Hypertension and sex related differences in mortality of COVID-19 infection: A systematic review and Meta-analysis

Methods: We have conducted a systematic review of published literatures that identifi ed the relationship between hypertension and mortality of COVID-19 infections. Nineteen articles were selected following structured inclusion and exclusion criteria for systematic review and analyses. A total of 21,684 hospital admitted COVID-19 patients were included in this review and metaanalysis from 19 studies. The studies covered the six months of the pandemic from December 2019 to May 2020.


Introduction
The Coronavirus Disease 2019 (COVID- 19) outbreak initially emerged in the early in December 2019 in Wuhan city, Hubei province, China [1]. It was declared a pandemic on 12 March 2020 by the World Health Organization (WHO) [2]. To date, this disease has spread to all ive continents and has caused a "Public Health Emergency of International Concern (PHEIC) since 30 January 2020" [3]. As of early August 2020, it has already infected more than 20 million people along with more than 0.7 million deaths. Among them, 99% of cases were mild, 72.50% cases which had an outcome and 5% of fatalities have occurred in 14.50 million known outcomes [4], with deaths rapidly emerging around the globe. Determining the potential factors related to increased mortalityin COVID-19 patients is required for competent resource allocation during the pandemic period, especially in the low-middle income countries.
The burden of chronic diseases has rapidly emerged and causes 41 million deaths each year around the world, in which cardiovascular diseases account for 17.9 million deaths [5]. Hypertension is one of the most severe medical conditions which increases the risk of cardiovascular diseases. Highest prevalence of this condition was observed among COVID-19 patients, followed by other comorbidities [6]. Globally, 1.13 billion people have hypertension and low-middle income countries account for 75% of the total hypertensive patients [7]. Hypertension also causes cognitive impairment in twothirds of elderly people [8,9]. In an in luenza pandemic, older men were more likely to have a severe outcome [10]. Previous studies have shown that virally infected advanced age men with other co-morbidities suffered from weak immune function, however uncertain, may increase susceptibility and mortality of COVID-19 infection [11][12][13]. Hence hypertension is an important contributor to COVID-19 mortality in relation to age and, and is consistent in different countries worldwide. We aimed to explore the relationship between hypertension disease and mortality due to COVID-19 infection and to assess the sex and age in luence on the relationship.

Eligibility criteria
We enrolled all the studies which included COVID-19 infection with hypertension disease, mortality and survival. The original research articles with sample size > 50 hospital admitted patients published in English language were included.

Search strategy and study selection
We carried out a comprehensive systematic peer-reviewed literature search by using PubMed, SCOPUS, EuroePMC and Google Scholar databases with the terms: "COVID-19" OR "SARS-CoV-2" AND ("Comorbidities" OR "hypertension"). We removed the replicas of the extracted articles. Two authors (AI and PKG) screened for relevant information in the abstracts of the remaining literature. Based on the eligibility criteria, we examined the full text of selected articles. We completed the article extraction on 31 July 2020followingthe guideline of "Preferred Reporting Items for Systematic Reviews and Metaanalysis (PRISMA)" [14] (Figure 1).
Quality assessment of the studies: Quality of the studies included for inal review was also assessed in terms of study size, sample selection methods and measurement of study outcomes. The studies were considered poor in quality if they have small sample size, non-random individuals and study outcomes were not measured objectively, moderate quality and better quality studies are those who used large sample, used randomization in patients section, and study outcomes were measured objectively.

Data extraction
Two independent authors (AI and PKG) extracted the 19 studies. We used standardized forms that included author, year, study design, sample size, age, sex, hypertension, diabetes and death of COVID-19 patients. The outcome of interest was mortality due to COVID-19 infection.

Statistical analysis
We performed Mantel-Haenszel formula to estimate the effect in risk ratio (RRs) with 95% con idence interval (95% CI) and performed Random-effects model to account for interstudy variability. All p-values in this study were two-tailed and statistical signi icance was set at ≤ 0.05. Meta-regression was performed to examine the in luence of covariates on the RR. We conducted further subgroup analysis to assess the more in luential group in the magnitude of RR based on median age and average proportion of hypertension in the studies. To evaluate the small-study effects, we performed a regressionbased Egger's test. We performed Funnel-plot analysis to identify the qualitative bias in the publication.

Characteristics of the included studies
While we extracted an initial 132 studies, there were no duplicate records and we excluded 109 studies after screening the abstracts. The remaining 23 studies were eligible for fulltext screening. We eliminated four studies because of no available data on sex, age, hypertension or diabetes. Finally, we included 19 articles for qualitative synthesis and metaanalysis ( Figure 1) ( Table 1).

