Seroprevalence and characteristics of Coronavirus Disease (COVID-19) in workers with non-specific disease symptoms

After emerging from Wuhan, the novel Coronavirus has rapidly transmitted throughout the world. It is now regarded as a pandemic [1], accompanied by a varying range of mild symptoms including cough, fever, cold and body ache and even severe, like Acute Severe Respiratory Distress (ASRD) become the leading cause of death among these patients [2‐4]. According to the World Health Organization (WHO) reports, we now have more than 296 million confirmed COVID-19 cases and more than 5 million deaths worldwide [5]. Locally in Pakistan, the first two cases were reported in February 2020, having a travel history from Iran. Although all preventive measures were taken to contain the disease, unfortunately, around 1,301,141 cases have been identified since then, accounting for 28,961 deaths [6]. But the concern regarding disease control, morbidity, and mortality in Pakistan remains; it is claimed that 415,352 cases have successfully recovered [6], possibly due to low testing rates in Pakistan compared to the rest of the world [7]. One of the significant reasons for limited testing facilities is the restricted healthcare budget and resources [7].

As COVID-19 has become a global threat, early detection and prevention of viral spread have become crucial. Reverse transcription-polymerase chain reaction (RT-PCR) is recognized as the gold standard diagnostic technique as it helps in the early recognition of COVID-19 [8]. Although, due to compromised test sensitivity in association with inadequate sample collection, the time between sample collection, the onset of symptoms, and the fluctuation in viral load [9] linger the duration. However, an easy, sensitive, and inexpensive alternate, i.e., Antibody Test, is now being used to identify SARS-COV-2. Antibody screening through serological assays helps determine the actual frequency and seroprevalence of the virus [10]. The antibodies IgG and IgM can be detected within 1–3 weeks after exposure to Coronavirus. A rapid increase in the antibody level and the seroconversion rate is observed during the first two weeks [11]. Presently the seroconversion rate is being widely studied in order to understand its dynamics, specifically from acute to convalescent phases [12]. It has been found that the IgM and IgA growth after exposure to SARS-CoV-2 is relatively slow compared to other acute viral infections contributing to its heterogeneous pathogenicity [13].

Old age, pre-existing comorbid conditions, low CD3 + CD8 + T-cells levels, high troponin I and d-dimer levels have been recognized as the major risk factors associated with seropositivity and mortality among the infected patients [14‐17]. More recently, studies conducted in Shenzhen, including 1286 close contacts (98 COVID positive cases) and Guangzhou including 2098 close contacts (134 COVID positive cases), discovered that significant risk factors for COVID-19 infection were among older age and traveling outside or inside the country, etc [18, 19]. Additionally, a Taiwanese study identified exposure to a confirmed or probable case of COVID-19 as a risk factor [20].

In the current study, we employed a large dataset, including 17,764 participants from Karachi, Lahore, Multan, Peshawar, and Quetta, to estimate the seroprevalence of SARS-CoV-2 antibodies, characteristics, and potential predictors of seropositivity. This serological investigation intends to elaborate Pakistan’s COVID infection statistics under the worldwide pandemic.

This is the first large-scale serological investigation of 2019-nCoV in Pakistan to the best of our knowledge. As previously published, local studies targeted a specific population or had a small sample size. A serosurvey from Lahore included 154 policemen [21]; similarly, the National Institute of Blood Diseases and Bone Marrow Transplantation from Karachi published the data on SARS-CoV-2 seroprevalence among 380 healthy blood donors [22], and another included 1,675 healthcare workers (HCWs) [23]. Moreover, the preprint data also represent seroprevalence in a limited sample size [24, 25]. The current findings revealed that the overall seroprevalence of anti-SARS-CoV-2 antibodies (IgG and/or IgM) was 16.0%, i.e., 2,842 out of 17,764 inhabitants had developed SARS-CoV-2 antibodies. This figure is significantly higher than that reported in Italy (11.0%), i.e., 6, 30,000 confirmed SARS-CoV-2 cases [26]. This high seroprevalence is consistent with other studies [27‐31], 9.7% reported among 2,766 subjects from Switzerland (2020) [27], 13.8% among 674 from Spain (2021) [28], 21.4% of 380 healthy blood donors from Pakistan (2020) [24] and 23% of 390 subjects from Italy (2020) [31].

