1987:flh:647-53

ANESTH ANALG

647

Caudal Morphine for Postoperative Analgesia in Children:

A Comparison with Caudal Bupivacaine and Intravenous Morphine
Elliot J. Krane, MD, Lawrence E. Jacobson, MD, Anne M. Lynn, Donald C. Tyler, MD
KRANE EJ, JACOBSON LE, LYNN AM, PARROT C, TYLER DC. Caudal morphine for postoperative analgesia in children: a comparison with caudal bupivacaine and intravenous morphine. Anesth Analg 1987;66:647-53.

MD,

Carol Parrot,

MD,

and

W e compared the efficacy, duration, and side effects of preservative-free morphine injected into the caudal space in children, with caudal bupivacaine and with intravenous morphine administration for relief of postoperative pain. Forty-six children, ages 1-16 yr, were randomly assigned to receive intravenous morphine (control group), caudal bupivacaine (0.25%, 1 mllkg), or caudal morphine (0.5 wixlml, 0.1 mglkg). In half the patients given caudal morphine, the morphine was mixed with a dose of lidocaine adequate to produce sacral analgesia, to confirm correct caudal injection of the morphine. Caudal injections were performed at the end of surgery. Time until the first required postoperative intravenous morphine dose was recorded for

each patient. The duration of analgesia U M S significantly greater with caudal morphine (median 12 hr, P < 0.02) than with caudal bupivacaine (median 5 hr), and both were greater than with intravenous morphine in control patienfs (median 45 min). Urinary retention, pruritis, and nausea appeared with slightly greater frequency in the caudal morphine group, but no delayed respiratory depression occurred. Caudal morphine (0.5 mglml, 0.1 mglkg) provided 8-24 hr of analgesia in children without a significantly greater incidence of side effects than caudal hupivacaine or intravenous morphine.
Key Words: ANALGESICS, NARCOTICS-morphine.
ANALGESIA-postoperative. ANESTHESIA-pediatrics. ANESTHETIC TECHNIQUES-caudal. ANESTHETICS, LOCAL-bupivacaine, PAIN-postoperative.

Orthopedic surgery and genitourinary surgery are often associated with appreciable postoperative pain in children. Although the management of acute postoperative pain has been the focus of many clinical studies in adults (1-3), less attention has been given to the management of pain in children, even though such pain is no less and is frequently undertreated (4,5). Recently, the use of preservative-free morphine injected into the lumbar epidural or subarachnoid space has proven successful in children (6-8). However, correct placement of a lumbar epidural needle may be a difficult undertaking in a small child, and one study has shown a high incidence of accidental dural puncture (8). In contrast, caudal injection is a simple
Presented in part to the American Academy of Pediatrics (1985) and the American Society of Anesthesiologists (1986). Funded by NIH Grant #507 RR 05655-16. Received from the Departments of Anesthesiology and Pediatrics, University of Washington School of Medicine and Children's Hospital and Medical Center, Seattle, Washington. Accepted for publication February 16, 1987. Address correspondence to Dr. Krane, Department of Anesthesiology RN-10, University of Washington School of Medicine, Seattle, WA 98195.
0 1987 by the International Anesthesia Research Society

technique to perform in children, thus explaining the recent popularity of caudal injection of local anesthetics for operative anesthesia and postoperative analgesia in children (9-11). Inadvertent intravascular administration of local anesthetics during caudal injection in adults may be associated with convulsions or cardiovascular collapse, and the latter has recently been reported in a child (12). The use of preservativefree morphine might offer several advantages over the use of local anesthetics when administered into the caudal space. Among these advantages would be longer duration of effect, absence of motor and sympathetic block, and less severe consequences from accidental intravascular administration. The epidural space in small children has spongy, gelatinous lobules and distinct spaces, as opposed to the densely packed fat lobules and fibrous strands that characterize the mature epidural space (13). This difference favors rapid longitudinal spread of drugs within the juvenile epidural space and may make caudally administered preservative-free morphine effective in treating postoperative pain in children. The purpose of this investigation was prospectively to

648

ANESTH ANAL(.
1987;66:647-53

KRANE ET AL.

