Exhaustive Treadmill Exercise Does Not Reduce Twitch Transdiaphragmatic Pressure in Patients with COPD MICHAEL I. POLKEY, DIMITRIS KYROUSSIS, SARAH E. J. KEILTY, CARL H. HAMNEGARD, GARY H. MILLS, MALCOLM GREEN, and JOHN MOXHAM Respiratory Muscle laboratory, Department of Thoracic Medicine, Kings College School of Medicine and Dentistry, Kings College Hospital, london, and Respiratory Muscle laboratory, Royal Brompton Hospital, london, United Kingdom Reduced diaphragm contractility has been described in normal subjects after whole body endurance exercise, and it indicates low frequency fatigue (LFF); it is unknown whether LFF is of clinical impor- tance. We therefore studied the effect of treadmill exercise to exhaustion on diaphragm contractility in six patients with severe chronic obstructive pulmonary disease (COPD) (mean FEV1, 0.71, 27% predicted). The subjects first performed a short (control), treadmill walk and then, after resting, a sec- ond walk to a state of severe dyspnea. Cervical magnetic stimulation of the phrenic nerve roots was performed at the start of the study and 20 and 30 min after each walk. The twitch transdiaphragmatic pressure (Tw Pdi) was reproducible (mean coefficient of variation, 5.3%; range, 2 to 12.5%). Mean Twi Pdis were 18.4 cm H20 at baseline and 19.6 cm H20 and 19.2 cm H20 20 and 30 min after the control walk. At the same times after the exhaustive walk, mean Tw Pdis were 19.6 and 20.4 cm H20. Tw Pdi was not reduced by exhaustive treadmill walking (p > 0.9), and a power calculation showed that the study had a 95% chance of detecting a 10% fall at the 5% significance level. We conclude that Tw Pdi is not reduced when patients with severe COPDwalk to a state of extreme breathlessness and that therefore low frequency fatigue of the diaphragm does not occur. Polkey MI, Kyroussis D, Keilty SEI, Hamnegard CH, Mills GH, Green M, Moxham I. Exhaustive treadmill exercise does not reduce twitch transdiaphragmatic pressure in patients with COPD. AM' RESPIR CRIT CARE MED 1995;152:959-64. Evidence that respiratory muscle fatigue is important in clinical situations is sparse. Patients with chronic obstructive pulmonary disease (COPD) might be expected to be vulnerable to respira- tory muscle fatigue during exercise because of hypoxia and aci- dosis as well as the mechanical disadvantage associated with hyperinflation (1). This possibility is further suggested by the ob- servation that sustained exercise may cause a fall in the high/low EMG ratio of the diaphragm (2) and a rightward shift of the force frequency curve of the sternamastoid (3). The technique of bilateral supramaximal phrenic nerve stim- ulation is currently considered to offer the best potential for de- tecting fatigue of the diaphragm (4), and a sustained reduction in twitch tension is characteristic of low frequency fatigue (5). Bilateral electrical stimulation (ES) has been successfully used to show fatigue of the diaphragm in normal subjects after repeated voluntary contractions (5), the combination of inspira- tory resistive loading and exercise (6), and heavy endurance ex- ercise alone (7, 8). However, ES may cause discomfort, and repeated stimulations are required to find the best site for stimu- lation; these factors limit application of the technique to patients. Moreover, if patients are unable to relax, then twitch potentiation is a confounding factor (9, 10). Cervical magnetic stimulation (Received in original form January 4, 1995 and in revised form March 7, 1995) Correspondence and requests for reprints should be addressed to Dr. Michael Polkey, Department of ThoracicMedicine, Kings CollegeHospital, Bessemer Road, London SE5 9PJ, United Kingdom. Am J Respir Crit Care Med Vol 152. pp 959-964, 1995 (CMS) is a recently described method of bilaterally stimulating the phrenic nerves that is less painful and comparable to the elec- trical technique (11, 12), and it is able to detect low frequency diaphragm fatigue in normal subjects (13). We therefore used CMS to evaluate the effect of exhaustive treadmill walking on diaphragm contractility in a controlled study of six patients with severe COPD. METHODS Six male patients with stable COPD were studied. All had been specifi- cally trained to relax during cervical magnetic stimulation and had prac- ticed exhaustive treadmill walking a mean of seven times before the study day. All patients were ambulant and free from overt cardiovascular or musculoskeletal disease; all complained that their exercise was limited by dyspnea. Patient 2 had had a partial lung resection in his youth. The protocol was approved by the ethical committee of Kings College Hos- pital, and all subjects gave written informed consent. Spirometry was obtained using the best of three efforts obtained on a wedge bellows spirometer (Vitalograph, Buckinghamshire, UK). Lung volumes were determined by whole body plethysmography (PK Mor- gan, Rainham, Kent, UK); the average of three representative measure- ments was used. Esophageal and gastric pressure (Pes and Pgas) were recorded using conventionally placed balloon catheters (PK Morgan). The catheters were connected to differential pressure transducers (Validyne MP45; Validyne Corp., Northridge, CA), carrier amplifiers (PK Morgan), a l2-bit NB- MIO-16analogue-digital board (National Instruments, Austin, TX), and a Macintosh Quadra Centris 650 personal computer (Apple Computer Inc., Cupertino, CA), running Labview" software (National Instru-