key: cord-0750695-arcyyfz2
authors: Mahdi, Ahmad; Khazen, Georges; Aziz, Nivine; Mina, Jonathan; Papazian, Aram; Daou, Leonardo; Ahmar, Jana; Assaf, Nour; Abdulkhalek, Anjy; Farhat, Hussein; Mokhbat, Jacques; Farra, Anna; Husni, Rola
title: Severe Acute Respiratory Syndrome Coronavirus 2 Antibody Seroprevalence in Lebanon: A Population-Based Cross-Sectional Study
date: 2022-01-29
journal: IJID Regions
DOI: 10.1016/j.ijregi.2022.01.011
sha: 7cc52ccf509dfbeba60b0ed07b2b611587985378
doc_id: 750695
cord_uid: arcyyfz2

Objectives The World Health Organisation promoted the use of serological testing as a rapid and accurate technique for detection of the immunity against Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). In Lebanon, a better understanding of the immune response against SARS-CoV-2 is needed to develop effective measures for prevention and plan an appropriate national vaccination program. This study aims to measure the immunity status in Lebanon. Methods In this cross-sectional study, the population consisted of male and female Lebanese and non-Lebanese residents of Lebanon between the ages 15 and 75. Exclusion criteria included same household, symptomatic individuals, and extremes of age (<15 and >75). Representative testing for SARS-CoV-2 antibodies (Anti-SARS-CoV-2 electrochemiluminescence immunoassay/ECLIA) was used to assess the prevalence of SARS-CoV-2 infection in Lebanon. Results Over 6 months, a total of 13,755 participants were recruited, and 3,168 (23.03%) individuals tested positive for Anti-SARS-CoV-2, with levels of positivity varying among districts. A higher level of seropositivity was detected in the female participants. Conclusion The seroprevalence against SARS-CoV-2 varied within Lebanon and was compatible with the levels seen in the MENA region at the time of the study. The seroprevalence documented in this study reflected a level of immunity that is not protective at the national level. Funding This study was funded by the Lebanese American University School of Medicine.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused a major spread of coronavirus disease 2019 (COVID-19) globally and was declared by the World Health Organisation (WHO) on March 21, 2020, as a pandemic. Several studies started assessing its viral pathogenesis, genetic shifts, and variations, in addition to the acquired immunity status against SARS-CoV-2. Such immunity is usually acquired at an individual level via natural infection or through vaccination (Randolph & Barreiro, 2020) .

Following the identification of the virus, many rapid tests became available, including antigen detection, polymerase chain reaction (PCR) testing, as well as highly specific and well-validated serologic assays (Beavis et al., 2020; GeurtsvanKessel et al., 2020; Meyer et al., 2020) . Highly specific assays were used to screen populations with low seroprevalence (Gudbjartsson et al., 2020) . Large-scale serology prevalence studies provided important estimates of the size of the population fraction that developed antibodies against SARS-CoV-2 (Kritsotakis, 2020) . The WHO indeed promoted the use of serological testing as a rapid and accurate technique to detect immunity against SARS-CoV-2(Laboratory Testing for 2019 Novel Coronavirus (2019-NCoV) in Suspected Human Cases, n.d.).

COVID-19 antibody testing is based on detecting antibodies against the Spike protein (S), the Nucleocapsid protein (N), and other proteins that elicit high antigenicity .

Several serology methods were developed to detect Immunoglobulin M (IgM) and IgG antibodies targeted against SARS-COV-2 antigens in the serum of patients (Xiang et al., 2020) .

In the summer of 2020, some studies raised concerns on the waning of antibody immunity, especially in patients who are mildly symptomatic (Ibarrondo et al., 2020; Long et al., 2020) .

However, it was later suggested that studies conducted after the first COVID-19 wave probably assessed transient wane in antibody titers, whereas studies performed after the second wave reflected a more factually correct state of a longer-lasting antibody titers (Alter & Seder, 2020) .

Overall, antibody assays remain a more cost-effective and reliable representation of collective immunity in populations (Alter & Seder, 2020) .

Since April 2020, seroprevalence studies were conducted in numerous countries including China, Xu et al., 2020) . Such studies provided insight into the immunity status of those countries and highlighted statistically significant variables that contributed to it. One metaanalysis studied SARS-CoV-2 seroprevalence globally and concluded that it varies across different regions in the world depending on several factors including human development indices, income levels, and geographic latitudes or/and climate (Rostami et al., 2021) .

