LAMP-Based Point-of-Care Nucleic Acid-Based Detection Method Can Be Useful for Quick Decision-Making for Diagnosis of Acute COVID-19 Emergency Cases in Hospital Settings

Abstract

Real-time RT-PCR is used as a gold standard method for the diagnosis of COVID-19. Since real-time RT PCR is nucleic acid-based, it is a highly sensitive and specific test. However, this test takes 4–8 h to generate results and, in emergency settings, this delay may prove fatal for certain patients. The frequent surge in COVID cases increases patient load in emergency settings. Thus, a nucleic acid-based rapid POC test is required that can generate results quickly as well as being comparable to real-time RT-PCR. In this study, comparison of real-time RT-PCR was carried out using the rapid nucleic acid-based LAMP method. Nasopharyngeal swabs were taken in duplicate from patients visiting the kiosk and were analyzed for the presence of the SARS-CoV-2 virus by both real-time RT-PCR and LAMP techniques ID NOW(bbott). Out of 14 positive and 31 negative samples tested by real-time RT-PCR, 13 samples were identified as positive and 31 were observed as negative with the LAMP-based test. Hence, the sensitivity and specificity of this method were found to be 92.9% and 93.5%, respectively. Therefore, LAMP-based point-of-care testing has the potential to be used in hospital emergency settings for quick diagnosis of critically ill patients, and the information generated here will further draw the attention of policymakers toward such nucleic acid-based rapid tests.

1. Introduction

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) known as Coronavirus disease 2019 (COVID-19) emerged in Wuhan, Hubei province, China in December 2019 [1]. The latest reports from WHO have shown 767,750,853 confirmed cases of COVID-19, including 6,941,095 deaths, worldwide [2]. To date, approximately 31,312 cases have been reported every day globally [2]. Due to the appearance of COVID-19 variants, recurrent epidemics or pandemics cannot be ruled out. Even rapid surges in COVID-19 cases from time to time also put health authorities on their toes. Consequently, the load of patients in emergency settings can also be a matter of concern during outbreak situations with different waves. It is observed that emergency services for critically ill patients are sometimes delayed, as PCR-based testing takes 4–8 h to obtain results. These delays can sometimes prove fatal to critically ill patients, and they also lead to poor patient flow through clinical areas, with suspected patients grouped into assessment areas until their results are available, which can lead to further spread of infection [3]. Such situations demand rapid point-of-care tests that can quickly detect the virus as efficiently as real-time RT-PCR. Rapid tests will not only help in the quick diagnosis of critically ill patients but could also help in controlling the spread of the disease.

Real-time PCR (RT- PCR), which is a nucleic acid-based detection method, is the only gold standard method currently available and is used by most hospitals [4]. Since nucleic acid-based methods are more sensitive and specific, they are considered to be the most reliable tests. To date, none of the other strategies have been able to replace nucleic acid-based testing specifically by real-time RT-PCR. As far as point-of-care methods are concerned, there is no reliable point-of-care test available. Even rapid point-of-care serological methods are being used only for surveillance, due to poor specificity [5,6].

Due to the low success of serological methods, scientists all over the world are trying to develop rapid tests that are nucleic acid-based and can generate results comparable to real-time RT-PCR. For the time being, such tests can be applied in emergency settings not only as rapid tests but also as bedside point-of-care tests [7,8]. Information is lacking about rapid tests that are nucleic acid-based and can be implemented in emergency hospital settings for timely admission of critically ill patients. One of the latest reports published in the Lancet states that point-of-care testing is associated with large reductions in time to results and could lead to improvements in infection-control measures and patient flow compared with centralized laboratory real-time PCR testing [3]. Therefore, there is an urgent need for well-established point-of-care rapid testing based on nucleic acid-based detection, which can provide results within minutes and should have sensitivity and specificity comparable to real-time RT-PCR. According to the literature, there are few potential tests available. These include LAMP-based, COVID Nudge, real-time PCR-based rapid tests, etc. [1,9]. Out of these, LAMP, or loop-mediated isothermal amplification, is a technology that has provided encouraging results and could have potential for the development of rapid nucleic acid-based point-of-care testing. However, limited evidence is available about the performance of this point-of-care test specifically in Indian health settings where the burden on emergency settings is high. For this reason, this method is not being used in emergency settings in many tertiary care hospitals in India. Therefore, in this study, we assessed for the first time the performance of a LAMP-based point-of-care testing machine (ID NOW, Abbott) in comparison to real-time PCR in one of the tertiary care hospitals of north India. This study has further added to evidence related to the success of this LAMP-based test for use in emergency hospital settings and also for point-of-care testing during community screening. The LAMP-based approach is not only suitable for the rapid detection of SARS-CoV-2 virus but also for the rapid detection of other viruses. The data generated here will add to the already available information and hopefully will draw the attention of policymakers to implement such point-of-care testing not only for COVID but also for other viruses which may be future pandemic threats.