Table 1. Pain Scores Used by Nurse Observers to Evaluate a Subject's Need for Supplemental Intravenous Morphine
1

3 Calm or asleep

Laughing, euphoric

Happy, contented, pl'lyful

Mild-moderate pain: crying, grimacing, restlessness; can distract with toy, food, parent

Severe pain: crying, screaming; inconsolable

evaluate the duration of postoperative analgesia and the incidence of side effects after caudal preservativefree morphine, compared with postoperative caudal bupivacaine and with intravenous morphine, using a double-blinded protocol and close, in-hospital observation.

Methods
Approval of our Institutional Review Board was obtained. After informed parental and (where appropriate to age) patient consent, 46 children, ages 1-16 yr, ASA physical status 1 or 2, and undergoing elective genitourinary or lower extremity orthopedic surgery, were randomly assigned to the control group, or to receive a postoperative caudal injection of either bupivacaine or preservative-free morphine, as described below. Preoperative or intraoperative narcotics were not administered to any patient. Halothane or isoflurane, with or without nitrous oxide, was used for operative anesthesia. In patients randomized to receive caudal injections, the injections were performed by one of the investigators at the conclusion of surgery, except for two patients receiving caudal morphine in whom injections were made before surgery because application of a spica cast followed surgery. Caudal injections were made with the child in the lateral position with the hips and knees flexed. The skin over the coccyx and sacrum was cleansed with povidone-iodine solution and alcohol. After palpation of the sacral cornua, a 22- or 23-gauge needle was placed into the sacral hiatus, identifying the epidural space by loss of resistance as the needle passed through the sacral ligament. After failure to aspirate cerebrospinal fluid or blood, and a negative test dose in the bupivacaine group, the drug solution was injected. Patients were then extubated while deeply anesthetized and were taken to the postanesthesia recovery room while asleep. ?'he patients, their families, and the nurses assessing pain were not aware of the treatment group assignment. Treatment groups were as follows: Intravenous Morphine Group (Control, n = 15): Children in this group did not receive caudal

injections. Morphine sulfate, 0.05-0.2 mgkg, was administered postoperatively every 2 hr as needed for analgesia. Caudal Bupivacaine Group ( n = 15): Children in this group received a postoperative caudal injection of 0.25% bupivacaine with 1:200,000 epinephrine, 1 ml/kg (maximum dose, 25 ml). Intravenous morphine was later given as described below. Caudal Morphine Group ( n = 16): Patients in this group were subdivided into two sections. Eight patients received postoperative caudal injection of preservative-free morphine alone (1 mg/ml diluted to 0.5 mg/ml with normal saline), in a dose of 0.1 mg/kg. In order to confirm proper placement of the injecting needle in the caudal epidural space, the remaining eight patients received postoperative caudal injection of preservative-free morphine, 1 mg/ml, in a dose of 0.1 mgkg mixed with 1%lidocaine, 0.25 ml/kg. The lidocaine dose was calculated as being adequate to produce anesthesia in sacral dermatomes (10). Anesthesia in sacral dermatomes in the postanesthesia recovery room confirmed caudal injection. Intravenous morphine was given as described below. For the first 24 hr after the operation each patient was monitored for respiratory depression with a chest wall impedance monitor to detect apnea, and with hourly determination of respiratory rates. Naloxone and resuscitation equipment were available at each bedside for 24 hr after surgery. When awake in the recovery room, patients in the bupivacaine and morphine-plus-lidocaine groups were evaluated by an investigator for anesthesia to pinprick in sacral dermatomes. Two patients given caudal bupivacaine and one patient given caudal morphineplus-lidocaine did not have a sensory block and were therefore excluded from the data analysis, thus reducing the sample sizes to 13 in the caudal bupivacaine group and 15 in the caudal morphine group. The efficacy of postoperative analgesia induced by caudal bupivacaine and caudal morphine was compared with the efficacy of intravenous morphine (in

CAUDAL MORPHINE FOR ANALGESIA IN CHILDREN

ANESTH ANALG
1987;66:6.17-53

649

Table 2 . Ages, Types of Operations, a n d Treatment G r o u p Assignment of Experimental Subjects Age (yr) Control group (intravenous morphine only)
1 2 3 1 5 6
I

