Section Editor: David J. Birnbach.
Characterization of the dose-response of small doses of intrathecal (IT) morphine for postcesarean delivery analgesia is incomplete. Authors have advocated doses of IT morphine as small as 25 [micro sign]g and as large as 200 mg for treating pain after cesarean delivery [1,2] [1,3-7] [8,9] [2,10,11]
Therefore, we designed this study in an effort to better characterize the analgesia of IT morphine administered along with other medications commonly prescribed after cesarean delivery. Specifically, we sought to determine a dose of IT morphine effective in providing analgesia without IV morphine or PCA. Such a determination might be useful to clinicians using "minidose" IT morphine as part of a multimodal approach to the management of pain after cesarean section.
We used an up-down method of sequential allocation of IT morphine doses. This method was chosen in part because it allowed for the administration of IT morphine over a broad range of doses adjusted according to patient response, while minimizing the number of subjects receiving inadequate or excessive doses of IT morphine.
We defined our criteria for adequate analgesia as an end point that we believed to be clinically relevant. Adequate postoperative analgesia after IT morphine has previously been defined as 24 h of pain relief without the requirement for other analgesics. Unfortunately, the doses required for this end point have been associated with a high incidence of side effects [1]
A second aim of this study was to analyze the acquisition costs and nursing care time of this study protocol, in which patients were able to request multiple medications in addition to IT morphine. We retrospectively compared our study protocol with a protocol using postoperative PCA after spinal anesthesia, the customary technique for providing analgesia after cesarean delivery at our institution.
Methods
After obtaining institutional ethical approval, 40 ASA physical status II, term (at least 37 completed weeks' gestation) parturients with singleton pregnancies who requested spinal anesthesia were scheduled for elective cesarean delivery on weekday mornings. Patients were enrolled over a 4-mo period. Written, informed consent was obtained from all patients. Women who weighed >114 kg, had fetuses with known anomalies, were taking any medications other than prenatal vitamins, were unable to understand the consent form, or were allergic to any of the study drugs were excluded. Demographic and obstetric variables (age, height, weight, length of gestation, gravidity, parity, indication for cesarean section) were recorded.
Spinal anesthesia was performed using of a 27 Whitacre needle or a 29 Quincke needle with the patient in the lateral position. All patients received 1.6 mL of 0.75% hyperbaric bupivacaine plus 20 [micro sign]g of fentanyl (0.4 mL) plus the study drug (diluted to 0.5 mL in saline) for a total volume of spinal injectate of 2.5 mL.
Using an up-down method of dose allocation, each patient received spinal morphine (Duramorph[registered sign]; Elkins-Sinn Inc., Cherry Hill, NJ) as the study drug. The first patient was assigned a dose of 200 [micro sign]g. Because this patient met our criteria for success, the next patient received 175 [micro sign]g. Doses for each subsequent patient were decreased by 25 [micro sign]g for success or increased by 50 [micro sign]g for failure of analgesia in the preceding patient. Doses <150 [micro sign]g were adjusted both up and down by 25 [micro sign]g. Doses were adjusted upward by 50 [micro sign]g, rather than 25 [micro sign]g, at the larger dose range in an effort to minimize our patients' exposure to inadequate analgesia. Because of the small doses used, only one such 50-[micro sign]g increment was necessary in the study. The predetermined maximal dose of IT morphine permitted was 500 [micro sign]g, and the minimal dose was 0 mg. Patients in whom the protocol was violated (e.g., in receipt of nonprotocol pain medications, pain medications for sedation rather than for pain, postoperative surgical complication requiring reoperation, failure of spinal block) were not included in our analysis. When a patient was excluded, the next patient enrolled received the same dose assigned to the excluded patient. Dose allocation was double-blinded at all times. Patients were not informed of the dose administered, a physician not providing intraoperative anesthetic care prepared the study drug, and nurses providing postoperative care were blinded to dose.
