Agreement and reliability of repeated bedside respiratory muscle strength measurements in acute and subacute stroke.

Abstract

Many stroke trials include maximal inspiratory pressure (MIP), maximal expiratory pressure (MEP), and sniff nasal inspiratory pressure (SNIP) outcome measurements. However, data on agreement and reliability of repeated MIP, MEP, and SNIP measurements in acute and subacute stroke patients are scarce. This study employed a test-retest design. Eighteen patients (seven female) with mean (SD) age 59 (14.5) years were recruited from neurological wards. Median (range) time since first stroke was 50.5 (21-128) days. MIP, MEP, and SNIP were measured repeatedly in three testing sessions (S1-3) conducted within 24h and following international standards. Intra-rater agreement between testing sessions was analyzed using the Bland-Altman method. Test-retest reliability was analyzed using intra-class correlation coefficient (ICC). Association between individual measurement variability, time poststroke, and level of stroke impairment was analyzed using Spearman's rho. Mean difference and 95% limits of agreement for MIP were -0.40 (-23.02, 22.22) cmH2 O between S1 and S2, and 2.14 (-12.79, 16.99) cmH2 O between S2 and S3; for MEP, -4.56 (-29.01, 19.90) cmH2 O between S1 and S2, and 0.29 (-24.28, 24.87) cmH2 O between S2 and S3; and for SNIP, -10.56 (-38.48, 17.37) cmH2 O between S1 and S2, and -6.06 (-27.32, 15.20) cmH2 O between S2 and S3. ICCs for MIP, MEP, and SNIP were ≥0.9 throughout. There were no strong correlations between individual measurement variability and time poststroke or level of stroke impairment. MIP, MEP, and SNIP in acute and subacute stroke patients show good test-retest reliability for group averages; however, absolute agreement can vary considerably for some individuals.