A new publication led by CHÉOS Statistician Dr. Terry Lee, Dr. Jim Russell at the Centre for Heart Lung Innovation, and UBC medical student Alessandro Cau has shown that a widely available medication may significantly reduce death from COVID-19 in hospitalized patients.
Angiotensin receptor blockers (ARBs) or angiotensin converting enzyme inhibitors (ACE inhibitors) are commonly used to treat high blood pressure by preventing excessive build-up of the hormone angiotensin II. COVID-19 is thought to cause illness and serious symptoms by triggering high levels of this same hormone, making ARBs and ACE inhibitors viable treatment options for COVID-19.
Despite the physiological plausibility of these treatments, research into their safety and effectiveness in COVID-19 has produced conflicting results; even meta-analyses, which combine results from multiple different studies, have not found consistent conclusions about these drugs.
But, Dr. Lee explains, these inconsistent results may be due to methodological differences and improper use of statistical analysis.
“This inaccuracy seems to be, in part, due to a failure to adjust for indication bias,” says Dr. Lee.
Indication bias refers to a situation where the severity of an underlying illness differs between people who are on a medication and those who are not. If the group on a medication is sicker to begin with, it makes the comparison with the non-medicated group unsound. In the case of ARBs/ACE inhibitors and COVID-19, if researchers don’t take into account that people on these drugs are often sicker, with conditions like high blood pressure, chronic kidney disease, cardiovascular disease, or diabetes, it can hide the protective effect of the medications.
While many studies in this area did control for this type of bias, others did not, and previously published meta-analyses of these studies didn’t effectively combine the data to produce accurate results.
“Ours is the first meta-analysis on this topic that combines these results in a methodologically sound way,” explains Dr. Lee, who led the analysis as part of his involvement in ARBs CORONA, a set of studies of these medications led by Dr. Russell.
The research team, which includes CHÉOS Clinical Trials Program Head Dr. Joel Singer and Scientist Dr. Karen Tran, used a statistical technique called meta-regression to combine the results of 30 previously published analyses and account for pre-existing conditions, among other factors.
The analysis showed that the odds of dying from COVID-19 was lower in hospitalized patients with chronic use of ARBs or ACE inhibitors compared with those who were not on these medications. But when the same data were analyzed without accounting for indication bias, the results were quite different, seemingly indicating the drugs didn’t make a difference. This highlights the importance of sound statistical analysis in this area.
Importantly, says Lee, this finding demonstrates only an association between these medications and better outcomes — the only way to see if the drugs are actually directly lowering the odds of dying from COVID-19 is through randomized controlled trials.
And the research team is doing just that.
The ARBs CORONA II trial is looking at the effect of losartan (an ARB) on mortality in people with COVID-19. The study is currently ongoing, with worldwide sites contributing data. This trial is one of several studies led by Dr. Russell on ARBs and ACE inhibitors in COVID-19. CHÉOS researchers and staff are providing statistical and data management support.
This publication appears in CJC Open, a journal of the Canadian Cardiovascular Society. CHÉOS Scientists Drs. Joel Singer and Karen Tran are co-authors on the study.
