Lumpy Skin Disease: A Systematic Review of Mode of Transmission, Risk of Emergence and Risk Entry Pathway

Abstract

The spread of lumpy skin disease (LSD) to free countries over the last 10 years, particularly countries in Europe, Central and South East Asia, has highlighted the threat of emergence in new areas or re-emergence in countries that achieved eradication. This review aimed to identify studies on LSD epidemiology. A focus was made on hosts, modes of transmission and spread, risks of outbreaks and emergence in new areas. In order to summarize the research progress regarding the epidemiological characteristics of LSD virus over the last 40 years, the Preferred Reporting Items for Systematic reviews and Meta-Analyses statement guidelines were followed, via two databases, i.e., PubMed (biomedical literature) and Scopus (peer-reviewed literature including scientific journals, books, and conference proceedings). A total of 86 scientific articles were considered and classified according to the type of epidemiological study, i.e., experimental versus observational. The main findings and limitations of the retrieved articles were summarized: buffaloes are the main non-cattle hosts, the main transmission mode is mechanical, i.e., via blood-sucking vectors, and stable flies are the most competent vectors. Vectors are mainly responsible for a short-distance spread, while cattle trade spread the virus over long distances. Furthermore, vaccine-recombinant strains have emerged. In conclusion, controlling animal trade and insects in animal transport trucks are the most appropriate measures to limit or prevent LSD (re)emergence.

1. Introduction

Lumpy skin disease (LSD) is an emerging infectious disease of cattle and buffaloes which until recently had been considered as a neglected disease. First reported in Zambia in 1920, it spread to other African countries and became endemic in most sub-Saharan areas [1]. The disease was contained within this region until Egypt reported its first case in 1988 [1]. Then Israel experienced outbreaks in 1989 [2]. Between the 1990s and 2010, it was reported in countries of the Arabic peninsula, i.e., Kuwait in 1991, Lebanon in 1993, Yemen in 1995, United Arab Emirates in 2000, Bahrain in 2003, Israel (with recurring outbreaks in 2006 and 2007) and Oman in 2010 [3,4,5,6]. In 2012, Israel had another epidemic, and the disease reached Jordan and Iraq, followed by Turkey in 2013. Turkey is an important crossroad between Asia and Europe; in 2014, Azerbaijan and Iran reported their first cases, followed by Armenia, Greece and Russia a year later [6]. The spread continued towards Europe, and Georgia, Kazakhstan, Albania, Bulgaria, Montenegro, North Macedonia and Serbia reported outbreaks or cases in 2016 [6]. Certain countries, in particular European Member States, contained the outbreaks and no additional countries reported LSD cases during the 2017–2018 period. In 2019, LSD emerged in central Asia; China, Bangladesh and India reported their first cases during this year. Afterwards, it continued spreading in the center of Asia as Bhutan and Nepal reported their first cases in 2020 [6]. That same year, it also moved towards South-East Asia, i.e., Hong Kong, Myanmar, Sri Lanka and Vietnam. In 2021, LSD continued to be reported in new Asian countries, i.e., Mongolia, Pakistan and Taiwan, and continued spreading towards South-East Asia as Cambodia, Thailand and Malaysia reported their first cases. Finally, in 2022, Afghanistan and Indonesia reported their first cases [6].

Globalization, which has made changes in trading patterns of animals and animal products, global climate change and civil conflicts occurring in certain countries have aided the continuous spread of LSD virus (LSDV). LSD is a threat to livestock health and food security especially in lower income countries. These threats include important production losses, loss of draught power, reduced feed intake, disease management, trade restriction, and long-term convalescence. For this reason, it is listed as a notifiable disease in bovines by the World Organization for Animal Health (WOAH) [6].

These characteristics of the disease and several factors related to the evolving epidemiology of the disease raise a great concern in terms of introduction and difficulty of eradication, i.e., (i) non-stop and rapid spread towards South-East Asia, (ii) reoccurrence in countries where control and preventive measures had achieved eradication such as Russia, (iii) endemicity in previously free-countries such as Turkey and (iv) spread to regions experiencing a colder climate. Such concern has renewed scientific interest and a lot of new information on LSD epidemiology has appeared in the scientific literature.

The aim of this literature review is to summarize the research progress regarding the epidemiological characteristics of LSDV over the last 40 years. It will analyze trends in the literature and the modes of transmission and spread, in order to establish the disease introduction pathway(s) and to assess the conditions of LSD (re)emergence. The final objective is to highlight future research directions that will contribute to the improvement of LSD prevention, control and eradication.

2. Materials and Methods

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA) guidelines [7] (Appendix A). The literature search was performed on 1st of September 2022 in the PubMed (www.ncbi.nlm.nih.gov/pubmed (accessed on 1 September 2022)) and Scopus databases (www.scopus.com (accessed on 1 September 2022)), with the search term “Lumpy Skin Disease”. Only English-written articles, with an available abstract, and published between January 1980 and September 2022, were extracted. Editorials and books were excluded.

These articles investigated LSD hosts, transmission modes, risk factors of an outbreak and disease spread, as well as analysis of a risk of introduction into a new area. After excluding duplicates resulting from the search in two different databases, the remaining papers underwent a double-stage screening process, considering several inclusion and exclusion criteria, as shown in

Table 1. Inclusion and exclusion criteria for peer-reviewed studies included in this review.

Inclusion Criteria	Articles published from 1980 to September 2022
	Studies focused on epidemiological characteristics of LSDV (i.e., hosts, animal reservoirs, vectors)
	Studies reporting LSD modes of transmission
	Studies analyzing historical or new outbreaks data with the purpose to highlight LSD risk factors
	Studies describing quantitative and/or qualitative risk modelling of LSD
	Studies reporting LSDV in ruminants other than cattle
Exclusion Criteria	First exclusion criteria
	Editorials, letters to the editor
	Studies related to a pathogen other than LSDV
	Studies concerning the investigation of LSDV molecular characteristic
	Studies on surveillance of LSDV
	Second exclusion criteria
	Articles describing modelling of economic impacts of LSD
	Studies reporting vaccine efficiency, molecular interaction of LSD, or LSDV characteristics
	Studies to evaluate test performance or surveillance systems
	Studies on outbreak control
	Reports on clinical signs
	Studies focusing on the prevalence of LSD and excluding its transmission and the risk factors of outbreaks
	General literature reviews of LSD

Legend: LSD, lumpy skin disease; LSDV, lumpy skin disease virus.

. The first exclusion criteria were applied to articles titles only, and the second exclusion criteria considered article titles and abstracts. Afterwards, articles were screened by reading them in full, and same second exclusion criteria were applied.

Articles included in this systematic review included different types of epidemiology studies. While some described certain characteristics of LSD epidemic, others focus on specifics of LSDV. Thus, in order to allow a proper analysis and create a better description of the articles, these were categorized according to the study design of study, i.e., experimental vs. observational (cross-sectional or descriptive), literature reviews, risk analysis of LSD introduction in a country. Afterwards, the following information was extracted and inserted into a summary table (see Appendix B): type of epidemiological study, methodology, modes of transmission, risk factors associated with LSD introduction/spread to a new location, vectors/wild animals involved, reservoir hosts, main conclusions and limitations of the studies.

3. Results

3.1. Selection Process

The results of the selection process are shown in Figure 1. The search made in the scientific databases returned 692 articles after the removal of duplicates. By applying the first exclusion criteria only to the title of the articles a total of 385 articles were selected for the second screening process. In the second screening round a total of 261 articles were excluded based on secondary exclusion criteria applied to title and abstracts. A total of 124 were selected. The full text was accessible and read for 121 of them (three of the articles had to be excluded as their full text could not be accessed). From the articles read in full, 35 were excluded based on the secondary exclusion criteria. When there was doubt, a consensus meeting between the first and last author was held to decide on final exclusion. Finally, a total of 86 articles were included in the review. The full details of the reviewed articles are summarized in Appendix B.

3.2. Description of the Retrieved Articles

The frequency of publications shows that, between 1982 and 2010, only eight articles were published; in some years, there were no publications on LSD transmission or risk at all. After 2010, there was at least one article published per year, most of them being published afterwards (Figure 2). The highest number of publications was recorded in 2022 (N = 14), followed by 2021 and 2019, with N = 12 and N = 11 articles, respectively.

Based on the articles selected in the literature review process, the classification of studies, per category, is presented in Figure 3: most of them were observational studies, equally distributed between cross-sectional and descriptive studies. Experimental studies were mostly related with research on vectors. Only one literature review focusing on the role of Stomoxys flies in LSD transmission was included in this review.

Table 2. Type, methodology and objective of the study from the articles retrieved in this systematic literature review.

