INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Respiratory muscle fatigue is hypothesized to play a major role in the pathogenesis of ventilatory failure, and rest is the most effective treatment, enabling the muscles to recover from fatigue. The duration of respiratory muscle rest is dependent upon the nature of fatigue. High frequency muscle fatigue requires shorter periods of rest (e.g., several minutes), whereas low frequency fatigue requires several hours of recovery (1). Low frequency fatigue is probably the more clinically relevant form of diaphragm fatigue; however, scant data examine the time required for recovery from human diaphragm fatigue.

Studies in healthy subjects in whom diaphragm fatigue has been induced experimentally, have demonstrated a reduction in the transdiaphragmatic twitch pressure (Pdi_T) and impaired force generation at low frequency stimulation, that persisted for at least 30 min (2) to greater than 1 h (2, 3), and recently, assessed by cervical magnetic stimulation, even after 24 h of rest (4).

In this study, we specifically used a selective measure of diaphragm strength to determine the recovery from diaphragm fatigue in normal subjects breathing against an inspiratory resistive load, and attempted to correlate the rate of recovery with endurance for performing the task of inspiratory resistive loading.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Seven healthy male subjects trained in respiratory maneuvers were recruited for this study. All subjects were in good health, were nonsmokers, and took no medications. This protocol was approved by our Institutional Review Board for human research. Informed consent was obtained in all subjects.

Measurement of Transdiaphragmatic Pressure

Following topical anesthesia (4% xylocaine), two thin walled balloon-tipped catheters were passed via the nares, one into the lower esophagus (esophageal pressure, Pes) and the other into the stomach (gastric pressure, Pga). These catheters were connected to pressure transducers (Validyne range ± 100 cm H₂O; Northridge, CA). Transdiaphragmatic pressure (Pdi) was continuously displayed as the electronic subtraction of Pes from Pga.

A molded plaster cast was placed over the anterior abdomen to prevent diaphragm shortening by minimizing outward displacement of the abdomen during stimulation. Voluntary Pdi_max was measured against an occluded airway using a combined expulsive-Mueller maneuver, during visual oscilloscopic feedback (5) while the subject was seated upright in a high-backed chair. The average of three values of Pdi_max, all within 5%, were reported as the Pdi_max.

Breathing Circuit and Recording Apparatus

During the study, subjects breathed through a pneumotachograph (Hans Rudolph, Inc., Kansas City, MO) connected to an in-line three-way valve thus allowing the subject to breathe spontaneously or against an occluded airway. Airflow signal, inspired and expired volumes via integration of the airflow signal, and airway pressure recorded at the mouth (Pm) were continuously displayed on a strip chart recorder (Gould ES 1000; Dayton, OH).

Electrophrenic Stimulation

Compound diaphragm action potentials (CDAP) were measured bilaterally by two, 3-mm EMG surface electrodes placed 2 mm apart in the 7th intercostal space in the anterior axillary line.

In each subject, the area for optimal phrenic nerve stimulation was located by using well determined neck anatomic landmarks (6). Once identified, an electrical stimulus was applied and the simultaneously produced CDAP was displayed on a recording oscilloscope to confirm phrenic nerve stimulation. The stimulus voltage was incrementally increased until there was no further increase in CDAP amplitude. Once maximum stimulus voltage was achieved, it was further increased by 20% to ensure supramaximal diaphragm activation.

A modified neck brace, which housed the right and left phrenic nerve stimulating probes, was used throughout the study to ensure consistency in phrenic nerve stimulation. The phrenic nerves were then stimulated transcutaneously (Grass S88 Stimulator; Quincy, MA) with 100-140 V (approximately 30 mA), 0.1 msec in duration to produce diaphragm twitch pressures.

Pdi Twitch at FRC

With the subject seated in an upright posture with the anterior abdomen casted, bilateral phrenic nerve stimulation was applied at functional residual capacity (FRC) after closure of the three-way valve at end expiration. Pes was continuously monitored to ensure that end expiratory lung volume had returned to a consistent baseline prior to valve closure. Six to 15 consecutive twitches at FRC were obtained at baseline, and at each time point during recovery.

Twitch Occlusion

In addition to Pdi_T at FRC, the twitch occlusion technique was performed in each subject. Each subject voluntarily contracted the diaphragm by making an inspiratory effort against an occluded airway to the targeted Pdi_max displayed on an oscilloscope. Once the subject achieved the targeted Pdi (20%, 40%, 60%, 80%, and 100% of Pdi_max), bilateral electrophrenic stimuli (5-6 twitches) were applied producing a superimposed twitch. This procedure was done at baseline, and then 30 min, once between 2-3 h, and once between 20-25 h post-fatigue.

