Statin Use and In-hospital Mortality in Patients with COVID-19 and Coronary Heart Disease

The worsening progress of coronavirus disease 2019 (COVID-19) is attributed to the proinflammatory state, leading to increased mortality. Statin works with its anti-inflammatory effects and may attenuate the worsening of COVID-19. COVID-19 patients were retrospectively enrolled from two academic hospitals in Wuhan, China, from 01/26/2020 to 03/26/2020. Adjusted in-hospital mortality was compared between the statin and the non-statin group by CHD status using multivariable Cox regression model after propensity score matching. Our study included 3133 COVID-19 patients (median age: 62y, female: 49.8%), and 404 (12.9%) received statin. Compared with the non-statin group, the statin group was older, more likely to have comorbidities but with a lower level of inflammatory markers. The Statin group also had a lower adjusted mortality risk (6.44% vs. 10.88%; adjusted hazard ratio [HR] 0.47; 95% CI, 0.29–0.77). Subgroup analysis of CHD patients showed a similar result. Propensity score matching showed an overall 87% (HR, 0.13; 95% CI, 0.05–0.36) lower risk of in-hospital mortality for statin users than nonusers. Such survival benefit of statin was obvious both among CHD and non-CHD patients (HR = 0.30 [0.09–0.98]; HR = 0.23 [0.1–0.49], respectively). Statin use was associated with reduced in-hospital mortality in COVID-19. The benefit of statin was both prominent among CHD and non-CHD patients. These findings may further reemphasize the continuation of statins in patients with CHD during the COVID-19 era.

Statistical analysis. Continuous variables were expressed as the median and interquartile range (IQR), and categorical variables were expressed as number and percentage (%). Statistical differences between the two groups were analyzed by the Mann-Whitney U test for continuous variables, while categorical variables were compared by Fisher's exact test or χ2 test. The risk of in-hospital mortality and the corresponding hazards ratio (HR) were calculated using the multivariable Cox regression model to compare the statin group versus the nonstatin group. A full model included demographics, clinical and laboratory variables independently associated with statin use or in-hospital mortality at a P < 0.05. The final multivariable adjustment including statin use (yes or no), age, sex, history of comorbidities, respiratory rate, severity classification, Lab results including serum B-type natriuretic peptide (BNP), Troponin I (TNI), aspartate aminotransferase (AST), Alanine transaminase (ALT), Creatinine (Cr), K, Na, Ca, glucose, Albumin, Lactate dehydrogenase (LDH), White blood cell (WBC), Lymphocyte, Neutrocyte, Platelet, CRP (C Reactive Protein), Procalcitonin (PCT), Interleukin 6 (IL-6), Interleukin 2 (IL-2), Tumor necrosis factor-α (TNF-α), activated partial thromboplastin time (APTT), international normalized ratio (INR), D-Dimer and Total Cholesterol (TC), vasoactive drugs, ARB and beta-blocker (BB) use during hospitalization. A two-side α less than 0.05 was considered statistically different. Data were analyzed in R-3.6.3 (R Foundation for Statistical Computing, Vienna, Austria) and SPSS Statistics (version 22.0, IBM, Armonk, NY, USA).
Propensity score-matched analysis. To minimize baseline differences between statin and non-statin groups, we performed propensity score-matched analysis (PSM). Baseline matching variables included age, sex, history of comorbidities, Na, Albumin, Glucose, Hemoglobin, Lymphocyte, Platelet, CRP, IL-6, TC, serum troponin I (TNI) levels, severity classification, use of vasoactive medicine, antiplatelet therapy, ACEI, ARB, betablocker, and CCB. To further balance the baseline difference between statin and non-statin groups in subgroups of the CHD and non-CHD populations, we performed PSM again in the CHD and non-CHD groups, respectively. Residual imbalance (p < 0.05) between statin and the non-statin group was further adjusted in the Cox regression model.
Statin users and nonusers were paired at 1:1 according to the propensity scores using exact matching with a caliper size of 0.05. The balance of covariates was evaluated by p-value before and after matching, and p-value > = 0.05 was considered qualified balancing between the two groups. For subgroup study of CHD or non-CHD population, the pair ratio was also 1:1 between statin and non-statin group, with the caliper size as 0.05.

