RESULTS
Our literature search yielded 1,690 unique abstracts. After the removal of duplication and applying the eligibility criteria, 218 relevant articles were examined for further consideration. Of these, 204 studies were excluded mainly for two reasons: no original data and combined mortality and severity endpoint (Figure S1 ). A total of 13 original studies (nine from China, one from Korea, one from the United States, and one multination database) were then included in this review; eleven studies reporting on mortality in ACEI/ARB exposed and unexposed groups and ten studies reporting on disease severity in ACEI/ARB exposed and unexposed groups.
Seven studies, Yang et al. [13], Meng et al. [14], Feng et al. [15], Peng et al. [16], Huang et al. [18], Li et al 20], and Liu et al [21] defined severity of COVID-19 according to Diagnosis and Treatment Protocol for Novel Coronavirus Pneumonia by Chinese National Health Commission. Zeng et al. [19] defined the severity of pneumonia using the guideline for community-acquired pneumonia. Zhang et al. [11] did not classify patients into severe and non-severe cohort but we considered patients with septic shock in the study as having severe/critical COVID-19.
Seven studies were included to calculate pooled relative risk for mortality, absolute risk difference, and number needed to treat to benefit [11,13,14,18-20,23] in COVID-19 patients with hypertension, exposed or unexposed to ACEIs/ARBs (Table 2 ). Lee et al. [12], Peng et al. [16], Chen et al. [17], and Mehra et al. [22] did not segregate the patients based on the presence of hypertension in their studies and therefore were excluded from our pooled analysis to ensure a uniform cohort of hypertensive patients being analysed. The age differences between ACEIs/ARBs and non-ACEIs/ARBs groups were also huge in studies conducted by Lee et al. [12] and Chen et al. [17]. The studies by Feng et al. [15], and Liu et al. [21], did not present mortality data by ACEIs/ARBs use and hence were also excluded from the analysis.