Pdi Contractile Characteristics

Twitch amplitude: maximal height of the twitch from baseline pressure to its peak.

Time to peak tension: time from twitch onset to the generation of peak pressure.

1/2 relaxation time: time from peak pressure to 1/2 of the peak during relaxation.

Maximal rate of relaxation: slope of a line drawn during the initial 1/3 of the relaxation portion of the twitch curve, normalized to peak pressure (MRR/Pdi_T).

Maximal contraction rate: slope of the line drawn during twitch contraction, normalized to peak pressure (MCR/Pdi_T).

Induction of Diaphragm Fatigue

Subjects breathed against a resistive load to achieve a targeted Pdi 80% of Pdi_max using oscilloscopic feedback. Inability to maintain the visual Pdi target for five consecutive breaths despite maximal encouragement determined the end of the fatigue run. To ensure fatigue, subjects underwent four consecutive fatigue runs each separated by one-min pause (2). Endurance time was calculated as the total time of all four fatigue runs.

Protocol

Subjects were acquainted to the apparatus and techniques on 1 d. On a subsequent day, baseline Pdi_max, Pdi_T at FRC, and Pdi twitch occlusion were measured. After these measurements, inspiratory muscle fatigue was induced, and the endurance time was recorded. Following fatigue induction, Pdi_T at FRC was measured 15-30 minutes, and 1, 2, 3, 4 h after fatigue. If total recovery (return of Pdi_T amplitude to within 5% baseline value) did not occur by 4 h, subjects were studied the following day (1-2 sets of 6-15 twitches each between 20-25 h post-fatigue) until return of values to baseline. Pdi_max estimated by twitch occlusion was measured at 30 min, once between 2-3 h, and once between 20-25 h post-fatigue.

Data Analysis

Analysis of variance was used for repeated measures of Pdi_T characteristics over time. All values are expressed as mean ± SE unless otherwise specified. p Value < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Table 1 shows individual and mean data for age, voluntary Pdi_max, Pdi_T, Pdi_T/Pdi_max, endurance time, and the Pdi_T obtained 15-30 min after fatigue induction. The mean Pdi_max produced by a voluntary combined expulsive-Mueller maneuver for the group before fatigue was 143 ± 7 cm H₂O. The Pdi_T was 23.6 ± 2.5 cm H₂O, approximately 16% of Pdi_max (Table 1).

The time to the development of fatigue ranged from 11.9 to 50.1 min, with a mean endurance time of 25.3 ± 12.3 min. In addition to the fall-off in Pdi during the fatigue runs, each subject exhibited variation in the compartmental contribution (Pes and Pga) to Pdi and throacoabdominal paradox (7) (i.e., the inward movement of the abdomen and outward movement of the thorax on inspiration) in the terminal portion of the fatigue run. The mean Pdi_T obtained no less than 15 min from the end of the fatigue run (in order to avoid twitch potentiation), was 11.9 ± 2.1 cm H₂O, approximately 50% less than the initial pre-fatigue value (p < 0.05). A representative twitch before (panel A) and after (panel B) fatigue is shown in Figure 1.

View larger version (45K): [in this window] [in a new window]   Figure 1.   Representative PdiT in one subject before (A) and after (B) diaphragm fatigue.

Figure 2 shows the Pdi_T as % initial Pdi_T value for each individual subject at baseline, and then at 15, 30 min, 1, 2, 3, 4 h after the onset of fatigue. By 2 h after fatigue, the Pdi_T in four of seven subjects returned to greater than 80% of the initial pre-fatigue value. All subjects underwent repeat testing on the following day. Two subjects had measurements made at hour 20, 3 at hour 21, 1 at hour 22, 2 at hour 24, and 1 at hour 25. The data from these times are grouped as hour 20-25, and are shown in Figure 3.

View larger version (19K): [in this window] [in a new window]   Figure 2.   Plot showing the reduction of PdiT amplitude expressed as a percent of the initial value in all seven subjects, after fatigue and up to 4 h into recovery.

View larger version (13K): [in this window] [in a new window]   Figure 3.   Plot showing group mean data for PdiT amplitude, maximum contraction rate normalized to PdiT (MCR/PdiT), and maximum relaxation rate normalized to PdiT (MRR/PdiT) before the induction of diaphragm fatigue, and then during the period of recovery following fatigue. All subjects were studied on the subsequent day (two of the seven subjects were studied at hour 20, three at hour 21, one at hour 22, two at hour 24, and one at hour 25).

