COVID-19, ACEI/ARBs, and Gastrointestinal Symptoms: The Jury Is Still Out on the Association

Dear Editors:

We read with great interest the article by Tan et al¹ investigating the association between the use of angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor II Blockers (ARBS), gastrointestinal (GI) involvement, and clinical outcome of coronavirus disease 2019 (COVID-19).¹ The authors concluded that ACEIs/ARBs treatment continuation was associated with a lower rate of GI manifestations (diarrhea, vomiting, nausea, and abdominal pain) and increased mortality. We tried to replicate their analyses in a similar cohort from a single tertiary center in Milan, Italy.

Our cohort included 325 consecutive patients with COVID-19 confirmed by reverse transcriptase polymerase chain reaction who were hospitalized between February 22 and March 30, 2020. The median patient age was 66 years (range, 24–93 years; interquartile range, 55–75), and 68.6% were males. A history of hypertension was reported in 51.3% (167/325) and coronary heart disease in 17.8% (58/325). At admission, a total of 114 patients (35.4%) were taking ACEIs/ARBs.

We observed no difference in terms of the severity of COVID-19 presentation between patients using ACEIs/ARBs or non-ACEIs/ARBs users: among the 2 groups, a similar percentage of patients were breathing in ambient air (34.2% vs 41.8%; P = .18), were receiving supplemental oxygen (50% vs 43.27%; P = .24) or needed mechanical ventilation (15.8% vs 14.9%; P = .83).

Despite our larger cohort, in univariable logistic regression analysis we could not find a statistically significant association between ACEIs/ARBs use and reduced GI involvement at admission (odds ratio [OR], 0.63; 95% confidence interval [CI], 0.37–1.08; P = .091). Considering diarrhea alone did not change the results appreciably (OR, 0.63; 95% CI, 0.36–1.11; P = .107). Similarly, we found no negative association between use of ACEIs/ARBs and liver injury, using the same cut-offs: aspartate aminotransferase of >40 UI/L, alanine aminotransferase of >40 UI/L, or total bilirubin of >20 mmol/L (OR, 0.94; 95% CI, 0.59–1.48; P = .782). We noted that the prevalence of diarrhea at admission in our cohort was higher than reported by Tan et al¹ (23.4% vs 12.0%), whereas prevalence of other GI symptoms was lower (6.5% vs 15.0%). We chose to restrict our analysis to GI symptoms present at admission only to minimize the confounding effect of other hospitalization-related causes of diarrhea, such as antibiotics or antiviral use.

The prognostic interpretation of these findings is not unequivocal and the interplay between sever acute respiratory syndrome coronavirus 2 (SARS-CoV-2) digestive involvement and overall clinical outcome remains unclear. In a previously published analysis of this same cohort, we found an association between GI symptoms and lower rate of clinical deterioration.² This finding is in contrast with a meta-analysis of Chinese studies, which concluded that patients with GI involvement tend to develop more severe COVID-19.³ Further prospective studies are needed to investigate the implications of SARS-CoV-2 digestive involvement.

In the secondary analysis, Tan et al found a protective effect of ACEIs/ARBs use on overall clinical outcome. We could not confirm this result in our cohort. Although in univariable logistic regression analysis the use of ACEIs or ARBs was associated with clinical deterioration, defined as death or intensive care admission (OR, 2.05; 95% CI, 1.28–3.28; P = .003), in multivariable analysis, after adjustment for potential confounding factors such as age, coronary artery disease, hypertension and diabetes mellitus, use of ACEIs/ARBs did not remain significantly associated with the outcome (adjusted OR, 0.99; 95% CI, 0.49–1.55; P = .975). Although a large retrospective Chinese study found a decreased risk of all-cause mortality among ACEI/ARBs users,⁴ the overall quality of evidence is still limited and often conflicting.⁵

Finally, Tan et al interpret their results according to the assumed protective effect of ACEIs/ARBs from endothelial damage, which could lead to less multiorgan involvement with milder GI manifestations and an overall more favorable outcome. This hypothesis lacks a strong biological background because the role of ACEIs/ARBs in the course of COVID-19 infection remains unclear. Moreover, it is in contrast with other studies reporting that multiorgan involvement such as GI or liver involvement is not associated with a more severe COVID-19 disease course.^{6 , 7} Consequently, the GI presentation of COVID-19 should not distract clinicians from giving patients the best level of medical care.

In conclusion, our findings are in contrast with those presented by Tan et al and suggest caution when interpreting clinical associations between outcome ad concomitant medications. Since ACEIs/ARBs are commonly prescribed in elderly and comorbid patients, any analysis of related outcomes must account for the potential confounders often found in this subset of patients.