• Antenor Rodrigues, Marianne L Da Silva, Danilo C Berton, Gerson Cipriano, Fabio Pitta, Denis E O'Donnell, and J Alberto Neder.
• Division of Respiratory and Critical Care Medicine, Queen's University, Kingston, ON, Canada; Laboratory of Research in Respiratory Physiotherapy, Department of Physiotherapy, State University of Londrina,... more Londrina, Brazil. less
• Chest. 2017 Jul 1; 152 (1): 32-39.

BackgroundSingle-point measurements of maximal inspiratory pressure (MIP) are frequently used to suggest muscle weakness in clinical practice. Although there is a large variability in "mean" predicted MIP depending on the chosen reference values, it remains unclear whether those discrepancies actually impact on the prevalence of weakness, that is, MIP below the lower limit of normal.MethodsA total of 1,729 subjects (50.1% men, aged 20 to 94 years) who underwent MIP measurements in a clinical laboratory comprised the study group. MIP was predicted according to the most frequently cited regression equations as of August 2015. Pretest probability of weakness was defined by a cluster of clinical and physiologic variables.ResultsPrevalence of weakness ranged from 33.4 to 66.9%. Set 2 equations agreed well in indicating weakness (κ [95% CI] ranging from 0.81 [0.79-0.83] to 0.83 [0.81-0.85]; P < .01). There was closer agreement between higher pretest probability of weakness and low MIP according to set 2 equations compared with set 1 equations. Thus, a significant fraction of subjects with abnormal MIP according to set 1 equations but preserved MIP according to set 2 equations had higher pretest probability of weakness (P < .05).ConclusionsThe choice of MIP reference values strongly impacts on the prevalence of weakness. Some specific equations relate better to clinical and physiologic indicators of weakness, suggesting that they might be particularly useful to screen subjects for advanced respiratory neuromuscular assessment.Copyright © 2016 American College of Chest Physicians. Published by Elsevier Inc. All rights reserved.

38 keywords...

hide…