Statin Use and In-Hospital Mortality
Among all admitted patients, 12.9% (n = 404) received a statin. Those on statin therapy were older (67 ± 15 vs. 61 ± 20 years; p < 0.001) and had more likely to have comorbidities such as CHD, diabetes, cerebrovascular disease, dyslipidemia, hypertension, heart failure, and atrial fibrillation. (All p < 0.001) Despite a higher comorbidity burden, we observed lower inflammatory markers such as at presentation a lower IL-2 (523. 5 (Table 1).
In the logistic regression model, after adjusting for age, sex, comorbidities, lab results on admission, including inflammatory markers, in-hospital medications, and severity classification, the use of statin was associated with lower all-cause mortality (adjusted HR, 0.47; 95% CI, 0.29-0.77) versus non-statin group. The results remained consistent and statistically significant with a propensity score matching analysis, demonstrating a lower risk of in-hospital mortality in the statin group (adjusted HR, 0.13; 95% CI [0.05-0.36]). (Table 2).

CHD and Statin Use
Among patients with CHD, 125 (50.4%) received, and 123 (49.6%) did not receive a statin. Those on a statin were at a similar age, with the similar possibility to have comorbidities such as hypertension, dyslipidemia, heart failure, diabetes, and pulmonary disease, as well as a similar COVID-19 severity classification. However, patients on statin were more likely to have concomitant use of aspirin, clopidogrel, beta-blocker, ARB, and CCB (all p < 0.05). As for in-hospital treatment, those on statin were more likely to have concomitant use of aspirin, clopidogrel, beta-blocker, ARB, and CCB, but less likely to use vasoactive drugs (12 [ (Table 3) 4. Statin Use and In-Hospital Mortality in CHD patients.

Discussion
In this retrospective study, statin use was associated with a lower risk of all-cause in-hospital mortality than the non-statin group among COVID-19 patients. The benefit was significant both among CHD patients and non-CHD patients. Although unmeasured confounding and indication bias may have contributed to the observed protective association, these data showed that statin use was associated with survival benefits in COVID-19. These findings provide additional clinical evidence supporting continuous statin use in CHD patients with COVID-19, both in CHD and non-CHD patients. Given this study's retrospective nature, these data need further validation in randomized controlled trials to determine the efficacy of statin use in patients and COVID-19.
Due to the conflicting considerations of statin use during COVID-19, several studies examined statin use during COVID-19. A study in China showed a survival benefit in statin users vs. non-statin users 17 . A most recent study examined diabetes patients with COVID-19 and showed that statin use was associated with reduced in-hospital mortality from COVID-19 in diabetes mellitus patients 3 . Such benefits did not show in non-diabetic patients. In diabetes or hyperglycemic status, virus entrance and replication are intensified. COVID-19 leads to inflammatory/oxidative stress and exaggerate cytokine storm, which is particularly enhanced in diabetes 12-14 . In COVID-19, serum concentrations of proinflammatory cytokines, including IL-2, IL-6, and tumor necrosis factor (TNF)-a, were markedly increased in severe cases than moderate cases, suggesting that cytokine storms www.nature.com/scientificreports/ could be associated with disease severity. Statin was known for its anti-inflammatory quality. Diabetes and cardiovascular disease were two major comorbidities that increased the mortality of COVID-19 2,18,19 , yet there is no study examining statin use in CHD patients. Also, our study examined the change of inflammatory marker levels at presentation, seven days post-admission, and their peak levels both in statin and non-statin groups among CHD patients. CHD is one of the most prevalent comorbidities in COVID-19 patients. Our COVID-19 population has a CHD prevalence of 7.9%, which echoes the previous COVID-19 report in Wuhan, and was lower than US studies 6 . The definition of CHD in our study was those who had a medical record shown CHD as one of the discharge diagnoses. In such case, CHD included all patients with coronary heart disease, such as those who had a history of CHD, treated with PCI, CAGB, or receiving medical therapy, as long as clinicians have once diagnosed it. The individual case fatality rate of COVID-19 patients with CHD was 10.5%, highest among those with any comorbidities, including chronic respiratory disease (6.3%), cancer (5.6%), diabetes (7.3%), and hypertension (6.0%) 20 . CHD predispose patients with proinflammatory and procoagulant status, together with COVID-19, increased the risk of thrombus formation, severe inflammatory response, and endothelial damage, leading to ACS, heart failure, cardiac arrhythmia, and sudden death [21][22][23] . COVID-19 can also directly trigger ACS by plaque rupture, coronary spasm, or microthrombi due to systemic inflammation or cytokine storm 24 . Meanwhile, the most common comorbidities associated with CHD, such as hypertension increased ACE2 receptor and facilitated   www.nature.com/scientificreports/ virus entry, diabetes enhanced cytokine storm, and leads to multi-organ failure 10,11,25 . Moreover, COVID-19 was associated with hypercoagulation status and could cause increased intracoronary thrombosis and worsen STEMI outcomes 26 . Therefore, COVID-19 patients with CHD were four times more likely to die than average COVID-19 patients, with an elevated level of proinflammatory cytokines such as IL-6, CRP, and procoagulant factors such as D-Dimer 20 .
Statin is known for its anti-inflammatory and anticoagulant effects 27 . SARS-CoV-2 is targeting the endothelium, one of the largest organs in the human body. Notably, endothelial cells express all the co-factors necessary for the internalization of SARS-CoV-2 in human host cells 28,29 . Moreover, the most common comorbidities in COVID-19 such as hypertension and diabetes were generally predisposed with endothelial dysfunction, which promotes virus entry, replication, clotting cascade, microvascular obstruction and multi-organ failure 25 . Dysregulated RAS system in CHD patients also increases ACE2 expression, results in pre-existing endothelial dysfunction which facilitates virus entry, finally causees subsequent cardiac damages [29][30][31] . On the other hand, SARS-CoV-2 per se induces cytokine storm, which further exacerbates endothelial injury and dysfunction, ultimately leading to microvascular thrombi, pro-thrombotic state, and cardiovascular ischemia status 25,29 . Expert in bioengineering and nanomedicine have tried to develop numerous innovative pharmaceutical approaches targeting endothelial dysfunction to tackle this problem 28 . Statin is one of the most promising drug classes for   www.nature.com/scientificreports/ treating endothelial dysfunction and preventing vascular damage in COVID-19. Similar to RAS inhibitors, statins improve endothelial function in patients with or at risk for cardiovascular disease. They improve endothelial function via reduction of LDL and suppression of pro-oxidant enzymes and other proinflammatory transcriptional and signal transduction pathways 32 . Previous studies showed that statin protected diabetic COVID-19 patients from in-hospital mortality by anti-inflammatory properties and antiplatelet effect 3 . Its acute administration improves survival in proinflammatory states such as myocardial infarction. Our study shows that statin use in CHD patients vastly improved in-hospital morality, with reduced IL-2, IL-6 level, and D-Dimer levels during the hospital stay.Similarly, such benefits are also seen in non-CHD patients and more critically ill patients. To avoid an unbalanced baseline, we used propensity score match to retest our result and yield a similar conclusion of statin benefit. However, due to this study's retrospective nature, a prospective design is warranted to prove the conclusion further.

