Sex differences in hospitalized COVID-19 patients

4.1 Key findings

We confirmed that in our center we had fewer female patients admitted to the emergency department or being hospitalized for COVID-19 than male (37.8% vs. 62.2%) and that they suffered from a less severe disease, as shown by 25% ICU admission rate. ICU admission was needed for 9.1% of hospitalized males compared to 5% of hospitalized females. Interestingly, overall hospital mortality was lower than reported during the first wave. Indeed, in our previous first wave case series overall hospital mortality was 19.5%, while during this second wave overall hospital mortality was 15.4% (176/901 vs. 187/1216, p = 0.012), thus suggesting an improvement in medical management over time. Furthermore, among elderly patients, both ICU admission rate and mortality rate were revealed to be increased in males compared to females.

Notably we did not observe an increase of admission rate in females over time, in contrast with what we observed during the first wave [8]. These findings are coherent with those recorded at the national level: the Italian national statistics found an equal percentage of female COVID-19 cases, 48.6% and 51.5% respectively before and after 23rd October 2020, as shown in Fig. 2[10].

Fig. 2.Overall percentage of female COVID-19 patients in Italy increase over time (from 48.6% to 51.5%). The percentage of female COVID-19 patients decreased in our hospital from 41.3% in the first study period to 36.3% in the second study period. In the intensive care unit, the percentage of females was only 25%.

4.2 Relationship with previous studies

Worldwide, in October 2020 females accounted for 48% of confirmed cases, 46% of hospitalizations, 31% of ICU admissions and 42% of deaths [7].

Since the beginning of the pandemic, in several countries there was the need to collect sex-disaggregated data in order to evaluate the incidence, the mortality rate and the disease treatment according to the sex. Importantly, demographic and clinical data collected globally demonstrated that the rate of COVID-19, including asymptomatic and mild forms, is similar between the sexes [11]. Instead, in our data, which include only patients hospitalized and who had attended emergency room, the percentage of females was lower than that of males both in the first and the second wave. Therefore, it is possible to hypothesize that females are affected by a less precipitating form of disease than men, hence requiring hospitalization and intensive care less frequently than men.

Regarding the other past coronavirus infections, sex differences between infected patients had been reported. For instance, both the 2003 severe acute respiratory syndrome (SARS) and the Middle East respiratory syndrome (MERS) reported a male predominance and a lower case fatality rate in females when compared with males [12, 13].

The first reports of sex bias in COVID-19 patients came from China’s Hubei province, where the pandemic has started. A more recent metanalysis by Li et al. [14] showed that females represented 40% of the COVID-19 cases.

European data also highlighted a significant male predominance in COVID-19 morbidity and mortality. Hereof, one of the first reports was a retrospective case series from Grasselli et al. [15] who reported that 82% of 1591 COVID-19 patients admitted to ICU were males.

In the United States, public sex-disaggregated data on COVID-19 morbidity and mortality showed that men are twice more likely to die from COVID-19 when compared to women [16].

There is a male predominance in almost every country reporting sex-disaggregated data on COVID-19 outcomes. The results reported by these studies are in agreement with those obtained in our center during the second wave. Therefore, we confirmed that sex is a risk factor for a higher severity and mortality in patients with COVID-19. In accordance with our results, the sex-based differences in lethality and ICU admissions may suggest that women are less prone to develop severe forms which lead to death.

4.3 Significance and interpretation of study findings

The reasons for this sex-based tolerance are still unknown. The possible mechanisms involved in this sex bias include social, clinical, immunological, and genetic factors. Generally, several diseases are known to have a higher incidence in men, namely cardiovascular diseases, diabetes, chronic pulmonary disease, hypertension and cancer [17], which were all found to be linked to a higher case fatality rate for COVID-19 [18].

There is a significant sex bias in the immune response to viral infections. This bias is considered the result of male-female differences in sex chromosomes and sex hormone milieu [11]. In general, female immune system responds to pathogens more efficiently, producing higher amounts of interferon (INF) and antibodies. This effect is primarily mediated by estrogens and attenuated in postmenopausal women [19].