2. Materials and Methods

2.1. Sample Size and Sampling Method

The study design was cross sectional and the sample size was calculated using the formula:

$$Sample size=\frac{4\times Sensitivity\times \left(1-Sensitivity\right)}{Precisio{n}^2\times Prevalence}$$

The sample size of approximately 45 was estimated at 89% confidence level, 1% absolute precision, and 88% expected prevalence according to the previous research available at the time of study design [10].

2.2. Patient Enrolment

Sample collection was done at the kiosk of the Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh, India. Participants visiting the kiosk for COVID testing were enrolled in the present study. Participants were informed about the study and its importance. Consent was taken from participants before taking swabs. Nasopharyngeal swabs were collected with proper precautions by the trained staff. The sample collection was completed wearing proper PPE kit. Participants included patients as well as staff members suspected of COVID-19 (

Table 1. Demographic detail of participants and validation of results.

Sr.no	Patient ID	Age	Sex	Symptomatic/Asymptomatic	LAMP (ID NOW)	Real-Time RT-PCR	Ct Value E/N/RDRP
1	Participant 1	23	F	Asymptomatic	Positive	Negative	N.A.
2	Participant 2	45	M	Asymptomatic	Negative	Negative	N.A.
3	Participant 3	59	F	Asymptomatic	Negative	Negative	N.A.
4	Participant 4	43	F	Symptomatic	Positive	Positive	23/20/21
5	Participant 5	32	F	Asymptomatic	Negative	Negative	N.A.
6	Participant 6	32	M	Symptomatic	Negative	Negative	N.A.
7	Participant 7	23	M	Symptomatic	Negative	Negative	N.A.
8	Participant 8	32	M	Asymptomatic	Negative	Negative	N.A.
9	Participant 9	29	F	Asymptomatic	Positive	Positive	25/22/24
10	Participant 10	31	F	Symptomatic	Negative	Negative	N.A.
11	Participant 11	31	F	Symptomatic	Positive	Positive	18/17/19
12	Participant 12	60	M	Asymptomatic	Negative	Negative	N.A.
13	Participant 13	38	M	Asymptomatic	Negative	Negative	N.A.
14	Participant 14	4	M	Asymptomatic	Negative	Negative	N.A.
15	Participant 15	11	F	Asymptomatic	Negative	Negative	N.A.
16	Participant 16	56	M	Symptomatic	Negative	Negative	N.A.
17	Participant 17	32	F	Symptomatic	Positive	Positive	25/22/24
18	Participant 18	34	M	Symptomatic	Negative	Negative	N.A.
19	Participant 19	53	M	Symptomatic	Positive	Positive	22/18/19
20	Participant 20	25	F	Symptomatic	Negative	Negative	N.A.
21	Participant 21	28	F	Symptomatic	Negative	Negative	N.A.
22	Participant 22	26	F	Symptomatic	Negative	Negative	N.A.
23	Participant 23	49	M	Symptomatic	Positive	Positive	17/21/19
24	Participant 24	25	F	Symptomatic	Negative	Negative	N.A.
25	Participant 25	29	M	Asymptomatic	Negative	Negative	N.A.
26	Participant 26	52	M	Symptomatic	Positive	Positive	16/17/17
27	Participant 27	36	M	Symptomatic	Negative	Negative	N.A.
28	Participant 28	45	M	Symptomatic	Negative	Negative	N.A.
29	Participant 29	60	M	Asymptomatic	Positive	Negative	N.A.
30	Participant 30	45	M	Symptomatic	Negative	Positive	20/17/18
31	Participant 31	24	M	Symptomatic	Negative	Negative	N.A.
32	Participant 32	36	M	Symptomatic	Positive	Positive	29/26/27
33	Participant 33	13	M	Asymptomatic	Positive	Positive	29/26/29
34	Participant 34	60	F	Symptomatic	Positive	Positive	21/16/18
35	Participant 35	30	M	Symptomatic	Negative	Negative	N.A.
36	Participant 36	30	M	Asymptomatic	Negative	Negative	N.A.
37	Participant 37	40	F	Symptomatic	Positive	Positive	24/23/23
38	Participant 38 *	36	M	Symptomatic	Negative Negative	Negative Negative	N.A. N.A.
39	Participant 39	32	F	Asymptomatic	Negative	Negative	N.A.
40	Participant 40	56	M	Symptomatic	Positive	Positive	Not available
41	Participant 41	52	F	Symptomatic	Positive	Positive	21/20/20
42	Participant 42	39	F	Symptomatic	Negative	Negative	N.A.
43	Participant 43	14	F	Symptomatic	Negative	Negative	N.A.
44	Participant 44	94	M	Asymptomatic	Negative	Negative	N.A.

