Age but not sex may explain the negative effect of arterial hypertension and diabetes on COVID-19 prognosis

Abstract

We read with great interest the article of Zheng and colleagues who summarized current evidence regarding risk factors for severe clinical forms of COVID-19.1Zheng Z. Peng F. Xu B. Zhao J. Liu H. Peng J. et al.Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis.J Infect. 2020 Apr 23; ([Epub ahead of print])https://doi.org/10.1016/j.jinf.2020.04.021Abstract Full Text Full Text PDF Scopus (1263) Google Scholar The authors concluded that patients who are males, aged over 65 and smoking might face a higher risk of developing into the critical or mortal condition by COVID-19.1Zheng Z. Peng F. Xu B. Zhao J. Liu H. Peng J. et al.Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis.J Infect. 2020 Apr 23; ([Epub ahead of print])https://doi.org/10.1016/j.jinf.2020.04.021Abstract Full Text Full Text PDF Scopus (1263) Google Scholar Besides, the authors highlighted that comorbidities such as hypertension, diabetes, cardiovascular and/or pre-existing respiratory diseases could also significantly affect the prognosis of the COVID-19.1Zheng Z. Peng F. Xu B. Zhao J. Liu H. Peng J. et al.Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis.J Infect. 2020 Apr 23; ([Epub ahead of print])https://doi.org/10.1016/j.jinf.2020.04.021Abstract Full Text Full Text PDF Scopus (1263) Google Scholar Collectively, the authors confirmed the negative effect of comorbidities on the natural course of COVID-19. Nevertheless, there are some aspects that need to be carefully assessed, including the influence of essential demographic confounding variables such as age and sex, which are also strongly associated with the above mentioned comorbidities but unfortunately were not thoroughly included in the analysis. Hence, to answer the question of whether age and sex may influence the effect size/s of pre-existing comorbidities on COVID-19 severe prognosis, we used the data of the Zheng and coworkers1Zheng Z. Peng F. Xu B. Zhao J. Liu H. Peng J. et al.Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis.J Infect. 2020 Apr 23; ([Epub ahead of print])https://doi.org/10.1016/j.jinf.2020.04.021Abstract Full Text Full Text PDF Scopus (1263) Google Scholar as input to perform a meta-regression analysis. Of note, we found that the negative effect of underlying arterial hypertension on COVID-19 critical illness significantly and positively correlated with the age difference between critical/mortal and non-critical patients (slope±SE: 0.0718±0.021, p=0.00066) but not with the difference in male sex proportion (slope:-0.010±0.023, p=0.66) (Fig. 1, A/B). Likewise, the negative effect of type 2 diabetes on severe COVID-19 infection significantly and positively correlated with the age difference between the two groups (slope: 0.079±0.025, p=0.00185) but not with the difference in male proportion (slope: -0.040±0.027, p=0.133) (Fig. 1, C). On the contrary, by meta-regression analysis, we found that the negative effect of pre-existing respiratory disease on COVID-19 critical illness significantly and positively correlated with the difference in male proportion (slope: 0.118±0.056, p= 0.034) but not with the age difference (slope: -0.030±0.048, p= 0.520). Finally, we observed that the negative effect of pre-existing cardiovascular disease on severe COVID-19 clinical course is not influenced either by sex (slope: 0.0343±0.033, p= 0.300) or by age (slope: 0.048±0.033, p= 0.143). In conclusion, there are three relevant messages to highlight. First, it is crucial to perform a meta-regression analysis to explain statistical heterogeneity in terms of study-level variables. Second, assessment of potential confounders, including essential demographic aspects, such as age and sex, are relevant to robustly demonstrate the putative association between variables, including assessment of disease risk and or severe prognosis. Most importantly, accurate assessment of confounding variables provides relevant information to stakeholders, including physicians and practitioners who need to take immediate action to reduce morbidity and mortality of COVID-19. Finally, assessment of the effect sizes of moderator variables on risk factors for severe COVID-19, including age and sex, may help to understand the biology of the disease in future larger studies. Carlos J Pirola: Writing - review & editing, Formal analysis, Conceptualization. Silvia Sookoian: Writing - review & editing, Conceptualization. We have no conflict of interest to declare. This study was partially supported by grants PICT 2014-1816, PICT 2015-0551, and PICT 2016-0135 (Agencia Nacional de Promoción Científica y Tecnológica, FONCyT), CONICET Proyectos Unidades Ejecutoras 2017, PUE 0055.