Comparison of Effect of Narcotic and Epidural Analgesia on

Postoperative Respiratory Function

Lance Miller, MD, FRCS (C), Montreal, Quebec, Canada

Morris Gertel, MD, FRCP (C), Montreal, Quebec, Canada

Gordon S. Fox, MD, FRCP (C), Montreal, Quebec, Canada

Lloyd D. MacLean, MD, FRCS (C), Montreal, Quebec, Canada

Spence and Smith [I] and Muneyuki et al [2] have demonstrated that 24 hours after upper abdominal surgery patients have higher oxygen tensions when postoperative pain relief is managed by continuous epidural rather than narcotic analgesia. However, these investigators have not shown early differ- ences in the measured mechanics of ventilation de- spite their findings regarding differences in oxygen tension. Therefore, a prospective study of postop- erative respiratory function was undertaken in twenty randomly selected patients older than thirty years of age admitted to hospital for elective cholecystectomy. Ventilatory function was mea- sured for the first 24 hours after surgery using sim- ple tests. A single method of anesthesia was used for all subjects. Postoperative pain relief was achieved with intramuscular meperidine in half the patients and the remainder received intermit- tent doses of epidural lidocaine.

Met hod

None of the patients had significant cardiovascular or respiratory disease as evaluated by history, electrocardi- ogram, and chest radiography.

Before the study was begun, a Wright peak flow meter was calibrated against flow as determined by flow through a bobbin type rotameter. A graph was plotted so that the actual flow could be calculated from the ob-

From Departments of Surgery and Anaesthesia. Royal Victoria Hospital, and McGill University, Montreal, Quebec. Canada. This work was sup- ported by a grant from the Medical Research Council of Canada.

Reprint requests should be addressed to Lloyd D. Maclean, MD, Sur- geon-in-Chief, Royal Victoria Hospital, 667 Pine Avenue, Montreal, Que- bec, Canada.

served flow. Our particular flow meter gave values con- sistently below actual peak flow when plotted at high and low flows on the graph. Vital capacity was measured with a previously calibrated Wright respirometer at- tached to a close-fitting face mask.

Preoperative assessment was made 24 hours before surgery. With the patient in the supine position three recordings of expiratory peak flow and vital capacity were performed. An average value of the three record- ings was then calculated and used as the preoperative value.

Postoperative pain relief was achieved with epidural injections of lidocaine and in group II patients with in- tramuscular meperidine. Respiratory function was as- sessed by measuring expiratory peak flow, vital capaci- ty, and arterial blood gases. These tests were used be- cause they are simple and easily and rapidly performed at bedside.

Immediately on return to the recovery room, the pa- tients were given 28 per cent oxygen by Campbell mask. An arterial blood gas sample was taken 10 to 15 minutes after admission to the recovery room. An arterial blood gas sample was taken during the third to fourth postop- erative hour at which time the patient’s pain had been relieved by either meperidine or an epidural injection of lidocaine. After this, the Campbell mask was removed and the means of three measurements of expiratory peak flow and vital capacity were recorded and ex- pressed as a percentage of the preoperative value.

All the measurements were repeated the following morning, the only difference in technic being that all the patients were breathing room air.

Respiratory function and blood gas samples were all taken by the same investigator (JLM); anesthetics were administered and epidural catheters were inserted by one of two anesthetists (MG and GSF). All blood gas

Vofuma 131, March 1976 291

Miller et al

analyses were performed on an Instrumentation Labora- tory model 113 blood gas system. Significance of the re-

sults was determined by Student t test.

Anesthesia and Analgesia Technic

All patients were premeditated with 100 mg pento-

barbitone, 0.4 mg atropine sulfate, and 50 to 75 mg mep- eridine intramuscularly one hour before scheduled oper- ation. In all group I patients an epidural catheter was in-

serted at a convenient intervertebral space between the twelfth thoracic and third lumbar interspace prior to in-

duction of general anesthesia. Anesthesia was induced with a “sleep dose” of thiopentone. Tracheal intubation

was facilitated by the use of 60 to 100 mg succinylcho- line intravenously. Anesthesia was maintained in all pa-

tients with nitrous oxide, oxygen, intravenous meperi- dine, and d-tubocurarine as required for muscle relaxa- tion. At the and of the operation, curarisation was re-

versed with 1.2 mg atropine and 2.5 mg prostigmine in-

travenously. Postoperatively, epidural analgesia was initiated and

maintained in group I patients with 1 or 2 per cent lido-

Caine with l/200,000 epinephrine given in hourly doses throughout the study period. Postoperative analgesia in

TABLE I Preoperative Status of Patients

in Groups I and II

Per Mean Cent

Mean Age Height Mean Weight Smok-

Group Age (vr) Range (vr) (inches) (pounds) ers

I 49* 11 33to68 64.9?3 141.4* 15 30

II 55 * 12 35 to 74 64.5 i 3 150.8 ? 35 30

Note: f = standard deviation.

