Survey of postoperative analgesia following ambulatory surgery.

BACKGROUND: The quality of pain relief during the first 48 hours following ambulatory surgery has been poorly documented. This questionnaire study was performed to evaluate the nature and severity of pain after the patient leaves the hospital. METHODS: 1100 patients in the age group 5-88 years who underwent ambulatory surgery during a period of 6 months were asked to complete a questionnaire 48 h after the end of the operation. In the case of children, parents were asked to complete a similar questionnaire. The questions were related to pain experienced during the first 48 h after surgery and to the nature and severity of postoperative complications. RESULTS: A total of 1035 out of the 1100 patients returned the questionnaire, 94.1%. Overall the majority (65%) of patients had only mild pain at home; however, patients undergoing certain types of surgery had moderate-to-severe pain: inguinal hernia surgery (62% patients), orthopaedic surgery (41%), hand surgery (37%) and varicose vein surgery (36%). In these patients the severity of pain did not decrease during the 2-day study period. About 10% patients had more severe pain than they had anticipated, and 20% had difficulty in sleeping at night due to severe pain. Despite this, over 95% of patients were satisfied with management of postoperative pain. Nausea (20%), tiredness (20%) and vomiting (8%) were the commonest complications reported during the first 48 h. A significant association was found between the administration of a general anaesthetic and the incidence of nausea postoperatively. A large number of patients were alone at home after the operation (28.4%); some (3.8%) had no access to a relative or friend in case of need. CONCLUSION: Our results show that about 35% of day-surgery patients experience moderate-to-severe pain at home in spite of analgesic medication. About 20% of patients had sleep problems due to severe pain. However, only 5% of patients were dissatisfied. Better analgesic techniques are necessary for patients undergoing certain types of surgery. Patient information and follow-up routines need to be improved.

Continuous epidural ropivacaine 0.2% for analgesia after lower abdominal surgery.

The purpose of this study was to determine whether a lumbar epidural infusion of ropivacaine 0.2% would provide effective analgesia with an acceptably low incidence of motor blockade and side effects after lower abdominal surgery. After combined general and epidural anesthesia and surgery, 125 patients were randomly assigned to receive either saline or ropivacaine 0.2% at a rate of 6, 8, 10, 12, or 14 mL/h (Groups R6, R8, R10, R12, and R14, respectively) for 21 h. Supplemental analgesia, if required, was provided with intravenous patient-controlled analgesia with morphine. Data were collected at 4, 8, and 21 h, and included morphine consumption, pain scores at rest and with coughing, motor and sensory block, and adverse events. Cumulative morphine consumption was less in Groups R10, R12, and R14 compared with the saline group. At 4 h analgesia was better among patients receiving ropivacaine, but at 21 h pain scores were identical. Sensory blockade at 8 and 21 h was greater in the ropivacaine groups compared with the saline group. Approximately 30% of R8, R10, and R12 patients, and 63% of R14 patients had demonstrable motor block of the lower limbs at 21 hours. We conclude that lumbar epidural ropivacaine 0.2% reduces parenteral morphine requirements but has little effect on pain scores and may be associated with motor blockade.

Evaluation of epidural sensory block by thermal stimulation, laser stimulation, and recording of somatosensory evoked potentials.

BACKGROUND AND OBJECTIVES: The existence of differential sensory block during epidural analgesia has been confirmed by some authors and disputed by others. This study attempts to elucidate this issue by using quantitative methods for evaluation of sensory block. METHODS: A single epidural injection of 20 mL 0.5% bupivacaine with epinephrine was administered at the L1-T12 level in 11 male volunteers. Sensory block was evaluated by two qualitative (pinprick and light touch) and two quantitative methods (thermal stimulation with Thermotest [Somedic, Stockholm, Sweden] and argon laser stimulation). For measurement of motor block in the lower extremities and in the rectus abdominis muscle, quantitative methods were used. Sensory block was also assessed by somatosensory evoked potentials recorded during electrical and laser stimulation at the most cranial analgesic dermatome (loss of sharpness in pinprick perception) and the anesthetic dermatome L2 (loss of light touch perception). RESULTS: The zone of anesthesia was smaller than the zone of any other investigated variable. The cranial spread of analgesia and motor block was lower than that of laser-assessed block. Partial block of laser perception and thermal perception lasted longer than analgesia and motor block. No consistent segmental or temporal differences were found between the Thermotest and laser methods. During epidural block, prolongation of latencies and reduction in amplitudes of somatosensory evoked potentials produced at the most cranial analgesic dermatome did not differ significantly from those produced at the anesthetic dermatome. CONCLUSIONS: No differential block of small nerve fibers was found during epidural analgesia by Thermotest and argon laser stimulation. Recording of somatosensory evoked potentials did not demonstrate significant difference between responses from the sites with most superficial and with most intense sensory block.

