Translate this page into:
Unexpected role of corticosteroids in coronavirus disease 2019: A Saudi Arabian cohort study
PhD, Marwah Afeef,2MSc, Zuhair S. Natto1PhD, Khalid A. Alzahrani3BSc, Samiyah O. Alharbi4BSN, Abdulrazaq I. Althobaiti5BSc, Sana M. Mubarak6MSN
*Corresponding author: Marwah Afeef, Department of Research and Studies, Al-Thagher Hospital, Jeddah, Saudi Arabia. mbinafee1@gmail.commafeef@moh.gov.sa
-
Received: ,
Accepted: ,
How to cite this article: Ashi HM, Afeef M, Natto ZS, Alzahrani KA, Alharbi SO, Althobaiti Al, et al. Unexpected role of corticosteroids in coronavirus disease 2019: A Saudi Arabian cohort study. Int J Health Sci (Qassim). 2025;19:25-30. doi: 10.25259/OJS_8681
Abstract
Objectives:
Multiple studies suggested that therapies used by individuals with respiratory illnesses to have a protective effect against coronavirus disease 2019 (COVID-19) infection. Our aim is to explore COVID-19 infection outcomes among patients with respiratory illnesses treated with corticosteroids.
Methods:
This was a retrospective cohort study. One hundred and twenty-six cases were recruited in the study. Among them, 63 were confirmed to be using any sort of corticosteroids to manage their respiratory illnesses, and 63 were confirmed to be non-corticosteroid users.
Results:
Logistic regression indicated receiving corticosteroids was a significant predictor of COVID-19 infection, with an odds ratio (OR) of 7.037 (95% confidence interval [CI]: 2.682–18.463; P < 0.001). In the overall sample, logistic regression showed that receiving corticosteroids did not significantly predict survival (OR: 4.361; 95% CI: 0.420–45.310; P = 0.218), meaning not receiving corticosteroids will lead to more death among all study samples. Among COVID-19 patients, not receiving corticosteroids was a significant predictor of mortality, with an OR of 383.423 (95% CI: 2.070–71015.113; P = 0.026), meaning not receiving corticosteroids will lead to more death once a patient has COVID-19. Age approached significance (P = 0.071), suggesting older age may be associated with higher mortality, but gender and medical history were not significant predictors.
Conclusion:
While not significantly associated with survival in the overall sample, the absence of corticosteroid therapy was a significant predictor of mortality among COVID-19 patients. This suggests a potential protective effect of corticosteroids in the context of COVID-19 infection.
Keywords
Coronavirus disease 2019
Corticosteroids
Mortality
Respiratory illnesses
Saudi Arabia
INTRODUCTION
Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is a coronavirus that, like severe acute respiratory syndrome and middle east respiratory syndrome (MERS), can cause severe respiratory illnesses.[1] SARS-CoV-2 is the primary causative agent of coronavirus disease 2019 (COVID-19).[2-5] On March 11, 2020, the World Health Organization announced that SARS-CoV-2 is a global pandemic.[3,4,6-8] Based on the COVID-19 Saudi Arabia Ministry of Health Dashboard, globally, as of April 5, 2021, there have been more than 131 million confirmed cases of the disease, including more than 2 million deaths. Nationally, as of April 5, 2021, there have been 392,682 confirmed cases, including 6,697 deaths.[9]
According to the U.S. Centers for Disease Control and Prevention, individuals with obesity, hypertension, diabetes, and respiratory illnesses are at heightened risk of experiencing severe complications from COVID-19.[10,11] Acute respiratory distress syndrome (ARDS) has been identified as a major cause of mortality in critically ill COVID-19 patients, often linked to the cytokine storm associated with the infection.[12] However, current evidence does not indicate that individuals with respiratory illnesses are at a significantly increased risk of adverse COVID-19 outcomes.[10,11,13] Surprisingly, among COVID-19 cases reported, asthma appeared to be very low compared to other comorbidities such as diabetes.[14,15] Multiple studies suggested that therapies used by individuals with respiratory illnesses to have a protective effect against the infection or its symptoms.[11,14,16]
The aim of our study is to explore the outcomes of COVID-19 infection among patients with respiratory illnesses treated with corticosteroids.
The objectives of the study are to assess the clinical information and disease manifestation, plus the demographic characteristics believed to have an influence on the disease characteristics and outcomes.
