BACKGROUND: Weak cough may result in the failure of noninvasive positive pressure ventilation (NPPV) in patients with AECOPD. However, no detailed descriptions have yet been published for the measurement of cough strength and associated outcomes in AECOPD patients. METHODS: This study prospectively enrolled 261 AECOPD patients who received NPPV. Semiquantitative cough strength score (SCSS, ranging from 0 = weak to 5 = strong) was recorded before NPPV. Patients who required intubation were defined as NPPV failures. RESULTS: NPPV failed in 55 patients (21.1%). Weak cough (SCSS ≤3, OR = 8.1), high disease severity (APACHE II score >19, OR = 3.8), and malnutrition (total proteins ≤58 g/L, OR = 2.8) were independent risk factors for NPPV failure. Patients with 1, 2, and 3 risk factors were 4.7, 13.6, and 21.6 times more likely, respectively, to experience NPPV failure compared with patients with no risk factors. The NPPV failure rates were 80%, 40%, and 10.2% in patients with SCSS of 0–1, 2–3, and 4–5, respectively (p < 0.001). Compared with NPPV success patients, NPPV failure patients stayed longer in ICU (10.1 ± 7.9 days vs. 6.5 ± 4.6 days, p < 0.001), and they had higher ICU costs (€2986 ± 1906 vs. €5680 ± 3,604, p < 0.001), higher hospital costs (€ 6714 ± 7025 vs. €10,399 ± 9,509, p = 0.009), and higher hospital mortality (72.7% vs. 4.4%, p < 0.001). Moderate accuracy to distinguish NPPV failure by APACHE II score, SCSS, and total proteins was evidenced by ROC curves, with areas under the curve of 0.71, 0.78, and 0.67, respectively. A combination of all three factors reached good accuracy, with an area under the curve of 0.86. CONCLUSION: AECOPD patients with weak cough had a high risk of NPPV failure. SCSS, APACHE II scores, and total proteins were predictors of NPPV failure. Combined, these factors increased the power to predict NPPV failure.
BACKGROUND: Weak cough may result in the failure of noninvasive positive pressure ventilation (NPPV) in patients with AECOPD. However, no detailed descriptions have yet been published for the measurement of cough strength and associated outcomes in AECOPD patients. METHODS: This study prospectively enrolled 261 AECOPD patients who received NPPV. Semiquantitative cough strength score (SCSS, ranging from 0 = weak to 5 = strong) was recorded before NPPV. Patients who required intubation were defined as NPPV failures. RESULTS: NPPV failed in 55 patients (21.1%). Weak cough (SCSS ≤3, OR = 8.1), high disease severity (APACHE II score >19, OR = 3.8), and malnutrition (total proteins ≤58 g/L, OR = 2.8) were independent risk factors for NPPV failure. Patients with 1, 2, and 3 risk factors were 4.7, 13.6, and 21.6 times more likely, respectively, to experience NPPV failure compared with patients with no risk factors. The NPPV failure rates were 80%, 40%, and 10.2% in patients with SCSS of 0–1, 2–3, and 4–5, respectively (p < 0.001). Compared with NPPV success patients, NPPV failurepatients stayed longer in ICU (10.1 ± 7.9 days vs. 6.5 ± 4.6 days, p < 0.001), and they had higher ICU costs (€2986 ± 1906 vs. €5680 ± 3,604, p < 0.001), higher hospital costs (€ 6714 ± 7025 vs. €10,399 ± 9,509, p = 0.009), and higher hospital mortality (72.7% vs. 4.4%, p < 0.001). Moderate accuracy to distinguish NPPV failure by APACHE II score, SCSS, and total proteins was evidenced by ROC curves, with areas under the curve of 0.71, 0.78, and 0.67, respectively. A combination of all three factors reached good accuracy, with an area under the curve of 0.86. CONCLUSION: AECOPD patients with weak cough had a high risk of NPPV failure. SCSS, APACHE II scores, and total proteins were predictors of NPPV failure. Combined, these factors increased the power to predict NPPV failure.