Correlation of maximal inspiratory pressure to transdiaphragmatic twitch pressure in intensive care unit patients

Gerald S. Supinski, Phillip Westgate, Leigh A. Callahan

Respiratory muscle weakness contributes to respiratory failure in ICU patients. Unfortunately, assessment of weakness is difficult since the most objective test, transdiaphragmatic pressure in response to phrenic nerve stimulation (PdiTw), is difficult to perform. While most clinicians utilize maximum inspiratory pressure (Pimax) to assess strength, the relationship of this index to PdiTw has not been evaluated in a large ICU population. The purpose of the present study was to assess both PdiTw and Pimax in ICU patients to determine how these indices correlate with each other, what factors influence these indices, and how well these indices predict outcomes. Studies were performed on adult mechanically ventilated patients in the University of Kentucky MICU (n = 60). We assessed PdiTw by measuring transdiaphragmatic pressure (Pdi) in response to bilateral twitch stimulation of the phrenic nerves using dual magnetic stimulators (Magstim 200). Pimax was determined by measuring airway pressure during a 30-second inspiratory occlusion. We also assessed the twitch and maximum force generation for diaphragms excised from control and septic mice. Both Pimax and PdiTw measurements were profoundly reduced for mechanically ventilated MICU patients when compared to normal reference values, e.g., Pimax averaged 56 % of the predicted value for normal subjects. For the ICU population as a whole, PdiTw and Pimax values correlated with each other (r(2) = 0.373, p < 0.001), but there was wide scatter and, as a result, PdiTw could not be reliably calculated from Pimax levels for individual subjects. Infection selectively reduced low-frequency force generation more than high-frequency force generation for both our mouse experiments (comparing muscle twitch to 150 Hz tetanic force) and for MICU patients (comparing PdiTw to Pimax). This effect of infection may contribute to scatter in the PdiTw to Pimax relationship. We also found that both PdiTw and Pimax were significantly correlated with both patient survival and the duration of mechanical ventilation, albeit statistically, PdiTw was the better predictor. While more difficult to measure, the PdiTw is a better predictor of outcomes in mechanically ventilated MICU patients than the Pimax. Nevertheless, for some clinical applications, the Pimax determination is a reasonable alternative.