In Lebanon, a total of 547, 497 infected cases were reported by July 11, 2021. As a result, 7,873 individuals lost their lives due to COVID-19. Despite the incremental increase in cured cases, the rapid spread and transmission of SARS-CoV-2 remain a major problem. Therefore, more information about the immune response against this virus is needed for better infection control and treatment management.

Up to this date, no national serology prevalence study for the residents of Lebanon has been published. This study measures the titer of the SARS-CoV-2 antibody in a cross-sectional sample of the population in Lebanon to identify their SARS-CoV-2 immunity. Greater insight into the collective immunity against SARS-CoV-2 can help steer public health policies for better disease control and will provide insight into the seroprevalence rates against SARS-CoV-2 in countries of the Middle East and North Africa (MENA) region.

In this population-based cross-sectional study, serological testing for anti-SARS-CoV-2 antibodies was used to assess the prevalence of SARS-CoV-2 infection in Lebanon across all eight Lebanese governorates, 26 Lebanese Districts, and 138 Lebanese municipalities. A minimum sample size of 12,372 participants was determined to be required for the significance of this study; however, 13,755 participants were tested. Male and female Lebanese and non-Lebanese residents, as well as Syrian and Palestinian refugees living in Lebanon between 15 and 75 years of age. Individuals of the same household, symptomatic individuals (for consistency and the safety of participants), individuals of extreme ages (<15 and >75 years old) due to sample acquiring difficulty, and vaccinated individuals were excluded. Collaboration was made with municipality officials within the Lebanese governorates and districts represented in this study. Municipalities selected and invited eligible participants, based on official records.

Participants were contacted via telephones to document any exclusion criteria and schedule appointments. Major municipalities with high population density were selected to be representative while maintaining spatial diversity.

Case ascertainment rates varied drastically over time in Lebanon. However, accurate estimates from this seroprevalence study were achieved by recruiting a representative sample of the population of interest. Furthermore, congregate settings (e.g., nursing homes, prisons, same household, or same building, etc…) were avoided to decrease bias.

Participants were screened and recruited based on the inclusion criteria. These individuals signed consent forms and were guided to fill an online questionnaire about their adherence to social distancing and mask-wearing, COVID-19-related symptoms since the beginning of the pandemic, exposures, and travel history. Five mL of venous blood was collected into a standard sampling tube. Elycses Anti-SARS-CoV-2 assay was used on serum samples. The mentioned assay manufacturer's internal study showed a specificity of 99.81% [at 95% CI] and a sensitivity of 65.5% at days 0-6, 88.1% at days 7-13, and 100% after day 14 of PCR confirmation (at 95% Finally, results are determined automatically by software comparing the electrochemiluminescence signal obtained from the reaction product with the signal of the cutoff value previously obtained by calibration. If COI is more than 1.0, the sample is considered reactive and positive for anti-SARS-CoV-2 antibodies.

The first date of blood collection was 21 September 2020 at the Jezzine district, whereas the last date of collection was 8 April 2021 at the Nabatieh district. On average, sample collection visits ranged between two to five trips weekly. The number of municipalities visited depended on the logistical feasibility and spread status of SARS-CoV-2 in the targeted municipalities. Our study took course across seven partial lockdowns and four complete lockdowns. During periods of lockdown, our trips were limited or adjusted as much as possible to comply with public health measures and to prevent super spreader events. Study design was not affected. Our blood sample collection took place over 103 days, with an average of 265 collected samples per day.

Patients' identities were kept confidential. Only the investigators and research team had access to this data. Data was not shared with an entity, person, or organisation outside of LAUMC-RH confidential records. Analysis and publications did not include any medical numbers or social security numbers. All study personnel have been trained to maintain the confidentiality of health information. All data with patient identifiers will be destroyed 5 years after publication of the final study report. was defined as exposure to individuals who had recent travels, infectious symptoms, or a confirmed positive COVID-19 PCR. Exposure was defined as any violation of social distancing and absence of face mask. If exposure was confirmed, patients were asked about the time of exposure and the total number of individuals they were exposed to. Individuals were also asked about their own recent travel history, time of travel and country of travel origin.

The minimum sample size was computed while assuming an infinite population size and using the following formula:

Where p is the seroprevalence rate and d is the precision.