Operation" Bunionectomy Osteotomy Osteotomy Osteotomy Hypospadias Tibial tumor excision Osteotomy Osteotomy Osteotomy Hypospadias Osteotomy Teno tomy Osteotomy Tibial tumor excision Hypospadias

Duration of analgesia (hr)

0.5 11.7 5. h 0.8 1.8 0.7 1.5 5.3 (1.2 >24 0.5 0.3 0.5 0.8 5.5 2.0 0.8

14 16 5 14 2 15

3
1 9 10 9 3 2 13 1 7.8

8 9

101' 11 12 13 14 15 Mean Median

Caudal bupivacaine group 16" 17 18 19 20 21 22 23 24 25 26 27 28 Mean Median Caudal morphine group without lidocaine 29 30" 31 32 33 34 35 36 Mean Median With lidocaine 37 38 39 40 41 42 43 Mean Median Morphine group mean Morphine group - . median

3 5 12 2 14 2 2 10 3 5 10 2 10 6.2 -

Hypospadias Hypospadias Osteotom y Osteotomy Osteotomy Osteotomy Hypospadias Osteotom y Osteotomy Hy pospadias Osteotomy Osteotomy Open fracture reduction

>24
3.8
9.8

9.9 4.5 50 3.9 3.7 10.0 10.3 11.0 5.2 3.5 6.7 5.2

13 7 9 1 11 1 6 16 -

Osteotomy Hypospadias Ten0 tomy Osteotomy Osteotomy Osteotomy Osteotom y Femoral bone graft

20.5 >24 9.7 5.2 4.0 6.8 12.1 10.9 9.9 10.3

2 9 4 9

5
11 12

7.7

Osteotom y Femoral Bailey rods Osteotomy Femoral bone graft Hypospadias Femoral Bailey rods Osteotomy

12.1 15.7 18.0 16.6 23.6 4 16.7 14.4 16.6 12.6 12.1

<Tenotomy denotes open adductor tenotomy; osteotomy denotes pelvic, femoral, tibia1 osteotomv or club foot repair with osteotomies. ''Patient excluded from calculation of mean duration of analgesia.

650

ANESTH ANAL(; 1937;66:647-53

KRANE ET AL.

1
v)

I-

z
LL

I V MORPHINE

1 a
a
I -

BUPIVACAINE
I -

0
0

z w

z w
a

a W

n
TIME (HOURS)

o!
0

12

16

20

24

TIME (HOURS)
Figure 1 . Percentage of patients in each group who did not require supplemental intravenous morphine during the 24-hr observation period. One patient in each group required no intravenous morphine; these three patients do not appear on the graph. The median times to first intravenous morphine dose for each experimental group are marked by an asterisk, which also denotes P < 0.05.

Figure 2. Proportion of patients from each group among those patients who did not require supplemental intravenous morphine 5,10, and 15 hr after their operations. N indicates the total number of such patients with continuing analgesia at each time period; the asterisk denotes P < 0.05 for caudal morphine compared with the other groups (x2test).

analgesia with patient age and weight. P < 0.05 was considered statistically significant.

Results
control subjects) by measuring the duration of time patients were pain-free after surgery, as reflected by the time before the first dose of supplemental parenteral narcotic was needed. In order to standardize criteria for the administration of the first and all subsequent doses of postoperative analgesics, nurses scored pain using a structured observation scale (Table 1)at 30-min intervals while in the recovery room, and every 2 hr thereafter for the first 24 hr after the operation. Intravenous morphine, 0.05 mg/kg, was administered for pain scores of 4 or 5, and repeated every 15 min as needed to achieve a score of less than 4. The time from admission to the postanesthesia recovery room until administration of the first dose of intravenous morphine was recorded for each patient, and cumulative morphine use was recorded for the first 24 hr after the operation. A log was kept at the bedside for noting the occurrence of possible complications, including respiratory depression, apnea, pruritis, urinary reten tion requiring bladder catheterization, or nausea and vomiting. Patients, their parents, and nurses were interviewed by one of the investigators after the 24-hr observation period to elicit subjective reactions to the control of postoperative pain. The Kruskall-Wallis test and Mann-Whitney test with Bonferroni corrections were used for statistical comparison of the tinie to the first intravenous morphine dose. Comparisons of the incidence of side effects and of patient demographics were made by y , square tests with Bonferroni corrections. Least-square regression analysis was used to correlate duration of There were no differences in ages, types of operations, or anesthetic management of patients in each of the three groups (Table 2). Most patients underwent pelvic, femoral, or tibia1 osteotomies.