Type of Study	Methodology	Objective of the Study	Count	References
Experimental studies	Experimental infections Molecular techniques to detected LSD. PCR, neutralization, gene sequencing	Vector competence of blood-sucking insects/ticks.	20	[8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27]
		Semen/oocytes: determine if there is LSDV in reproductive organs of cattle and bulls, semen, oocytes after experimental infection	6	[28,29,30,31,32,33]
		Direct transmission: detect if there is a direct transmission between experimentally infected animals and healthy animals in a vector-proof environment	2	[34,35]
		Establish the presence of LSDV in meat and offal products	1	[36]
		Establish the spill over from a vaccine	1	[37]
Observational studies
Cross-sectional studies	Multivariable logistic or regression modelling	Risk factors for LSD outbreaks, i.e., herd size, movement of animals, weather conditions	11	[38,39,40,41,42,43,44,45,46,47,48]
	Ecological niche models Bayesian hierarchical models	Identification of geographic locations and weather conditions which are suitable for the occurrence/spread of LSDV	3	[49,50,51]
	Mathematical modelling	Evaluation of modes of transmission; establish transmission parameters and the R0 between animals	2	[52,53]
	Thin-plate spline regression	Determine the spread rate	1	[54]
	Time series and spectral analysis	Temporal trends and seasonal effects	1	[55]
	Spatial temporal analysis	Evaluate the epidemic between different geographical areas	3	[56,57,58]
	Weather based model	Estimation of population dynamics of potential vectors	1	[59]
	Kernel-based modelling	Determine the force of infection based on distance and seasonality	1	[60]
	Hybrid single particle. Lagrangian-integrated trajectory model	Identify wind events that condition vector transport	1	[61]
Descriptive studies	Field sampling of animals/suspected vectors	Detecting LSDV in animals other than cattle	9	[62,63,64,65,66,67,68,69,70]
		Intrauterine transmission of LSDV in natural conditions	1	[71]
		Semen from naturally infected bulls	1	[72]
		Detection, isolation of vaccine strains	6	[73,74,75,76,77,78]
		Isolation of LSDV in field-collected vector	6	[2,79,80,81,82,83]
Risk Assessment
Qualitative	WOAH Risk analysis guidelines	Probability of introduction and/or spread into a country considering different pathways	4	[84,85,86,87]
	WOAH Risk analysis guidelines and trade data	Probability of introduction and/or spread into a country considering different pathways	1	[88]
Quantitative	Quantitative import risk analysis	Stochastic model for the probability of LSD introduction in a free country via a specific pathway	2	[89,90]
	Created a generic framework	A single pathway of introduction, i.e., live cattle trade	1	[91]
Literature Review	Literature review	Literature review of the Stomoxys fly with additional information of outbreak data	1	[92]

Legend: LSD, Lumpy skin disease; LSDV, Lumpy skin disease virus; PCR; R0, the basic reproduction number; WOAH, World Organization for Animal Health.

shows the different studies and methodologies used in the selected articles.

3.3. Host of Lumpy Skin Disease Virus

Ten articles reported LSDV infection, via antibodies, clinical signs and PCR, in animals other than cattle (

Table 3. Lumpy skin disease virus detected/isolated per animal species, country and year of sampling.

Animal (Species)	Type of Samples, Test and Location	Country Year of Sampling	Reference
African buffalo (Syncerus caffer)	- Sampling of free ranging African buffaloes living close to cattle holdings - 150 out of 254 African buffaloes were seropositive by IFAT to capripox virus - 85 seropositive to LSDV by microserum neutralization test	Kenya 1981	[62]
African buffalo (Syncerus caffer)	- Sampling of 248 wild African buffaloes living in a national reserve park - Indirect ELISA test IgG to LSDV detected in 28.2% of samples - Seroneutralization test antibodies to LSDV detected in 7.6% of samples	South Africa 2014	[63]
Egyptian buffalo (*)	- Asymptomatic farmed buffaloes in contact with clinically infected cattle were skin and blood samples - Skin biopsies were tested by real time PCR. Three samples all tested negative - Serum samples were examined using ELISA: 17 out of 96 samples were seropositive.	Egypt 2016 to 2019	[64]
Asian buffalo (Bubalus bubalis)	- Clinical examination of LSD suspected cases in buffaloes belonging to small holders. - Detailed findings recorded in a clinical register gave the diagnosis of LSD; two animals were considered positive to LSDV.	India 2020	[65]
Buffalo (*)	- Blood samples collected from buffaloes presenting clinical signs of LSD - 15.2% of blood samples were seropositive to LSDV (type of testing used, e.g., ELISA, seroneutralization not specified)	Egypt 2018	[38]
Buffalo (*)	- Confirm LSD from reported cases in Iraqi buffaloes- PCR: eight positive out of 150 samples - Histopathology of skin lesions of 13 suspected LSD cases: only 1 positive	Iraq 2021 to 2022	[66]
Arabian oryx (Oryx leucoryx)	- LSD clinical sign observed in a captive bred female Arabian oryx, at a National Wildlife Research Centre, Saudi Arabia - Neutralizing antibodies used to establish LSD diagnosis in two Oryx (one with -clinical signs and the other without) in a herd of 90 animals. Electron microscopy was used on a single sample from the clinical affected animal. Sample was considered positive	Saudi Arabia 1989	[67]
Southern eland (Taurotragus oryx)	- 40 nasal swabs collected from wild ruminants shot during a hunting season on a private farm - Asymptomatic eland tested (two samples only) positive by conventional PCR and real-time PCR for LSDV	Namibia 2019	[68]
Southern eland (Taurotragus oryx); Springbok (Antidorcas marsupialis); Impala (Aepyceros melampus); Wildebeest (Connachaetes gnou, C. taurinus)	- Serum samples of different free living wild animals in South Africa in the major vegetation zones, i.e., semi-desert, Cape shrub land, grassland, woodland and forest transition - ELISA: serum antibodies detected in 10% of Wildbeest- species Connachaetes gnou, 23% of C taurinus, 7% of southern eland, 23% of springboks and 20% of impalas	South Africa 1993–1995	[69]
Giraffe (Giraffa Camelopardalis)	- Genome detection and isolation of LSDV in a zoo giraffe with LSD clinical signs - Phylogenetic analysis: isolate closely related to the previous Vietnamese and Chinese LSDV cattle strains.	Vietnam 2021	[70]

Legend: (*) article did not specify the buffalo species; DNA = Deoxyribonucleic Acid; IFAT = Indirect Fluorescent Antibody Test; ELISA = Enzyme-Linked Immunosorbent Assay; IgG = immunoglobulin G; LSD = lumpy skin disease; LSDV = lumpy skin disease virus; PCR = polymerase-chain reaction.

). Specifically, these animals were mainly free-ranging African buffaloes (Syncerus caffer) [62,63], which were classified as LSDV positive via serological test, and the Asian water buffalo [38,64,65,66]. Other reported African wild ruminant species with antibodies against Capripox viruses included an Arabian Oryx (Oryx leucoryx) in a wildlife reserve [67], southern elands (Taurotragus oryx) [69], Springboks (Antidorcas marsupialis) [69], Impalas (Aepyceros melampus) [69], and wildebeests (Connachaetes gnou, C. taurinus) (Table 3) [69]. The southern eland was also reported positive to LSDV by PCR [68]. A captive giraffe (Giraffa Camelopardalis) [70] was confirmed to be positive to LSDV by genomic detection and virus isolation in a Vietnamese zoo.

3.4. Modes of Transmission

3.4.1. Direct Transmission

Direct transmission was investigated or reported in 12 articles: direct contacts between animals (N = 3) [34,35,52], seminal (N = 7) [28,29,30,31,32,33,72], intra-uterine transmission (N = 1) [71] and meat and offal (N = 1) [36].

The transmission via direct contact between animals was deemed as being ineffective. The 1995 experimental study [34] tested this route of transmission by performing seven separate experiments, in which one uninfected cow was housed in close contact with two infected animals for a month, in an insect-proof facility. The results showed that, although infected cattle excreted LSDV in saliva, nasal and ocular discharges, none of the healthy animals developed clinical signs or produced detectable levels of serum neutralizing antibodies (i.e., no infection occurred) [34]. In an Israeli study, mathematical modelling was applied to investigate three possible routes of transmission in a same herd: (i) indirect contacts between different groups in the same herd, (ii) direct contacts or contacts via common drinking water within each group and (iii) transmission by contact during milking. In that study, modelling was applied to data from an LSD outbreak reported in a dairy herd. In the presence of an infected cow, the basic reproduction number (R0) of indirect transmission was estimated at 15.7, compared to 0.36 for direct transmission. These results provided further evidence that indirect transmission was the only parameter that could solely explain the entire outbreak dynamics [52] and that indirect transmission is likely to be far more important than direct transmission.

However, a 2020 study [35] which conducted a similar experimental study established for the first time the transmission of LSDV between cattle via direct contact [35]. In that study, cattle were infected using a vaccine-derived virulent recombinant LSDV strain (Saratov/2017) and both infected and healthy animals were housed together for a 60 day-period, which means twice longer compared to the previous study.

Transmission of LSDV via bull semen was shown to be a possible route of transmission. Experimental studies highlighted that LSDV was present in semen from experimentally infected bulls and that bulls were positive to LSDV in all semen fractions, excreting the virus for prolonged periods (longer than 28 days) even when obvious clinical signs of the disease were no longer apparent [28,29]. Moreover, the virus has also been detected in the semen of naturally infected bulls [72]. The testis and epididymis were identified as sites of LSDV persistence [30]. Seminal transmission to uninfected heifers was reported [31]. Vaccination is effective in preventing the excretion of LSDV as the semen of vaccinated bulls tested negative to LSDV [28]. Regarding the presence of LSDV in cryopreserved semen and embryo production, experimental studies [32,33] showed that the virus could persist in semen even if it undergo standard treatments [33]; in vitro yield was significantly reduced by the presence of LSDV in frozen-thawed semen [32] with the resulting embryos testing positive to LSDV. Furthermore, when testing an LSD-infected herd, neutralizing antibodies were detected in a one-day old calf, providing evidence of intrauterine transmission [71].

