Keywords
COVID-19, pandemic, survivors, respiratory failure
This article is included in the Emerging Diseases and Outbreaks gateway.
This article is included in the Coronavirus collection.
COVID-19, pandemic, survivors, respiratory failure
The coronavirus disease 2019 (COVID-19) outbreak caused by the severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) infection was first identified in Wuhan, China, in December 2019. This evolved into a challenging pandemic disrupting public health worldwide, causing significant morbidity and mortality.1 Its transmission rate is also much higher than in the past coronavirus outbreaks.2,3 Nepal encountered the first wave in 2019 and the second one in April 2021 onwards.4 Being a developing country with inadequate health infrastructures, the country was not sufficiently prepared for its intensity, particularly for the second wave, and faced a huge shortage of hospital beds and oxygen supplies.5
The clinical manifestations of COVID-19 range from asymptomatic infection to severe respiratory failure. The common symptoms include fever, dry cough, dyspnea, and sore throat.6,7 Acute Respiratory Distress Syndrome (ARDS) and shock are the most common complications.8 Older age, male gender, pre-existing co-morbidities, lower oxygen saturation at admission, lymphopenia, increased C-reactive protein (CRP) and d-dimer levels have been found to be associated with critical illness and death.9–13 The in-hospital mortality varies from 18.9% to 20.3%, and is higher in patients admitted to the Intensive Care Unit (ICU) than in general wards14,15
The studies on outcomes of the pandemic in tertiary care centers are limited, especially in developing countries with limited health infrastructures. Likewise, there is a paucity of data on the second wave of COVID-19 outbreak in South East Asia. This study compares the clinical characteristics and outcomes between the survivor and non-survivor groups during the second COVID-19 spike. This could give valuable information in the early identification of high-risk groups, provision of appropriate supportive care, hence decreasing mortality.
The study was conducted in a tertiary care center, Kathmandu, Nepal, during the acute surge of the second wave COVID-19 pandemic. It is a teaching hospital with 635 beds in total. With the evolution of the pandemic, significant structural changes were made in our hospital's organization, which led to the formation of 100 bedded general COVID-19 ward, 30 bedded COVID-19 High Care Unit (HCU), and 20 bedded COVID-19 Intensive Care Unit (ICU).
This was a single-center, analytical, cross-sectional study conducted in the laboratory-confirmed COVID-19 patients admitted in a tertiary care hospital.
A non-probabilistic consecutive sampling method was adopted. All COVID-19 patients confirmed by real-time polymerase chain reaction (RT-PCR) were selected consecutively according to their admission to the hospital within the study period. The minimum sample size was calculated by using Cochran’s formula, resulting in 384.16, taking a prevalence of 50% with a confidence interval of 95% and a margin of error of 5%. However, considering the non-response rate of 10%, the final sample size was approximately 429.
The investigators themselves collected data using semi-structured questionnaires from patients’ hospital books and admission files. In addition, information related to vaccination status was collected from the attendants of the patients. Data collection was done during the peak of the second wave of COVID-19 in Nepal, starting from the second half of April 2021 and continuing until June 2021.
The collected data included socio-demographic information like age and gender. Baseline information on co-morbidities and vaccination status were also noted. The clinical characteristics at admission included presenting symptoms, pulse rate, respiratory rate, peripheral oxygen saturation in room air (SpO2), and severity of the disease. The severity at admission was categorized into four groups: mild, moderate, severe, and critical, based on the World Health Organization (WHO) COVID-19 Clinical Management Guidelines (Living Guidance 25 January 2021).16 Similarly, we also noted relevant laboratory tests like total leucocyte count, neutrophil and lymphocyte counts, d-dimer level, inflammatory markers (C-reactive protein, serum ferritin, and serum lactate dehydrogenase), as well as high-resolution computerized tomography (HRCT) scores under clinical characteristics.
All patients were followed up until discharge or death. Mortality (yes/no) was the primary outcome of interest. Based on primary outcome, we divided the cases into survivors and non-survivors. The secondary outcomes were the development of complications, need for mechanical ventilation, duration of hospital stay, and duration of ICU stay (for those who were initially admitted to ICU). Likewise, information regarding the types of mechanical ventilation (Continuous positive airway pressure [CPAP]; Bi-level positive airway pressure [BiPAP] and endotracheal intubation was also noted.