Publication bias
We found a qualitatively symmetrical association with mortality and other factors (male and hypertension) in the funnel-plot analysis ( Figure 5). A further Regression-based Egger's test showed that there was no effect of a small study on the association with mortality and other factors (male: p -value = 0.138; hypertension: p -value = 0.290).
patients was 58 years, and the proportion of male patients was 58.8%. In contrast, we estimated 33.26% of hypertensive and 19.16% of diabetes mellitus patients in the studies.

Association between male and mortality of COVID-19 outcome
The meta-analysis showed that male patients had a higher mortality due to COVID-19 infection compared to that with female patients ("RR = 1.28, [95% CI: 1.20-1.36]; I 2 = 58.7%, p -value = 0.001") ( Figure 2A). In the leave-one-out sensitivity analysis, we observed that elimination of the highest weight (Giacomo, et al.) study could not change the heterogeneity and the magnitude of association ("RR = 1.31, [95% CI: 1.21 -1.42]; I 2 = 60.2%, p -value = 0.001") ( Figure 3A). The association between being male and mortality was not affected by age (p -value = 0.199) in the meta-regression.

Association between hypertension and mortality of COVID-19 outcome
Hypertensive patients with COVID-19 infection were found to be at higher risk of mortality compared to non-hypertensive patients ("RR = 1.45, [95% CI: 1.35 -1.55]; I 2 = 77.1%, studies from China [22,23]. The mean score of all comorbidities in men (3.1) was signi icantly higher than women (1.8) (p -value < 0.05) [24] while the prevalence of hypertension is 25% in men and 20% in women worldwide [7]. However, the COVID-19 infected male patients had more risk of dying than the female patients probably due to weaker immune function in males with existence chronic condition [25][26][27][28].
The hypertensive patients were more likely to have a fatal COVID-19 infection compared to non-hypertensive patients in the present study. It is consistent with the prior studies which illustrated that hypertension causes death among COVID-19 patients [6,15,16]. However, the hypertensive patients were more adversely affected by SARS-CoV-2 virus that enters the epithelial cells passing through angiotensin-converting enzyme 2(ACE2). ACE2 decreases vascular in lammatory action of angiotensin 1-7 [29,30]. In lammatory storm in infected patients releases cytokines causes systemic immune function damage, which may be a signi icant cause of multiple organ failure and even death [31].

Discussion
In this systematic review and meta-analysis study, we observed that hypertension were associated with the increased risk of mortality in COVID-19 infections. Metaregression showed that the association between hypertension and mortality was in luenced by sex, which was addressed by the previous severe COVID-19 outcome studies [15]. Further subgroup analysis showed a signi icantly stronger association in the studies with a higher proportion of male and elderly patients, which is steady with severe and fatal patients in the prior analysis [16].
Although the association between sex and mortality was weaker in the ive studies [17][18][19][20][21], meta-analysis shows that the male COVID-19 infected patients were 1.28 times more likely to die. Even though the proportion of male patients was higher (58.8%) than the female patients in the present studies, a male patient had increased risk of dying independent of age. We found similar results with severely infected patients in  Subgroup analysis showed that the increased mortality with COVID-19 disease was more signi icant in a higher proportion of male > 55 years with hypertensive patient studies. However, sex and age were more explained in the association between mortality of COVID-19 outcome and hypertension because circulating protein expression of ACE2 is higher in older men with hypertensive condition [8,32].
This study has several limitations. Since angiotensinconverting enzyme inhibitors (ACEIs) and angiotensin II type 1 receptor blockers (ARBs) treatment are used in COVID-19 patients with hypertension, it may affect the outcome. Data on hypertensive medicine or therapy were not included in the selected studies. Almost all of the studies were retrospective and peer-reviewed, however, a few preprints before peer review were also included in the meta-analysis. Included studies were from Asia, Europe and North America, however, most of the studies were from China but from different hospitals, therefore, the sample might not represent COVID-19 patients globally.
In conclusion, hypertensive patients are strongly associated with higher odds of mortality due to COVID-19 infections which may account for the contradiction in the many studies. The effect of coexistence hypertension on the risk of mortality with COVID-19 patients was more in luential in elderly male patients. We should direct more attention to COVID-19 and hypertensive patients, especially in elderly men in the course of treatment to reduce mortality. Further, more comprehensive studies are needed to investigate more detailed information on the association between hypertension and death of COVID-19 patients, particularly in low-middle income countries.