB control that advocates for TBM can leverage. The End TB strategy envisions the world free of TB, with zero deaths, disease, and its catastrophic consequences [75]. It is well known that TBM has devastating sequelae and contributes to significant morbidity and mortality. The child and adolescent TB roadmap aims to draw attention to the childhood TB epidemic and has placed advocacy and fostering partnerships as one of its key actions points [6]. Within these frameworks, global partnerships and national TB programmes can be utilised to raise the profile of TBM and highlight its significant contribution of TB-related mortality in adults and children. This will ensure that strategic interventions are formulated and planned with the aim of increasing its detection, treatment, and prevention. These interventions should include promotion of operational research to better understand the burden of TBM disease, its outcomes and consequences outside what is routinely monitored in country TB programmes. Partnerships with the academic community, civil society organisations, community, and patient advocacy groups can play a key role in raising the profile of TBM and developing champions that serve as key leaders in advocacy for TBM.

Existing advocacy materials for TB can be included with enhanced messaging, raising awareness of communities about the signs and symptoms of TBM in order to improve early recognition and promote early healthcare-seeking. In addition, healthcare providers, who are fundamental to diagnostic and care pathways, should receive continued education to improve their ability to recognise and diagnose the disease, together with managing the sequelae and other neurological disabilities that often arise from TBM. Furthermore, TBM should be incorporated into advocacy for syndromic neuro-disabilities such as epilepsy and cerebral palsy. Research funders should be targeted to harness resources

for research in novel, non-invasive, affordable diagnostics and more effective or shorter treatment regimens.

Community messaging should highlight the importance of BCG vaccination and counter vaccine hesitancy that is increasing worldwide. In addition, advocacy for effective household contact tracing and provision and monitoring of TPT for high-risk groups, including individuals living with HIV, should be promoted and strengthened within national programs with communities recognising their rights to this. This combination of efforts has the potential to prevent severe forms of TB in vulnerable populations and the catastrophic costs associated with TBM (Table 6).

**Table 6.** Selected relevant strategic goals reproduced from Pillar 5 the "defeating meningitis" roadmap and suggested paediatric TBM-related activities [4].



**Table 6.** *Cont*.

#### **7. Conclusions**

Both the "Defeating Meningitis by 2030: global roadmap" and "End TB Strategy" are ambitious plans, whose timelines for success are likely to be hampered by the COVID-19 pandemic. Collateral impacts from the pandemic include impact on the resilience of health systems, reporting and surveillance structures, health-seeking behaviours, together with cuts in funding from major donors [4,76]. Nevertheless, the pandemic also brings opportunities and provides an opportune moment to consider children with TBM who have amongst the highest burdens of morbidity and mortality of any of the conditions encompassed by the two plans. Although there are evident constraints and limits to how much the "Defeating meningitis roadmap" can be applied to children with TBM, creative and collaborative working between clinicians, policymakers, public health, and community and advocacy organisations can help bring us closer to defeating paediatric TBM.

**Author Contributions:** Conceptualisation: R.B.R.; writing—original draft preparation: R.B.R., S.B.-K., C.C., J.H., H.M., N.S., S.S., V.S., E.W., S.T.A.; writing—review and editing: R.B.R., S.B.-K., C.C., D.M.G., J.H., H.M., N.S., J.A.S., S.S., V.S., E.W., S.T.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** The SURE trial is funded by the Joint Global Health Trials Scheme of the Department for International Development, UK (DFID), the Wellcome Trust, the Medical Research Council (MRC UK) and the National Institute for Health Research (NIHR). Grant reference number: MR/R006113/1. RB is funded by an NIHR Academic Clinical Lectureship (CL-2018-20-001). JAS is supported by a Clinician Scientist Fellowship jointly funded by the UK Medical Research Council (MRC) and the UK Department for International Development (DFID) under the MRC/DFID Concordat agreement (MR/R007942/1).

**Acknowledgments:** We thank and acknowledge the Deputy Director General (TB), Central TB Division, Ministry of Health and Family Welfare, Government of India for providing the data of notified cases of CNS TB.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the writing of the manuscript or in the decision to publish.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Microorganisms* Editorial Office E-mail: microorganisms@mdpi.com www.mdpi.com/journal/microorganisms

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-1808-4