Ethical approval was obtained from the Institutional Review Committee (IRC Regulation number 427, Reference number 245) of the Nepalese Army Institute of Health Sciences. Before conducting the study, permission was obtained from the hospital authority and the COVID-19 “In-charge”, i.e. the medical doctor responsible for the COVID-19 wards at our facilities. The informed verbal consent was taken from patients’ attendants (next of kin). In fact, all COVID-19 patients were treated in isolation. Only the on-duty doctors and nurses visited them in their daily rounds and according to need. No other persons were allowed inside. Therefore, in this scenario, we had to seek consent from the patients’ attendants. Regarding the type of consent, we took only verbal consent from them. The main reason behind this was that during the COVID-19 pandemic, we had to strictly follow safety guidelines. We had to maintain a safety distance while talking to them. Due to this, we found that verbal consent was the most appropriate method of taking consent. Likewise, the privacy and anonymity of patient information were well-maintained.
The statistical analysis was run using Statistical Package for the Social Sciences (IBM-SPSS), version-23. The dependent variable was the mortality outcome (yes/no), while the other variables affecting mortality were independent variables. A Shapiro-Wilk W test was performed to check the normality of continuous data. Mean/standard deviation and median/interquartile range (IQR) were calculated for normally and non-normally distributed variables, respectively. A student’s t-test was applied to check the association between normally distributed continuous variables whereas a Mann Whitney U test was used for non-normally distributed continuous variables. Likewise, a Chi square test was applied for categorical variables. However, in case of >20% cells having expected count less than five, Fisher’s exact test was applied.
All the variables which were significant in the bivariate analysis and did not show collinearity among themselves were further tested by multivariable analysis using a logistic regression model. The significance level was set as p < 0.05, with a 95% confidence interval considering a 5% standard error throughout the analysis.
Only univariate analysis was performed for variables related to laboratory tests and HRCT scores because data related to these were not available for all cases.
A total of 429 COVID-19 patients admitted to our center were selected and analyzed.
The mean age of patients was 59.21±15.19 years, and 256 (59.67%) were males. The survivors were significantly younger, with a mean age of 57.41±14.93 years than non-survivors, whose mean age was 66.61±14.03 years. In addition, the mortality in males (21.48%) was higher compared to females (16.76%) though it was not statistically significant.
Hypertension (129, 30.07%) was the most common comorbidity, followed by diabetes mellitus (82, 19.11%) and chronic obstructive pulmonary disease (44, 10.26%). Among hypertensive patients, 102 (79.07%) survived, and 27 (20.93%) died.
Vaccination status was known in only 293 cases. Among these, 83 (28.33%) cases were vaccinated with either single or both doses of COVID-19 vaccine (Covishield/Vero cell). The mortality was slightly higher in non-vaccinated compared to vaccinated (21.43% vs 19.28%). However, this was not statistically significant (Table 1).
The clinical characteristics at the time of admission to the hospital are shown in Table 1. The common symptoms at presentation were fever (290, 67.60%), cough (263, 61.31%), and shortness of breath (261, 60.84%). Mortality was significantly higher in those who presented with shortness of breath (61 [23.37%], p=0.014) and anorexia (10 [40.00%], p=0.016). The time of hospital admission after symptom onset was similar across both groups (5.00, IQR=3.00-7.00).
The median pulse and respiratory rate were 90.00 (IQR, 82.00-100.50) and 22.00 (IQR, 21.00-24.00) respectively. Non-survivors had significantly higher pulse rate (95.50, IQR=86.00-103.50 vs 90.00, IQR=80.00-100.00; p = 0.013) and respiratory rate (24.00, IQR=22.00-28.00 vs 22.00, IQR=20.00-24.00; p<0.001).
The average SpO2 at admission was 89.00% (82.00-92.00). The non-survivors were more likely to have lower oxygen saturation at admission (80.00, IQR=68.00-86.75 versus 90.00, IQR=85.00-93.00; p<0.001). A total of 227 (52.91%) patients were severe to critically ill at admission. The mortality in this group was significantly higher (75 [33.04%], p<0.001) compared to those who were mild to moderately sick (9 [4.66%], p<0.001).
Complete blood count at admission showed polymorphonuclear leukocytosis (80.00% [71.00% - 87.00%] and lymphopenia (16.00% [9.90% - 24.00%]). Neutrophil count was higher and lymphocyte count was lower in non-survivors compared to survivors.