An unchecked inflammatory response and the subsequent cytokine storm is the main cause of severe COVID-19, which damages the lung tissue [20]. A more recent study [21] documented sex differences in immune response during the disease course of COVID-19. Indeed, females have a more robust CD4 and CD8 T cell activation than males during SARS-CoV-2 infection. By contrast, males have higher plasma levels of innate immunity chemokines and cytokines, such as interleukin 8 (IL-8), interleukin 18 (IL-18) and C-C motif chemokine ligand 5 (CCL5), and a poor T cell response. In addition, this poor T cell response was negatively correlated with patients’ age, and it was associated with worse disease outcomes in male patients than in female patients.

Emerging evidence suggests that type I IFNs may reduce the virus replication [22]. Notably, a poor INF-I release, associated with an increased production of inflammatory genes, such as interleukin 6 (IL-6), C-C motif chemokine ligand 2 (CCL2), and C-C motif chemokine ligand 8 (CCL8), was observed in response to coronavirus infection [23]. Therefore, the lack of an adequate INF response may represent a deficit in the ability of the immune system to fight COVID-19 infection and prevent severe illness [24].

Toll-like receptor 7 (TLR7) was identified as an important receptor for recognizing and responding to coronavirus infections. In addition, this receptor leads to the production of IFN. Notably, TLR7 seems to be more intensively expressed in females. Consequently, females are more likely to express a higher number of TLR7 which improves the likehood of generating a strong INF response, limiting virus evasion, and accelerating viral clearance [25].

Moreover, high IL-6 level seems to be a good predictor of severity and death in patients with COVID-19 [26]. Interestingly, females express lower levels of IL-6 in response to viral infections [27].

X chromosome inactivation and female mosaicism are particularly relevant to the sex bias observed in COVID-19 outcomes because most of the immune regulatory genes are encoded by X chromosomes, e.g., TLR, costimulatory molecules, and transcription factors. As reported by Raza HA et al. [28], there are some Y-linked genes, such as transmembrane protease serine 2 (TMPRSS2), which role is to promote virus entry into the host cells and may highlight sex-based disparity in immune response associated with illness severity. Indeed, the cytokine storm might be influenced by genetic variability and mosaicism.

The sex difference in inflammatory response is largely driven by sex hormones. Physiological levels of estrogens in premenopausal women modulate the immune system, with an anti-inflammatory effect. Instead, low levels of estradiol and high levels of proinflammatory cytokines (IL-6, IL-8, and TNF-a) are reported in postmenopausal women [19, 29].

Firstly, SARS-CoV-2 enters in the pneumocytes via the angiotensin-converting enzyme type 2 (ACE2) receptor. It seems that females have a lower expression of ACE2 in the lungs which may possibly reduce their susceptibility to this coronavirus [30]. Interestingly, Sward et al. [31] have reported that females have lower levels of soluble ACE2 when compared to age-matched males. Furthermore, the serum activity of ACE2 is higher in older women (age 55 or older) than in younger women [5]. It is possible to speculate that the increase in ACE2 activity with age might be related to changes in sex hormones, particularly estrogen.

Secondly, ACE2 binds to type 1 angiotensin receptor and activates the nuclear-factor $\kappa$B pathway, thus inducing vasoconstriction and inflammation [32]. Furthermore, activated estrogen receptor-$\alpha$ mitigates the inflammatory response induced by the nuclear-factor $\kappa$B pathway and the cytokine production via immune cells (neutrophils, lymphocytes and macrophages) [33].

Finally, SARS-CoV-2 uses TMPRSS2 and a disintegrin and metalloproteinase 17 (ADAM-17) to cleave viral and cellular proteins. Estrogens can also inhibit the TMPRSS2 mRNA translation, reducing its availability to drive the infection of SARS-CoV-2 [34].

4.4 Limitations of the study

Being a single-center study is the primary limitation. However, our findings are supported by numerous studies worldwide. Another limitation is the lack of confounding variables in the database that did not allow us to proceed with a multivariate analysis.

4.5 Future perspectives

Despite this limitation, this observational study stresses again the sex gap in COVID-19 patients and the importance of more adequate treatments empowering precision medicine. Moreover, future studies may be necessary to investigate sex-based differences in immune response to COVID-19 vaccine.