The effect of continuous epidural infusion of ropivacaine (0.1%, 0.2% and 0.3%) on nerve conduction velocity and postural control in volunteers.

BACKGROUND: Continuous epidural infusions of local anaesthetics have become increasingly popular in postoperative pain treatment, especially as they permit early mobilisation. Ropivacaine is a promising new agent which induces more pronounced sensory than motor blockade. This study was focused on the influence of continuous epidural infusion of ropivacaine on impulse conduction in large nerves (by measurement of F and H latencies), and on the subjects' ability to maintain postural control during mobilisation. METHODS: Healthy male volunteers received 0.1%, 0.2% or 0.3% ropivacaine, and bupivacaine 0.25% was used as reference. A bolus epidural injection of 10 ml of the drug, at L2/3 level, was followed by continuous infusion at 10 ml/h for 21 h. Motor blockade was assessed by mechanical measurements of force during big toe flexion and by recording of F latency. Sensory blockade was monitored by pin-prick and Thermotest methods, and by H latency recording. The subjects' ability to perform a postural test was evaluated by posturography. RESULTS: The F and H latencies became prolonged/abolished dose-dependently. With ropivacaine, F latency recovered significantly later than motor function (P = 0.0002), and H latency recovered later than normal pin-prick perception (P = 0.0006). However, the duration of partial blockade of thermoperception was comparable to that of H latency prolongation. Posturographically, the subjects receiving 0.1% ropivacaine differed significantly from all others (P < 0.001) in that they were able to maintain postural control during the infusion. The recovery period after termination of infusion was significantly shorter with ropivacaine than with bupivacaine for all measured variables. CONCLUSION: Recovery of postural control with 0.2% and 0.3% ropivacaine is significantly faster than with bupivacaine 0.25%. H latency recording allows detection of epidural blockade intensity that does not prevent subjects from performing postural tests.

The effect of continuous lumbar epidural infusion of ropivacaine (0.1%, 0.2%, and 0.3%) and 0.25% bupivacaine on sensory and motor block in volunteers: a double-blind study.

BACKGROUND AND OBJECTIVES: In animal studies, ropivacaine has shown more pronounced sensory block than motor block, which makes it an interesting drug for postoperative pain relief. The aim of this study was to investigate the dose response of sensory and motor block during continuous epidural infusion of 0.1, 0.2, or 0.3% ropivacaine in volunteers in a double-blind manner. Bupivacaine 0.25% and isotonic saline were used as reference and control, respectively. METHODS: Each treatment group consisted of eight healthy men. After a bolus dose of 10 mL at the L2-L3 interspace, the solution in question was infused at 10 mL/h for 21 hours. Sensory block was evaluated by the pinprick, light touch, and Thermotest methods. Motor block was measured by the Bromage scale, by average rectified electromyography in abdominal muscles, and by mechanical measurement of isometric muscle force in the lower extremities. Mobilization of the subjects was attempted throughout the investigation. RESULTS: The number of blocked dermatomes (evaluated by pinprick) with 0.1% ropivacaine was significantly smaller than with the other test solutions (P = .002-.0008). Motor block was minimal with 0.1% ropivacaine, so that all subjects could be mobilized; it was moderate with 0.2 and 0.3% ropivacaine and most intense with 0.25% bupivacaine. The regression phase was significantly shorter with all three concentrations of ropivacaine than with bupivacaine (P < .01). CONCLUSIONS: Ropivacaine 0.1% produced limited analgesia and minimal motor block, so that ambulation was possible throughout the investigation. With 0.2 and 0.3% ropivacaine, analgesia was more extensive, and motor block was considered moderate. Ropivacaine 0.2% should be evaluated for future postoperative pain treatment.

Pharmacokinetics of ropivacaine and bupivacaine during 21 hours of continuous epidural infusion in healthy male volunteers.

