Closing the window on COVID-19 case-control studies

While the official definition of post-acute sequelae of SARS-CoV-2 infection (PASC) evolves as a multisystem condition, cardiovascular complications continue to be a major concern (1). PASC cardiovascular disease (PASC-CVD) includes a wide range of conditions, such as myocarditis, pericarditis, myocardial infarction, arrhythmias, and venous thromboembolism. Furthermore, an additional subset of patients with PASC also reports experiencing PASC cardiovascular syndrome (PASC-CVS), defined as experiencing adverse cardiovascular symptoms (exercise and/or orthostatic intolerance, chest pain, dyspnea, etc.) despite no diagnosis of underlying cardiovascular disease. These long-term complications have occurred in a wide range of demographics and have been observed from weeks to years postinfection. Although several theories have been suggested, the long-term implications of COVID-19 on the cardiovascular system, including PASC-CVD and PASC-CVS, pathophysiology, as well as prevention and treatment modalities have yet to be identified. In a recent article in the American Journal of Physiology-Heart and Circulatory Physiology, Van der Sluijs et al. (2) completed a cardiovascular and physical function examination in a large sample of middle-aged adults who were 4–7 mo post-COVID-19 and were not hospitalized during their COVID-19 infection (n = 101 subjects; 59 males/42 females) and age- and sex-matched controls (n = 101 subjects; 59 males/42 females) who reported no known COVID-19 infections (2).

Like many other COVID-19-related studies, the post-COVID-19 cohort was cross sectionally compared with age- and sex-matched adults who self-reported that they had not experienced “signs, symptoms, or suspicions of a SARS-CoV-2 infection, nor a lifetime positive test of any sort for SARS-CoV-2” (2). A limitation of cross-sectional COVID-19 investigations is the lack of an objective biomarker-based measure to indicate whether a COVID-19 infection has occurred in controls, as up to 40% of known COVID-19 cases are asymptomatic (3). In addition, as pandemic-related shutdowns have been lifted, it is near impossible to confidently report all known COVID-19 exposures. As the window of “COVID-19-naïve” individuals is closing, this will be a continued challenge in this area of research. Future investigations should use nucleocapsid antibody testing in matched control groups to support negative COVID-19 history. Another alternative approach for future research is using patient cohorts with confirmed diagnoses of non-COVID-19 viral or acute respiratory illnesses, such as influenza or respiratory syncytial virus, as a comparison group. These and other respiratory illnesses have relatively undescribed and potentially different timelines of inflammation and subsequent impacts on the cardiovascular system. Future research could also consider the comparison between those who have PASC-CVD and/or PASC-CVS and individuals who report a full COVID-19 recovery.

A strength of the current study is the use of both self-reported and objective physiological measures. Generally, there were no differences in the investigated cardiovascular parameters between adults who had COVID-19 and controls. Physiological and functional markers, such as arterial stiffness, blood pressure, grip strength, and gait speed, did not differ between groups. Cardiac biomarkers (e.g., amino-terminal pro-B-type natriuretic peptide, high-sensitive cardiac troponin I, and C-reactive protein) were also similar between groups. Not only were there no physiological markers of cardiovascular risk in adults post-COVID-19, but there were also no self-reported functional differences observed. For example, there were no differences in self-assessed quality of life or perceived general health status between post-COVID-19 adults and controls. Similar to other findings (4, 5), these data suggest that middle-aged adults 4–7 mo after experiencing mild COVID-19 do not display obvious differences in markers of cardiovascular risk relative to their matched cohort. One exception was that post-COVID-19 adults slept on average ∼1 h less per night compared with controls. A 1-h sleep difference is clinically significant in the inverse relationship between sleep duration and cardiovascular disease risk; however, both groups on average were within or greater than the recommended 7–9 h of sleep a night (mean [interquartile range]: post-COVID-19, 8.8 [7.7–9.4] vs. controls, 9.8 [8.9–10.3] h/day) (6). Furthermore, the directionality of this relation warrants future investigation as sleep duration and immunity have a complicated bidirectional interdependence (7). Immune system activation can result in disturbed sleep patterns, and conversely, abnormal sleep patterns can alter immune system efficacy. For example, observations have been reported linking COVID-19 to increased sleep disturbance and lower sleep quality (8, 9). Chronic fatigue is also a common PASC-CVS symptom (8, 9). However, lower sleep duration has also been correlated with a greater risk of COVID-19 infection (10). The pathophysiological mechanisms underlying COVID-19, sleep, and cardiovascular disease have yet to be clarified.

Given the heterogeneity of the COVID-19 clinical spectrum and the lack of a standardized symptom quantification method, it can be difficult to fully characterize a patient’s COVID-19 experience and make comparisons across studies. In the present study, participants classified 33 symptoms as either none, minor, moderate, major, or severe. Ninety-eight percent of participants with post-COVID-19 reported experiencing symptoms at the time of infection. The majority of COVID-19-related symptoms were reported mild, with the exception of fatigue. Fifty percent of the post-COVID-19 group reported moderate-to-severe fatigue. A third of the adults with post-COVID-19 reported experiencing PASC symptoms at the time of testing (∼4–7 mo postinfection). To determine the impact of PASC on cardiovascular health, Van der Sluijs et al. (2) performed a subanalysis on these individuals. As predicted, adults with PASC reported poorer quality of life. However, while adults with PASC were expected to be in poorer health, there were no differences in markers of cardiovascular risk between the control and patients with PASC. In fact, the patients with PASC had on average lower pulse wave velocities than controls, suggesting less arterial stiffness. To our knowledge, these are the first data to report less central vascular stiffness in adults with PASC compared with controls. Questions regarding how COVID-19 symptomology, including asymptomatic cases, initial severity, intermittent symptoms, and symptom duration (i.e., long-haul symptoms), relate to post-COVID-19 cardiovascular complications remain unanswered.

Additional confounding factors and their roles in PASC-CVD and PASC-CVS etiology that remain enigmatic include preexisting comorbidities, fitness, COVID-19 vaccination status (type and doses), previous viral infections (including repeated COVID-19 infections), COVID-19 variant, and mental health status (including pandemic-induced psychosocial stress). Furthermore, PASC is a condition that encompasses several interdependent organ systems. Future COVID-19-related research could benefit from multidisciplinary physiology collaborations to achieve comprehensive investigations. Extensive high-resolution phenotyping of COVID-19 survivors, especially in patients with PASC-CVD and PASC-CVS, is warranted to discover COVID-19’s long-term clinical outcomes. Understanding the mechanisms of PASC is critical for the development of effective predictive risk assessments, as well as prevention and treatment approaches.

GRANTS

This work was funded by National Institutes of Health Grants T32-DK07352 (to G.A.D.), K01 HL148144 (to S.E.B.), and R35-HL139854 (to M.J.J.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

G.A.D. drafted manuscript; G.A.D., M.J.J., and S.E.B. edited and revised manuscript; G.A.D., M.J.J., and S.E.B. approved final version of manuscript.