Out of 252 cases, CRP was positive in 206 (81.75%) cases, of which 23.30% died. However, mortality in CRP-negative cases was only four (8.70%). Likewise, the average serum lactate dehydrogenase (LDH), serum ferritin, and d-dimer levels were also raised at admission, and these values were comparatively higher among the non-survivors than survivors.
HRCT was done in 139 cases. Among them, the median CT severity score was 15.00 (11.00-18.00). The score was higher in non-survivors (19.00 [16.50-23.00]) compared to survivors (14.00 [11.00-17.00]) (Table 2).
The overall mortality was 84 (19.58%). The median duration of hospital stay was 11.00 (8.00-14.00) days. Among patients initially admitted to ICU, the mortality was 36 (58.06%), and the average duration of ICU stay was 8.00 (6.00-12.00). The most common complications were hypoxemic respiratory failure (69, 16.08%) and ARDS (37, 8.62%). The time of development of the first complication after symptom onset was 8.00 (5.00-11.00), and it was lower in survivors than non-survivors (7.00, IQR=5.00-11.00 vs 8.00, IQR=5.00-11.00). In total, seven (1.63%) patients required invasive mechanical ventilation, and none of them survived (Table 1).
Table 3 shows a binary logistic regression analysis taking mortality (yes/no) as an outcome of interest. A total of 10 variables found to be significant in bivariate analysis, namely age, shortness of breath (yes/no), anorexia (yes/no), respiratory rate, pulse rate, SpO2 in room air, severity at admission, hypoxemic respiratory failure (yes/no), mechanical ventilation (yes/no) and total duration of hospital stay were taken for multivariable analysis. Other significant variables like hypercapnic respiratory failure (yes/no), ARDS (yes/no), sepsis/septic shock (yes/no), and AKI (yes/no) were excluded because of the insufficient number of cases (<10) in each comparison groups. Likewise, mechanical ventilation (CPAP, BiPAP, and endotracheal intubation) were omitted because of collinearity. The pseudo R2 value for multivariable logistic regression analysis was 0.682, indicating that our model predicted a similar outcome in about 68% of observations.
In this model, only four variables (age, clinical severity at admission, mechanical ventilation, and total duration of hospital stay) were found to be significantly associated with mortality outcome at a 5% level of significance. After adjusting for other covariates, it was seen that every one-year increment in age had a 7.7% higher odds of occurrence of mortality event (adjusted OR, 1.077; 95% CI, 1.045-1.110, p<0.001). Similarly, the patients who were severe critically ill at admission had approximately six times higher risk of mortality compared to mild to moderate cases (adjusted OR, 5.861; 95% CI, 1.989-17.270, p=0.001). Patients who were under mechanical ventilation had a significantly higher risk of mortality than those who were not ventilated (adjusted OR, 39.059; 95% CI, 5.249-290.655, p<0.001). Likewise, the duration of hospital stay was also significantly associated with mortality (adjusted OR, 0.795; 95% CI, 0.731-0.863, p<0.001).
This was a single-center cross-sectional study conducted during the peak of the second wave of COVID-19 pandemic in a developing country. We investigated different clinical, laboratory and radiological features of RT-PCR-confirmed COVID-19 patients, their treatment patterns, and outcomes. In our study, the mean age of the patients was 59.21±15.19 years. Hypertension was the most common co-morbidity, and the survivors tend to be younger compared to non-survivors. In addition, the death rate was higher in non-vaccinated cases (21.43%) than vaccinated ones (19.28%).
Patients presenting with shortness of breath and anorexia had higher mortality rates. The non-survivors tended to have higher pulse rate, respiratory rate, and lower oxygen saturation at admission. Half of the patients (52.91%) were severe to critically ill with higher mortality than mild to moderate cases. The median CT severity score was higher in non-survivors compared to survivors. The former had a higher neutrophil count, lower lymphocyte, and increased ferritin, LDH, and d-dimer at the time of admission. Hypoxemic respiratory failure (69, 16.08%) and ARDS (37, 8.62%) were the most common complications associated with higher mortality. None of the patients who required invasive mechanical ventilation survived. The average duration of hospital stay was 11.00 (8.00-14.00) days. The overall in-hospital mortality was 84 (19.58%), and ICU mortality was 36 (58.06%).
