INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Respiratory muscle dysfunction is well documented after upper abdominal surgery (1). Maximum static inspiratory and expiratory pressures are reduced after laparotomy (1) and even after laparoscopy (5), and recovery may take several days. Multiple factors have been implicated in such respiratory muscle dysfunction, including irritation and inflammation or trauma next to the diaphragm, leading to local mechanical failure, reflex inhibition, and pain. However, the role of pain in respiratory muscle dysfunction after upper abdominal surgery remains controversial. Simonneau and colleagues (3) excluded any role of pain, on the basis of the failure of epidural analgesia to restore both the decreased relative contribution of the diaphragm to tidal breathing and the reduced maximal transdiaphragmatic pressure induced by upper abdominal surgery. Manikian and coworkers (4) found that epidural analgesia partly restored the decreased relative contribution of the diaphragm to tidal breathing induced by upper abdominal surgery, suggesting that pain contributed to the impaired diaphragmatic function. The latter investigators, however, did not assess the pressure-generating capacity of the respiratory muscles, and considered the role of pain to be minor (4).

Dysfunction of the respiratory muscles induced by upper abdominal surgery has been implicated in the development of postoperative complications such as atelectasis, with a concomitant increase in morbidity and hospital length of stay. Therefore, delineation of the role of pain in respiratory muscle dysfunction after upper abdominal surgery is clinically important. Accordingly, we conducted a prospective, randomized, controlled double-blind study to delineate the role of pain in the development of respiratory muscle dysfunction induced by upper abdominal surgery.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Fifty patients (28 males, 22 females) undergoing elective upper abdominal surgery and 10 normal subjects (seven males and three females) participated in the study. The age, height, and weight of the 10 normal subjects were 51 ± 12 yr (mean ± SD), 170 ± 9 cm, and 73 ± 11 kg, respectively. The anthropometric characteristics of the patients and types of surgery they had are presented in Table 1. The study protocol was approved by the Institutional Ethics Committee of Evangelismos Hospital, and informed consent was obtained for all subjects.

Study Design

At the time of study entry, patients were randomly assigned to receive either analgesia (analgesia group, n = 25) or placebo (placebo group, n = 25) in a double-blinded manner. Patients were studied on three different occasions: (1) the day before surgical operation (Session 1); (2) 24 h after upper abdominal surgery (Session 2); (3) 1 h after receiving either analgesia or placebo at the end of Session 2 (Session 3). At each session, inspiratory and expiratory muscle strength and the magnitude of pain were assessed. Analgesia was provided by intramuscular injection of pethidine (meperidine) in a dose of 1 mg/kg, whereas placebo consisted of an intramuscular injection of normal saline. To exclude the possibility of a direct positive inotropic effect of pethidine on respiratory muscle strength, we also studied 10 normal subjects before and 1 h after giving them pethidine in a dose of 1 mg/ kg intramuscularly.

Measurements

Inspiratory muscle strength was assessed through the sniff mouth pressure (Psniff). On the day before surgery, all patients were instructed on how to perform the sniff maneuver. Psniff was measured at FRC, using a flexible, open-tipped catheter (I.D. = 1.8 mm) (6, 7). The tip of the catheter was positioned in the oral cavity, as close as possible to the posterior wall of the pharynx (6, 7). The 95% response time of the catheter system was 30 ms. The patients were asked to take short and sharp maximal sniffs through the nose, with the mouth closed. Between sniffs there was a pause of at least 30 s. Sniffs were repeated until Psniff did not increase further, and the maximum value obtained was used for analysis. Only sniffs with a total duration of < 0.5 s were accepted. Pressure measurements were made with a differential pressure transducer (MP 45; Validyne, Northridge, CA), and were recorded on a polygraph recorder (Gould ES 1000; Gould Instruments, Cleveland, OH).