Total number of samples tested = 45; results matched for = 42 samples; result = 13 were positive by LAMP technique (ID NOW) testing out of which 14 were positive by real-time RT-PCR testing. False positive = 2; false negative = 1. * Participant number 38 was sampled twice, so the total number of samples analyzed was 45 from 44 participants.

). Samples were collected till the required sample size was obtained. Demographic details were taken from all the participants (Figure 1 and Figure 2). The symptoms of patients were also recorded (Table 1).

2.3. Collection of Swabs

In this study, 45 samples from 44 enrolled participants (1 participant was sampled twice) were collected from 2 July 2022 to 26 July 2022.

Two nasopharyngeal swabs were collected from each participant by trained staff taking proper precautions. One swab was used for the LAMP-based (ID NOW) COVID-19 rapid test and the other one for real-time RT-PCR. Swabs to be used for real-time RT-PCR were transferred in a viral transport medium to the laboratory of the Virology Department of PGIMER, Chandigarh. Ct (cycle threshold) value was also noted for real-time RT PCR to obtain information about the viral load and its correlation with the performance of LAMP-based testing.

2.4. Rapid Point-of-Care Testing

For the isothermal nucleic acid amplification test, ID NOW COVID-19 assay (Abbott Diagnostics, Chicago, IL, USA) was used and was located at the kiosk. Nasal swabs were immediately and directly used for testing the instrument. The dimensions and weight of the instrument are user friendly, which makes it ideal for use as a bedside test. The run time was as little as 5 min for positive cases and 13–15 min for negative results. The results obtained from LAMP-based point-of-care testing (ID NOW) were compared with real-time RT-PCR. Sensitivity and specificity were calculated using statistical formulas.

2.5. Statistical Analysis

The data was entered into Microsoft Excel and the analysis was carried out in SPSS. The sensitivity, specificity, positive percent agreement (PPA), and negative percent agreement (NPA) were calculated using the formulas mentioned below [11]. These were calculated by comparing the test results of the LAMP-based point-of-care test with the gold standard real-time RT-PCR method.

Sensitivity was calculated using the formula:

$$Sensitivity=\frac{True Positive}{True Positive+False Negative}$$

Specificity was calculated using the formula:

$$Specificity=\frac{True Negative}{True Negative+False Positive}$$

Positive percent agreement (PPA) was calculated using this formula:

$$PPA=\frac{True Positive}{True Positive+False Positive}$$

Negative percent agreement (NPA) was calculated using this formula:

$$NPA=\frac{True Negative}{True Negative+False Negative}$$

where True Positive—participants tested positive by both tests, True Negative—participants tested negative by both tests, False Positive—participants tested positive by LAMP-based test but negative by the gold standard method, and False Negative—participants tested negative by LAMP-based test but positive by the gold standard method.

3. Results

Out of 45 samples from 44 participants, 13 samples were positive (true positive) with both LAMP and real-time RT-PCR whereas only one sample was false negative with the LAMP-based technique. Out of 14 samples that came out positive with real-time RT-PCR, 13 were positive with the LAMP test, indicating its sensitivity of 92.9%. No correlation of Ct value with false negativity could be observed. Out of 31 samples tested negative by RT-PCR, 29 were identified as negative (true negative) by the LAMP rapid test whereas only 2 samples were false positive. So, overall, 15 samples came out positive with LAMP-based ID NOW. The specificity of the LAMP test was observed as 93.5%, and 86.67% PPA (positive percent agreement) and 96.67% NPA (negative percent agreement) were also observed.

Among the 44 participants enrolled, 16 were asymptomatic and 28 had COVID-19-like symptoms. Major symptoms observed were fever, cough, sore throat, cold, and body aches (Figure 3). Out of 13 true positives (positive by both real-time RT-PCR and LAMP technique), 10 were symptomatic and 3 were asymptomatic (Table 1 and Figure 4). Among the false positives by the LAMP test, one participant was symptomatic whereas the second one was asymptomatic. Similarly, only the false negative sample observed by LAMP was asymptomatic. The overall Ct values of positive participants according to real-time RT-PCR ranged from 16 to 29, indicating the reach of diagnostic sensitivity of the LAMP-based approach.

