Profile of Nonresolving Pneumonia in a Tertiary Care Center in South India: A Prospective Study

Introduction: Nonresolving pneumonia (NRP) or slowly resolving pneumonia is a major concern among clinicians. The definition and approach toward diagnosis and management are unclear. Herein, we conducted a prospective observational study to analyze the comorbidities, etiology, and mode of diagnosis of NRP at our center. Methodology: All subjects admitted with a diagnosis of NRP (who received at least 2 weeks of antimicrobial therapy without any clinical improvement) were enrolled in the study. Patients with poor general condition, hemodynamic instability, and uncooperative patients were excluded from the study. After noting clinical and demographic details, contrast-enhanced computed tomography (CT) scan was done for all patients. All subjects were assessed to undergo flexible bronchoscopy, image-guided sampling, surgical lung biopsy, or pleural fluid analysis based on the imaging findings on CT scan. The details of the mode of diagnosis and the yield of procedures were noted. Results: We included 102 subjects; an infective etiology was diagnosed in 64 (62.7%) of the cases and 38 (37.2%) were diagnosed to be of noninfective etiology. Among the infections, pulmonary tuberculosis [21 (20.6%)] and invasive fungal diseases [20 (19.5%)] were the most common diagnoses. Diabetes mellitus (57.8%) and recent COVID-19 (26.5%) were the common comorbid conditions predisposing to infection. Among the noninfectious etiology, pulmonary adenocarcinoma was the most common diagnosis seen in 14 (13.7%) cases. Flexible bronchoscopy was most instrumental in obtaining the diagnosis as seen in 71 (69.6%) cases, followed by image-guided biopsy in 17 (16.6%) and surgical lung biopsy in 7 (6.9%) patients. Overall yield of bronchoscopy in our study population is 77.2% (71/92) and image-guided sampling is 94.4% (17/18). The infectious group had shorter disease duration, more comorbid illnesses, mucopurulent secretions on bronchoscopy, and higher mortality at 1 month as compared to noninfectious group. After adjusting the covariates, the presence of fever [odds ratio (OR) 12.75; 95% confidence interval (CI), 2.74–59.26] and recent history of COVID-19 (OR 12.29; 95% CI, 1.43–105.6) were independently associated with infectious etiology. Conclusion: Infections, particularly tuberculosis and invasive fungal diseases, are the major causes of NRP. Diabetes mellitus is the predominant underlying comorbid illness, and recent infection with COVID-19 is an emerging risk factor for NRP. Flexible bronchoscopy and image-guided sampling, when used in rational approach, are helpful in establishing the diagnosis of NRP.