A sample size of 5,379 was initially required for our study. This was calculated while considering a type I error rate of 5%, a precision (margin of error) of 1% and assuming a seroprevalence rate of 20%. The 26 districts and the Refugee Camp (combined camps) were then considered as the 27 clusters in the study and used to compute the design effect while assuming an intraclass correlation of 0.05, as follows:

Applying the design effect to correct the estimated sample size, resulted in a total of 12,372 

Baseline characteristics, sex and age-group were described for each district. The crude SARS-CoV-2 seroprevalence was computed for the general population, sex, district, and age-group. The overall crude frequencies of positive tests were calculated and stratified by age and sex. The 95%

CIs for seroprevalence were constructed using the bootstrap method in ci_proportion from the 'confintr' package in R, with 10,000 repetitions. We also computed the seroprevalence rate adjusted by the test sensitivity and specificity using the following formula from Sempos et al. (Sempos & Tian, 2021) , adjusted prevalence = (crude prevalence + specificity -1) / (sensitivity + specificity -1)

The specificity and sensitivity were assumed to be 99.81% and 99.5% respectively for the ECLIA test. Given that the sensitivity changes relatively to the number of days since infected, we computed the weighted average based on the number of samples reporting being exposed, within 6 days, between 7 and 13 days and after 14 days.

The chis-squared test of independence was used to compare the seroprevalence rate between the different sex and age groups and districts. The Pearson correlation coefficient was computed to assess the relationship between our daily SARS-CoV-2 seroprevalence rates, and the daily percentage of COVID-19 PCR positive cases reported by the Lebanese Ministry of Public Health. A p-value <0.05 was statistically significant. We also compared the quantitative antibody titers of the SARS-CoV-2 among the different sex and age Groups. We first used the Shapiro test to check for normality and then used the Wilcoxon Rank Sum and Kruskal-Wallis tests to assess the differences between the sex and age groups respectively. The quantitative measures are reported as mean ± standard deviation. The odds ratios (OR) of having a positive seroprevalence, along with their 95% confidence intervals, were computed using the logistic regression function glm and were adjusted to the Sex, Age group, being Exposed and Symptomatic covariates. The

ORs were computed relative to 15 to 19 years old, Female, non-Exposed and non-Symptomatic individuals.

Response rate was very high in our study which limited response bias. However, there may have been potential for mild sampling bias as the municipalities had different reach to their citizens in times of quarantine. This bias is mostly negligible due to the efforts to maintain a representative sample. Also, recruited individuals were subjective in reporting their symptoms. This might have added to the limitation of this study since there was no documentation of symptoms by clinicians at the time of disease manifestation.

The sponsor of this study was the Lebanese American University (LAU). LAU administration had no intervening role in the study design, data analysis or interpretation, or writing of the report. In addition, the corresponding author had full access to the entire data in this study and had the final responsibility for the decision to submit this study for publication.

In this population-based cross-sectional study, we initially recruited 13,843 individuals, but 88 were excluded for not meeting the eligibility criteria. Blood samples were collected from the 13,755 remaining individuals to conduct the Elycses Anti-SARS-CoV-2 assay (Figure 1 Minieh-Danieh, and the lowest seropositivity was detected with 1.76% (0.50-3.01) in Jubail and 1.04% (0.00-2.59) in Marjaayoun.

The seroprevalence rate was found to be significantly associated with the three covariates: sex, age group and district (chi-squared P<0.0001 for the three variables). In fact, 24.68% (1618/6555) in were tested positive for SARS-CoV-2 antibodies in females while only 21.43% (1550/7200) in males ( Table 1) .

A statistically significant difference was recorded between males and females with respect to positive seroprevalence (chi-squared p-value = 1.24e-05), 24.68% (1618/6555) in females and 21.43% (1550/7200) in males tested positive for SARS-CoV-2 antibodies ( Table 1) .

Out of all positively tested participants, 51% (1618/3168) were females and 49% (1550/3168) were males ( Table 3) . Additionally, 87% reported applying quarantine and social distancing, 57% reported having prior symptoms since the beginning of the pandemic, 54% reported an exposure history to patients with COVID-19, 15% reported an exposure history to symptomatic individuals and 11% reported contact with a recent traveller. Moreover, 47% of participants that tested negative for Anti-SARS-CoV-2 antibodies were females and 53% were males, wherein, 86% reported applying quarantine and social distancing, 12% reported having prior symptoms since the beginning of the pandemic, 27% reported an exposure history to patients with COVID-19, 31% reported an exposure history to symptomatic individuals and 17% reported contact with a recent traveller. A statistically significant positive correlation (correlation coefficient = 0.402; P<0.0001) was also found between our daily SARS-CoV-2 seroprevalence rates and daily percentage of COVID-19 PCR positive cases as reported by the Lebanese Ministry of Public Health ( Figure   2 ).

The positive seroprevalence adjusted Odds Ratios (OR) are presented in Table 4 . There is a statistically significant increase in the positive seroprevalence among individuals that were exposed to an infected individual. Among those exhibiting symptoms, it was 1.76 and 7.78 times higher in the Exposed and Symptomatic individuals, respectively, as opposed to non-exposed (Table 4 ).