Efficacy
The median times to the first supplemental intravenous morphine dose were 45 min for the intravenous morphine group, 5 hr for the caudal bupivacaine group, and 12 hr for the caudal morphine group; differences between groups were statistically significant (Fig. 1). Ten of 15 (67%) intravenous morphine (control) patients required morphine in the recovery room, and 14 (93%) received supplemental intravenous morphine within 6 hr after surgery. One patient in the control group did not require analgesics during the 24-hr observation period after hypospadias repair. Of the 13 patients in the caudal bupivacaine group who had successful sacral block, all had cutaneous sensory blockade to the T4-10 level. Most experienced motor block of varying degree, ranging from paralysis to mild weakness of the lower extremities. All patients given epidural bupivacaine with demonstrable sensory blocks were analgesic upon awakening from general anesthesia: none required morphine while in the recovery room. Children who received bupivacaine blocks had pain scores of 1-3 for 3.5-10.5 hr; the majority (54%) of bupivacaine blocks receded by 6 hr. One patient in the bupivacaine group did not require analgesics during the 24-hr observation period after hypospadias repair.

CAUDAL MORPHINE FOR ANALGESIA IN CHILDREN

1987;66:647-53

ANESTH ANALG

651

Table 3. Frequency of Postoperative Complications in Each Group

Intravenous morphine (n = 15) Respiratory depression Urinary retention Nausedvomiting Pruritis
0 3 5 0

Bupivacaine (n = 13)
0

Caudal morphine (n = 15)

0 4 7 2

x'
1.03 1.36 2.14

P
NS NS NS NS

2 4 1

Abbreviation: NS, no statistically significant difference.

In contrast, the range of duration of analgesia after caudal morphine was 4-24 hr. The onset of analgesia after caudal morphine was apparently rapid; upon awakening from general anesthesia, every patient in the caudal morphine group was analgesic, with pain scores of 1-3, and none required intravenous morphine in the recovery room. Indeed, only four of 15 (27%)caudal morphine patients required supplemental intravenous morphine within the first 8 hr after the operation: two of these patients received intravenous morphine after about 4 hr, neither of whom had confirmation of correct caudal placement by injection of lidocaine; a third patient had otalgia believed to be caused by operative use of nitrous oxide, and may have required morphine supplementation for this reason. Eleven patients who received caudal morphine (73%) maintained pain scores less than 4 for 9 to about 24 hr. One patient required no analgesics in the 24-hr observation period, also after hypospadias repair. After the first 4 postoperative hr, most patients who had not needed parenteral morphine were in the caudal morphine group (Fig. 2). The duration of analgesia after caudal morphine did not correlate with the age or weight of the patients. There was a tendency for patients who received the mixture of preservative-free morphine with lidocaine to have a longer duration of analgesia than the patients who received preservative-free morphine alone (mean 14.4 and 9.9 hr, respectively), but this was not statistically significant (P = 0.22). Patient and family acceptance of caudal morphine was enthusiastic. Among patients who had undergone operations of a similar nature previously, enthusiasm for the quality of analgesia provided by caudal morphine was especially high, even at the expense, in the instance of one 16-yr-old boy, of bladder catheterization. In contrast, many parents of children in the control group complained of inadequate analgesia during the first 24 hr after the operation.

patients who underwent hypospadias repair had routine postoperative bladder drainage. Among the other patients, intravenous morphine alone was associated with a 25% incidence of urinary retention requiring bladder catheterization (3/12 patients). Urinary retention was slightly more common in the caudal morphine group (413 patients, 31%), but two of these four patients received intravenous morphine prior to their inability to void. One patient given caudal bupivacaine required catheterization more than once. Nausea and vomiting were common in all groups, and occurred in the caudal morphine patients prior to the administration of intravenous morphine. Treatment with antiemetics was variably successful. Mild nasal pruritis occurred in two caudal morphine patients, but did not require treatment. Respiratory depression (respiratory rate less than lO/min) or apnea did not occur in any patient, although monitor false alarms were not uncommon.

