Sepsis is associated with a preferential diaphragmatic atrophy: a critically ill patient study using tridimensional computed tomography.

Background: Diaphragm and psoas are affected during sepsis in animal models. Whether diaphragm or limb muscle is preferentially affected during sepsis in the critically ill remains unclear.

Methods: Retrospective secondary analysis study including 40 patients, comparing control (n = 17) and critically ill patients, with (n = 14) or without sepsis (n = 9). Diaphragm volume, psoas volume, and cross-sectional area of the skeletal muscles at the third lumbar vertebra were measured during intensive care unit (ICU) stay using tridimensional computed tomography scan volumetry. Diaphragm strength was evaluated using magnetic phrenic nerve stimulation. The primary endpoint was the comparison between diaphragm and peripheral muscle volume kinetics during the ICU stay among critically ill patients, with or without sepsis.

Results: Upon ICU admission, neither diaphragm nor psoas muscle volumes were significantly different between critically ill and control patients (163 ± 53 cm vs. 197 ± 82 cm for the diaphragm, P = 0.36, and 272 ± 116 cm vs. to 329 ± 166 cm for the psoas, P = 0.31). Twenty-five (15 to 36) days after admission, diaphragm volume decreased by 11 ± 13% in nonseptic and by 27 ± 12% in septic patients, P = 0.01. Psoas volume decreased by 11 ± 10% in nonseptic and by 19 ± 13% in septic patients, P = 0.09. Upon ICU admission, diaphragm strength was correlated with diaphragm volume and was lower in septic (6.2 cm H2O [5.6 to 9.3]) than that in nonseptic patients (13.2 cm H2O [12.3 to 15.6]), P = 0.01.

Conclusions: During the ICU stay, both diaphragm and psoas volumes decreased. In septic patients, the authors report for the first time in humans preferential diaphragm atrophy compared with peripheral muscles.