The seroprevalence of anti-SARS-CoV-2 total antibody was significantly higher among males (OR 1.220, 95% CI 1.110–1.340) (Table 4). In contrast, no significant differences in the seroprevalence of COVID antibodies have been reported among males than females (OR 1.2, 95% CI 0.8–1.8), which is also evidently proven by the existing literature [26, 32, 33]. Furthermore, we found no significant increase in the odds of SARS-CoV-2 antibody seropositivity with respect to age (OR 1.003, 95% CI 1.000–1.006) (Table 4). In contrast, a study reported that the risk of seropositivity was lower among individuals ≤ 20 years of age than those aged > 60 years (OR: 0.72, 95% CI: 0.60–0.87) [34].

In addition to age and gender, close unprotected contact with a confirmed or probable COVID-19 case was also identified as a potential factor affecting the odds of seropositivity (OR 2.437, 95% CI 2.221–2.675) (Table 4) and also confirmed by the forward Wald stepwise logistic regression analysis, including the significant predictors of positive antibody in univariate analysis (Tables 5 and 6). As per the final observed model, number of dependents, symptomatology, closed unprotected contact with the COVID-19 probable or confirmed case, travelling history, and proximity with someone who travelled were the independent significant predictors of seropositivity (Table 7). A study conducted by Rudberg et al. reported that the seropositivity was significantly affected by the patient-related work, i.e., 21% of the 1,764 subjects in patient contact vs. 9% of the 305 subjects without patient contact (OR 2.9, 95% CI 1.9–4.5) [35]. Moreover, the subjects with confirmed COVID patient contact had 1.4 times higher seropositivity odds than those who had non-COVID patient contact (p < 0.05) [35].

As for clinical symptoms, we found that symptomatic participants had 2.18 times higher odds of IgG seropositivity and 1.2 times higher IgM seropositivity than the asymptomatic participants (Table 3). The presence of symptoms was significantly associated with seropositivity of IgG antibody and insignificant for the IgM antibody. Interestingly, 9% of the asymptomatic subjects showed IgM antibodies, and 5.2% had IgG antibodies. Also supported by multiple other studies [24, 26], Vena et al. reported seropositivity among 8.6% of the asymptomatic subjects, and a similar local study reported that 53.2% of COVID positive HCWs were completely asymptomatic [26]. These findings indicated that non-apparent infections are common among healthy, active individuals. However, the seroprevalence of SARS-CoV-2 antibodies was reportedly high among symptomatic individuals (p < 0.05). The most commonly reported symptoms of the IgM seropositive patients were fever, cough and flu. Whereas headache, flu, cough and sore throat were the most frequently reported symptoms among IgG-positive cases (Table 2). In contrast, a study described the strongest association of symptoms like anosmia, ageusia, and fever with seropositivity of SARS-CoV-2 antibody [35].

The present study's findings have several implications for pandemic management since a large proportion of COVID infected subjects remain asymptomatic, also supported by the current study findings. Hence, the actual numbers of confirmed SARS-CoV-2 cases might be higher than those reported. Thus, we need to employ stringent screening strategies targeting the infectious spread beyond the current symptom-driven approach [36]. Furthermore, the study limitations must also be discussed; we analyzed the serum samples of only those who voluntarily decided to be tested (workers from selective organizations), suggesting selection bias. Moreover, the prevalence estimates might have varied due to false serology test results and among elderly or immunosuppressed subjects. Hence, the data cannot be generalized to whole population and the objective was to determine the significant characteristics which are still difficult to predict due to varying and non-specific nature of Coronavirus disease.

In conclusion, the overall SARS-CoV-2 antibodies (IgG and/or IgM) seroprevalence was 16.0%. The multivariate model predicted that number of dependents, symptomatology, close unprotected contact with a confirmed or probable case of COVID-19, travelling history and proximity with someone who traveled are the significant independent predictors of total antibody seropositivity in the studied population. More efforts concerning public health investigations for COVID-19 on a regional and national level should be made to identify the particular risks and outcomes related to SARS-CoV-2 seroprevalence.