In our study, the most common presenting symptom was fever (67.60%) followed by cough (61.3%) and shortness of breath (60.84%). This is similar to previous studies by Rodriguez-Molinero et al.,17 Du et al.10 and Chen et al.11 However, other studies showed shortness of breath18 and cough19 as the most common symptoms. The patients in the deceased group in our study had a significantly higher pulse rate, which corroborates with Du et al.10 In the present study, hypertension was the most common co-morbidity, which is in line with other studies.10,11 On the contrary, a study by Ali et al.18 reported diabetes mellitus as the most common co-morbidity. Our findings and past studies10,20 depicted that patients in the non-survivor group were older than survivors.
Many previous studies found that COVID-19 patients present with neutrophilia, lymphopenia, raised C-reactive protein, LDH, and d-dimer levels.11,14,17,21 These laboratory findings also correspond to the present study. However, in our study sample, the total leucocyte count was within normal limits, which contradicts studies by Rodriguez-Molinero A et al.17 and Zhang G et al.14 that reported leukocytosis and leucopenia respectively.
Hypoxemic respiratory failure and ARDS were the most common complications in our patients. Similar complications were observed in Rosenthal N et al.15 and Chen R et al.11 Other complications seen in these studies were secondary infection, septic shock, and acute kidney injury. In our case, sepsis/septic shock and acute kidney injury were the least common.
The overall in-hospital mortality in the present study was comparable with that depicted by United States (US) based research by Rosenthal N et al.15 On the contrary, studies conducted in London by Goodall JW et al.21 showed higher mortality. On the other hand, lower mortality has been reported by different studies in China from where the pandemic originated.10,14,20 Likewise, a study from another South Asian developing country, Pakistan, showed comparatively higher mortality than ours.18 So, the mortality pattern is inconsistent in different parts of the world. These differences could be due to different study populations, sample size, study time, and unique variants of coronavirus. Another reason for the discrepancy could be different governmental health policies, crisis management strategies, and health expertise. In addition, patients’ age was significantly associated with mortality in our study. A similar association was reported by Rosenthal N et al.,15 Goodall JW et al.,21 and Xun Li et al.22
Unlike other studies from the US,15 and China,20 the ICU admission rate was comparatively lower in ours. Similarly, the need for invasive mechanical ventilation was found comparatively higher in those from developed countries like the US15 and the United Kingdom (UK).21
There are some limitations of our study to be mentioned. It was a single-center study with small sample size; therefore, the results may not be generalized for the whole country or community. However, the study site was a tertiary teaching hospital with high patient flow. Cases from different parts of the country are referred here. We included all the COVID-19 patients admitted consecutively within our study period regardless of their socio-demographic and severity profile. Most of the case-related information was recorded from patients’ admission files. In this situation, it is possible that patients’ clinical history and laboratory values may not have been recorded accurately and timely, possibly giving rise to information bias. To reduce this to a minimum, we cross-checked clinical history from patients’ socio-demographic information and laboratory parameters from the available Health Management Information System (HMIS) computer database.
Since mild cases were also admitted for observation in our hospital, all laboratory and radiological investigations were not performed in these groups. This gave rise to missing data in the laboratory and radiological variables. Therefore, we did not run bivariate and multivariate analyses on these parameters. However, we still performed univariate analysis from available information, the results of which are significant to the medical literature and case management in the future. Similarly, another limitation is that we only selected RT-PCR-confirmed COVID-19 patients. This could potentially exclude those cases infected with coronavirus but being false negative on laboratory investigation. However, there was a practice of repeat PCR in suspected patients in our center, which might have contributed to some extent in reducing this selection bias.
Alongside the limitations, we have a lot of strengths worth mentioning. There are a very few studies on COVID-19 vaccination published in our setting. In this study, we have compared the outcomes of vaccinated and unvaccinated individuals in terms of mortality. To our knowledge, this is the first study from a developing country in South-East Asia which reports on clinical characteristics and outcomes of COVID-19 patients during the second wave of the pandemic. Likewise, we have included cases of different severity ranging from mild to severe to critical, thus possibly reducing the overestimation of mortality and morbidity.
The non-survivors of COVID-19 tended to be of older age, severe to critically ill at presentation, require mechanical ventilation, and have a shorter duration of hospital stay, compared to survivors. Hence, these factors should be considered during the initial assessment and then appropriate care should be provided.
Figshare: Comparison of clinical characteristics and short-term outcomes among COVID-19 patients in a tertiary care center during second wave pandemic in Nepal: a cross-sectional study, https://doi.org/10.6084/m9.figshare.16941103.23
This project contains the following underlying data:
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Views | Downloads | |
---|---|---|
F1000Research | - | - |
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)