In this population-based cross-sectional study, the seroprevalence of Anti-SARS-CoV-2 antibodies was estimated in all Lebanese governorates and districts including refugee camps.

This was performed through 13,755 tests conducted at 138 randomly selected municipalities. No seroprevalence study of Anti-SARS-CoV-2 antibodies has been published to estimate the immunity status in Lebanon. A predictive model estimating a 27% immunity status by May 1 including social distancing measures and vaccination rate (Health Data, 2021) . In our study, 23 .037% of the tested population was determined to be immune. This rate is higher than other countries' reported rate for the same time period in-spite of early lockdown in our country (Health Data, 2021) . To note, mask-wearing and social distancing was not properly practiced because of social norms.

Further analysis of the results reveals that females were more exposed, infected, and immune to SARS-CoV-2 than males (24.68% vs 21.53%). However, different age groups have different immunity statuses. Younger age groups (15-19 years and 35-39 years) had the highest positive rates, whereas older age groups (55-64 years) had the lowest. This might be caused by the belief that the disease is worse selectively for older individuals, making the young less committed to mask-wearing and social distancing measures, whereas older individuals took stricter precaution measures. In addition, the majority of those who tested positive denied exposure to symptomatic individuals. Hence, probably most exposure happened to asymptomatic patients prior to their disease manifestation. Moreover, only 11% of individuals that tested positive had a recent exposure to a recent traveller. Thus, most of SARS-CoV-2 spread was local. In addition, most of those who tested positive reported practicing quarantine or social distancing measures, which were most likely practiced inefficiently and failed to yield the desired disease protection.

As for the geographic distribution, results showed that the highest rate of SARS-CoV-2 antibodies was found in the Hasbaya (60%) and Minieh-Danieh Districts (40%), which may be attributed to potential socioeconomic confounding factors at the mentioned locations where members interact more closely and disregard preventive measures in the face of social norms that encourage physical contact. In addition, the positive correlation between the daily SARS-CoV-2 cases reported by the Lebanese Ministry of Public Health and SARS-CoV-2 immune individuals in this study confirm the correlation between the rate of infectivity and rate of immunity development against the virus (Figure 2. ). It also serves to externally validate our study.

Seroprevalence studies, serve to estimate COVID-19 infections since most asymptomatic and presymptomatic patients do not usually get tested (Kronbichler et al., 2020) . One study in the US revealed a total number of cases 6-24 times the previously confirmed number after accounting for the untested asymptomatic and presymptomatic patients (Havers et al., 2020) . Seroprevalence studies are also important in predicting second and third waves in all countries, one metaanalysis suggests (Rostami et al., 2021) . When compared to global results, the seroprevalence rate in Lebanon (23.03% at 95% CI) is one of the highest in the world. It is significantly above the global average (3.38% at 95% CI) and very similar to previously reported results in Central and Southern Asia, (22.16% at 95% CI in Iran for instance) (Rostami et al., 2021) .

To our knowledge, no other SARS-CoV-2 seroprevalence study in Lebanon has given insight on the immunity status against SARS-CoV-2 (Bizri et al., 2021) . This study was conducted at a time were the Institute for Health Metrics and Evaluation (IHME) noted a three-fold increase in the death rate in the MENA region between September and December 2020(Institute for Health Metrics and Evaluation, 2020). Therefore, our study is essential for the understanding the initial spatial spread of SARS-CoV-2 and the immunity status in Lebanon for different subpopulations.

This will help guide necessary interventions where needed, such as allocating vigorous testing and vaccination for the virus in districts where these are most needed. This, along with adjusting quarantine measures, would increase the efficiency of lockdown measures without heavily stressing the economy.

Limitations of this study include the possibility that some participants may have belonged to the same household. This task was allocated to contacted municipalities, some of which could have allowed few violations. However, our research team double checked on this issue and made sure that no members of the same family were included. Also, our selection criteria excluded individuals <15 years of age and >74 years of age as well as symptomatic individuals regardless of their age. This may neglect a portion of the society that contributes to the overall seroprevalence status, but this was compensated by adding more numbers in the other age groups similar to other epidemiological studies done at national levels. Also, recruitment frequency was partially limited during lockdown periods, all efforts were done to compensate for the time of the lockdown. The study period serves also as a limitation to this study, however the study was completed within 6 months, decreasing bias and serving as an acceptable estimation of the crosssectional seroprevalence. Limitations of this study also include the hook-effect of the test used which cannot be adjusted for. Also, limited medical history appropriate for the study was collected. The results can however be generalized to the studied population as the sample size was representative of the population of interest, response rate was very high in recruited individuals, and the technique used has high sensitivity and specificity.