Based on one single study the transmission through bovine meat and offal products would be very low [36]. Following experimental infection, it appeared that lymph nodes and testicles of clinically and sub-clinically infected animals were reservoirs of live LSDV whilst live virus was not detected in deep skeletal meat [36].

3.4.2. Indirect Transmission via Vectors

The only route of indirect transmission retrieved in this literature review is via arthropod vectors. Twenty-nine articles focused on identifying possible vectors of LDS, and their potential role as mechanical (the vector simply “transport” the pathogen from one host to another), or biological vector (the pathogen undergoes replication and/or transformation inside the vector before transmission to other animals through subsequent blood meals) [93]. The identification of vectors potentially responsible of reported outbreaks was also assessed. These number of studies included 20 experimental studies (i.e., in laboratory conditions) [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27], six observational studies [11,12,13,14,15,59,79] and one systematic review that focused on the role of stable flies [92]. There were four studies [18,19,39,66] in which the primary objective was not sampling LSDV from vectors in the field, but they were part of the study and thus included in the results. The groups of vectors cited in the selected articles were: the stable fly Stomoxys calcitrans (N = 12), mosquitoes (N = 6), biting midges Culicoides spp. (N = 6), ticks (N = 12), horse flies (N = 2) and non-biting flies (N = 2). Thus, two classes of arthropods were identified as potential vectors of LSDV, i.e., Insecta and Arachnida (ticks).

Blood sucking vectors-Insects

Experimental studies focused on establishing the competence and/or capacity of transmitting LSDV by different blood-feeding insect vectors. Parameters investigated for each vector are shown in

Table 4. Parameters of insect vectors investigated in the experimental studies and LSDV in field-collected vectors.

Vector Investigated	Detection of LSDV (a) on the Vector	Detection of LSDV in a Specific Body Part of the Vector	LSD Viral Retention on the Insect	Evidence of LSDV Replication in the Insect	Transmission Attempts of LSDV	Basic Reproduction Number (R0)	Detection of LSDV in Field-Collected Samples
Stable fly
Stomoxys calcitrans	[8,9,10,12,13,14,15]	[10,12,14,15]	[8,9,10,12,13,14,15]	[8,9,10,12,13,14,15]	[8,12,15]	[9,11]	[59,79,80]
Stomoxys sitiens	[12,13]	[12]	[12,13]	[12,13]	[12]
Stomoxys indica	[12,13]	[12]	[12,13]	[12,13]	[12]
Mosquitoes
Aedes aegypti	[9,10,16,17]	[10,17]	[9,10,16,17]	[9,10,16,17]	[16]	[9,11]
Anopheles stephensi	[8]		[8]	[8]	[8]	[11]
Culex quinquefasciatus	[8,9,10]	[10]	[8,9,10]	[8,9]	[8]	[9,11]
Culex pipiens	[17]	[17]	[17]	[17]
Aedes japonicus	[17]	[17]	[17]	[17]
Biting midges
Culicoides nubeculosus	[8,9,10,17]	[10,17]	[8,10,17]	[8,10,17]	[8]	[9,11]	[39]
Culicoidess spp., C. punctatus	[17,39]	[17]	[17]	[17]			[17,39]
Horseflies
Haematopota spp.	[15]		[15]		[15]
Tabanus bromiums							[79]
Non biting flies
Musca domestica L.	[81,82]						[81,82]
Muscina stabulans	[81]						[81]

Legend: ^(a) LSDV = lumpy skin disease virus; R0, the basic reproduction number.

.

Four experimental studies [8,9,10,11] assessed the potential role of stable flies, mosquitoes and biting midges as vectors of LSDV. These studies allowed comparing the different potential vectors. A first experimental study carried out in 2003 [8] intended to reproduce the mechanical transmission of LSDV by several blood-feeding insects, i.e., Stomoxys calcitrans, Culex quinquefasciatus and Anopheles Stephensis (mosquitoes), and Culicoides nubeculosus. The transmission attempt was made 24 h after feeding. None of the susceptible animals seroconverted or showed any reaction to exposure (i.e., no transmission was achieved). Furthermore, there was no evidence of viral replication in any of the aforementioned species. The virus was detected by PCR in S. calcitrans up to one day post-infective feed, only immediately post-feeding in C. nubeculosus, after 8 days in Anopheles stephensis and after 6 days in Culex quinquefasciatus [8].

Two studies [9,10] focused on Stomoxys calcitrans, C. nubeculosus, and mosquitoes Culex quenquefasciatus and Aedes aegypti. Authors quantified the acquisition and retention of LSDV in different anatomical locations of these species. Neither study included experimental transmission to healthy animals, and insects were not tested for the virus beyond 8 days post-infection. The probability of vectors acquiring LSDV from a subclinically infected animal was very low (0.006) compared with the probability of infection from an animal with clinical signs (0.23). An insect feeding on a sub-clinically-infected animal was 97% less likely to acquire LSDV than one feeding on a clinically affected animal. The probability of acquiring LSDV was substantially greater when feeding on a lesion compared with feeding on normal skin or blood from a clinically affected animals [9]. There was no evidence of virus replication in the vector and the mean duration of viral retention differed among the four insect species, being the longest for Ae. aegypti (5.9 days) and S. calcitrans (5.5 days), followed by Cx. quinquefasciatus (4.5 days) and C. nubeculosus (2.4 days) [9]. After feeding on a skin lesion, LSDV was retained on the proboscis for the longest period (mean duration: 6.4 to 7.9 days), followed by the head/thorax (5.2 to 6.4 days), and for the shortest time in the abdomen (2.1 to 3.3 days) [10].

The basic reproduction number (R0) for the same aforementioned species of insects was determined in two studies [9,11]. The first study published in 2019 [11] used a transmission model that considered the underlying process involved in the vector-borne transmission to cattle. The parameters included in the model were estimated by reanalyzing data from published transmission studies and using Bayesian methods to quantify uncertainty. Sensitivity analysis allowed for calculating R0 and determining the parameters with the greatest influence. The other study [9] used data from their quantification study, combined with data from the earlier study [11] to recalculate the R0 values. The results of both studies were relatively consistent, but the wide prediction intervals should be noted. The estimated R0s were the following: 19.1 (95% predictive interval of 2.73–57.03) [9] and 15.5 (95% prediction interval of 1.4–81.9) [11] for S. calcitrans; 7.4 (95% prediction interval of 1.3–17.6) [11] and 2.41 (95% credibility interval of 0.50–5.22) for Ae. aegypti [9]; 0.8 (95% predictive interval of 0.9–3.5) [11] and 0.55 (95% credibility interval of 0.06–2.37) for Cx. quinquefasciatus [9]; 1.8 (95% prediction interval of 0.06–13.5) [11] and 7.09 (95% credibility interval of 0.24–37.10) C. nuberculosis [9]. An R0 for An. stephensis was only estimated in the earlier study and reached 1.6 (95% predictive interval of 0.2–6.0) [11].

When considering all these studies [8,9,10,11], it appears that S. calcitrans is likely to be the vectors with most capacity of transmitting LSDV, as well as the mosquito species Ae. aegypti. By contrast, C. nubeculosus, An. stephensi, and Cx. quinquefasciatus are likely to be inefficient vectors of LSDV.

Stomoxys calcitrans was the most studied vector in the present review, through (i) four observational studies which investigated or inferred its role in LSD outbreaks [2,59,79,80], (ii) eight experimental studies that determined its vector competency [8,9,10,11,12,13,14,15] and (iii) one literature review that discussed its role in the LSD epidemic in the Russian Federation [92].

In field settings, S. calcitrans was suspected to be responsible for the first known LSD outbreak in 1989, in an Israeli dairy farm. Authors suggested that LSDV-infected S. calcitrans were carried by the winds from Egypt which was experiencing LSD outbreaks at that time. Such a hypothesis was based on the circumstantial evidence that there was no cattle trade with countries experiencing LSD outbreaks, strict control measures were implemented at the border and winds were adequate to carry infected S. calcitrans from Egypt [2]. A later work also performed in Israel [59] calculated the monthly relative abundance of each dipteran in each farm that had been affected by LSD 1–2 year previously. The relative abundances of S. calcitrans in the month parallel to the outbreaks (December and April) were significantly higher compared to other Diptera, and their populations peaked in the months of LSD onset in the studied farms. Using a stable fly population model based on weather parameters to validate these finds showed that the peaks in S. calcitrans numbers matched the peaks in monthly numbers of newly affected dairy farms in the study area. However, the observations and model predictions revealed a lower abundance of stable flies during October and November, when LSD affected adjacent grazing beef herds. Authors inferred that these results suggest that another vector was probably involved in LSDV transmission in grazing beef herds [59].

In 2021–2022, two observational field studies which sampled different blood sucking possible vectors [79,80] reported that LSDV was isolated from stable flies. In both studies the number of samples was very small; from an LSD outbreak in Kazakhstan only two Stomoxis flies were tested with just one being positive [79] and from sampling a south African feedlot out of the 53 samples collected, eight were positive [80].

The single literature review [92] used a compilation of information regarding the entomology of Stomoxys calcitrans, the spread of LSD of cattle in Russia in the years 2015–2019, and the climatic conditions of the regions where LSD cases were recorded. With this data reviewed in the study, the authors concluded that the peak incidence of infection occurred in the warm month indicating the significant role of the Stomoxys fly in the epidemiology of the disease, fitting the hypothesis that this fly was the culprit for the occurrence of LSD outbreaks. However, it was noted that there were cases registered of LSD during the autumn-winter period of Russia when the intensity of the Stomoxys was minimal or completely absent and some of the outbreaks occurred at distances longer than the fly’s flying ability. Thus, authors indicated that there were other factors that influence the spread of LSD in Russia during this period of study [92].