Discussion
The caudal administration of 0.1 mg/kg of preservative-free morphine (with or without 1%lidocaine, 0.25 mykg) after lower extremity or genitourinary surgery was a safe and effective method of achieving prolonged postoperative analgesia in children. Caudal morphine provided approximately twice the duration of analgesia as caudal bupivacaine, without motor or sensory block. The onset of effect of caudal morphine in children was sufficiently rapid to provide analgesia in the recovery room a short time after the postoperative caudal injection, such that all the children who received caudal morphine were analgesic upon arousal from general anesthesia, which was within 15-20 min of the caudal injections in most instances. This is a notable departure from experience with adult patients, in whom the onset of analgesia after lumbar epidural morphine administration requires about 1 hr (14). Lidocaine was added to the preservative-free morf half the caudal morphine group patients to phine o confirm correct placement of the caudal needle by means of a short-acting drug that was not likely to

Cornplications
There was no statistical difference in the frequency of complications among the three groups (Table 3). All

652

ANESTH ANALG 1987;66:617-53

KRANE ET AL.

influence the outcome of the study. In so doing, we identified one failed injection in the group of eight morphine-plus-lidocaine patients. Although the addition of lidocaine to the caudal morphine injectate tended to increase the duration of postoperative analgesia, this increase did not reach statistical significance, and may have been a result of the inclusion of two patients in the group that received preservativefree morphine alone who may not have had true caudal injection of the study drug but in whom we did not have the ability to make that determination. Elimination of these patients from the comparison of those who received morphine alone vs those who received morphine with lidocaine further narrows the difference in mean durations of analgesia between those subgroups of subjects (13.9 and 14.4 hr, respectively). We made no attempt to compare the quality of analgesia conferred by these three methods of controlling postoperative pain. Presently there are no validated and accepted tools for measuring pain in small children, and such a comparison must await the development of reliable pediatric pain scales. However, the subjective opinions of the hospital nursing staff who care for postoperative children and of our surgical colleagues, all of whom were blinded to treatment group assignment, were that there was no decrement in the quality of analgesia when comparing caudal morphine to bupivacaine, and that both were superior to control patients, in whom pain followed a cyclic pattern. The frequency of side effects tended to be higher in caudal morphine patients, though a proportion of the side effects might have been attributable to subsequent administration of intravenous morphine. Though naloxone has been reported to be effective in treating urinary retention and nausea associated with epidural morphine (15,16), we did not evaluate this drug in our patients. We observed no cases of delayed respiratory depression, but our series is a small one. More experience with spinal narcotics in children will need to be gathered to assess fully the risk of respiratory depression after epidural morphine, to identify subpopulations at risk, to adequately inform families of risk, and to guide selection of appropriate drug dosage and monitoring technique. Until the full extent of such risk has been defined in children, respiratory monitoring should be used and equipment and drugs for resuscitation should be at hand. Although it may be a laudable goal to eliminate postoperative pain in children, there are some instances in which pain is a valuable symptom that alerts the physician to the existence of a problem. During the course of this study, one orthopedic surgeon believed that the motor and sensory block as-

