Page 44 - Physical activity in recipients of solid organ transplantation - Edwin J. van Adrichem

Chapter 3
to determine total lung capacity, residual volume, and di

using capacity of the lung for
carbon monoxide (DLCO) for some of the recipients (of the total of 346 lung function
measurements, 247 included measurement of total lung capacity and residual volume and
212 included measurement of DLCO). Measurements were obtained in accordance with
American

oracic Society/European Respiratory Society guidelines.
20
Peripheral muscle strength
Maximal voluntary isometric strength was measured for the quadriceps, biceps brachii, and
triceps brachii muscles and for the 3-point grip with a handheld dynamometer (MicroFET
II, Hoggan Health Industries, West Jordan, Utah) by experienced physical therapists.
Measurements were obtained bilaterally 3 times for each muscle group and in the testing
positions described earlier.
21

e break method was applied; with this method, muscle force
was gradually overcome, and the measurement ceased at the moment of “giving way”.

ree-point grip strength was tested in a seated position with the upper arm in a neutral
position and the elbow in 90 degrees of

exion.

e index and middle

ngers were placed
on the force pad of the dynamometer, and the thumb was placed on the opposite side.
With the thumb facing the
strength, as measured, were not available.
Data Analyses

oor, the patient was instructed to pinch as hard as possible.

e mean measured peak force (in newton) of the dominant side of the patient was used
for further analysis. Predicted values were calculated
medians and interquartile ranges for variables with a skewed distribution. For categorical
23
; however, reference values for grip
22
Descriptive statistics were applied for baseline characteristics. Normality assumptions were
model with the restricted maximum-likelihood method and an unstructured correlation
model on the basis of the minimum Akaike information criterion.
results were subsequently pooled to gain further insight into the robustness of the predicted
42.
tested with the Shapiro-Wilk test. Data are presented as means and standard deviations or
data, proportions are shown. To predict changes in 6MWD over time and to account for
intra-recipient correlations between multiple measurements, we applied a linear mixed
structure. With this repeated measures analysis, all available data points were used; if a
single score was missing, then other data from that patient were not omitted from the
analysis. Measurement moment and diagnosis category were set as factors.
explanatory variables (including age, sex, quadriceps force, grip strength, lung function
24
Possible
values, type of transplantation [single versus double], length of hospital stay, rejection, and
body mass index) were selected on the basis of clinical relevance and were explored by
being entered into the linear mixed-model analysis. Variables were maintained in the
estimates,
To increase clinical interpretability and to avoid large correlations between
24
we centered quadriceps strength at 250 N, grip strength at 91 N, FFV, at 65%
of the predicted value, and length of hospital stay at 39 days (means of the study
population). All variables were included as

xed e

ects. Interaction e

ects and a random
intercept were explored. Additional linear mixed-model analyses were performed after 5
separate imputations of missing data (multivariate imputation with chained equation).
e