In experimental studies, LSDV was isolated from different body parts of S. calcitrans, but mostly from the proboscis [10,12]; the fly excretes the virus both by regurgitation and defecation [12]. No evidence of virus replication was found in the vector [8,12,13,14]. Additionally, transmission was successful when it occurred immediately [12,15], but not 24 h after feeding [8]. These findings suggest that the stable fly is a competent mechanical vector of LSDV. Furthermore, another experimental study demonstrated the incompetence of three Stomoxys spp., i.e., S. calcitrans, S. sitiens and S. indica, as biological vectors after inoculation with LSDV [12].

The role of mosquitoes was experimentally studied in six selected articles [8,9,10,11,16,17]. The mosquito species of concern were: Ae. aegypti, Cx. quinquefasciatus, An. stephensi, Ae. japonicus, and Cx. pipiens. All species were shown to harbor viable LSDV in their bodies for 4 to 10 days after oral exposure [8,9,10,16,17], although LSDV retention in Ae. aegypti, Cx. quinquefasciatus and An. stephensi varied among the studies. As previously mentioned, only Ae. aegypti was able to transmit LSD to susceptible cattle under experimental conditions [16]. Thus, retention of LSDV in mosquitoes might be a general feature but the mechanism remains unknown. All experimental studies reported that the mosquito acts as a mechanical vector, however the mode is not as simple as “dirty-pin” type of virus transfer.

The potential role of biting midges in the transmission of LSDV was investigated in six selected articles [8,9,10,11,17,39]. Four experimental studies focused on C. nubeculosus [8,9,10,17]. The transmission of LSDV to susceptible cattle by collected field Culicoides spp. and laboratory-reared C. nubeculosus could not be reproduced, although LSDV was detected in their body parts and virus was retained for some days [17]. Moreover, there was no evidence of virus replication in C. nubeculosus [8,9,10]. These studies concluded that biting midges are not competent mechanical vectors of LSDV. A single field study found that C. punctatus [39] collected from a Turkish outbreak were harboring LSDV, and authors suggested that it could play a role in the transmission of the virus.

Regarding the potential role of horseflies in the transmission of LSD, no pre-2019 publication was retrieved from this systematic literature review. One experimental study [15] tested the transmission of the virus to cattle by the horsefly species Haematopota spp. Transmission from infected to healthy animals was achieved. Authors established that their large mouthparts are in favor of mechanical transmission, as they can retain a high volume of blood, and thus inoculate higher viral doses during interrupted feeding on several hosts [15]. Finally, they suggested that horse flies could be more competent than the stable fly, since there were less of the former than the latter in the experiment.

Only one observational field study investigated the LSDV infection rate of horse flies: LSDV was isolated from 14.29% of horseflies Tabanus bromiums sampled during an LSD outbreak [79]. Although they could not confirm the transmission, the authors did not discard the potential implication of horseflies in the outbreak.

Non-biting flies have never been investigated experimentally, given that they have never been inferred as LSDV carriers. However, two recent observational studies [81,82] trapped different insects within the frameworks of surveillance campaigns after LSD outbreaks in Russia and in the West Chinese border; authors isolated LSDV DNA in Musca domestica and Muscina stabulans.

Indirect transmission via ticks

Thirteen articles [18,19,20,21,22,23,24,25,26,27,66,79,83], all of them published from 2011 onwards, investigated the vectorial capacity of hard ticks to be vectors of LSDV. Only five studies [18,19,66,79,83] sampled ticks obtained from the field. The authors of one study sampled ticks (species not specified) from LSD infected buffaloes, but the virus was not detected [66]. Within the frameworks of another field study that relied on the sampling of different vectors from an outbreak in Kazakhstan [79], authors isolated LSDV in four Dermacentor marginatus and nine Hyalomma asiaticum ticks. A single study used a large sample size of ticks (4000 adult ticks). Three pools of infected ticks out of 20 were found positive to LSDV, which extrapolates to 15% of the whole specimens were positively infected (i.e., 600 positive ticks) [83]. A study which obtained samples from both Egypt and South Africa found viral DNA in four out of four collected Rhipicephalus spp. from Egypt; and of the 52 samples collected from South Africa, 11 were R. appendiculatus, four R. Boophilus, seven A. hebraeum, four H. truncatum, two Amblyomma sp. and six Rhipicephalus Boophilus sp. [19].

From 2011 to 2015, experimental studies focused on the role of ticks as either mechanical or biological vectors of LSDV. Thus, the main focus of experimental studies (

Table 5. Type of transmission researched and achieved in tick species in experimental studies.

Type of Infection/Transmission	Tick Species
	Amblyoma hebraeum	Rhipicephalus appendiculatus	Rhipicephalus decoloratus	Rhipicephalus annulatus
Intrastadial infection. Either nymphs or adult ticks without LSDV were allowed to feed on LSD infected cattle and then tested for the presence of the virus (body, or specific organs, e.g., salivary gland, gut)	[20,26,27]	[20,26,27]	[27]	[18] *
Intrastadial/mechanical transmission. Adult ticks are interrupted in their feeding from a cow experimentally infected with LSDV and placed onto susceptible cows which are later tested for LSDV infection (i.e., transmission occurred)	[23]	[25]
Transstadial persistence. Ticks at the larvae or nymphal stage are fed to repletion on cattle experimentally infected with LSDV. Nymphs then are incubated for molting into adults which are later tested for LSDV presence	[20,21,26,27] **	[20,26,27]
Transstadial/mechanical transmission. Ticks at the larvae or nymphal stage are fed to repletion in cattle experimentally infected with LSDV. Emerging adult ticks are transferred onto healthy cattle to check if they were infected (i.e., transmission occurred)	[23]	[20]
Transovarial passage. Female ticks were allowed to feed on LSDV experimentally infected cattle and later incubated to oviposit and for eggs to hatch. Eggs and/or mature larvae were tested for LSDV infection	[22]	[22]	[21,22,27]	[18]
Transovarial transmission. Female adult ticks or larvae were allowed to feed on LSDV infected cattle and later incubated to oviposit and for eggs to hatch. Hatched larvae were place into healthy cows which are later tested to check if they were infected (i.e., transmission occurred)	[22]	[22]	[22,24]

Legend: * [18] In this study, ticks were collected on naturally infected cattle. ** [21] In this study the LSDV was directly inoculated into the nymphs or adult ticks.

) was to determine if the tick would get infected after feeding (intrastadial infection), if it could persist in the tick’s life stages and progeny (transstadial, transovarial persistence), which tick stage(s) could infect an animal (transstadial, transovarial transmission).

The three tick species of interest were Amblyomma hebraeum, Rhipicephalus appendiculatus, and R. decoloratus. All the three species of ticks had intrastadial infection [20,26,27], transovarial passage and transmission [20,21,22,23,24,26,27]. Intrastadial transmission and transstadial persistence was demonstrated only by A. hebraeum, R. appendiculatus ticks [20,21,26,27]. An additional species which was investigated was Rhipicephalus annulatus [18]. These ticks were collected from cows in farms which were having LSD infections (i.e., naturally infected ticks) and incubated for oviposition to test the eggs and hatched larvae for the presence of LSDV. Thus, transovarial passage was observed [18].

One study [21] demonstrated the transstadial and transovarial transmission of LSDV by A. hebraeum nymphs and R. decoloratus female adults after a two-month exposure to night and daily temperatures of 5 °C and 20 °C, respectively, suggesting possible over-wintering of the virus in these ticks (i.e., possibility of these ticks being a reservoir for LSDV).

The latest study reported investigated the possibility of the tick being a biological vector. It attempted the in vitro growth of the virus in Rhipicephalus spp. tick cell lines and examined in vivo the presence of the virus in ticks collected from cattle during LSD outbreaks in Egypt and South Africa [19]. No evidence was obtained for replication of LSDV in tick cell lines although the virus was remarkably stable, remaining viable for 35 days at 28 °C in tick cell cultures, in growth medium used for tick cells and in phosphate buffered saline.

3.5. Emergence of Vaccine-Like Recombinant Strains

Between 2018 and 2022, nine articles [37,56,73,74,75,76,77,78,81,82] concluded that Russian and Chinese outbreaks were caused by a vaccine-like LSDV strain. For the first time, a vaccine-like strain (Neethling type) was identified during the 2017 Russian outbreak, in a region sharing a border with Kazakhstan [81]: the aforementioned strain was isolated in cattle and in house flies (Musca domestica) [81]. Although the route of introduction in Russia remains unclear, authors suggested that it was most likely due to the illegal use of the live attenuated homologous vaccines or the illegal movements of animals from Kazakhstan. It was highlighted in the Russian studies that while the use of homologous LSDV vaccines is not authorized in Russia, the Lumpivax vaccine (KEVEVAPI) was used in Kazakhstan shortly before the emergence of the vaccine-like strains [81]. This fortuitous finding led to a follow-up study on the epidemiological situation of LSD in Russia since 2016 [73]. The authors examined samples containing vaccine-like LSDV strains, collected in 2017 in the Privolzhsky Federal District, a Russian region that is geospatially outside the zone affected in 2016 and where live vaccines against LSDV had never been authorized or knowingly used. The study reported the widespread presence of vaccine-like LSDV strains in Russian cattle [73]. Following that first finding, sequential articles established the presence of vaccine-like strains. In 2018, the re-emergence of LSD was reported in Kurgan Oblast, Russia. The named ‘Kurgan/2018’ strain was neither from the vaccine nor from the field groups, strongly suggesting a novel recombinant profile [74]. In early March 2019, the Republic of Udmurtiya experienced an outbreak of LSD, while temperatures remain permanently below 0 °C, thus with no insect activity [75]. The causative LSDV (LSDV_Udmurtiya_Russia_2019) was shown to be a recombinant composed of a live attenuated Neethling-type vaccine strain (dominant parental strain) and a Kenyan KSGP/ NI-2490-like virus (minor parental strain) [75]. Furthermore, a recombinant vaccine-like LSDV from a 2019-outbreak in the Russian region of Saratov (Saratov/2019), where the first recombinant Saratov/2017 was documented, was described [76]. Even though both strains were isolated two years apart, Saratov/2019 seemed to be clonally derived from Russia/Saratov/2017, thus suggesting overwintering of the LSDV in the region since 2017.