4. Discussion

Molecular diagnostics has undergone a period of rapid development and growth in the past decade. The latest pandemic of COVID-19 has further attracted the attention of policymakers toward molecular diagnostic tests such as real-time RT-PCR. This pandemic has led policymakers to think that not only clinical but also public health labs need to be equipped with molecular tests. Data generated during emergency situations help to implement future policies. Similar to molecular diagnostic tests, it is now time to generate evidence about the success of nucleic acid-based point-of-care testing. Although rapid nucleic acid-based molecular point-of-care tests are available and under development, the data related to their success are still insufficient to win the confidence of policymakers as well as end users to implement such tests on a large scale. The burden of COVID-19 cases during the peak of the pandemic increased the demand for a rapid, cost-effective, reproducible, and technically convenient diagnostic test specifically for emergency settings to save the lives of critically ill patients and also to control the rapid spread of the disease.

The scientific community has achieved extraordinary progress in response to the limitations of real-time RT-PCR, resulting in the development of rapid diagnostic tools for improving SARS-CoV-2 diagnosis and surveillance. Since nucleic acid-based tests are considered to be more sensitive and specific, real-time RT-PCR have remained the test of choice for diagnosis of COVID-19. Along with nucleic acid-based rapid testing, efforts were also made to develop serology-based tests but none of them could be proved as reliable as real-time RT-PCR. For example, rapid antigen tests (RATs) offer results in approximately 15–30 min, are easy to perform, do not require highly trained staff or specialized laboratory equipment, and are user friendly and cost effective. However, rapid antigen tests are not highly successful for diagnosis and are mostly used for surveillance purposes only, due to their low specificity and sensitivity. The FDA recently issued an alert about the potential for false positive results that can occur with antigen tests. Low viral protein produced false negative results by rapid antigen testing, thus making these tests less sensitive and less accurate than molecular tests. In addition, because antigen tests are qualitative, they can be inaccurately interpreted due to reader error. That is why if an antigen test returns negative or positive a confirmatory real-time RT-PCR test is still recommended [12,13]. This is the reason that rapid antigen tests were not accepted for travel nor for hospital admissions both for emergency as well as general wards. For general ward admission, waiting for real-time RT-PCR does not cost much, but in emergency settings, a delay of 3–4 h of real-time RT PCR report may prove fatal for critically ill patients.

In the recent past, for diagnosis of COVID-19 infections, IgM/IgG testing has also been conducted. However, it is observed that some individuals do not develop detectable IgG or IgM antibodies after infection, reflecting the drawback of antibody testing. Currently, over 100 antibody tests are available or in the development stage. They do not all have the same level of sensitivity and specificity. The lack of standardization for validating antibody tests in the United States makes their quality variable and test interpretation complex. Relying on antibody tests to make decisions about individual immunity or back-to-work orders is problematic [12].

Due to the drawbacks of antigen or antibody detection methods, the research towards the development of nucleic acid-based rapid testing has attracted considerable attention from scientists all over the world. Real-time RT-PCR, which is a nucleic acid-based test, proved to be the gold standard test and it has remained so till now. However, real-time RT-PCR testing demands trained personnel along with transportation medium, expensive equipment, and a sophisticated laboratory setting, and it takes 4–6 h to generate results.

Therefore, due to the limited success of serological tests and the limitations of real-time RT PCR, rapid nucleic acid-based testing with performance as good as real-time RT-PCR is urgently required. Rapid nucleic acid-based testing can not only play a central role in rapid diagnosis in hospital emergency centers but can also help in the effective prevention and control of the pandemic by allowing on-site testing to be performed. Few potential tests are available, such as LAMP-based, COVID Nudge, real-time PCR based rapid tests [1,8,9] and these nucleic acid-based point-of-care tests are either still in the validation stage, or scarce data are available about their success. Among these tests, LAMP-based testing is showing convincing performance [14,15,16,17] but more information about their utility and success is still required from countries such as India. Most of the studies available are from the USA, whereas Indian health settings are very different in comparison with developed countries, so data regarding the performance of this point-of-care test from Indian settings are required to expand the usage of such tests to save the lives of critically ill patients in emergency settings. To our best knowledge, this is the first study to provide a comparison of the performance of the LAMP-based technique in Indian hospital settings where the load of admissions in emergency wards is quite high in comparison to developed countries [18].

