Dr. Michael Fulks explores two studies on ECG abnormalities in low-risk populations, revealing surprising disparities in relative mortality risks.
Two fascinating (at least for life medical directors trying to rate screening ECG findings), freely available ECG mortality studies in low-risk populations were published this year in Jour. American Heart Assoc. and JAMA Internal Medicine using similar populations but showing rather different outcomes. The first in JAHA by Sung Ho Lee et al. looked at an employed insured Korean population receiving annual health exams including ECG, with average age of 39, looking at all-cause mortality over a median of almost 9 years. It provided hazard ratios (HR) for all-cause mortality associated with ECG abnormalities using the Minnesota coding and adjusting for age, sex, smoking and health parameters.
The second in JAMA Int. Med. by Ryuichiro Yagi et ai, looked at an apparently similar cohort of Japanese employed persons also having insurance who were provided annual exams with ECG. Average age was older at 47 and follow-up was shorter at a median of 5.5 years. ECGs were coded in a similar manner but outcome for this cohort was the combination of all-cause mortality plus cardiovascular hospital admission. The HRs relative to those with normal ECG was similar for mortality, for admission and for the combination suggesting the combo could be used as a surrogate for just mortality.
Prevalence of minor ECG abnormalities was similar (17 & 20%) with more major abnormalities in Korea (3.6% in Korea and 1.5% in Japan). The HR for each abnormality in the Korean study was fully adjusted for age, sex, history and physical findings. I presume (but not stated for those 2 tables in the article supplement.) that the HR for each ECG abnormality was similarly adjusted in the Japanese study. ECGs in both studies were machine read with physician review.
Given the apparent similarity of the cohorts, it was a surprise to see the relative risks for ECG abnormalities in the Korean study were much lower. Minor ECG abnormities together had a HR of <1 and major had an HR of 1.1 with RBBB at 1, LBBB at 1.3, and A fib at 1.5. In the Japanese study, minor ECG abnormalities together had an HR of 1.26 and major at 1.97 with RBBB at 1.2, LBBB at 2.3, and A fib at 6.5. In the Korean study, results seemed a bit lower than or on par with the ratings that might be found in a variety of life insurance manuals and Japanese results suggested risks considerably higher than existing ratings.
I had hoped these two studies would go a long way to solving the dilemma faced by medical directors trying to assign risk to ECG findings because of the paucity of similar studies in low-risk populations. Unfortunately, the disparity in results and the use of a composite endpoint (and uncertainty regarding level of adjustment for variables) in the Japanese study serve to increase rather than decrease uncertainty. Could the disparity be explained by the Korean cohort being younger (average ages 39 vs. 47 and both younger than those currently receiving screening ECGs in the U.S.) or different endpoints, adjustments, etc. or are there other differences between the two cohorts or other methodologic factors? Still, given the lack of available mortality data for screening ECG results, these two studies are worth looking at.
About the Author
Michael Fulks, MD, Consulting Medical Director, is board-certified in internal and insurance medicine. After leaving practice, he served as a medical director, creating or editing several underwriting manuals and preferred programs. More recently, Mike has consulted for CRL participating in its mortality research on laboratory test results, BP and build, and in the development of risk-scoring tools for laboratory and non-laboratory data.