Age-independent predictors of Long-COVID in females. Data from the Polish Long-COVID Cardiovascular (PoLoCOV-CVD) Study

Abstract

Abstract Background The long-Covid (LC) is now a great challenge for the healthcare systems worldwide and one of the main issues now is to select the risk factors of LC. Female gender is one of the LC risk factors that have been described in the available literature, however nobody has looked what are LC predictors in females themselves. Purpose To assess the age-independent predictors of long COVID (LC) in female population. Methods Patient’s information, course of the disease with symptoms, post-COVID-19 complaints were collected in women at least 12 weeks after the COVID-19 recovery. Next, the women were followed for 12 months. ECG, 24-hour ECG monitoring; 24-hour blood pressure (BP) monitoring, echocardiography, and selected biochemical tests were performed. LC was recognized based on the WHO definition. Results We identified 1913 consecutive female patients, including those with LC (n=1195) and without LC/control group (n=718). We also divided the women based on the median age (53 years old) and presence of LC to assess the predictors, regardless of age, of the LC occurrence. Women with LC compared to the control group had higher BMI (27 vs 26 kg/m2; p<0.001) and had more often one of the ECG abnormalities: HR >100/min, QRS >=120ms, ST-T changes, T inversion, arrhythmia, QRS fragmentation [29 vs 23%; p=0.006] and myocardial damage assessed in late gadolinium enhancement cardiac magnetic resonance (CMR) [20 vs 12%; p=0.021]. Weight, HDL cholesterol concentration and 24-hour average pulse pressure may be related to the occurrence of long COVID, but mainly in the group of younger women <53 years old [p=0.02, p=0.02 and p=0.04, respectively]. In the same group, the course of COVID-19 with hospitalization due to pneumonia had an independent effect on the long-COVID diagnosis (6 vs 3%; p<0.001). In the entire female population regardless of age, LC diagnosis was significantly associated with the severe course of the COVID-19 at home or with the need of hospitalization (33% vs 22% for women <53 and 40% vs 29% for women >53, respectively; p<0.0001 for both). The presence of comorbidities: hypertension, diabetes, coronary artery disease, did not affect the occurrence of LC. Lifestyle parameters before the onset of COVID-19: smoking, alcohol abuse, and physical activity had no impact on the occurrence of LC regardless of age (p>0.05) (tab. 1). In a multivariate analysis in females <53 years old, HDL<59 mg/dl [OR 1.41;95% CI 1.03-1.95; p=0.03] and 24-h PP>52 mmHg [OR 1.57;95% CI 1,05-2.34; p=0.03] are independent LC predictors (fig. 1). In females >53 years old we observed only a trend for LC occurrence for BMI>27,5 [OR 1.52; 95% CI 0.84-2.7; p=0.1] and LVEF<62% [OR 1.6;95% CI 0.9-2.9;p=0.1]. Conclusion The independent predictors of long COVID in younger females are HDL cholesterol and 24-h PP . The presence of comorbidities and lifestyle before COVID-19 had no impact on the occurrence of Long COVID in females regardless of age.