In contrast to real-time RT-PCR, the LAMP-based point-of-care method has the advantage of amplifying genes at a constant temperature of approximately 65 °C and is relatively easy to operate [19]. In addition, this method can produce results in less than 30 min. It has also been demonstrated that it can detect low copy numbers of SARS-CoV-2 RNA [20]. With LAMP-based technology, on-site or bedside testing is also possible as the machine is easy to carry. Nasopharyngeal swabs can be inserted directly into the machine and extraction of RNA need not be carried out separately. This also saves time and resources involved in the transportation of samples to the laboratory. A sophisticated laboratory setting is not required as RNA extraction takes place within the instrument itself. Positive results are reflected within 5–10 min and for negative results the maximum run time is 30 min. In this study, we also observed similar observations for the LAMP-based approach (ID Now, Abbott), where samples collected at the kiosk were directly inserted into the machine without the need of transporting the swabs to a sophisticated lab. The positive results were observed within 5–10 min.

Since the Indian health system works in resource-limited settings and policymakers demand evidence before the implementation of tests on a large scale, the above observations regarding the success of the LAMP-based approach fits well with the expectations of a point-of-care test. Furthermore, to validate the performance of this approach, the performance of point-of-care testing was directly compared with the gold standard method in a north Indian tertiary care hospital. In addition, due to the appearance of region-specific COVID-19 virus variants, it becomes important to check the success of this technology at ground level. Therefore, for on-site experience, the LAMP machine was kept at the sample collection kiosk situated at PGIMER Hospital, Chandigarh, India.

The LAMP-based approach (ID Now) was found to have sensitivity and specificity of 92.8% and 93.53%, respectively. The sensitivity and specificity found in this study fall within previously reported ranges [16,17]. PPA and NPA were 86.67% and 96.67%, comparable to earlier studies [17]. As expected, the SARS-CoV-2 positivity rate was higher in symptomatic patients compared with asymptomatic participants. The Ct values of positive participants ranged from 16 to 29. The detection of the virus by the LAMP-based approach from samples which showed Ct values as high as 29 in RT-PCR indicates its high sensitivity. However, false negativity could not be correlated with the Ct value. Moreover, one previous report mentioned that the diagnostic accuracy of LAMP testing decreases beyond Ct of 32. [21]. Earlier studies also showed the presence of false negative results, similar to our study. However, no study could be found where false positive results were obtained with this approach, as observed in the present study [16,17]. So, the overall results from this study suggest that ID NOW can be a reliable tool for COVID-19 screening for both symptomatic and asymptomatic patients, at least for emergency admissions in hospital settings.

Apart from ID Now, other LAMP-based point-of-care instruments are also in the research stage [22,23,24,25]. Although LAMP-based Abbott ID NOW is available on the market, limited information about its performance, specifically in the Indian context, is available. Therefore, the current research can be helpful for clinicians as well as policymakers to support the implementation of LAMP-based technology in at least emergency settings in hospitals.

Furthermore, the present study can be replicated in future with more samples (or participants), not only for COVID-19, but to assess the performance of LAMP-based approaches for rapid diagnosis of other viral infections [26,27]. Many viruses including Ebola, influenza, Zika, and other vector-borne viruses are emerging and causing health threats and may be causes of pandemics in the future. Therefore, we can conclude that a LAMP-based approach can be useful not only for the detection of the SARS-CoV-2 virus but also for the diagnosis of other viral infections in emergency settings. This approach should not remain limited to viral infections but should include rapid diagnosis of life-threatening bacterial infections such as Streptococcus pneumoniae, etc. This approach may potentially include a rapid diagnostic test. However, their use in the community setting for preventing the rapid spread of infections is still far from reality due to the high cost per test. The limitations of this strategy are the high cost per test and a lack of high-throughput output and multiplexing options in currently available models. For that reason, this strategy can be used only in emergency settings. These limitations need to be addressed before it can be used for routine testing and surveillance.

5. Conclusions

In conclusion, we can say that rapid and reliable nucleic acid-based COVID-19 testing can allow more effective diagnosis of patients, especially in emergency settings where a slight delay in the admission of patients may prove fatal. This study reveals that LAMP-based ID Now provides excellent sensitive and specific testing comparable to real-time RT-PCR. Swabs can directly be used within the instrument, which saves the time of transportation as well as resources and manpower. Furthermore, the RNA isolation step takes place in the machine itself so a sophisticated laboratory setup is not required. The dimensions and weight of the instrument make it easy to carry to the bedside, so it can be considered an excellent screening tool to support clinical decision-making in emergency hospital settings. This technology not only has the potential to be used in hospital settings but it can be used in the community for the detection of viruses or bacteria during outbreak situations or for surveillance. However, the cost per test needs to be brought down before it can be used for routine testing and surveillance. Further studies are required with more samples and other viruses as well as bacterial samples.