A molecular epidemiology study conducted in Russia from 2015 to 2018 concluded that LSDV epidemiology had split into two independent waves. The 2015–2016 epidemic was attributable to a field isolate, whereas the 2017 epidemic, and in particular the 2018 epidemic, represented a disease importation, as the strain was not genetically linked [77]. A 2022 study analyzed the epidemiological evolution of LSD in Russia over a 6-year period, i.e., from 2015 to 2020 [56]. The results showed the disease tended to form spatiotemporal clusters in 2016–2018. These were associated with genetic changes in the virus and they were vaccine-like recombinant isolates; while the early clusters (2015–2106) were only formed by the field LSDV isolate [56]. Authors concluded that the LSD epidemiology could be affected severely by the use of homologous live-attenuated vaccines.

In 2019, China reported the isolation of a recombinant vaccine strain in the Xinjiang province, which borders Kazakhstan. That strain, named GD01/2020, was distinct from the two recombinant strains previously isolated in Russia [37]. Its origin remains unknown, but it was more probably introduced in the country in 2019 and responsible for the first outbreaks of that year, and eventually spread to other regions in the year 2022 [37]. This prompted to investigate insects as potential vectors involved and in 2022, a field study relying on the trapping of LSDV vectors was performed: the vaccine-like LSDV strain was isolated in two species of non-biting flies, i.e., Musca domestica L. and Muscina stabulans [82].

Given all the circumstantial evidence which pointed to the Lumpivax vaccine as the culprit of the emergence of these new recombinant vaccine strain, a study [78] analyzed the composition of two batches of the Lumpivax (KEVEVAPI) vaccine. Additionally, it investigated the possible link between the vaccine and the recent vaccine-like recombinant LSDV strains. By directly analyzing the genomes present in the vaccines they found that although labelled as a pure Neethling-based LSDV vaccine, the Lumpivax had a combination of at least three different Capripoxvirus strains: a Neethling-like vaccine strain, a Kenyan-like sheep and goat pox virus (KSGP) as well as an LSDV vaccine strain and a Sudan-like goatpox virus vaccine strain [78]. The genomic data of these finding indicated that the exchange of genetic material did not occur in co-infected animals but during vaccine production. The authors then concluded that the latest emergence of vaccine-like LSDV strains in a large part of Asia was therefore most likely the result of a spill-over from animals vaccinated with the Lumpivax vaccine which was poorly manufactured [78].

3.6. Risk Factors of Lumpy Skin Disease Outbreaks and Spread

Table 6. Risk factors that were identified with LSD occurrence or reoccurrence in the articles retrieved from the systematic literature review.

Identified Main Risk Factors	Country/Region of Study	Reference
Seasonality
Risk of outbreaks increases with higher temperature and/or rainfall	Egypt, Middle East, Balkans, Iran, Ethiopia, Albania, Eurasia, Uganda, Eastern and central Asia, Turkey, Russia	[40,41,42,49,50,51,53,54,55,56,57,60]
Animal movements or trade	Egypt, Balkans, Ethiopia, Turkey, Kazakhstan	[39,40,41,43,44,45,54]
Herd characteristics
Type of holdings, i.e., backyard, commercial farms	Turkey, Middle East, Russia	[39,50,56]
Herd size	Ethiopia, Kazakhstan	[43,44]
Cattle characteristics
Age	Mongolia, Egypt, Uganda, Ethiopia, Turkey	[38,40,41,42,46,48]
Breed	Turkey, Egypt, Bangladesh	[38,39,40,47]
Sex	Turkey, Uganda, Mongolia, Bangladesh	[41,42,46,47]
Farm location/landscape
Urban and mixed rain-fed arid livestock system	Middle East	[50]
Areas mostly covered with croplands, grassland or shrub land	Eurasia	[57]
Presence of a water body near the farm (e.g., lake, river, pond, well)	Turkey, Ethiopia, Mongolia	[39,46,48]
Type of agro-climate	Ethiopia	[45,55]
Type of herd management
Water sources: communal or located in farm	Egypt, Uganda, Ethiopia, Mongolia	[42,43,45,46]
Grazing: private or communal/pastoral	Uganda, Egypt, Ethiopia	[40,42,43,45]
Contact of cattle with other animals (e.g., buffaloes, sheep)	Egypt, Uganda, Ethiopia	[40,42,48]
Cattle density	Eurasia, Middle East	[50,57]

summarizes the selected cross-sectional studies (N = 17) which identified the main herd level risk factors for LSD cases and what geographic and climatic conditions are favorable to the disease occurrence and spread.

Farm level risk factors were assessed using multivariable logistic regression models in ten studies [38,40,41,42,43,44,45,46,47,48].

Table 7. Odds ratio retrieved from the studies that used multivariable logistic regression models.

Category Factor	Risk Factor		Odds Ratio (95% C.I.)	Reference
Herd characteristics	Genus/breed	Buffalo Cattle	Reference 4.08 (1.98–8.4)	[38]
		Baladi Mixed Holstein	Reference 4.59 (1.83–11.48) 4.58 (1.73–12.12)	[40]
		Local Cross breed	Reference 3.58 (1.40–9.17)	[47]
	Sex	Male	Reference
		Female	19.29 (2. 46–151.32)	[41]
			1.72 (1.02–2.92)	[42]
			2.40 (1.11–5.16)	[46]
			3.96 (2.16–7.27)	[47]
	Age	<1 year 1–2 years >2 years	Reference 2.35 (1.48–3.7) 1.33 (0.88–2.01)	[38]
		<1 year 1–3 years >3 years	Reference 1.41 (0.63–3.11) 2.49 (1.17–5.32)	[40]
		>24 months <24 months	Reference 21.1 (8.83–50.43)	[41]
		0–12 months 13–24 months >25 months	Reference 1.24 (0.63–2.44) 1.96 (1.15–3.34)	[42]
		0.5–1 year 1–4 years ≥4 years	Reference 1.38 (0.90- 2.09) 2.44 (1.67- 3.55)	[48]
		Calf Young Adult	Reference 0.21 (0.02–1.71) 0.05 (0.01–0.37)	[46]
	Herd size	Small (2–11 animals) Medium and large (>12 animals)	Reference 19.3 (1.4–50)	[43]
		Small Medium Large	Reference 0.68 (0.54–0.84) 0.63 (0.49–0.81)	[44]
Management	Grazing system	Communal/pastoral Fenced farm Zero grazing	Reference 5.26 (2.64–10.48) 0.28 (0.06–1.44)	[42]
		Separate Communal Both	Reference 1.55 (0.91–2.60) 0.75 (0.39–1.42)	[40]
	Communal water sources	No Yes	Reference 3.28 (2.11–5.09)	[40]
			3.31 (1.42–7.71)	[42]
	Grazing and water sources	Separate/Private Communal	Reference 4.1 (2.02–6.18)	[45]
			14.44 (2.23–94.0)	[43]
	Water source	River Pond Tube well	Reference 0.18 (0.06–0.53) 0.16 (0.05–0.47)	[46]
Management	Free animal movement	No Yes	Reference 0.36 (0.24–0.52)	[40]
	Contact with other animals	No Yes	Reference 3.40 (1.62–7.10)	[40]
			0.41 (0.23- 0.74)	[48]
	Contact with buffalo	Never Daily Weekly/monthly	Reference 1.78 (0.50–6.31) 0.49 (0.29–0.85)	[42]
	New introduction of cattle in the herd	No Yes	Reference 8.5 (6.0–11)	[45]
			2.22 (1.32–3.71)	[40]
			4.43 (2.6–7.5)	[43]
	Purchase of animals	No Yes	Reference 11.67 (8.87–15.35)	[44]
	Sale(s) of animals during LSD outbreaks	No Yes	Reference 1.24 (1.06–1.45)	[44]
	Vaccination	No Yes	Reference 0.13 (0.05–0.34)	[41]
Environment	Season	Winter Autumn Spring Summer	Reference 0.19 (0.02–1.50) 0.87 (0.29–2.51) 7.30 (3.97–13.42)	[40]
	Mean annual rainfall	800–1000 mm 1001–1200 mm 1201–1400 mm	Reference 5.60 (2.35–13.34) 4.58 (2.23–9.40)	[42]

summarizes the odds ratio (OR) obtained from such models. LSD positivity (i.e., outcome variable) was determined through blood sampling or clinical signs. There were different reported risk factors, being the three main reported risk factors (i.e., higher odds of presenting LSD): female cattle [41,42,46,47], animal movements (introduction of new cattle and sales) [40,43,44,45] and communal watering/grazing systems [40,42,43,45]. Other identified risk factors were genus and breeds (local breeds and buffaloes less likely to present LSD clinical signs) [38,40,47], and contact with other animals (sheep, goats, buffalo) [40,42,48]. Age and herd size showed different results as their group categories differed in the studies. One study showed a higher risk for medium and large size herd [46], and another the contrary [44]. Likewise, age showed various results, young cows had higher risk [38,46] and in others older ones were at risk [40,42,48]. Two studies which included weather conditions in their models found that higher risk was found the summer season [40], and a mean annual rainfall of 1001–1200 mm [42].

