Early electrocardiographic changes associated with anticancer chemotherapy in a Sub-Saharan African population: a prospective cohort study

Abstract

Cardiovascular disease (CVD) is a leading contributor to morbidity and mortality among people living with cancer. In sub-Saharan Africa, due to the overall cost of the treatment, resource limitations, and a lack of specialized cardio-oncologists, cardio-oncology care is limited. As such, access to advanced cardiac imaging and biomarkers is often limited, making the 12-lead electrocardiogram (ECG) a pragmatic tool for early detection of chemotherapy-associated electrophysiological abnormalities. We investigated early ECG abnormalities following chemotherapy in an African cohort. We conducted a five-month prospective cohort study at the Douala General Hospital, Cameroon. Adults (≥ 19 years) with cancer initiating first-ever chemotherapy were enrolled. A 12-lead ECG was obtained at baseline (pre-chemotherapy) and 48 h after each chemotherapy cycle. A “new ECG abnormality” was defined as an abnormality present post-cycle that was absent at baseline. We described the frequency and types of ECG abnormalities and examined factors associated with a new ECG abnormality after cycle 1 using multivariable logistic regression. A total of 102 participants (74 women, 72.5%) were included; mean age was 48.7 ± 13.5 years. At baseline, 71/102 participants (69.6%) had a normal ECG. During follow-up, 53/102 participants (51.9%) had at least one ECG abnormality after chemotherapy. The most frequent abnormalities were prolonged corrected QT interval 19/53 (35.8%), T-wave/repolarisation abnormalities 17/53 (32.0%), and sinus bradycardia or tachycardia 10/53 (18.8%). Atrial fibrillation occurred in 2/53 participants (3.7%). In univariable analysis, hypertension was associated with ECG abnormalities after chemotherapy (OR 4.03, 95% CI 1.22–13.28). In the cycle 1 multivariable model, no covariate was statistically significantly associated with a new ECG abnormality; hypertension showed a positive but non-significant association (aOR 2.96, 95% CI 0.79–11.03. We found that ECG abnormalities were common after chemotherapy in this Cameroonian cohort, with QT prolongation and repolarisation changes most frequently observed. Hypertension showed a positive association with new ECG abnormalities after the first chemotherapy cycle. In resource-constrained settings, baseline ECG and early post-cycle ECG surveillance may support detection of potentially clinically relevant electrical disturbances. While this study suggests ECG changes were common, these ECG findings should not be interpreted as synonymous with cardiotoxicity, but should be corroborated with imaging and biomarkers. Larger studies with cycle-level follow-up are needed to clarify predictors and prognostic implications of early ECG changes.