MATERIALS AND METHODS
Study design
This study employed a retrospective cohort design to investigate the association between corticosteroid use and the risk of COVID-19 infection among patients with respiratory illnesses. This study was approved by the Institutional Review Board holding the National Registration number with NCBE-KACST, KSA: (H-02-J-002) based at Jeddah Health Affairs. IRB Log No (1525) August 04, 2021. It is a retrospective cohort study that took place at public hospitals in Jeddah city of Saudi Arabia.
Participants
Patients were identified from the electronic medical records of King Fahad General Hospital between February 2020 and May 2021. Inclusion criteria included: (1) diagnosis of a respiratory illness (e.g., asthma, chronic obstructive pulmonary disease, and pneumonia); (2) receipt of corticosteroids or non-corticosteroid medications for the treatment of their respiratory illness; and (3) availability of complete medical records, including information on COVID-19 testing and outcomes.
Sample size calculation
A sample size calculation was conducted using GPower (Version 3.1). Assuming the expected proportion of patients with COVID-19 among the corticosteroids group is 10%, and among non-corticosteroids 30%,[11,14,16] sample size of n = 124 (62 in each group) is adequate to obtain a Type I error rate of 5% and a power 80%.
A total of 126 cases, among them 63 confirmed to be using any sort of corticosteroids to manage their respiratory illnesses, and 63 confirmed using non-corticosteroid medications to manage their respiratory illnesses.
All patients were contacted by telephone and verbally consented to take part in the study. Consented participants reported their previous infection with COVID-19. Then medical records were reviewed to extract the following data: patient demographics (age, sex, comorbidities); corticosteroid or non-corticosteroid medication used; COVID-19 testing results (positive or negative); COVID-19 severity (mild, moderate, and severe); hospitalization status; and mortality.
Data management
A web-based survey tool used to collect the gathered data, Google Forms (Google, LLC), used to generate a standard digitally secured questionnaire link that was in the possession of the data collector and entry individual to fill out each patient form.
Statistical analysis
For the statistical analysis, we conducted a comprehensive descriptive and inferential analysis. Descriptive statistics were performed using means and standard deviations for continuous variables and frequencies and percentages for categorical variables. We employed a forward stepwise logistic regression model to identify predictors of COVID-19 infection and mortality among the study participants. The steroid treatment status was categorized for binary logistic regression to assess the impact of corticosteroids on the outcomes. Odds ratios (OR) and 95% confidence intervals (CIs) were calculated for each predictor to quantify the strength and precision of the associations. All analyses were conducted using IBM Statistical Package for the Social Sciences Statistics, version 29.
RESULTS
Demographic characteristics
The study sample consisted of 126 individuals divided equally into two groups: Those who received corticosteroids (n = 63) and those who did not (n = 63) [Table 1]. The mean age for the corticosteroids group was significantly lower (47.38 ± 18.36 years) compared to the non-corticosteroids group (61.97 ± 17.94 years), with P < 0.001 indicating statistical significance. The body mass index was similar between the two groups (29.26 ± 7.26 vs. 28.98 ± 7.01; P = 0.862). In terms of gender distribution, there were more females in the corticosteroids group (49.2%) compared to the non-corticosteroids group (30.2%), while the majority in the latter group were males (69.8%), showing a statistically significant difference (P = 0.029). The occupation of healthcare workers did not significantly differ between the groups (P = 0.717). Medical history revealed significant differences (P < 0.001). A larger portion of the noncorticosteroids group had a combination of cardiovascular diseases and diabetes (41.3%), compared to 14.3% in the corticosteroids group. Moreover, 60.3% of the corticosteroids group had no medical history, in contrast to 23.8% in the non-corticosteroids group. COVID-19 infection was more prevalent in the corticosteroids group (66.7%) compared to the non-corticosteroids group (28.6%), with a significant P-value (<0.001). Symptomatic cases were more common in the corticosteroids group (90.7%) compared to the noncorticosteroids group (80.0%), though this difference was not statistically significant (P = 0.251). Regarding the type of isolation, most patients were hospitalized, with no significant difference between the groups (76.2% vs. 83.3%; P = 0.736). The need for oxygen therapy was also similar between the groups (P = 0.478). The outcome showed a significant difference, with no deaths reported in the corticosteroids group compared to 33.3% mortality in the non-corticosteroids group (P < 0.001) [Table 1].