The aim of the present study was to evaluate the pharmacokinetics of ropivacaine and to compare the results with those of bupivacaine during prolonged epidural infusion. Ropivacaine 1, 2, or 3 mg/mL (0.1%, 0.2%, or 0.3%), bupivacaine 2.5 mg/mL (0.25%), or placebo (sodium chloride 0.9%) was given randomly and in a double-blind manner to five parallel treatment groups (37 healthy volunteers) as a continuous epidural infusion for 21 h. A 10-mL epidural bolus dose was first given, and the epidural infusion was started immediately afterward. The subjects received 10 mL/h corresponding to infusion rates of 10, 20, or 30 mg/h ropivacaine and 25 mg/h bupivacaine, respectively. Peripheral blood samples for measurements of ropivacaine or bupivacaine were taken during a 25-h period. The total plasma concentration increased continuously but seemed to reach a plateau (C5-10h) after approximately 5 h infusion, remaining fairly constant up to approximately 10 h after the start of administration. The C5-10h values were proportional to the dose of ropivacaine and were estimated as 0.3, 0.6, and 0.9 mg/L, and for bupivacaine as 0.7 mg/L. During the subsequent infusion the plasma concentration increased, with maximum plasma levels at the end of the infusion and with corresponding values of 0.4, 0.9, 1.2, and 0.9 mg/L. The highest individual plasma concentration was 1.7 mg/L (20 mg/h), and no patient showed signs of toxic systemic plasma levels. The free concentrations also increased continuously during the infusion. The free fraction was independent of the dose (6.1% for ropivacaine and 4.8% for bupivacaine).(ABSTRACT TRUNCATED AT 250 WORDS)

Risk of catheter migration during combined spinal epidural block: percutaneous epiduroscopy study.

Combined spinal epidural (CSE) block with the needle-through-needle technique has become increasingly popular during recent years. However, the risk of epidural catheter penetrating dura mater through the hole made by the spinal needle (migration) is a major concern. In 15 fresh cadavers a percutaneous epiduroscopy technique with a rigid epiduroscope and video recording was used to assess the risk of catheter migration when a CSE block is performed. The experimental sequence included (a) one dural hole made by the spinal needle, (b) multiple (five) dural holes made by the spinal needle, and (c) a dural hole made by Tuohy needle. Twenty-four experimental sequences were performed in the lumbar region. Four sequences were failures due to technical problems. In the remaining 20 cases, the anatomic structures in the epidural space were recognized easily. The epidural space appears to be only a potential space, kept open either by epiduroscope or by repeated injections of air or saline. The dural holes made by Tuohy and spinal needles, and the ease of difficulty of catheter penetration through these holes, were clearly visible. Extensive tenting of the dura was seen in all subjects. It was impossible to force an 18-gauge epidural catheter through the dural hole after a single dural puncture made by a 25-gauge spinal needle. After multiple (five) dural punctures with the spinal needle, the epidural catheter penetrated the perforated dura in 1 of 20 cases. The epidural catheter penetrated the dural hole made by the Tuohy needle in 9 of 20 cases. The distribution of fat, rather than any dorso median connective tissue band, influences the course of epidural catheter in epidural space.(ABSTRACT TRUNCATED AT 250 WORDS)

F response and H reflex for monitoring nerve block during epidural analgesia with ropivacaine.

The effect of a single epidural injection of ropivacaine on the motor and sensory function controlled from the L5/S1 level was investigated in 28 male volunteers. Concentrations of 1%, 0.75%, or 0.5% ropivacaine, 20 mL, administered at the L2/3 level were studied. Motor function was assessed quantitatively (measurement of muscle force by mechano-transducers), and sensory function by the pinprick method. In addition, F response and H reflex, tests which measure the conduction velocity in the central parts of peripheral nerves, were used. Epidural ropivacaine caused dose-dependent prolongation of the latencies of both these variables. F response latency recovered significantly later than motor function measured by mechano-transducers in the two lower concentration groups. H reflex latency recovered significantly later than sensory function assessed by the pinprick method in all three concentration groups. The time needed for recovery of F and H latencies was not significantly longer than the time from epidural injection to mobilization. At the time when the subjects could go through the mobilization procedure, 12 of 28 subjects were not completely recovered. In 5 of these 12 subjects, the H reflex latency was persistently prolonged at the end of the investigation, long after the subjects felt "normal" again. On follow-up recordings 5 mo later, the baseline latency had been regained in all five subjects. We conclude that F response and H reflex latencies are good indicators of the inhibition of nerve impulse conduction induced by epidural analgesia.

