The Long and the Short of It: Is "Long COVID" More Than Slow Resolution of the Acute Disease?

To date, the coronavirus disease (COVID-19) pandemic has taken millions of lives and sickened well over 100 million (1). We cannot yet measure the full impact of this ongoing outbreak on the health of the world, but it certainly has been impactful. Although the acute phase of COVID-19 represents a fearsome health threat, this disease can also cause symptoms that persist long after the acute phase of illness has resolved (2–10). This set of symptoms has been termed “long COVID” when persisting beyond 4 weeks, and distinctions between 4–12 weeks and greater than 12 weeks after diagnosis have also been proposed (11). Nomenclature aside, these syndromes are of the utmost importance; as was seen with postencephalitis Parkinsonism after the 1918 Spanish Flu pandemic (12), this aspect of COVID-19 could be a major factor long after the pandemic resolves. However, medium- and long-term outcomes from COVID-19 remain poorly defined, and persistent COVID-19 symptoms are poorly understood with respect to risk factors, prognosis, and etiology.

In this issue of AnnalsATS, Townsend and colleagues (pp. 997–1003) report a well-conducted cross-sectional study of a cohort of survivors of COVID-19 that adds important information on this topic (13). This cohort included survivors across disease strata with respect to the acute phase of illness. Of the entire group, 56% did not require hospitalization and the remaining were hospitalized, including 14% who required intensive care unit (ICU) admission for their acute illness. This cohort is from early in the pandemic, requiring positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing between March and May 2020 followed by evaluation in clinic at least 6 weeks after either the cessation of the acute COVID-19 syndrome or hospitalization. Of note, having been evaluated an average of 75 days from acute illness resolution, this cohort would be defined as having “ongoing symptomatic COVID-19,” as opposed to “post–COVID-19 syndrome” per National Institute of Health and Care Excellence guidelines (11). The authors go on to describe factors associated with persistent fatigue, breathlessness and other respiratory sequelae of COVID-19 infection across different disease strata.

To approach the question of whether initial disease severity predicts persistence of symptoms, the severity of initial disease was determined retrospectively (by chart review), including the degree of oxygen support, the peak C-reactive protein, length of stay, and presence or absence of abnormal chest imaging. At follow up, recruited subjects completed a battery of testing on various parameters, including functional status (e.g., 6-min-walk testing [6MWT]), persistent pulmonary infiltrates on chest imaging, dyspnea severity (e.g., modified Borg dyspnea scale), and fatigue (e.g., Chalder Fatigue scale), among others. Patients who were never hospitalized, those hospitalized without requiring ICU care, and those who were admitted to an ICU for their acute COVID-19 illness were then analyzed. Similar to prior reports (2–4, 9, 10), fatigue was a common symptom in this cohort. The principle finding of this study is that persistent dyspnea and exertional limitation were very common (reported in 62% of the cohort), but, perhaps surprisingly, inasmuch as could be determined, these were not associated with the initial disease severity (e.g., admission to an ICU or requirement of high degree of respiratory support). These data add a significant and important dimension to our understanding of these persistent COVID-19 symptoms.

There are noteworthy caveats to the study’s conclusions. This single-center post–COVID-19 cohort is fairly small (n = 153) and, importantly, represents only a portion of the studied population (31% of a COVID-19–positive population who were invited to follow up in a post–COVID-19 clinic). Also, this cohort includes primarily a Caucasian population and may not be generalizable to other groups, especially as Black and Latino populations have been shown to have increased risk from SARS-CoV-2 infection (14, 15). Furthermore, several analyses undertaken in this study may have been underpowered (e.g., testing for association between peak CRP and 6MWT distance). This and the absence of cross-sectional pulmonary function testing limits the biological inferences that can be made from this study. Nonetheless, these early data would strongly suggest that the prevalence of persistent symptoms in COVID-19 survivors is at least in part driven by factors other than the severity of the initial illness, at least with respect to the degree of respiratory failure.

The study reveals an abnormal imaging rate of only 4% in those for whom convalescent imaging was obtained, with only 19% of the hospitalized patients having abnormal imaging at follow up. Reassuringly, the observed persistent abnormalities resolved in most patients on repeat imaging 6 weeks later. Though intriguing, the low rate of radiographic abnormalities seen in this cohort should be considered cautiously. Other studies have identified much higher rates of abnormal imaging (>50%) months after disease resolution, especially when evaluating convalesced patients who had required hospitalization for their acute illness (6, 7). The true incidence and significance of radiographic abnormalities and fibrosis in the post–COVID-19 period remain uncertain, but it seems likely that plain film chest radiographs used for screening in this study are not sensitive enough to visualize subtle changes. Thus, the latter may be more common than the rate of detection in this study would suggest. However, the significance of such putative subtle changes would be unclear, and given the authors’ other findings, they would seem unlikely to account for much of the prevalent dyspnea and exertional limitation seen in this cohort.

Despite these limitations, Townsend and colleagues provide crucial insight into long COVID, identifying that significant morbidity persists even when impairments are not clearly identified by available objective testing, such as 6MWT. Their results also demonstrate persistent low-grade inflammation after infection, hinting at a potential underlying mechanism for some sufferers of long COVID. Knowledge gaps in our understanding of the sequelae of COVID-19 persist. However, the significant respiratory morbidity identified by Townsend and colleagues at the various strata of disease severity highlights the importance of this line of inquiry and carries significant implications for future research. The authors compare their findings to the those seen in survivors of the 2003 severe acute respiratory syndrome (SARS) outbreak. Survivors of SARS reported impaired health-related quality of life at 6 months, including persistent fatigue, myalgias, weakness, and depression even up to 4 years after acute illness (16). Given the much larger scale of the present pandemic relative to the SARS outbreak, study of long COVID will indeed have enormous societal implications as well.

Future studies of persistent health issues after COVID-19 should build on this work by examining the interplay between the biology of SARS-CoV-2 infection and baseline patient-level factors. Although Townsend and colleagues focused primarily on respiratory sequelae similar to the acute phase of this disease, post–COVID-19 symptoms are multisystem (11), and, thus, comprehensive evaluations of patients will be necessary to identify physical function impairments and target appropriate rehabilitation interventions. Additional research is also needed to study the progression from ongoing symptomatic COVID-19 to post–COVID-19 syndrome per se, identify and test potential interventions to improve outcomes (perhaps including cardiopulmonary rehabilitation), and to delineate the trajectory of recovery.

Though the COVID-19 pandemic may show signs of waning, an “epi-pandemic” of persistent symptoms from this novel disease is upon us. It is now clear that this is a concern not only to those who had life-threatening complications early in the course of infection but also to patients who were never hospitalized. Long COVID may be with us for a long time. Despite the danger and saliency of the acute phase of COVID-19, the longer-term aspects of this disease cannot be overlooked.