TABLE I I Preoperative Respiratory Assessment

of Groups I and II

Mean Expiratory Peak Flow

Rate (L/ Range Mean Vital Group min) (L/min) Capacity (L) Range (L)

I 396i 78 310 to 573 2.25 + 0.7 1.37 * 3.4 II 387 * 93 275 to 530 1.90 f 0.7 1.18 * 2.4

Note: ? = standard deviation.

TABLE I II Duration of Operations and Narcotic Dose

Mean Mean Dose Operative Meperidine

Group Time (min) Range (min) (mg) Range (mg)

I 104*44 55 to 195 67 * 21 40 to 120 II 128 f 54 65 to 245 75 * 26 40 to 120

group II patients was achieved with intramuscular mep- eridine in doses judged adequate by the anesthetist in charge of the recovery room. Group I patients remained in the recovery room until after the study period. Group II patients returned to the ward after the usual period in

the recovery room.

Results

Both groups were comparable in age, height, weight, and smoking habits as shown in Table I. In group I there were six females and four males and in group II there were seven females and three males. Preoperatively there were no significant differences in expiratory peak flow and vital ca- pacity between the two groups. (Table II.) Table III demonstrates that there was no significant dif- ference between the two groups as to the duration of operation or the amount of meperidine received intraoperatively. The operative time was consid- ered as the interval between induction of anesthe- sia and the time the patient left the operating room.

Although the patients were admitted for elective cholecystectomy, some did undergo additional op- erative procedures, which as noted in Table III made no significant difference between the two groups with reference to the operative time or the amo’unt of narcotic required. The operative proce- dures for group I include: cholecystectomy (4 pa-

tients); cholecystectomy and operative cholangio- gram (3); cholecystectomy with common bile duct exploration (1); cholecystectomy with right ingui- nal herniorrhaphy (1); and cholecystectomy with incidental appendectomy and umbilical hernior- rhaphy (1). The operative procedures for group II include: cholecystectomy (7 patients); cholecystec- tomy with common bile duct exploration (2); and cholecystectomy and Nissen fundoplication (1).

Arterial oxygen tension (PaOs), per cent oxygen saturation, and carbon dioxide tension (PaCOs) were comparable in both groups during the imme- diate postoperative period and 3 to 4 hours after surgery. However, on the morning after surgery, patients in group I had significantly higher arterial oxygen tensions (p <O.Ol) and per cent oxygen sat- uration (p <O.OOl) and lower carbon dioxide ten- sions (p <0.04). (Table IV.)

The results comparing the two groups of postop- erative respiratory function tests are illustrated in Table V. There is a significant difference between the two groups in the vital capacity at 3 to 4 hours (p <0.05). The epidural group had vital capacities which were closer to preoperative volumes than

Note: + = standard deviation. the narcotic group. The next morning there was no

292 The American Journal of Surgery

Postoperative Respiratory Function

difference in vital capacity. There was no differ- ence between the groups in expiratory peak flow rate at 3 to 4 hours postoperatively, nor on the fol- lowing morning. The postoperative analgesia re- quirement for group I were 958 f 481 mg (SD) lid- ocaine (mean number of doses, 12.8 i 7.3 [SD]) and for group II, 167 f 99 mg (SD) meperidine (mean number of doses, 2.7 f 1.3 [SD]). The final assessment in the postoperative period, comparing complications and length of hospital stay, showed the mean duration of hospital stay to be 9.2 f 1.8

days (SD) for group I and 11.2 f 5.4 days (SD) for group II. The mean number of days in hospital in- cluded the day of operation and the day of dis- charge. One patient in the narcotic group (group II) who had cholecystectomy and common bile duct exploration remained in hospital for twenty- six days postoperatively but was awaiting place- ment in a convalescent home during the latter part of his hospitalization. One patient in each group had pulmonary complications: a suspected, un- proved, untreated pulmonary embolus in one pa- tient in the epidural group and atelectasis with a small pleural effusion in one patient in the narcot- ic group. Five patients in group I and one in group II required a urinary catheter. There is no signifi- cant difference between the length of hospital stay in the two groups.

TABLE IV Postoperative Blood Gas Analyses

Pa02 (mm Hg) -

Comments

Expiratory peak flow has been shown to be a good measure of ventilatory function [3]. Shortly after surgery, both groups of patients had similar ventilatory measurements. There was no statisti- cal difference in expiratory peak flow in 4 hours after return to the recovery room and after the es- tablishment of postoperative analgesia. The signif- icant difference in vital capacity indicates that those patients with epidural analgesia were capa- ble of taking a deep inspiration on command or at will. However, the difference in ability to deep breathe had no appreciable effect on oxygen ten- sion at that time.

Twenty-four ‘hours after surgery, group I pa- tients had expiratory peak flow and vital capacity measurements that were closer to preoperative levels than were group II measurements; however, although no significant difference existed between the two groups, the epidural group had greater ar- terial oxygen tensions and lower carbon dioxide tensions than did the narcotic group. The lower carbon dioxide tension implies greater minute al- veolar ventilation in group I. The differences in oxygen saturation are of importance since the oxy- gen tension falls on the steep part of the oxygen hemoglobin dissociation curve. These differences

Per Cent Oxygen Saturation PaC02 (mm Hg)

Time Group I

Immediately 128.9 5 57.4 postopera- tively

3 to4 hr 133.4 f 65.9 postopera- tively

24 hr post- 80.2 f 13.6” operatively

Note: t = standard deviation. *p < 0.01. i p < 0.001. z p < 0.04.

