Discussion
In this study, which was carried out in the first wave of COVID-19, the effects of demographic data and drugs utilized in the last 6 months before contracting COVID-19 on the probability of hospitalization, ICU admission or mortality were examined. Increasing age was associated with all three main endpoints of the study, which were listed as hospitalization, ICU admission, and mortality. In addition, the presence of COVID-19 thoracic CT findings and polypharmacy was associated with an increased probability of hospitalization and being alone and the presence of COVID-19 thoracic CT findings were associated with an increased probability of ICU admission. When the effects of drugs used in the last 6 months before contracting COVID-19 were examined, it was seen that DM medications increased the probability of hospitalization, and CCBs increased the probability of ICU admission and mortality.
Increasing age and the presence of comorbidity have been shown to be an important risk factor for the severe clinical course of COVID-19 since the beginning of the COVID-19 pandemic [1,2,4]. We also obtained similar results in our study; HT, DM, hyperlipidemia, asthma/COPD, cardiovascular, kidney and cerebrovascular diseases were associated with the clinical course of COVID-19. It was observed that HT increased the probability of ICU admission (Table 5), while CVD increased the probability of mortality 73 times (Table 6).
Numerous studies have been conducted since the beginning of the pandemic that ACEI/ARBs are associated with COVID-19 in harmful or protective ways. According to data from most large-scale studies and meta-analyses, RAS blockers do not change the clinical course of COVID-19 [7,19,20]. Similar findings were obtained in our study; it was observed that ACEI and ARBs had no effect on the clinical course of COVID-19.
In our study, it was also observed that DM medications increased the probability of hospitalization three fold (Table 4). However, it had no effect on ICU admission and mortality. In the logistic regression analysis when DM medications were divided into subgroups as metformin, sulfonylureas, meglitinides, thiazolidinediones, α-glucosidase inhibitors, GLP-1 agonists, DPP4 inhibitors, SGLT-2 inhibitors and insulin, no significant difference was observed between the drug groups in terms of their effects on hospitalization. However, chi-square analysis revealed that all DM drug subgroups were associated with hospitalization, and SGLT-2 inhibitors were also associated with mortality and admission in the ICU. In the literature, it is stated that the use of metformin is generally associated with a decrease, while the use of insulin is associated with an increase in mortality due to COVID-19 [11,12,21]. Regarding other DM drug groups, there are publications that show different results from each other. In a study involving 2.85 million patients with type 2 DM in the UK, statistical evidence was presented that patients receiving metformin, SGLT2 inhibitors and sulfonylurea treatments had a lower risk of death than those who did not take these drugs [11]. In the same study, it was determined that the risk of death was higher in those whose given insulin and DPP-4 inhibitors than those who did not [11].
According to the results of Yang et al.’s meta-analysis, which included 17 studies and analyzed data from a total of approximately 21.000 patients, metformin was associated with significantly reduced mortality in COVID-19 patients with DM [12]. Regarding insulin therapy; in Yu et al.’s study, in addition to matching other medical characteristics, propensity score matching was applied to HbA1c levels, and insulin therapy was associated with increased mortality. It has been stated that the use of insulin therapy at a more advanced stage in type 2 DM may be a residual confounding factor, although pairings have been made [21]. Although age and comorbidity were added to the logistic regression models as confounding factors when analyzing the effects of chronically used drugs on hospitalization, ICU admission, and mortality, we could not make a match based on the clinical level of DM in the case of DM drugs. In our study, which included 525 COVID-19 patients in total, 64 of these patients were diagnosed with DM and 64 were receiving DM medication. When divided DM medications into subgroups, there was no difference between the subgroups in terms of their effects on the three main endpoints of this study, which may due to the small patient groups. According to two meta-analyses examining the relationship between DPP4-inhibitors and mortality due to COVID-19, contrary to the results of Khunti et al.’s study on DPP4-inhibitors [11], Rakhmat et al. stated in their study that the use of DPP-4 inhibitors was associated with lower mortality in COVID-19 patients [15]. In the study by Hariyanto et al., they stated that DPP-4 inhibitors did not change the results of COVID-19 [16].
According to the results of our study, CCB use is closely related to ICU admission and increased mortality. There were studies showing parallel findings with our study regarding CCBs [22,23]. In the study of Mendez et al., 245 patients hospitalized due to COVID-19 and diagnosed with HT were included; the data of 75 patients using CCBs and 175 patients not using CCBs were compared. Those who received dihydropyridine group CCBs had a significantly increased risk of intubation or death compared to those who did not [22]. In the study conducted by Jackson et al., which included 297 patients, the need for mechanical ventilation and mortality were investigated in patients hospitalized with the diagnosis of COVID-19. ARB or CCB use before hospitalization for COVID-19 was found to double the probability of death (aORs, 2.02 [95% CI, 1.03-3.96] and 1.91 [95% CI, 1.03-3, 55]) [23]. According to a population case-control study examining the development of symptoms and contracting COVID-19 with chronically used drugs, the risk of developing COVID-19 symptoms in people with HT who received CCBs was significantly increased (OR) = 1.73; 95% CI 1.2–2.3), disease risk was significantly lower in ARB and diuretic users (OR = 0.22; 95% CI 0.15–0.30, and OR = 0.30; 95% CI, respectively). 0.19-0.58) [24]. On the other hand, studies indicating that CCBs reduce the mortality of COVID-19 or the possibility of serious illness have also been published [25,26]. In a study by Chouchana et al., which included 3686 patients with HT hospitalized for COVID-19, demonstrated that CCBs reduced the probability of mortality [25]. In a meta-analysis, it was stated that there was a significant reduction in all-cause mortality and disease severity in CCB users [26].