| Variables | Total n=126 n(%) | Corticosteroids group n=63 n(%) | No corticosteroids group n=63 n (%) | P-value |
|---|---|---|---|---|
| Age (mean±SD) | 54.67±19.50 | 47.38±18.36 | 61.97±17.94 | <0.001* |
| BMI (mean±SD) | 29.12±7.11 | 29.26±7.26 | 28.98±7.01 | 0.862 |
| Gender | ||||
| Females | 50 (39.7) | 31 (49.2) | 19 (30.2) | 0.029* |
| Males | 76 (60.3) | 32 (50.8) | 44 (69.8) | |
| Occupation healthcare worker | ||||
| Yes | 8 (6.3) | 5 (7.9) | 3 (4.8) | 0.717 |
| No | 118 (93.7) | 58 (92.1) | 60 (95.2) | |
| Medical history | ||||
| Cardiovascular | 18 (14.3) | 7 (11.1) | 11 (17.5) | <0.001* |
| Diabetes | 9 (7.1) | 5 (7.9) | 4 (6.3) | |
| Cardiovascular+Diabetes | 35 (27.8) | 9 (14.3) | 26 (41.3) | |
| Others | 10 (7.9) | 3 (4.8) | 7 (11.1) | |
| None | 53 (42.1) | 38 (60.3) | 15 (23.8) | <0.001* |
| COVID-19 | ||||
| Yes | 60 (47.6) | 42 (66.7) | 18 (28.6) | NA |
| No | 66 (52.4) | 21 (33.3) | 45 (71.4) | |
| Symptomatic** | ||||
| Yes | 55 (87.3) | 39 (90.7) | 16 (80.0) | 0.251 |
| No | 8 (12.7) | 4 (9.3) | 4 (20.0) | |
| Type of isolation** | ||||
| Hospital | 47 (78.3) | 32 (76.2) | 15 (83.3) | 0.736 |
| Home | 13 (21.7) | 10 (23.8) | 3 (16.7) | |
| Received oxygen** | ||||
| Yes | 51 (82.3) | 33 (78.6) | 18 (90.0) | 0.478 |
| No | 11 (17.7) | 9 (21.4) | 2 (10.0) | |
| Outcome** | ||||
| Died | 6 (9.7) | 0 (0) | 6 (33.3) | <0.001* |
| Survived | 56 (90.3) | 44 (100.0) | 12 (66.7) |
Logistic regression analysis for predictors of COVID-19
The logistic regression analysis indicated that receiving corticosteroids was a significant predictor of COVID-19 infection, with an OR of 7.037 (95% CI: 2.682–18.463; P < 0.001) [Table 2]. Age, gender, and medical history were not significant predictors in this model.
| Variables | B | S.E. | Wald | Sig. | OR (95 CI) |
|---|---|---|---|---|---|
| Receive corticosteroid | |||||
| Yes | 1.951 | 0.492 | 15.718 | <0.001 | 7.037 (2.682, 18.463) |
| No | Ref | ||||
| Age | 0.014 | 0.013 | 1.087 | 0.297 | 1.014 (0.988, 1.041) |
| Gender | |||||
| Males | Ref | ||||
| Females | 0.409 | 0.419 | 0.953 | 0.329 | 1.505 (0.662, 3.423) |
| Medical history | |||||
| Cardiovascular | Ref | ||||
| Diabetes | 0.531 | 0.668 | 0.631 | 0.427 | 1.700 (0.459, 6.301) |
| Cardiovascular+Diabetes | 1.765 | 1.023 | 2.974 | 0.085 | 5.839 (0.786, 43.394) |
| Others | −0.125 | 0.948 | 0.017 | 0.895 | 0.882 (0.138, 5.652) |
| None | 0.172 | 0.669 | 0.066 | 0.797 | 1.187 (0.320, 4.402) |
The P-value for Hosmer and Lemeshow test is 0.226. Bold indicate P<0.05. COVID-19: Coronavirus disease 2019, OR: Odds ratio, CI: Confidence interval, S.E.: Standard error, Sig.: Significance, B: Unstandardized coefficient, Ref: Reference category
Logistic regression analysis for predictors of outcome (survived vs. died) among all study sample
In the overall sample, the logistic regression analysis showed that receiving corticosteroids did not significantly predict survival (OR: 4.361; 95% CI: 0.420–45.310; P = 0.218), which means not receiving corticosteroids will lead to more death among all study sample [Table 3]. Age, gender, and medical history were also not significant predictors of survival.