Limitation
Our study should be viewed considering several limitations inherent in retrospective analyses. First, although we had detailed information on medical history, our observational study cannot exclude the potential impact of unmeasured confounders such as social status, the duration of statin use, and the reason for statin use. Second, we did not examine pre-hospital statin use. Such information was missed in the electronic data collection system because of the urgent COVID-19 time. However, we defined statin use as patients received statin within 3-day post-hospitalization. We choose this definition to include the statin group was patients who were regular statin users. Also, it has not been our routine practice for clinicians to initiate new statin therapy in patients with acute COVID-19 illness. In addition, clinical scenarios leading to statin cessation such as liver enzyme elevation were not significantly different between the statin and non-statin groups. Meanwhile, our data showed only two patients who initiated statin after three-day post-hospitalization. Both of them were fully recovered and discharged. As a result, excluding the two late-use statin cases may have underestimated the benefit of statin treatment on COVID-19 patients but would not change the direction of our results. Third, due to all patients were hospitalized with COVID-19 in Wuhan, generalizability may be limited. Finally, due to the observational study's inherent limitation, our study did not examine the causality between statin use and reduced in-hospital mortality, which should be tested only by a randomized control trial.

Conclusion
Among patients hospitalized with COVID-19, statin use was associated with reduced in-hospital mortality from COVID-19. The benefit of statin was both prominent among patients with CHD and non-CHD patients. These findings may further reemphasize the continuation of statins to patients with CHD during the COVID-19 era.
Ethical approval. All procedures performed in studies involving human participants were following the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent. This study was approved by the Ethics Commission of Tongji Hospital. Written informed consent was waived for the retrospective study by the Ethical Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology.
Consent for publication. All authors have also reviewed and approved the manuscript.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.