Expiratory muscle strength was assessed as the maximal static expiratory pressure (MEP), measured according to the method of Black and Hyatt (8), near TLC. A flanged plastic mouthpiece was connected to a closed, rigid plastic tube, and the pressure inside the tube was measured and recorded as previously described. The minimum duration for an accepted effort was 1 s. At least three efforts with less than 5% variability were recorded, and the maximum value was used for analysis. All measurements were made with subjects in a semirecumbent position. On the day before surgical operation, patients were trained to perform the sniff and MEP maneuvers in order to maximally reduce learning effects. The magnitude of pain before and during the sniff and MEP maneuvers was quantified with a visual analog scale (VAS) (9), the difference from before to during each maneuver being considered as representing pain attributed to the maneuver.

Anesthesia and Analgesia

Anesthesia was induced with thiopental 5 mg/kg given intravenously and fentanyl 5 to 10 µg/kg intravenously. Tracheal intubation was facilitated with vecuronium 0.1 mg/kg intravenously. Anesthesia was maintained with a mixture of N₂O/O₂ and 0.5 to 1% isoflurane under controlled ventilation. Incremental doses of vecuronium at 0.03 mg/kg intravenously were given to maintain adequate muscle relaxation. At the end of surgery, residual neuromuscular blockade was reversed with atropine 0.02 mg/kg intravenously and neostigmine 0.04 mg/kg intravenously. Postoperative pain control was achieved with pethidine 1 mg/kg given intramuscularly every 4 h. Before Session 2, no analgesic drug had been delivered for the three preceding hours unless the patient was experiencing severe pain (VAS score > 4).

Statistical Analysis

Results are expressed as mean ± SD. Two-way analysis of variance (ANOVA) was used, with the presence or absence of analgesia as one factor and the different times of measurement (sessions) as the second factor. Post hoc comparisons were made through Scheffe's test. A value of p < 0.05 was considered significant. A mixed-effects model with random intercept, having either Psniff or MEP as the dependent variable and both the occurrence of surgical operation and the level of pain as fixed effects, was used in the analysis.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Analgesia Group

Psniff changed through the different sessions (F = 19.23, p < 0.05; Figure 1). More specifically, Psniff decreased from 70 ± 15 cm H₂O at Session 1 to 42 ± 11 cm H₂O at Session 2 (p < 0.05). Pethidine caused an increase in Psniff to 56 ± 14 cm H₂O at Session 3 (p < 0.05). Pain increase attributed to the sniff maneuver also varied through the different sessions (F = 9.87, p < 0.05; Figure 2). The pain VAS score was 0.3 ± 0.6 at Session 1, 4.4 ± 1.5 at Session 2 (p < 0.05), and 2.1 ± 1.0 at Session 3 (p < 0.05).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Average ± SD values of Psniff in the analgesia (A) and placebo (B) groups before surgery (C ), 24 h after surgery (24 H), and 1 h later, after administration of either pethidine or placebo (25 H). Psniff decreased in both groups after surgery, but improved only after administration of pethidine.

View larger version (11K): [in this window] [in a new window]   Figure 2.   Average ± SD values of the increase in VAS pain scores during the sniff maneuver in the analgesia (A) and placebo (B) groups before surgery (C ), 24 h after surgery (24 H), and 1 h later, after administration of either pethidine or placebo (25 H). Although the increase in VAS pain score was comparable in both groups after surgery, it was diminished only after administration of pethidine.

MEP did not change significantly through the study sessions (F = 1.68, p = 0.2). Although MEP decreased from 82 ± 54 cm H₂O at Session 1 to 52 ± 39 cm H₂O at Session 2, and then increased to 60 ± 43 cm H₂O at Session 3, the changes did not reach statistical significance. Pain increase attributed to the MEP maneuver also varied through the different sessions (F = 9.46, p < 0.05). The pain score for MEP maneuver was 0.3 ± 0.6 at Session 1, 5.1 ± 1.4 at Session 2 (p < 0.05), and 2.2 ± 1.0 at session 3 (p < 0.05).

