Coughing correlates: insights into an innovative study using cough peak expiratory flow to predict extubation failure

Main text We read the innovative study by Duan et al. [1] with great interest. However, key results need to be interpreted carefully to reach the proper conclusions. First, their primary finding that patients with low cough peak expiratory flow (CPEF) have significant benefit from non-invasive positive pressure ventilation (NIPPV) in the prevention of re-intubation and 90-day mortality is not unsurprising given how CPEF represents the severity of underlying respiratory pathology. In addition, the strength of their study lay in the methodology. Each precise detail regarding the protocol of weaning and re-intubation mirrors that of previous landmark studies [2, 3]. These careful design choices help to bridge the methodological differences and heterogeneity among preceding studies. However, their non-standardized use of CPEF cutoffs makes external validity difficult to achieve. Previous studies have studied extubation failure at various CPEF cutoffs (e.g., ≤35 L/min in Beuret et al. [4] and ≤60 L/min in Salam et al. [5] and ≤70 L/min in Duan et al. [1]). Consequently, it is not possible to determine if a subgroup of patients within the weak cough group may have derived more benefit from NIPPV. Conversely, this arbitrary cutoff may have obscured a beneficial effect of NIPPV among patients with strong coughs. This design choice segregates the two arms asymmetrically in that the baseline demographics of patients above the CPEF cutoff appear to be younger,