LAUSR

We read the article by Sun et al. published in Digestive Diseases and Sciences with great interest. It finds famotidine has no statistically significant difference in reducing the risk of serious disease in COVID19 patients [1]. Meanwhile, it was found that famotidine increased the risk of severe disease in COVID-19 patients after the Propensity Score Matching in Zhou’s article [2]. Because of these contradictory conclusions, we would like to further explore the effect of Propensity Score Matching on famotidine through meta-analysis. A total of 158 articles were searched from Pubmed, Embase, and Web of Science databases up to 21/05/2021. The following terms were used: ((SARS-CoV-2) or (COVID19) or (COVID-19) or (2019-nCoV) or (Coronavirus Disease-19)) AND ((Ranitidine) OR (Ometidine) OR (Famotidine) OR (Cimetidine)). The following criteria were used to include studies: 1. Famotidine use versus famotidine non-use. 2. COVID-19. 3. Human study. 4. Propensity Score Matching study. The following criteria were used to exclude studies: 1. Meta-analysis. 2. duplicate. The last 4 articles were included [2–5]. The HR values of Propensity Score Matching from 4 articles were extracted, the patient's serve disease or death or intubation data of famotidine and non-famotidine were converted into the corresponding log (RR) and SE (considering the low mortality of COVID19, it can be considered that the HR value is equivalent to the OR value) (Table 1). Finally, statistical analysis was performed by Revman 5.3, including forest plot, funnel plot, heterogeneity test, etc. (Figs. 1, 2). Considering the large heterogeneity observed in the results (I2 = 91%, heterogeneity = 0.28), the random-effects model was performed. The pooled RR value was 0.82, CI (0.46, 1.45) P = 0.49 indicated that after Propensity Score Matching, the correlation between the use and non-use of famotidine in COVID-19 patients with the serious illness was not statistically significant, and the P-value is over 0.05 (Figs. 1, 2). Finally, the article by Freedberg et al. was an observational study that cannot exclude the possibility of unmeasurable confounding factors or hidden biases and performed insignificant differences between use and non-use of famotidine reducing the risk of serious disease in COVID-19 patients [3]. This was a single-center study, which may limit the generality of the findings [3]. Zhou et al. indicated that the use of famotidine may increase the risk of severe COVID-19 disease after Propensity Score Matching [2]. However, Mather et al. believed that the matching of propensity scores to adjust for age differences between famotidine and non-famotidine patients did not change the significance of the differences in mortality or combined mortality and intubation in the analysis [4]. Similarly, Shoaibi et al. stated that with real-world data from a large multi-agency hospital database, no evidence has been found that famotidine used in hospitalized COVID-19 patients reduced the risk of death compared with non-users [5]. To better understand these contradictions, we analyzed the data of these articles through meta-analysis. Our work shows no significant statistical difference between use and non-use of famotidine reducing the risk of serious disease in COVID-19 patients after Propensity Score Matching. One more thing worth noting is, due to the suddenness of the epidemic, many studies cannot be performed ordinarily. This is susceptible to many complex factors. Therefore, more studies with Propensity Score Matching are needed in the future to further shed light on this problem.