Three studies used ecological niche modelling to investigate the association between environmental factors (e.g., climate and land cover) and location data on disease outbreaks [49,50,51]. These associations were then used to predict the geographic distribution of LSDV in underreporting regions. Two of those studies focused on used land geography, not borders [49,50], thus including several countries, while the other used data from an Iranian region [51]. These studies concluded that environmental predictors contributing to the ecological niche of LSDV were: annual rainfalls, land cover, higher mean diurnal temperature range, type of livestock production system and global livestock densities. One study [49] identified wind speed as an important driver explaining the observed distribution of LSDV; higher wind speeds were negatively associated with LSDV incidence.

Another study used spatial regression model to predict the risk of LSD spread in neighboring free-countries of Europe and Central Asia [57]. They reported a significant effect of land cover, cattle density of the area, as well as higher annual mean temperature and higher mean diurnal temperature range on the occurrence of an LSD outbreak [57]. Using time series analysis and spatial distribution to detect seasonality and cyclical patterns in LSD outbreaks reported that LSD incidences were registered in warm and humid highlands [55]. Likewise, when analyzing the LSD epidemic from 2015 to 2020 in Russia the seasonality of LSD for that period showed that outbreaks occurred during warm months between May and October with the highest peak of incidence in July. It also reported cases in November 2018 and March 2019 when there were winter conditions (snow and freezing temperatures) [56]. It also showed that the distribution of outbreaks tended to occur at higher levels in backyard cattle compared to commercial farms [56].

A study using mathematical models [53] reported the daily transmission rate between animals was slightly lower in the crop–livestock production system (0.072; 95% CI 0.068–0.076) compared to an intensive production system 0.076 (95% CI 0.068–0.085) [53]. Similarly, a 1.07 R0 (95% CI 1.01–1.13) was estimated between animals in the crop–livestock production system (95% CI 1.01–1.13), vs. 1.09 between animals in the intensive production system (95% CI 0.97–1.22) [53].

Regarding the spread modalities, the studies included in this literature review [54,58,60,69] reported that short-distance spread (i.e., between herds) was most likely attributed to a dispersal by arthropod vectors, whereas long-distance spread (i.e., transboundary, introduction into new geographical areas) was related to livestock movements. Both short- and long-distance spreads are associated with climatic conditions, especially a high temperatures and rainfalls. A study performed in the Balkans suggested that LSD was mostly transmitted at a rate of about 7.4 km/week and was due to a local, vector-borne spread [54]. However, a faster transmission at longer ranges, i.e., around 54.6 km/week, which is less frequent, was attributed to movements of infected animals [69]. Another study used a Kernel-based approach to describe the transmission of LSDV between herds in Albania [60]. All transmission routes were combined in a single generic mechanism with the probability of transmission from an infected to a non-infected herd assumed to depend on the distance between them (i.e., transmission). The authors inferred that transmission occurred over <5 km distances, which can be attributed to vectors, but with an appreciable probability of transmission over longer distances, that can be related to livestock movements [60]. Spatio-temporal analysis of LSD outbreaks that affected dairy farms in north-eastern Thailand discovered that these outbreaks occurred in numerous dairy farms over a short period of time, and that several affected farms were concentrated in the area [58]. Based on these findings and on the fact that cattle movements between dairy farms are few, the spread was attributable to vectors. A geographic information system (GIS) software [41] concluded that the introduction of the disease in Turkey may have originated from Syria and Iraq, as movements of live animals are reported across the Syria–Iraq border; furthermore, the first outbreak was recorded near the border.

Another climatic factor that has been under consideration of long-distance spread by carrying infected vectors is winds. Following the previous study [2] which proposed the hypothesis that the first LSD outbreak in Israel was most likely caused by the Stomoxys carried from winds of Egypt, Klausner et al. 2017 [61] identified relevant synoptic systems that could have allowed long-distance dispersal of infected vectors by wind from Egypt to Israel in the month preceding the 1989 and 2006 outbreaks [61]. However, this is conditioned by the vector’s survival.

3.7. Risk Analysis of Introduction of Lumpy Skin Disease to a Free-Area

Eight studies assessed the risk of LSD introduction in a country, i.e., five qualitative [84,85,86,87,88] and three quantitative risk assessments [89,90,91] (Table 2). With the exception of one study [87] conducted in Turkey, all risk analyses related to importation were performed in three historically LSD-free European countries, i.e., United Kingdom [84,85], Ukraine [86] and France [88,89,90].

All qualitative assessments [84,85,86,87,88] determined that the risk of introduction and/or spread of LSD in a country by pathways others than animal movements or vector transmission (excluding the tick) was “negligible”. Although considered slightly higher, the risk of introduction via animal movements or arthropod vectors (excluding the tick), was still estimated as “low”.

As LSD is endemic in Turkey [87], the following risk question was raised: “What is the probability of cattle LSD being introduced in the animal market?” Based on different release scenarios, the risk was considered as “high”. In the overall exposure assessment, the authors considered two different pathways, i.e., the probability of cattle being exposed to LSDV during seasonal migration—risk considered as “high”—and the probability of exposing cattle to LSDV from veterinary equipment—risk considered as “medium” [87].

Regarding the quantitative approach, stochastic models assessed the risk of LSD introduction in France [89,90]. One study considered the risk of introduction by arthropod vectors through animal transport trucks [89]. The annual risk of LSDV being introduced by St. calcitrans travelling in animal trucks was between 6 × 10⁻⁵ and 5.93 × 10⁻³ (median: 89.9 × 10⁻⁵); it was mainly related to the risk that insects transported in vehicles come from high-risk areas to enter French farms. The risk associated with the transport of cattle to slaughterhouses or horse transport was much lower (between 2 × 10⁻⁷ and 3.73 × 10⁻⁵, and between 5 × 10⁻¹⁰ and 3.95 × 10⁻⁸, for cattle and horses, respectively). The other risk analysis [90] focused on the importation of cattle in France. Authors estimated that the probability of the first LSD outbreak to occur after importation of infected live cattle for breeding or fattening was 5.4 × 10⁻⁴ (95% probability interval (PI): 0.4 × 10⁻⁴; 28.7 × 10⁻⁴) in the summer and 1.8 × 10⁻⁴ (95% PI 0.14 × 10⁻⁴; 15 × 10⁻⁴) in the winter [90].

A generic framework for spatial quantitative risk assessments of infectious disease used LSD as a case study. Such an approach was carried out to assess the risk of LSDV spreading to other European countries after its introduction in the Balkans, in 2016 [91]. One single pathway of introduction was considered, i.e., registered movements of cattle: the highest mean probability of infection was in Croatia, followed by Italy, Hungary and Spain. Figure 4 illustrates a summary of the main modes transmission and spread which were established in this literature review.

4. Discussion

The aim of this paper was to review the general epidemiological characteristics of LSD described over the last 40 years in order to better understand the continuous emergence and spread of LSD to new areas. Unlike other reviews, which have usually focused on specific aspects of the disease in determined locations/regions, this systematic review is the first that aimed to cover aspects of epidemiological data related specifically to LSD modes of transmission, pathways of introductions and conditions of (re)emergence.

During the last 5 years, the research on LSD modes of transmission and risk factors or areas at risk of an outbreak has substantially increased, which confirms that the disease is becoming a global concern. Such increased interest is correlated to the arrival of LSD in Eastern Europe, Russia and Asia. The methodologies used have also evolved, as analyses have focused on finding additional LSD vectors and on geographical niches suitable for LSD to become endemic.

LSDV is host-restricted, similarly to other viruses of the genus Capripoxvirus. Although diagnosis of LSDV was performed mostly by serological methods, of which the main limitation is the lack of distinction between all Capripoxviruses, it is safe to assume that besides cattle, the other affected species are African and Asian water buffaloes, and just a few additional wild ruminant species [62,63,64,65,66,67,68,69,70]. Buffaloes seem to be more resistant to the disease than domestic cattle as studies reported less seropositivity, although it should be considered that the number of tested samples which studies reported were usually small. It was also suggested that the African buffalo could maintain the LSDV during non-epidemic periods [63]. This inference however was made only on the basis of positive samples with no additional information given to the context of when the samples were taken (i.e., time of year, other LSD outbreaks in the area). Thus, the role of the buffalo in the epidemic of LSD still remains to be elucidated. Indeed, to date, no experimental infection has been conducted in buffaloes, to establish the clinical signs or viraemic periods. This is of particular importance as, in some countries, buffaloes live close to or are part of the herd; they could represent a source of LSD infection in cattle herds. Moreover, they live also in countries which are still LSD-free, so their infection might go unnoticed until an outbreak occurs in cattle. Thus, understanding the biology of LSDV with the buffaloes would give a better insight of its role in the epidemiology of LSD.