The mixed-effects model showed that Psniff had a statistically significant relationship to pain (p < 0.001). Adjusting for the occurrence of surgical operation did not affect this result. In contrast, MEP had no statistically significant relationship to pain after adjusting for surgical operation.

Placebo Group

Psniff decreased from 69 ± 15 cm H₂O at Session 1 to 42 ± 10 cm H₂O at Session 2 (p < 0.05). Placebo did not cause any further change in Psniff, which was 43 ± 12 cm H₂O at Session 3 (p = NS) (Figure 1). Pain increase by VAS score was 0.4 ± 0.5 at Session 1, 4.3 ± 1.5 at Session 2 (p < 0.05), and 4.7 ± 1.7 at Session 3 (p = NS) (Figure 2).

MEP also did not change significantly among the different sessions, despite the decrease in MEP from 89 ± 48 cm H₂O at Session 1 to 56 ± 41 cm H₂O at Session 2. MEP at Session 3 was essentially the same as at Session 2, at 53 ± 35 cm H₂O. Pain attributed to MEP maneuver by VAS score was 0.3 ± 0.5 at Session 1, 4.9 ± 1.5 at Session 2 (p < 0.05), and 5.0 ± 1.5 at Session 3 (p < 0.05).

Comparison of Analgesia and Placebo Groups

Psniff behaved differently in the analgesia and placebo groups (F = 5.52, p < 0.05). It was nearly the same in both groups at Sessions 1 and 2, but different at Session 3 (56 ± 14 cm H₂O analgesia group, versus 43 ± 12 cm H₂O placebo group, p < 0.05). Pain increase also behaved differently in the analgesia and placebo groups (F = 16.2, p < 0.05). It was nearly the same in both groups at Sessions 1 and 2, but different at Session 3.

Effect of Pethidine in Normal Subjects

Administration of pethidine had no effect on respiratory muscle performance in normal subjects; Psniff was 75 ± 8 cm H₂O before and 77 ± 9 cm H₂O after pethidine (p = NS, paired t test), and MEP was 136 ± 28 cm H₂O before and 138 ± 30 cm H₂O after administration of pethidine (p = NS, paired t test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major finding of our study was that pain is an important contributor to inspiratory muscle dysfunction after upper abdominal surgery, and that analgesia may partly reverse this dysfunction.

Respiratory muscle dysfunction is well documented after upper abdominal surgery (1). This was confirmed in the present study, in which Psniff was found to be significantly reduced at 24 h after upper abdominal surgery. The nature of the postoperative muscle dysfunction remains elusive. Irritation, inflammation, and trauma in the vicinity of the diaphragm, leading to local mechanical failure, have been advocated as causing the dysfunction (1), but there are no data that either support or refute this hypothesis. The intrinsic contractility of the diaphragm assessed during bilateral electrical stimulation of the phrenic nerves is not altered after upper abdominal surgery (10). Studies with animal models (11) suggest that a reflex inhibition of the diaphragm plays a role in postoperative dysfunction of this muscle, but direct proof of this in humans is lacking.

Pain has been either excluded (3) or has been considered of minor importance (4) in the genesis of diaphragmatic dysfunction after upper abdominal surgery. Our results indicate that pain is an important contributor to this dysfunction. In fact, analgesia was able to partially restore respiratory muscle function, whereas patients in the placebo group showed no such improvement. Both groups of patients had a comparable decrease in Psniff and MEP at 24 h after upper abdominal surgery, and the observed improvement in the analgesia group therefore cannot be attributed to a less severe and therefore more readily reversible dysfunction in this group. A learning effect may lead to better performance of tests of respiratory muscle strength, but this effect was minimized by special care in training our patients in the performance of these tests before surgery. Furthermore, the total number of tests performed by the analgesia and placebo group patients was similar. Thus, the different behavior of the analgesia and placebo groups is not likely to have been due to a more developed learning effect in the analgesia group. A direct inotropic effect of pethidine could also have accounted for the observed improvement in the analgesia group, but to our knowledge no such effect has been described. Furthermore, we found no increase in respiratory muscle performance in 10 normal subjects after pethidine administration. Consequently, the improvement in inspiratory muscle strength after pethidine in our analgesia group of patients was due to the reduction of pain. This is further supported by the finding of a significant relationship of Psniff to pain, and by the failure of adjustment for the effect of surgical operation to affect this relationship.