Blockade of the abdominal muscles measured by EMG during lumbar epidural analgesia with ropivacaine--a double-blind study.

A single shot of 20 ml of 1%, 0.75% or 0.5% ropivacaine was administered epidurally (at L2/3 level) to 30 volunteers, in a double-blind manner. The blockade of the rectus abdominis muscle was measured quantitatively by registration of the average rectified electromyographic signal (AREMG) at the T7, T9 and T11 motor segmental levels and with a qualitative test for blockade of the rectus abdominis muscle (the so-called RAM test). The maximal cranial spread of analgesia, evaluated by the pin-prick method, was not significantly different for the three concentrations (T8-T10 dermatome; median value). The intensity of motor blockade, measured by the AREMG method, increased progressively from the T7 segment and caudally with all three concentrations. The blockade was partial (i.e. 85-25% of baseline AREMG activity was present at its maximum) in all subjects. When the effect of the three concentrations of ropivacaine was compared at the same segmental level, the intensity and duration of maximal motor blockade seemed to be dose-dependent, but the difference was not statistically significant. The total duration of motor blockade was shorter with the 0.5% solution than with the higher concentrations. The AREMG method gave a more exact and graded picture of blockade of the rectus abdominis muscle than the RAM test. The duration of sensory blockade did not outlast motor blockade at any level. In half of our subjects the maximal spread of sensory blockade was either equal to or higher than the spread of partial motor block. In the other half, this relationship was reversed--the maximal cranial level of partial motor block was 1-4 segments higher than the maximal level of analgesia.

Epidural anesthesia with 0.5% bupivacaine: influence of age on sensory and motor blockade.

To determine the influence of age on epidural blockade, 10 elderly male patients (mean age, 73 yr) and 11 young male volunteers (mean age, 25 yr) received epidural anesthesia with 20 mL of 0.5% bupivacaine with epinephrine at the L2-3 interspace. Median cephalad analgesia level (pinprick) was T-4 in the elderly and T-10 in the young subjects (P less than 0.01). The duration of analgesia at the L-1 to S-3 segments was about 60 min shorter in the older patients than in the young volunteers (P less than 0.05). Motor blockade was evaluated quantitatively by isometric muscle force measurements in the lower extremities and by electromyographic recordings from the abdominal and quadriceps muscles. Motor blockade in the lower extremities was also assessed qualitatively by the Bromage scale. The intensity of motor blockade was not influenced by age. The mean duration of maximal motor blockade of the lower abdominal muscles was 69 min in elderly patients and 140 min in young subjects (P less than 0.05). The corresponding mean duration in the lower extremities was approximately 1.5 h in the older patients and 3 h in the young subjects (P less than 0.01). The elderly patients could be mobilized 2 h earlier than the young volunteers. Duration of Bromage grade 1 was about 1.5 h shorter in the elderly than in the young subjects (P less than 0.01). Not until 1.5-2 h after attainment of Bromage grade 0 was the muscle force in the lower extremities restored to 90% of baseline value. The authors conclude that cephalad spread was more extensive and duration of epidural blockade was shorter in elderly patients versus younger volunteers.

Sensory and motor blockade during epidural analgesia with 1%, 0.75%, and 0.5% ropivacaine--a double-blind study.

Levels of sensory (pinprick) and somatic motor blockade were measured in a double-blind study of 30 volunteers given single epidural injections of 1%, 0.75%, and 0.5% ropivacaine. Onset of analgesia was rapid with all concentrations (7-10 min). Maximal levels of analgesia were established 60 min after injection, with no significant differences in the maximal median cephalad spread. Duration of analgesia at the T-12 level and total duration were significantly longer with 1% and 0.75% than with 0.5% ropivacaine. Motor blockade was assessed by a quantitative method (measurements of isometric muscle force) and a qualitative method (modified Bromage scale). Onset of motor blockade measured by the quantitative method was significantly slower with 0.5% ropivacaine than with the higher concentrations. Maximal muscle weakness occurred 1-1.5 h after injection with all three concentrations. With increase in ropivacaine dose from 100 to 200 mg, the intensity and duration of motor blockade increased. Muscles involved in knee extension were blocked most, those of plantar flexion least. Recovery of motor function, assessed by the above-mentioned quantitative method, occurred simultaneously with the recovery of pinprick perception. Motor blockade registered by Bromage scale showed a slower onset for 0.5% ropivacaine than for the higher concentrations. Mean durations of grade 1 and 2 block were longest for the 1% solution. Motor blockade described by the Bromage scale showed only the first part of the regression phase. Full recovery of muscle strength (Bromage scale = 0) was attained 1.5-2.5 h earlier than assessed by the quantitative method. No adverse effects were registered.