| Variables | B | S.E. | Wald | Sig. | OR (95 CI) |
|---|---|---|---|---|---|
| Receive corticosteroid | |||||
| Yes | Ref | ||||
| No | 1.473 | 1.194 | 1.52 | 0.218 | 4.361 (0.420, 45.310) |
| Age | 0.033 | 0.029 | 1.277 | 0.259 | 1.033 (0.976, 1.093) |
| Gender | |||||
| Males | Ref | ||||
| Females | 0.115 | 0.847 | 0.018 | 0.892 | 1.122 (0.213, 5.904) |
| Medical history | |||||
| Cardiovascular | Ref | ||||
| Diabetes | 0.516 | 1.203 | 0.184 | 0.668 | 1.675 (0.158, 17.710) |
| Cardiovascular+Diabetes | 1.161 | 1.573 | 0.545 | 0.460 | 3.194 (0.146, 69.721) |
| Others | NA | ||||
| None | 0.023 | 1.66 | 0 | 0.989 | 1.024 (0.040, 26.493) |
The P-value for Hosmer and Lemeshow test is 0.880. COVID-19: Coronavirus disease 2019, OR: Odds ratio, CI: Confidence interval, S.E.: Standard error Sig.: Significance, NA: Not applicable, B: Unstandardized coefficient, Ref: Reference category
Logistic regression analysis for predictors of outcome (died vs. survived) among COVID-19 patients only
Among COVID-19 patients, not receiving corticosteroids was a significant predictor of mortality, with an OR of 383.423 (95% CI: 2.070–71015.113; P = 0.026), which means not receiving corticosteroids will lead to more death once a patient has COVID19 [Table 4]. Age approached significance (P = 0.071), suggesting that older age may be associated with higher mortality, but gender and medical history were not significant predictors.
| Variables | B | S.E. | Wald | Sig. | OR (95 CI) |
|---|---|---|---|---|---|
| Receive corticosteroid | |||||
| Yes | Ref | ||||
| No | 5.949 | 2.664 | 4.987 | 0.026 | 383.423 (2.070, 71015.113) |
| Age | 0.283 | 0.157 | 3.255 | 0.071 | 1.327 (0.976, 1.805) |
| Gender | |||||
| Males | Ref | ||||
| Females | −3.315 | 2.589 | 1.639 | 0.200 | 0.036 (0, 5.811) |
| Medical history | |||||
| Cardiovascular | 3.228 | 2.534 | 1.622 | 0.203 | 25.226 (0.176, 3623.598) |
| Diabetes | −2.781 | 2.144 | 1.684 | 0.194 | 0.062 (0.001, 4.137) |
| Cardiovascular+Diabetes | Ref | ||||
| Others | NA | ||||
The P-value for Hosmer and Lemeshow test is 0.839. Bold indicate P<0.05. COVID-19: Coronavirus disease 2019, OR: Odds ratio, CI: Confidence interval, S.E.: Standard error, Sig.: Significance, NA: Not applicable, B: Unstandardized coefficient, Ref: Reference category
DISCUSSION AND CONCLUSION
The findings of this study indicate that the use of corticosteroids was significantly associated with an increased risk of COVID-19 infection. Our findings are in line with the evidence supporting the immunosuppressive properties of corticosteroids to impair both innate and adaptive immune responses, elevating the individuals’ risk of systemic infections.[17] Furthermore, it is possible that patients receiving corticosteroids were more likely to be immunocompromised or have underlying conditions that predisposed them to infection.[18] Furthermore, the timing of corticosteroid administration in relation to COVID-19 infection could be a confounding factor.
Our analysis revealed that the absence of corticosteroid therapy was a significant predictor of mortality among COVID-19 patients. Nevertheless, the evidence suggested that the administration of methylprednisolone appeared to be associated with a decreased mortality rate in patients with ARDS.[19] This suggests a potential protective effect of corticosteroids in the context of COVID-19 disease.
Age emerged as a trend toward significance in predicting mortality among COVID-19 patients, aligning with previous studies demonstrating increased risk in older populations. However, the lack of significant associations for gender and medical history in both models suggests that these factors may have limited influence on COVID-19 infection and outcome in our study population.
It is important to acknowledge the limitations of this study, including the potential for confounding factors, the relatively small sample size, and the absence of detailed information on corticosteroid dosage and duration of treatment. Further research is needed to explain the complex relationship between corticosteroids, COVID-19 infection, and patient outcomes. Prospective studies with larger sample sizes and comprehensive data collection are essential to establish causality and inform clinical practice.
Our analysis revealed that while not significantly associated with survival in the overall sample, the absence of corticosteroid therapy was a significant predictor of mortality among COVID-19 patients. This suggests a potential protective effect of corticosteroids in the context of COVID-19 infection.
Authors’ contributions:
HMA, MAA, and ZSN: Conceived and planned the study, and carried out the study in their workplace. MAA and ZSN: Contributed to data analysis, interpretation of the results, and took the lead in writing the manuscript. All authors HMA, MAA, ZSN, KAA, SOA, AIA, and SMM have contributed to the study literature review, data collection and management. All authors provided critical feedback and helped shape the research, analysis, and manuscript. I certify that we have participated sufficiently in the intellectual content, conception, and design of this work. All authors have critically reviewed and approved the final draft and are responsible for the content and similarity index of the manuscript.
