Epidemiological characteristics of 1212 COVID-19 patients in Henan, China

Pei Wang; Junan Lu; Yanyu Jin; Mengfan Zhu; Lingling Wang; Shunjie Chen

is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/2020.02.21.20026112 doi: medRxiv preprint ( Fig.2 B) . We guess that there are three possible reasons that may result to such gender difference. Firstly, males may be more active and with wider social activities than females, which increases their risk to infect COVID-19. Secondly and may be the most reasonable, an existing medical investigation reported that the expression and distribution of angiotension converting enzyme 2 (ACE2) was wider in male patients than those in females [5] . Similarly to SARS-CoV, it is reported that COVID-19 invades human body through the ACE2 receptor. Based on single cell RNA sequencing technology, researchers have investigated the expression profiles of ACE2 in two male and six female patients with COVID-19 from single cell resolution, and they found that the expression of ACE2 was correlated with gender. The proportion of ACE2 expressed cells is higher in males than in females (1.66% vs. 0.41%) [5] . Besides, the distribution of ACE2 was also wider in males than that in females. Zhao et al. [5] reported that there were at least five different types of cells in the lungs of males with the expression of the ACE2 receptors, while the number was about 2 ∼ 4 in female patients. This may be one of the deep reasons that why male patients were higher than the female ones. Thirdly, one may doubt that the distribution difference may be caused by population structure. To verify whether the male to female ratio (MFR) affects the finding, based on the statistical yearbook report of Henan province in the year 2019, the MFR in the whole province was 1.0126 : 1, which has great difference with the male to female patients ratio (MFPR) (637 : 521 = 1.2226 : 1). Moreover, we perform correlation analysis for the two ratios in the 18 regions, the two ratios show very weak correlation, with Pearson correlation coefficient r = 0.1555 ( Fig.2 C) . Kolmogorov-Smirnov (KS) test [31] on the MFR and MFPR in the 18 regions also shows that the two have significant difference (KS = 0.6111, p = 0.0018 < 0.05). Thus, we speculate that the MFR has no apparent effect on the MFPR. In conclusion, the former two reasons may be the driven force that there was gender difference in confirmed patients.