Abstract
OBJECTIVES: Due to the rapid rate of severe acute respiratory syndrome coronavirus 2 transmission and the heterogeneity of symptoms of coronavirus disease 2019, expeditious and effective triage is critical for early treatment and effective allocation of hospital resources. DESIGN: A post hoc analysis of respiratory data from non-invasive venous waveform analysis among patients enrolled in an observational study was performed. SETTING: Vanderbilt University Medical Center. PATIENTS: Peripheral venous waveforms were recorded from admission to discharge in enrolled coronavirus disease 2019-positive patients and healthy age-matched controls. INTERVENTIONS: Data were analyzed in LabChart 8 to transform venous waveforms to the frequency domain using fast Fourier transforms. The peak respiratory frequency was normalized to the peak cardiac frequency to generate a respiratory non-invasive venous waveform analysis respiratory index. Paired Fisher exact tests were used to compare each patient's respiratory non-invasive venous waveform analysis respiratory index at admission and discharge. A nonparametric one-way analysis of variance was used for multiple comparisons between patients with coronavirus disease 2019 and healthy controls for respiratory non-invasive venous waveform analysis respiratory index. MEASUREMENTS AND MAIN RESULTS: Fifty coronavirus disease 2019-positive patients were enrolled between April 2020, and September 2020, and 45 were analyzed; 34 required supplemental oxygen and 11 did not. The respiratory non-invasive venous waveform analysis respiratory index was significantly higher for the 34 patients with coronavirus disease 2019 who received supplemental oxygen (median, 0.27; interquartile range, 0.11-1.28) compared with the 34 healthy controls (median, 0.06; interquartile range, 0.03-0.14) (p < 0.01). For patients with coronavirus disease 2019 who received supplemental oxygen, respiratory non-invasive venous waveform analysis respiratory index was significantly lower at hospital discharge (p = 0.02; 95% CI, 0.10-1.9) compared with hospital admission (median = 0.12; interquartile range, 0.05-0.56). For patients with coronavirus disease 2019, a respiratory non-invasive venous waveform analysis respiratory index of 0.64 demonstrated sensitivity of 92%, specificity of 47%, and positive predictive value of 93% for predicting requirement of supplemental oxygen during the hospitalization. CONCLUSIONS: Respiratory non-invasive venous waveform analysis respiratory index represents a novel physiologic respiratory measurement with a promising ability to triage early care and predict the need for oxygen support therapy in coronavirus disease 2019 patients.