Mechanical power normalized to lung-thorax compliance indicates weaning readiness in prolonged ventilated patients

Ghiani, Alessandro; Paderewska, Joanna; Walcher, Swenja; Tsitouras, Konstantinos; Neurohr, Claus und Kneidinger, Nikolaus (2022): Mechanical power normalized to lung-thorax compliance indicates weaning readiness in prolonged ventilated patients. In: Scientific Reports, Bd. 12, Nr. 1, 6

Abstract

Since critical respiratory muscle workload is a significant determinant of weaning failure, applied mechanical power (MP) during artificial ventilation may serve for readiness testing before proceeding on a spontaneous breathing trial (SBT). Secondary analysis of a prospective, observational study in 130 prolonged ventilated, tracheotomized patients. Calculated MP's predictive SBT outcome performance was determined using the area under receiver operating characteristic curve (AUROC), measures derived from k-fold cross-validation (likelihood ratios, Matthew's correlation coefficient [MCC]), and a multivariable binary logistic regression model. Thirty (23.1%) patients failed the SBT, with absolute MP presenting poor discriminatory ability (MCC 0.26;AUROC 0.68, 95%CI [0.590.75], p=0.002), considerably improved when normalized to lung-thorax compliance (LTCdyn-MP, MCC 0.37;AUROC 0.76, 95%CI [0.680.83], p<0.001) and mechanical ventilation PaCO2 (so-called power index of the respiratory system [PIrs]: MCC 0.42;AUROC 0.81 [0.730.87], p<0.001). In the logistic regression analysis, PIrs (OR 1.48 per 1000 cmH(2)O(2)/min, 95%CI [1.241.76], p<0.001) and its components LTCdyn-MP (1.25 per 1000 cmH(2)O(2)/min, [1.061.46], p<0.001) and mechanical ventilation PaCO2 (1.17 [1.061.28], p<0.001) were independently related to SBT failure. MP normalized to respiratory system compliance may help identify prolonged mechanically ventilated patients ready for spontaneous breathing.

• Dokumententyp: Zeitschriftenartikel
• Fakultät: Medizin
• Themengebiete: 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit
• ISSN: 2045-2322
• Sprache: Englisch