Group II

154.4 f 78.0

136.5 2 62.3

63.3 i 16

Group I

97.4 ? 0

97.8 * 0.32

95.0 t 0.67t

Group I I

96.6 f 0.7

97.4 * 0.5

90.8 + 0.9

Group I Group I I

40.1 ? 4.2 40.7 + 3.3

37.4 + 4.5 40.3 + 4.3

33.4 k 3.0$ 36.4 k 3.1

-

TABLE V Changes in Expiratory Peak Flow and Vital Capacity

Internal

Expiratory Peak Flow (per cent of control)

Group I Group II

Vital Capacity (per cent of control)

Group I Group II

3 to 4 hr post- 48.3 ? 13.4 43.6 k 20.1 49.9 r 15.2 36.5 + 12.6

operatively (NS) (p = < 0.05)

1st postopera- 48 i 10.4 42.1 ? 10.2 45.6 2 14.0 42.1 ?- 20.2

tive day (NS) (NS) ______

Note: + = standard deviation; NS = not significant.

Volume 131. March 1976 293

Miller et al

in oxygen saturation occurred in patients who had no preoperative respiratory disease.

We did not compare the alveolar-arterial oxygen difference between the groups as this would pro- vide no more useful comparison than arterial oxy- gen tensions. Since mixed venous blood was not obtained, a comparison of total shunt was not cal- culated because we could not assume comparable arteriovenous differences in oxygen content in the two groups.

The patients in group II were at a disadvantage early in the initial recovery period due to their in- ability to inspire deeply and to open previously closed alveoli. This could have hastened the onset of atelectasis which resulted in decreased ventila- tion/perfusion ratios and decreased arterial oxygen tensions. Epidural analgesia allowed group I pa- tients to breathe deeply or sigh and theoretically minimized the chance of atelectasis. Narcotics are known to decrease the frequency of sighing [4]. These findings are in agreement with those of Spence and Smith [I] who studied a group of pa- tients very similar to ours. They demonstrated the onset of ventilation/perfusion abnormalities as early as 45 minutes after the administration of a narcotic analgesic for pain after cholecystectomy.

Twenty-four hours after surgery, the measured mechanics of ventilation were impaired in both groups to a similar degree, but arterial oxygen ten- sions were superior after epidural analgesia. Spence and Smith [1] did not measure ventilation in the early postoperative period. Our findings demonstrate that 24 hours after surgery, simple tests of respiratory function will not help to identi- fy those patients who might have arterial oxygen desaturation. Patients in group I who maintained an adequate vital capacity in the early postopera- tive period had higher arterial oxygen saturation the day after surgery. Expiratory peak flow was of no value in predicting changes in arterial oxygen tension. The changes in expiratory peak flow, vital capacity, and blood gases could possibly be more pronounced in the elderly or in patients with preexisting lung disease. Evidence for this might be obtained by another study of older patients

with higher closing volumes [5], or a study of pa- tients with diffuse pulmonary disease. An obvious disadvantage of this technic, which might be more troublesome in the elderly, is urinary retention.

Summary

A prospective, randomized comparison was made of the value of meperidine versus epidural analgesia when used for the relief of pain after cholecystectomy in twenty patients without car- diopulmonary disease. Respiratory function was assessed the day before surgery and at 3 to 4 hours and 24 hours after operation by the bedside mea- surement of expiratory peak flow, vital capacity, and arterial blood gases. The two groups of pa- tients were comparable as to age, height, weight, smoking habits, preoperative peak flow, vital ca- pacity, and duration of operation. The arterial oxygen tension and oxygen saturation were signifi- cantly greater and carbon dioxide tension lower in the epidural analgesia group 24 hours after opera- tion. At this time peak flow rates and vital capaci- ty were not different. However, at 3 to 4 hours postoperatively, vital capacity was significantly greater in the epidural anesthesia group. This might account for the differences in arterial blood gases the following day. These findings suggest that epidural analgesia is valuable in the early postoperative period after upper abdominal sur-

gery.

References

1. Spence AA, Smith G: Postoperative analgesia and lung func- tion: a comparison of morphine with extradural block. Br J Anaesth 43: 144. 1971.

2. Muneyuki M. Veda Y, Urabe N. Takeshita H. lnamoto A: Post- operative pain relief and respiratory function in man: com- parison between intermittent intravenous injections of meperidine and continuous lumbar epidural analgesia. An- esthesiology 29: 304, 1966.

3. Wright, BM, McKerron CB: Maximum forced expiratory flow rate as a measure of ventilatory capacity. Br Med J 21: 1041,1959.

4. Egbert LD, Bendixen HH: Effect of morphine on breathing pat- tern; possible factor in atelectasis. JAMA 168: 485, 1964.

5. Anthonisen NR, Danson J, Robertson PC, Ross WRD: Airway closure as a function of age. Resp Physiol8: 58, 1969.

294 The Amdcan Journal of Surgery