Table 2 shows the contractile characteristics of the twitch at baseline, post-fatigue and at various time points into recovery. The values for pre-fatigue and 20-25 hour recovery twitch amplitude are significantly (p < 0.05) different from the post-fatigue and 1 hour recovery values (Figure 3).

Pdi_max determined by the twitch occlusion technique was 75 ± 6% of the pre-fatigue Pdi_max, and 87 ± 8% by 3 h into recovery. Pdi_max was 100 ± 8% by 25 h (Figure 4).

View larger version (12K): [in this window] [in a new window]   Figure 4.   Plot showing group mean data for Pdimax before the induction of diaphragm fatigue and they during the period of recovery following fatigue.

Endurance was inversely related to recovery within the first 3 hours after fatigue (r value = 0.79 , p = 0.03, Figure 5).

View larger version (13K): [in this window] [in a new window]   Figure 5.   Plot showing the relationship between endurance and the % recovery in PdiT (r = 0.79, p = 0.03).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our data demonstrate that the induction of diaphragm fatigue by breathing against high exhaustive resistive loads in normal subjects produces a 50% reduction in Pdi twitch at FRC, with a return to 72% of baseline by 3 h, and complete recovery by 20-25 h. The reduction of Pdi_max after fatigue was 25% of initial value, and returned to 87% of baseline within 3 h, and was nearly 100% by 25 h. In addition, we found a significant inverse correlation between the rate of Pdi_T recovery and time before fatigue occurred.

The time course of recovery from low frequency fatigue in nondiaphragmatic skeletal muscle has been shown to be long lasting (1, 8, 9). The diaphragm, unlike peripheral skeletal muscle, however, differs in that its function involves continuous, rhythmic contraction throughout life. Information regarding recovery from low frequency fatigue specifically of the human diaphragm is limited. Aubier and colleagues showed that low frequency fatigue induced in normal male subjects breathing through a resistive load persisted for at least 30 min after fatigue (10), and probably longer than 1 hour (2). Yan and coworkers showed in five normal male subjects in whom diaphragm fatigue was induced, that there was a decrease in the Pdi twitch uniformly at different lung volumes, specifically, about 40% reduction at FRC, and recovery in the twitch amplitude was incomplete at 1 hour (3).

Other studies involving whole body exercise have also shown a reduction in Pdi twitch (11, 12), however, the reduction in Pdi twitch after exercise was short-lived. The rapid rate of recovery of the Pdi twitch in these studies may have been related to the lack of a specific resistive load to the respiratory muscles, and the relatively modest reduction in the Pdi twitch amplitude presumably produced by whole body exercise to exhaustion.

Recently, Laghi and coworkers (4) showed in eight normal subjects breathing against a resistive load until task failure, a significant reduction in Pdi twitch amplitude with only complete recovery in two of eight subjects within 24 h. This study differed from the current study, however, in the use of a magnetic stimulator which some suggest is less specific in activating the diaphragm alone (13). Activation of extra diaphragmatic muscles (muscles of the pectoral girdle and neck) which may be preferentially more susceptible to fatiguing loads (16), when reassessed during a recovery period, may underestimate the recovery of the Pdi twitch.

Our data are consistent with findings of Aubier and coworkers (10) who found the Pdi_T/Pdi_max ratio to be 17%, while we found it to be 16%. Similarly, they reported an endurance of 27 minutes again similar to our findings, using the same fatigue protocol. They found also a similar degree of Pdi_T reduction post fatigue at approximately 50% on average. Our study, however, importantly differs from those previous studies in that we provide a clearer time course of recovery from human diaphragm fatigue, supporting the contention that like extradiaphragmatic muscle, low frequency diaphragm fatigue is of long duration, and requires several hours of recovery.

In each of our subjects, the Pdi_max derived by the twitch occlusion technique showed a similar pattern of recovery to the Pdi_T. The post-fatigue Pdi_max on average for all subjects was 75% of the initial value, similar to data provided by Gandevia and McKenzie who showed a Pdi_max 73% of the pre-fatigue Pdi_max (17) in normal subjects in whom diaphragm fatigue was induced by prolonged expulsive effort with an open glottis.

A inherent limitation of this study includes the applicability of our findings to clinical practice. Artificially induced muscle fatigue in healthy motivated subjects cannot easily be applied to the many factors that contribute to respiratory muscle fatigue in patients with ventilatory failure. Moreover, the role of fatigue in such patients remains to be determined and such speculations regarding recovery time may be misleading. However, these data represent preliminary attempts to determine the rate of recovery of diaphragm fatigue so as to guide future examination of the role of ventilatory muscle rest (e.g., mechanical ventilation) in the treatment of patients with respiratory failure.