Ethical approval:
This study protocol was reviewed and approved by the Institutional Review Board holding the National Registration number with NCBE-KACST, KSA: (H-02-J-002) based at Jeddah Health Affairs. IRB Log No (1525) August 04, 2021.
Declaration of patient consent:
The authors certify that they have obtained all appropriate patient consent.
Financial support and sponsorship:
Nil.
Conflicts of interest:
There are no conflicts of interest.
Availability of data and material:
The datasets generated and/ or analyzed during the current study are available from the corresponding author upon reasonable request. Data sharing subject to institutional and ethical regulations to ensure participant confidentiality and compliance with data protection policies.
References
- The short-term effect of different chlorhexidine forms versus povidone iodine mouth rinse in minimizing the oral SARS-CoV-2 viral load: An open label randomized controlled clinical trial study. Medicine (Baltimore). 2022;101:e28925.
- [CrossRef] [PubMed] [Google Scholar]
- SARS-CoV-2 infection among travelers returning from Wuhan, China. N Engl J Med. 2020;382:1476-8.
- [CrossRef] [PubMed] [Google Scholar]
- High prevalence of olfactory and taste disorder during SARSCoV-2 infection in outpatients. J Med Virol. 2020;92:2310-1.
- [CrossRef] [PubMed] [Google Scholar]
- Cutaneous manifestations in COVID-19: A first perspective. J Eur Acad Dermatol Venereol. 2020;34:e212-3.
- [CrossRef] [Google Scholar]
- COVID-19: Present and future challenges for dental practice. Int J Environ Res Public Health. 2020;17:3151.
- [CrossRef] [PubMed] [Google Scholar]
- Olfactory and oral manifestations of COVID-19: Sex-related symptoms-a potential pathway to early diagnosis. Otolaryngol Head Neck Surg. 2020;163:722-8.
- [CrossRef] [PubMed] [Google Scholar]
- Clinical characteristics of hospitalized patients with SARS-CoV-2 infection: A single arm meta-analysis. J Med Virol. 2020;92:612-7.
- [CrossRef] [Google Scholar]
- Features, Evaluation, and Treatment of Coronavirus (COVID-19). 2023 In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025.
- [PubMed] [Google Scholar]
- COVID19. 2020. Available from: https://covid19.moh.gov.sa [Last accessed on 2020 Jun 25]
- [Google Scholar]
- Asthma in COVID-19 hospitalizations: An overestimated risk factor? Ann Am Thorac Soc. 2020;17:1645-8.
- [CrossRef] [PubMed] [Google Scholar]
- Prevalence and characterization of asthma in hospitalized and nonhospitalized patients with COVID-19. J Allergy Clin Immunol. 2020;146:307-14.e4.
- [CrossRef] [PubMed] [Google Scholar]
- Mortality in COVID-19 patients with acute respiratory distress syndrome and corticosteroids use: A systematic review and meta-analysis. . 2020;14:1149-63.
- [CrossRef] [PubMed] [Google Scholar]
- Asthma in a large COVID-19 cohort: Prevalence, features, and determinants of COVID-19 disease severity. Respir Med. 2021;176:106261.
- [CrossRef] [PubMed] [Google Scholar]
- Do chronic respiratory diseases or their treatment affect the risk of SARSCoV-2 infection? Lancet Respir Med. 2020;8:436-8.
- [CrossRef] [PubMed] [Google Scholar]
- Characteristics of oral manifestations in symptomatic non-hospitalized COVID-19 patients: A cross-sectional study on a sample of the Saudi population. Int J Gen Med. 2021;14:9547-53.
- [CrossRef] [PubMed] [Google Scholar]
- Inhaled corticosteroids and COVID-19: A systematic review and clinical perspective. Eur Respir J. 2020;55:2001009.
- [CrossRef] [PubMed] [Google Scholar]
- The risk of mycobacterial infections associated with inhaled corticosteroid use. Eur Respir J. 2017;50:1700037.
- [CrossRef] [PubMed] [Google Scholar]
- Corticosteroids for COVID-19 therapy: Potential implications on tuberculosis. Int J Mol Sci. 2021;22:3773.
- [CrossRef] [PubMed] [Google Scholar]
- Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern Med. 2020;180:934-43.
- [CrossRef] [PubMed] [Google Scholar]