Influence of intravenous local anaesthetics upon quadriceps muscle function.

This study was undertaken to assess the effects of intravenous administration of mepivacaine and etidocaine on muscle function. Seven male volunteers were given mepivacaine (5 mg/kg) and etidocaine (50 mg) intravenously, on separate occasions. A reference group of 11 male volunteers received 0.9% saline solution intravenously. Muscle function was tested by measurements of isometric muscle force of knee extension and by quantitative electromyographic (EMG) recordings from the quadriceps muscle during knee extension at different degrees of isometric muscle force. At the end of the mepivacaine and etidocaine infusions, the mean venous plasma concentrations of the two anaesthetic agents were 2.9 and 1.2 micrograms/ml, respectively. The muscle strength remained unchanged during infusion of the two local anaesthetics. Mepivacaine had a minor effect on the mean rectified EMG amplitudes at the end of the infusion at maximal voluntary muscle contraction, but no such effect was observed at submaximal knee extension force. However, at the plasma concentrations mentioned above, the clinical influence of intravenous infusion of the local anaesthetics on muscle function was negligible.

Motor and sensory blockade after epidural injection of mepivacaine, bupivacaine, and etidocaine--a double-blind study.

In a double-blind study of epidural anesthesia, 30 young volunteers were given either 2% mepivacaine (400 mg), 0.5% bupivacaine (100 mg), or 1.5% etidocaine (300 mg), all solutions containing epinephrine (1:200,000). The spread of analgesia was equal in the groups, whereas the longest duration was noted in the etidocaine and bupivacaine groups. With use of a method for determining muscle force, motor blockade during anesthesia was recorded quantitatively for hip flexion, knee extension, and plantar flexion of the big toe. Onset of motor blockade was significantly more rapid with etidocaine than with bupivacaine and mepivacaine. All subjects given etidocaine developed complete motor blockade, but with the other local anesthetics 5%-33% of the initial muscle force remained. The least motor blockade was found in the L5-S2 segment (plantar flexion of the big toe). The duration of maximal motor blockade varied between 60 min (mepivacaine) and 360 min (etidocaine). With each of the three local anesthetics, motor function returned simultaneously in the three muscle groups tested. Complete restoration of muscle function occurred significantly later for etidocaine (600 min) than for bupivacaine (360 min) and mepivacaine (180 min). With etidocaine, the motor blockade outlasted the sensory blockade by 150 min. The Bromage scale corresponded to the motor blockade only during the first half of the regression phase. Not until 1-3 h after attainment of Bromage grade 0 was the muscle force of all movements restored (90% of control values).

Motor blockade and EMG recordings in epidural anaesthesia. A comparison between mepivacaine 2%, bupivacaine 0.5% and etidocaine 1.5%.

In a double-blind study young volunteers randomly received 20 ml of mepivacaine 2%, bupivacaine 0.5% or etidocaine 1.5% epidurally, all solutions with adrenaline. The mean cephalad spread of pin-prick analgesia was equal (T10) in the groups, but the duration was longest for bupivacaine and etidocaine. The motor blockade of the rectus abdominis muscles was assessed quantitatively by rectified integrated electromyographic recordings (RIEMG) and as number of turns in EMG recordings [changes in the direction (rise/fall) of the EMG; TURNS] from three different segmental levels, T7, T9 and T11. The motor blockade of the quadriceps muscles was estimated by EMG recordings simultaneously with muscle force measurements at maximal isometric knee extension. Motor blockade was also evaluated by the Bromage scale. There was good correlation (correlation coefficient 0.91) between RIEMG values and muscle force in knee extension during epidural anaesthesia. TURNS showed a non-linear relationship to isometric force during epidural anaesthesia and added no further information. At the lower parts of the abdomen (T11), etidocaine gave more profound and longer motor blockade than mepivacaine. For quadriceps muscle function, motor blockade was almost complete with all three local anaesthetics; the duration of maximum motor blockade was short (45-60 min) for mepivacaine, but about 5 h with etidocaine. At the time when the Bromage scale indicated complete regression of motor blockade, the muscle force of knee extension was only 30% and the quadriceps RIEMG 35% of control values and 1-3 h remained until the time of mobilization.(ABSTRACT TRUNCATED AT 250 WORDS)

Evaluation of motor blockade by isometric force measurement and electromyographic recording during epidural anaesthesia--a methodological study.