In conclusion, this study examines the nationwide and regional estimates of SARS-CoV-2 spread in Lebanon. It portrays different seroprevalence rates in different districts, age groups and sexes.

It highlights how the transmission mode is most likely through asymptomatic locals. In addition, it shows that the seroprevalence rate is not sufficient to achieve a high immunity status, thus, it reemphasizes the importance of adhering to social distancing and of wearing face masks to avoid transmission and infection.

Ahmad Mahdi was responsible for data curation, investigation, methodology, project administration, resources supervision, writing-original draft, and writing-review and editing.

George Khazen responsible for data curation, formal analysis, methodology, software, validation, visualization, writing-original draft, and writing-review and editing. Nivine Azziz was responsible for data curation, formal analysis, methodology, investigation, software, validation, visualization, writing-review, and editing. Jonathan Mina was responsible for data curation, investigation, methodology, writing review and editing. Aram Papazian was responsible for data curation, formal analysis, methodology, investigation, software, validation, and visualization. Leonardo Daou was responsible for data curation, formal analysis, methodology, investigation, software, validation, and visualization. Jana Ahmar was responsible for data curation and investigation. Nour Assaf was responsible for data curation and investigation. Anjy Abdulkhalek was responsible for data curation and investigation Hussein Farhat was responsible for methodology and resources. Jacques Mokhbat was responsible for writingreview & editing. Anna Farra was responsible for writingreview & editing. Rola Husni was responsible for conceptualization, funding acquisition, methodology, project administration, resources, supervision, and writing-review and editing. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.

All the individual participant data collected during the trial, after deidentification, along with the study protocol will be available starting 3 months and ending 5 years after publication for anyone who wishes to access the data for any purpose. Data will be available on a link to be later revealed.

We declare no competing interests.

This study was funded by the Lebanese American University School of Medicine (LAU SOM).

Thank you for the Lebanese American University Medical Center-Rizk Hospital laboratory technicians, and LAU drivers for helping us conduct our research investigation. Thank you to all municipality employees for collaborating with us to conduct our participant recruitment events.

This study was reviewed and approved by the Lebanese American University (LAU) Institutional Review Board (IRB) with the reference number "LAUMCRH.RH5.24/Apr/2020" 

The Power of Antibody-Based Surveillance

Seroprevalence of anti-SARS-CoV-2 antibodies in Iquitos

Evaluation of the EUROIMMUN Anti-SARS-CoV-2 ELISA Assay for detection of IgA and IgG antibodies

COVID-19 Pandemic: An Insult Over Injury for Lebanon

Lower prevalence of antibodies neutralizing SARS-CoV-2 in group O French blood donors

Elecsys® Anti-SARS-CoV-2. Diagnostics

An evaluation of COVID-19 serological assays informs future diagnostics and exposure assessment

Humoral Immune Response to SARS-CoV-2 in Iceland

Remarkable variability in SARS-CoV-2 antibodies across Brazilian regions: Nationwide serological household survey in 27 states

Seroprevalence of Antibodies to SARS-CoV-2 in 10 Sites in the United States

Rapid Decay of Anti-SARS-CoV-2 Antibodies in Persons with Mild Covid-19

Institute for Health Metrics and Evaluation

On the importance of population-based serological surveys of SARS-CoV-2 without overlooking their inherent uncertainties

Asymptomatic patients as a source of COVID-19 infections: A systematic review and metaanalysis

Laboratory testing for 2019 novel coronavirus (2019-nCoV) in suspected human cases

Evaluation of nucleocapsid and spike protein-based enzymelinked immunosorbent assays for detecting antibodies against SARS-CoV-2

Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections

Validation of a commercially available SARS-CoV-2 serological immunoassay

Prevalence of SARS-CoV-2 in Spain (ENE-COVID): A nationwide, population-based seroepidemiological study. The Lancet

Herd Immunity: Understanding COVID-19

SARS-CoV-2 seroprevalence worldwide: A systematic review and meta-analysis

Adjusting Coronavirus Prevalence Estimates for Laboratory Test Kit Error

Seroprevalence of COVID-19 virus infection in Guilan province, Iran

Evaluation of enzyme-linked immunoassay and colloidal gold-immunochromatographic assay kit for detection of novel coronavirus (SARS-Cov-2) causing an outbreak of pneumonia (COVID-19)

Seroprevalence of immunoglobulin M and G antibodies against SARS-CoV-2 in China