In this review, little evidence was reported regarding the role of other wild ruminant species as LSDV hosts or sources of outbreaks. This is expected as studies on wildlife prevalence require economic and human power resources. All but two studies [68,70] reported wild animals positive to LSDV using serological testing. Although they reported them as LSD positive, this type of testing has the main limitation that current serological tests for LSDV cannot differentiate antibodies (Abs) to the virus from Abs towards other Capripoxviridae, i.e., Sheeppox virus and goatpox virus. Thus, it cannot be known with certainty that it was the LSDV causing the immunology response. Another important consideration is that animals with a mild or asymptomatic LSDV infection do not always develop a level of Abs detectable by a neutralization assay. Additionally, serological positivity does not necessarily imply that the virus replicates in the animals and that there is excretion; thus, they may not be able to transmit the virus. This could explain why clinical signs were only reported in one captive Arabian Oryx [67] and one giraffe [70]. Wild animals showing clinical signs of LSD are likely to be more susceptible to predators, which could explain the lack of reports of clinical disease in wild species. In addition, the presence of LSD clinical signs in wildlife might be easily missed, as the monitoring of skin lesions is difficult or impossible in their geographical settings. With all these considerations taken into account, it could be possible that the actual number of LSDV-infected wild ruminants may be considerably higher. Regardless of the difficulties mentioned, studies on LSD prevalence in wildlife should be encouraged as the virus may affect other Asian or European wild life, particularly those of the Bovidae family such as the European bison (Bison bonasus). Indeed, if LSD is introduced in a new geographical area where different wild ruminants coexist (either farmed or free ranging) and are naïve to LSDV, they could be infected transmit and maintain the disease. This could modify the dynamics of LSD epidemiology, making future outbreaks harder to control.

Regarding the modes of transmission, evidence from studies included in this literature review shows that direct or indirect transmission without the intervention of vectors is ineffective. The latest study that tested this route [35] managed to achieve a direct transmission between animals. Although there were important differences compared to the previous study [34] (virulent recombinant field strain and longer period of co-housing), such a finding highlights the importance of establishing further studies on LSDV biology. It is a priority to gain insights into whether the transmission achieved in this study is a de novo-created feature absent from both parental strains of the novel (recombinant) LSDV isolate used, or whether it was dormant but unlocked after genetic recombination. The study [52] which used mathematical modelling to estimate parameters of transmission modes also established that direct transmission was unlikely. However, the data used in the latter study came from an Israeli LSD outbreak in which all animals showing severe clinical signs were removed from the herd immediately, which may have artificially reduced the consequences of animal-to-animal contact.

Regarding other modes of direct transmission, the only plausible mode seems to be via seminal pathway. Experimental studies showed that LSDV is present in semen and seminal transmission was also achieved [28,29,30,32,33,72]. LSDV was detected in frozen semen samples which were collected from naturally infected bulls [31]. However, the effectiveness of such mode of transmission in the field still needs to be assessed. Given that laboratory conditions are controlled (e.g., infection of bulls with a virulent LSDV strain, the sample being collected during the viraemic period), the scenario differs from that which occurs in the field. The same comment is worth making for intrauterine transmission as a report included in this literature review mentioned that one single calf was considered as LSD-positive based on neutralizing Abs concentration [71]. It is unknown at what stage of pregnancy the cow was infected by the virus, and only one single calf was considered. Thus, these routes are still considered as unimportant when considering the spread of LSDV into new geographic areas (in contrast to other viruses of the genus Capripoxvirus, i.e., sheep and poxviruses in which direct contact or via aerosol are important).

Mechanical indirect vector-borne transmission is still considered as the main mode of transmission of LSDV, thus vector capacity and competence were extensively investigated, both in experimental and field studies. It important to distinguish the terms “vectorial capacity” and “vector competence”. Vectorial capacity is a measure of the transmission potential of a vector borne pathogen within a susceptible population. Vector competence, a component of the vectorial capacity equation, is the ability of an arthropod to transmit an infectious agent following exposure to that agent [94]. This distinction was not always made in the articles retrieved in this literature review as these terms are often used interchangeably to describe the ability of a vector to transmit a disease. Although this distinction was not always clarified in the research articles, it can be concluded that experimental studies focused mainly on the competence of hematophagous insects and hard ticks. Regarding vector competence, it is safe to assume that from the tested vector the stable fly Stomoxys spp. is the most competent vector of LSD as it could transmit LSDV in more than one of the experiments and presented the longest LSDV harboring time [8,9,10,11]. Given that it is the vector with the highest competency, it is also the vector with the highest vectorial capacity, as some observational descriptive and cross-sectional studies and the literature review determined they were the most abundant and inferred as the culprit of LSD outbreaks [2,59,79,80,92].

Moreover, studies reported it with having the highest R0 within the blood sucking insects studied (i.e., stable fly Stomoxys calcitrans, mosquitoes Ae. Aegypti, Cx. quinquefasciatus, C. nubeculosus) [9,11]. Furthermore, this insect is ideally suited to this type of virus transmission as it has a painful bite, which results in animals taking defensive actions such as tail switching, thus preventing the completion of a full blood-meal (i.e., interrupted feeding) and moving into the next animal [8]. This characteristic increases their vectorial capacity.

Given their importance as biological vectors in several diseases, mosquito species were among the blood sucking insects studied in experimental conditions. From the species studied, Ae. aegypti seems to be the most probable competent as it harbored the virus for the longest period [9], presented the highest R0 among the three mosquito species [9,11] and was shown to be fully capable of LSDV mechanical transmission [16]. By contrast, An. stephensi, and Cx. quinquefasciatus are more likely to be inefficient vectors of LSDV. However, considering that, on one side, in laboratory experiments, mosquitoes are fed via spiked blood through artificial membranes or cotton pads soaked in blood spiked with LSDV and, on the other side, its anthropophilic character (not relevant in a farm environment), its capacity as an LSD vector is mostly likely reduced in natural field conditions.

The biting midges have been proposed as vectors for LSD as they play a major role in the spread of other important ruminant pathogens, i.e., Bluetongue and Schmallenberg virus. However, the results show that they should be considered as incompetent vectors for LSD. Indeed, under laboratory conditions, C. nubeculosus was not able to transmit the virus to susceptible animals, no viral replication was observed and they were already negative to LSDV 24 h post-feeding [8,9,10,17]. Given its poor vector competency, although the virus was isolated from C. punctatus collected on infected farms [39], it is probable that its capacity to transmit the disease is low.

Until recently, there was no direct evidence of the role of tabanids in the transmission of LSDV although they are able to mechanically transmit a wide range of pathogens (e.g., Trypanosoma evansi, Besnoitia besnoiti) and are regularly found around cattle. A recent study achieved the transmission of LSDV by tabanids, and even inferred that they could be more efficient than stable flies in transmitting the virus, given their large mouth. Thus, tabanids could be competent mechanical vectors. Given that this was the only experimental study which used tabanids [15] and only a single study reported LSDV in field collected tabanid [79], their vector capacity is not clear. However, they are contained to outdoor cattle and do not enter buildings or vehicles, if a horse fly enters a truck, it rapidly wrecks its wings, loses its flying ability and dies within a few hours [89]. Thus, more experimental and field studies focusing on tabanids are necessary to establish their role in transmitting and spread LSDV (i.e., evaluate its vectorial capacity).

The role of the non-biting flies Musca domestica and Muscina stabulans in the LSD epidemic only until recently came under questioning when DNA of LSDV was isolated in the aforementioned flies collected in new LSD outbreaks in Russia (2019) [81] and China (2020) [82]. Such an observation raises questions on whether they had been the culprits of introducing LSD in these new areas, as these flies are well-known mechanical vectors of numerous viruses and bacteria and feed off ocular discharges and skin lesions [95]. Further competence and surveillance studies on non-biting flies are necessary in order to establish their eventual role in the transmission and spread of LSD.

Ticks transmit several viruses, e.g., Flaviviridae, that cause encephalitis-like diseases (e.g., tick-borne encephalitis virus, Kumlinge virus and louping ill virus), and Bunyaviridae, responsible of hemorrhagic fevers (e.g., Nairobi sheep disease virus and Crimean-Congo hemorrhagic fever virus). Thus, the role of ticks as biological vectors of LSDV has always been of interest. The results of this systematic review show that only hard ticks were associated with LSDV transmission [18,19,20,21,22,23,24,25,26,27,79,83]. However, their role in outbreaks or epidemics is not clear. In this systematic review, only four field studies sampled ticks in search of LSDV [19,66,79,83], so the virus infection rate remains unknown in ticks. Experimental studies focused on the competency of ticks to act as biological vectors, and as such, to be reservoirs of the virus, and transmit it to their progeny and to recipient cattle [18,19,20,21,22,23,24,25,26,27,83]. Experimental studies achieved mechanical intrastadial, transstadial and transovarial transmission of the virus in both A. hebraeum and Rh. appendiculatus tick species, under cold temperatures. Although the passage of the LSDV between tick stages was achieved, studies could not establish that the tick could act as a biological vectors. Studies only determined mechanical transmission. As for their role in the epidemiology of LSD, ticks remain attached to the host for a long period, and thus one could discard their responsibility in a rapidly spreading epidemic. It is more likely that, if ticks are involved in the disease epidemiology, they act as a reservoir of the virus, and possibly maintain it during cold seasons. This may explain the capacity of the virus to overwinter outside the arthropod period of activity which has been reported in Russia [76].

