# Prevalence of Sleep Apnea and Echocardiographic Correlates in a Community-Based Cohort of Cardio-Oncology Patients

## TL;DR

This study finds that obstructive sleep apnea is common in cardio-oncology patients and linked to early signs of heart dysfunction before cancer treatment.

## Contribution

First to show a strong association between untreated sleep apnea and abnormal heart strain in cardio-oncology patients.

## Key findings

- 38.5% of cardio-oncology patients had obstructive sleep apnea, higher than other cardiovascular risk factors.
- Untreated sleep apnea was most strongly linked to abnormal global longitudinal strain in multivariable models.
- Abnormal heart strain was observed even when ejection fraction remained normal in sleep apnea subgroups.

## Abstract

Left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS) predict cancer therapy-related cardiac dysfunction (CTRCD). Obstructive sleep apnea (OSA) contributes to LV dysfunction, abnormal GLS and heart failure, but has not been well-studied in cardio-oncology (CO) patients and is absent from CTRCD risk assessment tools.

The purpose of this paper was to evaluate the prevalence of OSA in a community-based CO population and the association to echocardiographic measures for CTRCD.

Cross-sectional data including prevalence of conventional cardiovascular (CV) risk factors and sleep apnea status were collected in CO patients (n = 218). Pretreatment LVEF and GLS were obtained in these patients. Multivariable adjusted ORs (logistic regression) were calculated to assess the association between OSA status and pretreatment LVEF and GLS.

In CO patients, prevalence of OSA (38.5%, 84/218) was high and more frequent than many other CV risk factors. Baseline abnormal GLS, not an abnormal LVEF, was significantly different between the four subgroups: low OSA risk (prior negative sleep study or a low snoring, tiredness, observed apnea, and high blood pressure-body mass index, age, neck circumference, and gender score), treated OSA, untreated OSA, high OSA risk (a high snoring, tiredness, observed apnea, and high blood pressure-body mass index, age, neck circumference, and gender score) but a sleep study was not performed due to patient preference. In multivariable models, the presence of untreated OSA showed the strongest association with abnormal GLS (OR: 4.13; 95% CI: 1.32-12.93) when compared to other CV risk factors in CO patients.

These observations suggest OSA be considered in risk algorithms for CTRCD. Larger studies are needed to corroborate these findings and assess the impact of OSA prevalence and treatment on outcomes in this high-risk population.