Considering why CCBs may worsen the COVID-19 clinic, as in our study; CCBs can inhibit type II pneumocyte secretion, leading to alveolar collapse [27]. In addition, precapillary vasodilation due to CCBs may cause alveolar edema [28–31]. Another reason may be that CCBs may cause hypoxic pulmonary vasoconstriction in patients with pulmonary disease, leading to profound hypoxemia [22,32–35]. On the other hand, when we evaluate why CCBs may improve the clinical course of COVID-19; since calcium is necessary for virus penetration into the cell, viral gene expression, processing of viral proteins, and viral maturation and extracellular release, CCBs cause depletion in intracellular calcemia, thus negatively interfering with the life cycle of the virus. CCBs can inhibit viral replication by reducing intracellular calcium levels [36,37]. On the other hand, in vitro studies suggest that CCBs can be used in therapy by reducing intracellular calcium levels, which provide the environment for virus entry [38]. Additionally, some studies have demonstrated the anti-inflammatory and anticoagulant effects of CCBs [37,39].
Thiazolidinediones, aspirin and famotidine, which are among the medications whose effects on the clinical course of COVID-19 were examined in previous studies, were included in the logistic regression models within their main groups (diuretics, anticoagulants, stomach drugs; respectively) in our study. However, they were found to have no effect on all three study endpoints. A meta-analysis of aspirin suggests that aspirin use is associated with a lower risk of death from COVID-19 [40]. In a case-control study, however, aspirin had no effect on the COVID-19 clinic [41]. Regarding famotidine, in the study of Yeramaneni et al., 7158 patients were included and it was stated that famotidine had no effect on mortality due to COVID-19 [42].
When we examined the effect of marital status on the clinical course of COVID-19, we found that being married reduced the possibility of ICU admission. In the literature, we could not find any other study examining the effect of marital status on the clinical course of COVID-19. In addition, there are publications investigating the relationship between being married or living alone with the frequency of anxiety and depression during the COVID-19 pandemic; in these studies, it was stated that the frequency of depression was higher in the COVID-19 pandemic in those who were single [43,44]. This may lead to suppression of the immune system and be associated with a worse clinical course of COVID-19. It needs to be confirmed by larger studies.
As far as we know, our study was the first study in Turkey investigating the relationship between drug utilization in chronic diseases and COVID-19 clinical course using primary care data. In the study by Senkal et al., 611 hospitalized COVID-19 patients were included in the study and they aimed to reveal the possibility of severe COVID-19 clinics in patients under ACEI or ARB treatment. It was concluded that chronic ACEI exposure was associated with a reduced likelihood of serious disease [45]. Since our study presented primary care data, we had the opportunity to compare the effects of chronically used drugs on the clinical course of COVID-19 between outpatients diagnosed with COVID-19 and patients who were admitted to the hospital, ICU or dead. It is known that some genetic and ethnic factors under investigation, especially ACE2 and TMPRSS2 expression, may predispose to COVID-19. Accordingly, it had been reported that some patients may experience the disease more severely or mildly [17,18,46–49]. In this context, studies conducted in different countries that reveal the factors affecting the clinical course of COVID-19 gain importance.
One of the limitations of our study is that it was not possible to exclude all confounders, since this study was retrospective. For example, since the antropometrics ​​of the patients were not known, the presence of overweight status was unknown if the obesity diagnosis was not stated in the patient records. Other than that, when the study population of 525 was subdivided, there was a small number of patients for comparison. Regarding CCBs; although the established logistic regression models have been adjusted for age and comorbidity, it should be considered that they are mostly used in the elderly hypertensive population.
To conclude, we have demonstrated that increasing age, HT, and CVD are associated with the severe clinical course of COVID-19; being married reduces the probability of ICU admission due to COVID-19. In addition, DM medications increased the probability of hospitalization three fold; while CCBs increased the probability of ICU admission 155 fold and the probability of mortality 295 fold. On the other hand, it was observed that RAS blockers (ACEIs and ARBs) did not affect the clinical course of COVID-19. In conclusion, larger cohort studies and meta-analyses as with ACEI and ARBs, are needed for CCBs and DM medications.