sociated with caudal bupivacaine delayed the recognition of a compartment syndrome in one patient who had undergone a tibial osteotomy, a recognized complication of this operation. After that experience, we chose not to administer caudal bupivacaine for postoperative analgesia after tibial osteotomies. It remains to be seen whether caudal morphine confers analgesia sufficient to blunt the intense pain of the tissue ischemia associated with compartment syndromes, but because caudal morphine does not ordinarily result in motor blockade, the loss of voluntary movement distal to a compartment syndrome should remain a diagnostic tool for the surgeon. In 1981 Jensen (17) published the first and, to date, only report of caudal morphine for control of postoperative pain in children. In that study, caudal morphine (0.05 mg/kg) was compared to caudal bupiva0.5 mVkg) in young boys after outpatient caine (0.2570, circumcision or inpatient hypospadias repair. Only the inpatient group was evaluated for duration of analgesia. As in our experience, all patients except one with a failed block were analgesic in the recovery room. Jensen reported a duration of analgesia of 4-8.5 hr (mean 6 hr) after caudal bupivacaine in five children, and 5-37 hr (mean 20 hr) after caudal morphine in seven children. The incidence of nausea was similar in the two groups, but other side effects or complications were not studied. Unlike our study, a control group was not included, respiratory monitors were not used, nor is it specified how efficacy or duration of effect were determined. Therefore, our data confirm the results of Jensen, further demonstrate the efficacy of caudal morphine for analgesia after both genitourinary and orthopedic surgery in a wider age group, and demonstrate the absence of significant side effects compared to a control group receiving conventional parenteral narcotics. In summary, caudal preservative-free morphine provided prolonged analgesia of longer duration than caudal bupivacaine and without side effects greater than those seen with conventional parenteral morphine, in children after lower extremity orthopedic and genitourinary surgery. Because a needle may be placed within the caudal space more easily than in the lumbar epidural space in children, and because of the present unavailability of appropriately sized epidural needles for pediatric use, caudal injections or caudal catheters may be the preferred route for administration of epidural narcotics in young children.

Duramorph provided by the A. H. Robins Company.

CAUDAL MORPHINE FOR ANALGESIA IN CHILDREN

ANESTH ANALG
1987;66:647-53

653

References
1. Magora F, Olshang G, Eimerl D, Shorr J, Katzenelson R, Cotev S, Davidson JT. Observations on extradural morphine analgesia in various pain conditions. Brit J Anaesth 1980;52247-32.

2. Cousins MJ, Mather LE. lntrathecal and epidural administration of opioids. Anesthesiology 1984;61:276-310. 3. Cohen SE, Woods WA. The role of epidural morphine in the postcesarean patient: efficacy and effects on bonding. Anesthesiology 1983;58:500-4. 4. Schechter NL. Status of pediatric pain control. Pediatrics 1986;m 11-5. 5. Mather L, Mackie J . The incidence of postoperative pain in children. Pain 1983;15:271-82. 6. Jones SEF, Beasley JM, Macfarlane DWR, Davis JM, Hall-Davies G. Intrathecal morphine for postoperative pain relief in children. Br J Anaesth 1984;56:137-40. 7. Shapiro LA, Jedeiken RJ, Shaler D, Hoffman S. Epidural morphine analgesia in children. Anesthesiology 1984;61:210-2.
8. Dalens B, Tanguy A, Haberer J-P. Lumbar epidural anesthesia

for operative and postoperative pain relief in infants and young children. Anesth Analg 1986;65:1069-73.
9. McGown RC. Caudal analgesia in children. Five hundred cases

for procedures below the diaphragm. Anaesthesia 1982;37: 805-18. 10. Arthur DS, McNicol LR. Local anaesthetic techniques in paediatric surgery. Br J Anaesth 1986;58:760-78. 11. Kay B. Caudal block for postoperative pain relief in children. Anaesthesia 1974;29:6104. 12. Matsumiya N, Dohi S, Takahashi H, Kondo Y, Naito H. Cardiovascular collapse in an infant after caudal anesthesia with lidocaine-epinephrine solution. Anesth Analg 1986;65:1074-6. 13. Schulte-Steinberg 0. Neural blockade for pediatric surgery. In: Cousins MJ, Bridenbaugh, PO, eds. Neural blockade in clinical anesthesia and management of pain. Philadelphia: Lippincott, 1980:508. 14. Bromage PR, Camporesi E, Chestnut D. Epidural narcotics for postoperative analgesia. Anesth Analg 1980;59:473-80. 15. Rawal N, Mollefors K, Axelsson K, Lindgardh G, Widman B. An experimental study of urodynamic effects of epidural morphine and of naloxone reversal. Anesth Analg 1983;62641-7. 16. Rawal N, Wattwil M. Respiratory depression following epidural morphine. An experimental and clinical study. Anesth Analg 1984;63:8-14. 17. Jensen BH. Caudal block for post-operative pain relief in children after genital operations. A comparison between bupivacaine and morphine. Acta Anaesthesiol Scand 1981;25:373-5.