Cluster analysis unveils a severe persistent respiratory impairment phenotype 3-months after severe COVID-19

Abstract

Background The mid-term respiratory sequelae in survivors of severe COVID-19 appear highly heterogeneous. In addition, factors associated with respiratory sequelae are not known. In this monocentric prospective study, we performed a multidisciplinary assessment for respiratory and muscular impairment and psychological distress 3 months after severe COVID-19. We analysed factors associated with severe persistent respiratory impairment, amongst demographic, COVID-19 severity, and 3-month assessment. Methods Patients with severe SARS-CoV-2 pneumonia requiring ≥ 4L/min were included for a systematic 3-month visit, including respiratory assessment (symptoms, lung function, CT scan), muscular evaluation (body composition, physical function and activity, disability), psychopathological evaluation (anxiety, depression, post-traumatic stress disorder-PTSD) and quality of life. A cluster analysis was performed to identify subgroups of patients based on objective functional measurements: DLCO, total lung capacity and 6-min walking distance (6MWD). Results Sixty-two patients were analysed, 39% had dyspnea on exercise (mMRC ≥ 2), 72% had DLCO < 80%, 90% had CT-scan abnormalities; 40% had sarcopenia/pre-sarcopenia and 31% had symptoms of PTSD. Cluster analysis identified a group of patients (n = 18, 30.5%) with a severe persistent (SP) respiratory impairment (DLCO 48 ± 12%, 6MWD 299 ± 141 m). This SP cluster was characterized by older age, severe respiratory symptoms, but also sarcopenia/pre-sarcopenia, symptoms of PTSD and markedly impaired quality of life. It was not associated with initial COVID-19 severity or management. Conclusions and clinical implication We identified a phenotype of patients with severe persistent respiratory and muscular impairment and psychological distress 3 months after severe COVID-19. Our results highlight the need for multidisciplinary assessment and management after severe SARS-CoV-2 pneumonia. Trial registration The study was registered on ClinicalTrials.gov (May 6, 2020): NCT04376840

Lung pathophysiology in patients with long COVID-19: one size definitely does not fit all

“…So far, the origin of dyspnoea and poor exercise performance in l-COVID patients has been investigated in numerous physiological and clinical studies; contrasting findings were obtained mostly because of the heterogeneity of patients, small sample size, variety of applied procedures and methodology [ 4 – 13 ]. Persistence of dyspnoea in patients that experienced COVID-19 pneumonia or milder forms appears to be unrelated to the degree of disease severity or the residual impairment in lung function, which often corresponds to a mild/moderate reduction in lung diffusion capacity for carbon monoxide ( D LCO ) and, although to a minor extent, of total lung capacity (TLC) [ 5 , 9 , 11 – 13 ].…”

Post-acute sequelae of COVID-19 symptom phenotypes and therapeutic strategies: A prospective, observational study

“…This study differs from other COVID-19 cluster analyses [15][16][17][18][19][20] in that, not only did we identify phenotypes of PASC symptoms, but we also mapped these clusters to corresponding therapeutic strategies and their subjective effectiveness. Disaggregating PASC, which is a highly heterogeneous condition, and identifying useful treatment strategies is critically needed, particularly since we found that 50% of hospitalized COVID-19 patients have PASC symptoms.…”

“…We further aimed to determine the rates at which patients reported subjective improvement with a given treatment program. While there are cluster analyses evaluating symptoms in the acute phase of COVID-19 illness [14][15][16][17], and others focused on specific post-acute symptoms such as pulmonary complaints [18], brain fog [19], or neuropsychiatric symptoms [20], we chose to evaluate the entire post-acute COVID-19 symptom report form published by the World Health Organization (WHO) [21] and map symptom clusters to therapeutic interventions. These analyses may provide insight into the efficacy of various symptom-based treatment strategies and underscore the need for a multi-disciplinary holistic approach to post-COVID-19 care.…”

Exploring patient medication adherence and data mining methods in clinical big data: A contemporary review

“…Research on COVID‐19, such as Perotin et al., 268 used cluster analysis to discern risk factors from respiratory impairments for psychological disorders. Yang et al.’s analysis 269 unveiled spatiotemporal clustering patterns in 366 cities in China, influenced by factors like importation risk and humidity.…”

“…[263][264][265] As deep learning has progressed, an increasing number of researchers have harnessed neural networks for feature extraction and optimized the process of feature extraction and clustering by combining deep neural networks (DNNs) with clustering algorithms. 266,267 Research on COVID-19, such as Perotin et al, 268 Nevertheless, cluster analysis is challenged by determining the optimal cluster number, the potential for suboptimal clustering due to random initialization, and difficulty in distinguishing genuine data from noise. [272][273][274] Data mining, especially cluster analysis, is instrumental in decisionmaking.…”