Routine statin treatment is significantly associated with increased mortality in type 2 diabetes patients hospitalized for coronavirus disease 2019 (COVID-19), according to a study published in Diabetes & Metabolism.
The findings came from the Coronavirus–SARS-CoV-2 and Diabetes Outcomes (CORONADO) study – a nationwide observational study aiming to describe the phenotypic characteristics and prognosis of type 2 diabetes patients with COVID-19 admitted to 68 French hospitals between 10 March and 10 April 2020. Out of the entire CORONADO study population, 2449 patients with both type 2 diabetes and COVID-19 and with available information on routine statin use were included in the analysis.
Of these patients, 1192 (48.7%) had been treated with statins before hospitalization. Compared with statin non-users, patients taking statins were more likely to be male (67.8% vs 60.5%). Additionally, statin users were more likely to have comorbidities, including hypertension (86.6% vs 74.1%), macrovascular (53.8% vs 26.6%) and microvascular (49.5% vs 41.1%) diabetes complications, heart failure (14.2% vs 9.9%) and treated obstructive sleep apnoea (14.5% vs 8.1%). Meanwhile, estimated glomerular filtration rate (eGFR) was slightly lower in statin users (66.2 vs 70.0 mL/min) compared to non-statin users. On the other hand, long-term glycaemic control, admission plasma glucose concentrations and inflammatory biological parameters on admission such as C-reactive protein levels, lymphocyte count did not differ between statin users and non-users.
In unadjusted analyses, study data showed that patients using statins had similar rates of composite primary outcome, which comprised tracheal intubation and/or death within 7 and 28 days of admission, compared to non-users within both 7 (29.8% vs 27.0%, respectively; P = 0.1338) and 28 days (36.2% vs 33.8%, respectively; P = 0.2191) of admission. However, mortality rates were significantly higher in statin users within 7 (12.8% vs 9.8%, respectively; P = 0.02) and 28 days (23.9% vs 18.2%, respectively; P < 0.001).
Further, the researchers estimated the association between statin use and outcomes by logistic regression analysis after applying inverse probability of treatment weighting (IPTW) using a propensity score-weighting approach. Significant associations were observed with statin use and the primary outcome within 7 days (odds ratio [OR], 1.38; 95% confidence interval [CI], 1.04–1.83) and with death within 7 (OR, 1.74; 95% CI, 1.13–2.65) and 28 days (OR, 1.46; 95% CI, 1.08–1.95). In contrast, routine statin use was found to be not significantly associated with any increased risk of tracheal intubation for mechanical ventilation. Similar results were obtained in sensitivity analyses performed after additional adjustments for HbA1c and eGFR values.
“These results are somewhat surprising as previous observational studies of the general population had highlighted the potentially beneficial effects of statins on COVID-19 prognosis, including its in-hospital use,” wrote Bertrand Cariou, Université de Nantes, Nantes, France, and colleagues. “One potential explanation for this observed discrepancy with our present study is that CORONADO was focused only on patients with [type 2 diabetes] hospitalized for COVID-19 and was therefore associated with more severe prognosis … The possibility that the effect of statins on COVID-19 prognosis varies according to the stage and severity of disease cannot be excluded. In accordance with this speculation, the mortality [hazard ratio] was also lower at 0.4% in non-statin users in [a study in England], with cardiorenal morbidity affecting only 21.7% of the studied population, a considerably lower rate than in CORONADO, where 42% of participants exhibited cardiovascular complications.”
“Albeit observational, our present results do not support the hypothesis of a protective role of routine statin use against COVID-19, at least not in hospitalized patients with [type 2 diabetes]. Indeed, the potentially deleterious effects of routine statin treatment on COVID-19-related mortality demands further investigation and, as recently highlighted, only appropriately designed and powered randomized controlled trials will be able to properly address this important issue,” the authors concluded.
The authors acknowledged that there remained a possibility that some residual confounding factors were still persistent in the propensity score analysis even though many covariates were captured. In addition, they said there was a lack of information regarding the continuation (or not) of statin treatment after hospital admission.