Methods for assessing motor blockade by means of isometric force measurements and surface electromyographic (EMG) recordings in the lower extremities and abdominal wall were evaluated in 30 volunteers. The coefficients of variation were 10% for force measurements and 14% and 20% for average rectified EMG (RIEMG) recordings over the quadriceps muscle and abdominal muscles, respectively, and 8% overall for TURNS (the number of changes in the sign of the direction of the EMG signal). Seven of the 30 volunteers received epidural anaesthesia with 20 ml of mepivacaine 2% with adrenaline. The mean maximal cephalad analgesic level was T9. At that abdominal segment, RIEMG showed a reduction of 50% and TURNS of 20%. Isometric force and RIEMG recorded simultaneously in the quadriceps muscles during epidural anaesthesia displayed a linear relationship with a correlation coefficient of 0.91. TURNS was insensitive to force variations above 60% of maximum voluntary contraction. During the regression phase, 90% of both the initial force and RIEMG value was noted 180 min after the epidural injection. It is concluded that recording of RIEMG is a good method for quantitative assessment of motor blockade during epidural anaesthesia.

Venous blood concentration of lidocaine after nasopharyngeal application of 2% lidocaine gel.

In 34 subjects undergoing topical anaesthesia with lidocaine gel (2% xylocaine gel), the concentrations of lidocaine base were measured in repeated venous blood samples. Twenty-three patients (Group I) were given 20 ml of 2% lidocaine gel (400 mg of lidocaine) and six subjects (Group II) 40 ml of 2% lidocaine gel (800 mg of lidocaine) nasopharyngeally, the gel reaching the pharynx being swallowed. Five minutes after administration of the gel, a gastric tube was introduced via the nasal cavity. In six patients (Group III), the bladder was catheterised in addition to insertion of a gastric tube. Before the catheter was introduced, these patients were given lidocaine gel nasopharyngeally (400 mg) and endourethrally (400 mg). After the introduction of the gastric tube, all subjects (except four volunteers in Group II) were given general anaesthesia. Before intubation, the patients in Group III also received lidocaine spray laryngotracheally (50 mg; 10% xylocaine spray). The initial absorption of lidocaine was rapid, although the blood concentrations were low. The mean peak concentrations (Cmax) of lidocaine in the three groups were 0.57, 1.39 and 0.73 micrograms/ml, respectively. The blood concentration in Group II was significantly higher than those in Groups I and III. The mean length of time between the nasopharyngeal application of lidocaine gel and the time when Cmax was reached (tpeak) was the same in all three groups (60-70 min).(ABSTRACT TRUNCATED AT 250 WORDS)

Epidural morphine for postoperative pain relief: a comparative study with intramuscular narcotic and intercostal nerve block.

The relatively new technique of epidural morphine analgesia was compared with two well established method of pain relief in 90 patients undergoing gallbladder surgery and divided randomly into three groups of 30 patients each. The first group received intramuscular narcotic analgesic ketobemidone, the second group was given 0.5% bupivacaine-epinephrine intercostal nerve block, and the third group received a single dose of 4 mg of epidural morphine for postoperative pain relief. The mean duration of analgesia after ketobemidone was 5.5 hours, and after intercostal block 11 hours. Of the patients given epidural morphine, 40% did not require further analgesia after the initial injection; the remaining patients in this group were pain free for a mean duration of 19 hours. The mean reduction in postoperative peak expiratory flow was most marked following ketobemidone and least after epidural morphine. Postoperative changes in PaO2 and PaCO2 reflected the changes in peak expiratory flow. Plasma levels of morphine after epidural injection were so low that a regional spinal analgesic action of epidural morphine appeared more likely than a systemic effect. Delayed respiratory depression was not encountered after epidural morphine. It is concluded that a single dose of 4 mg of epidural morphine provides excellent regional analgesia of long duration without drowsiness or circulatory of respiratory depression thus facilitating early ambulation. The technique is superior to more common methods of pain relief after gallbladder surgery, e.g., intercostal nerve block and intramuscular narcotics.