Our results are not directly comparable to those in previous reports that have questioned the role of pain in the development of inspiratory muscle dysfunction after upper abdominal surgery (3, 4). Although measurements in these studies (as in our study) were made 24 h after upper abdominal surgery, analgesia was provided by thoracic extradural block, which induced motor blockade of muscles affecting respiration. The results obtained in the studies in which this was done were discrepant: fentanyl produced no effect (3), whereas bupivacaine had a partial minor effect (4). Furthermore, a control group was lacking in these studies. In contrast, we provided systemic analgesia that has no motor effect, and we also included a control group of patients who received no analgesia.

Our experimental design does not allow us to pinpoint the mechanism(s) by which pain contributes to inspiratory muscle dysfunction. However, two mechanisms, not mutually exclusive of one another, might account for pain-induced respiratory muscle dysfunction. Pain may cause a submaximal voluntary effort activation of the respiratory muscles in anticipation of an increase in pain with further effort. This is supported by previous anecdotal observations that some patients can voluntarily revert to predominantly abdominal breathing after upper abdominal surgery if asked to do so (1), indicating increased voluntary activation of the diaphragm. The fact that Psniff was reproducible postoperatively does not guarantee full effort, since even submaximal efforts can be quite reproducible (14). Alternatively, pain may be elicited by activation of the same small afferent nerve fibers that are considered the afferent limb of the reflex pathway that inhibits the respiratory muscles (especially the diaphragm) after upper abdominal surgery (15). Analgesia could then lead to improved performance of the respiratory muscles by partly blocking the afferent pathways of this reflex inhibition. In this regard, Pansard and colleagues (16) reported reduced electrical activity of the diaphragm (recorded from intraoperatively inserted electrodes) on the first postoperative day that was considerably enhanced after thoracic epidural block, which would have interrupted afferent inhibitory inputs. Similarly, the beneficial effect of aminophylline on diaphragmatic function reported after upper abdominal surgery (2, 17), despite normal intrinsic diaphragmatic contractility, may be related to the stimulatory effect of this drug on the respiratory centers, which can increase phrenic nerve output (18).

Interestingly, endogenous opioids have been shown to diminish the ventilatory response to resistive loads (19, 20), whereas in our study the exogenous opioid pethidine increased respiratory muscle performance (presumably through increased ventilatory effort) after upper abdominal surgery. This discrepant effect of opioids might be due to a differential relative efficacy of endogenous and exogenous opioids on the various types of opiod receptors. It could also be due to a preferential "central" action of endogenous opiods as compared with a preferential "peripheral" action of pethidine, with the former effect leading to diminished respiratory output and the latter to either partial blocking of afferent pathways of the reflex inhibition of the respiratory muscles, or simply to increased effort as a result of decreased pain perception.

Although MEP in our study showed the same tendency as Pniff, the observed differences in this variable did not reach statistical significance. This may have been due to the relatively small number of patients in our study, in association with the well known variability of MEP. Furthermore, upper abdominal surgery results in a direct mechanical insult to the abdominal muscles from the surgical incision, which reduces the likelihood of a significant improvement in MEP after pain relief.

In conclusion, we have shown that pain is an important factor contributing to the development of inspiratory muscle dysfunction after upper abdominal surgery, and that systemic analgesia can partly reverse this dysfunction.