As for the modes of spread, these are associated with modes of transmission. Risk factors studies at a herd level (i.e., short distance spread) using logistic regression [38,40,41,42,43,44,45,46,47,48], had differences on how they defined a herd or animal as being positive to LSDV. Some studies relied on serological tests (ELISA) while others considered LSD clinical signs reported by the cattle holder or veterinary services to consider if an animal or herd positive to LSD. This may affect the number of positive animals as it could be under- or overestimated. Indeed, serological tests could give false positive results (cows may develop Abs after exposure to sheep and goat poxviruses). On the other hand, the reliability of a person observing clinical signs depends on his/her knowledge and ability to clinically diagnose LSD. Additionally, the sample size and strategy were not systematically conducted and/or reported. The chosen risk factors to be considered in the logistic regression model varied among the studies; indeed, some studies lacked important variables (risk factors) such as climate, geographical location and herd vaccination status. Despite these important differences and the geographical diversity of study locations, three herd risk factors were consistent. Cattle trade, i.e., purchases, sales, introduction of new animals in the herd, increased the risk of LSD prevalence in the herd. Females are more likely to develop LSD than males, as it is the case for foreign breeds compared to buffaloes and local breeds. As for the breed of cattle, studies in endemic countries reported that local breeds of dairy cattle, i.e., Bos indicus, may present some natural resistance to the virus compared to foreign breeds such as Holstein cattle [39,40,47]. Although these results need to be taken with caution given their differences in methodology, it is important to take them into account as many countries that are currently experiencing or reporting new outbreaks of LSD (e.g., Thailand, Indonesia) may have herds mainly composed of foreign breeds, which could lead to higher number of cases and more outbreaks over time. Other mentioned risk factors were directly related to herd management, such as the sharing of pastures and water sources. Although three studies reported these factors as having a higher risk [40,42,43,45], they did not specify how the sharing was organized, e.g., shared among different farms or shared by the same herd. Consequently, another study reported that fenced farms were at higher risk of reporting LSD compared with farms sharing pastures [42]. Age and herd size was a risk factor included in most of these studies [38,40,41,42,43,44,46,48]. However, each study categorized them with different cut-offs. Thus, results differed and the effect of age and herd size on risk for presenting LSD cannot be determined.

All studies agreed that blood-feeding insects are responsible for short-distance spread while long-distance spreads are related to animal movements. The spread through blood-feeding vectors is also conditioned by climatic conditions: indeed, higher temperatures and rainfalls are correlated with a higher vector activity, and thus the risk of outbreak increases [40,41,42,43,44,45,46,47,48,49,56]. Field studies supported that statement: most LSD outbreaks occur in the summer, after the rainy season, by the time of peak arthropod activity. Animal movements, via legal or illegal transports, are associated with long-distance spread. Additionally, the risk analyses included in this systematic review showed that animal transport, along with the vector-borne character, pose the highest risk of LSD introduction in a country [84,85,86,87,88,89,90,91]. Although these studies each have their own limitation (Appendix B), it safe to establish that animal trucks can transport not only cattle, but vectors as well. The spread is also conditioned by the geographic origin of animals and the duration of transport (with or without interruption). It is also important to consider that the control of transboundary animal movements (higher transhumance) is lacking in low income or politically unstable countries (conditions which pose difficulties to include when formulating a risk analysis of introduction model) which favor the illegal or uncontrolled movement of cattle. Other modes of spread, such as the trade of animal products or sub-products, are not a viable mode for LSDV, given the results of experimental studies [36]. Indeed, the qualitative risk analysis always deemed this route as ‘null’ [84,85,86,88].

As for the conditions favoring the (re)emergence of LSD, studies based on different modelling methods showed seasonality as an influence factor. Indeed, the risk is positively associated with higher diurnal/annual mean temperatures and annual rainfalls, i.e., geographical areas experiencing a humid and warmer weather are more at risk of emergence of LSD [49,50,51,53]. Geographical areas with higher cattle density were reported of being at higher risk of LSD occurrence [57]. Likewise, global livestock densities were one of the most important environmental predictors that contributed to the ecological niche of LSDV [50]. The type of livestock production system was also considered an environmental predictor when using this type of model [50]. Additionally, the daily transmission rate (R0) between animal was found to be slightly higher in intensive production systems [53] than in crop-livestock production systems, although the differences reported in this study were insignificant. Regardless of the differences in type of epidemiological model used, these results show that higher number of livestock and concentrated in an area pose a risk for emergence of LSD. This is most likely related to the reason of the mode of transmission of LSD, i.e., a higher concentration of livestock is correlated with a higher number of vectors.

This systematic review showed that novel vaccine-like strains have emerged and were responsible of some LSD outbreaks in Russia and China [56,73,74,75,76,77,81,82]. This has raised concerns as the reversion to virulence of a strain included in a live inactivated vaccine has been previously cited in the case of bluetongue vaccination in Europe (e.g., [96]). However, the emergence of this vaccine-like strain in Russia was most likely due to a poorly manufactured Lumpivax vaccine (KEVEVAPI) [37], which was widely used in neighboring Kazakhstan. Nevertheless, since these first reports, the epidemiological situation has become more complicated, as some countries such as Vietnam, Thailand and Mongolia reported that the newly emerged outbreaks were not caused only by field strains but also by novel recombinant vaccine-like strains. Thus, these newly emerged strains have spread to other countries and the effects on the epidemiology of LSDV are yet to be elucidated. Given that vaccination is the most efficient way to control and eradicate the disease, with successful examples in the Balkan region and Israel and emergency situations warrant their use, regulatory measures concerning vaccine manufacturing need to be implemented with strict rigorous controls and vaccination campaigns to be conducted using proper protocols.

The transmission of LSDV by contaminated needles used during vaccination campaigns has been suggested as a potential mechanism for the spread of infection within a herd [97]. However, no study retrieved in this literature review reported this mode as a risk factor, and thus it could be safely said that the risk is very low.

The spread through blood-feeding vectors is also influenced by climatic conditions: indeed, higher temperatures and rainfalls are correlated with a higher vector activity, and thus the risk of outbreak increases [40,41,42,43,44,45,46,47,48,49,56]. Another climatic condition that needs to be highlighted is winds. Long distance spread of LSDV-infected vectors carried by winds started to raise concern when Israel experienced outbreaks in 1989 and 2006. The author of this theory concluded that although it is a viable route, it depends of the vector’s capacity [61]. Given that there are some examples of possible transmission of other viruses through wind-assisted travel of vectors, e.g., it was proposed that Japanese encephalitis virus was introduced to Australia by wind-blown Culex spp. [98], and wind assisted in the spread of bluetongue virus in Europe [99], this route merits further investigation as LSD could reach countries by crossing geographical areas in which animal trade is easier to control (e.g., an island). Moreover, a study using ecological niche models to quantify the potential distribution of pathogens by correlating environmental abiotic conditions (e.g., temperature, precipitation and wind speed) with disease occurrence location, determined that wind speed was negatively associated with LSDV incidence [50]. Thus, wind is a climatic condition that may have effects on the epidemiology of LSD, but confirmation is needed.

In summary, the most efficient pathways for the emergence of LSDV in a country are the introduction of infected animals (in particular for long-distance spread) and the active transport of flying vectors to a naïve country (short-distance spread, e.g., from infected areas close to the borders). The risk of emergence is conditioned by: (i) climatic factors, i.e., warm weather promotes a higher vector activity and thus increases the risk of emergence, (ii) adverse economic situation, as border control is lacking, (iii) illegal or uncontrolled cattle movements, (iv) poor disinfection practices, (v) small cattle holdings and (vi) the use of poorly manufactured vaccines.

5. Conclusions

In conclusion, this systematic review reveals an increasing number of studies in countries where the disease is not endemic yet. Modelling LSD field data has become more specific and complex, thus broadening the epidemiological knowledge on the disease. Additionally, biotechnology has also advanced and research does not rely only on serology to confirm the diagnosis of LSD. Field and experimental studies have shifted towards the investigation of vectors others than stable flies. These conditions are positive, as the ultimate goal is to understand LSD epidemiology and stop its introduction in free-countries. The emergence in the Balkans, Europe, and Russia, where outbreaks are still reported, have required the rapid implementation of vaccination campaigns to control disease outbreaks and prevent its further spread. Indeed, vaccination is the only effective control and preventive strategy and remains the main approach to protect animal health and prevent economic losses. However, when considering the vaccine-associated outbreaks, there is a need to improve vaccine manufacturing standards, and to ensure quality control and traceability. Recent findings, i.e., new potential vectors, LSDV overwintering and new vaccine-recombinant strains, illustrate the multiple gaps in understanding the epidemiology, genetic features and transmission mechanisms of LSDV, which significantly impede the development of control strategies. A better understanding of LSDV will improve control programs in newly infected but also endemic countries. Insect control in cattle herds and transport vehicles is a crucial measure to prevent the emergence of LSD. Vaccination campaigns immediately after the emergence in a free country are easier to implement in high-income countries. In low-income areas, mitigation measures such as farmer education to detect LSD clinical signs, so they can identify the disease and notify the authorities, and insect control should be encouraged, along with vaccination during the period of vector activity. The control of LSD in endemic countries will reduce the risk of introduction and spread in neighboring nations.