Postoperative pain was treated first with oral hydrocodone 5 mg and acetaminophen 500 mg one or two tablets every 4 h as needed. If this was not effective, 4-mg increments of IV morphine were administered until pain was <or=to4 on a 10-point verbal pain scale, then administered every 2 h. Patients chose one or two oral hydrocodone/acetaminophen tablets at their discretion and received IV morphine only at their request. While in the postanesthesia care unit (PACU), patients were offered IV fentanyl in a dose of 25 [micro sign]g every 5 min up to a maximum 100 [micro sign]g as needed for pain persisting after the administration of oral hydrocodone/acetaminophen. Postoperatively, patients were permitted to request medications for nausea and pruritis. On patient request, nausea was treated with nalbuphine 5 mg IV every 4 h, and pruritus was treated with 10 mg of IV nalbuphine every 4 h. Patients were offered 0.625 mg of IV droperidol every 4 h as needed for nausea unrelieved by nalbuphine and 25 mg of diphenhydramine IV every 4 h as needed for pruritis unrelieved by nalbuphine. Nonsteroidal antiinflammatory drugs (NSAIDs) are not used by all surgeons at our institution and were specifically excluded by protocol. No medications for sedation, nausea, pruritis, or analgesia were permitted outside of the protocol medications.
In this study, the administration of morphine IV at any time before 12 h after spinal blockade was defined as the criterion for failure. The administration of IV fentanyl (up to 100 [micro sign]g) during the brief period in which the patient remained in the PACU and oral hydrocodone/acetaminophen administered postoperatively were not considered criteria for failure.
Respiratory rate and sedation scores were recorded by nursing staff every hour for 12 h, then every 2 h for the next 12 h. Sedation was scored as follows: 1 = wide awake, 2 = drowsy, 3 = mostly sleeping, 4 = only awakes when aroused. Pain was scored every 2 h for 24 h on a 0-10 verbal scale (0 = no pain, 10 = the worst possible pain). Episodes of nausea, emesis, or pruritis requiring treatment were noted.
Within 72 h of cesarean delivery, each patient completed a six-question survey administered by one of the investigators who was blinded to morphine dose. In this survey, they were queried regarding pain expectations ("none," "little," "some," "a lot"), pain intensity ("none," "little," "some," "a lot"), quality of pain relief ("poor," "fair," "good," "excellent"), side effect ("what one thing bothered you the most"), comparison to previous cesarean section (if applicable), and willingness to have IT morphine again.
Patients who were enrolled to receive IT morphine were retrospectively compared with a historical group of patients who received IV morphine PCA. This historical group consisted of 15 patients who received PCA after cesarean delivery under bupivacaine spinal anesthesia in the 4 mo before our study. PCA morphine was delivered with a 1.5-mg dose, a lockout time of 6 min, a 10-mg 1-h limit, and no basal infusion [12]
Morphine consumption and pain scores were obtained prospectively as above (for IT morphine patients) and retrospectively from chart review (for PCA patients). Acquisition costs for postoperative analgesia therapy were calculated by multiplying 1997 United States wholesale drug prices (for all analgesics and all medications used to treat side effects) by the actual number of treatments administered to each patient in each group over 24 h. Nursing time spent on analgesic therapy was estimated as follows. Ten experienced registered nurses who cared for both IT morphine and PCA patients individually completed an anonymous survey in which they estimated the time required to perform each of 12 specific patient-care interventions (e.g., instruction in PCA use, treating nausea). These time estimates were averaged and multiplied by the actual number of interventions made for each patient in each group over 24 h.
Data are presented as mean +/- SD or percentage (number). Data were analyzed using Fisher's exact test, the Mann-Whitney U-test, or Kruskal-Wallis test followed by a Dunn posttest, as appropriate. Attempts to determine a 50% effective dose (ED (50 )) of IT morphine were made by using probit analysis. A P value <0.05 was considered significant.
Results
Four patients were excluded from the study after providing their consent. One patient was excluded for absent local anesthetic block immediately after placement of her spinal anesthetic before cesarean section. Three other patients were excluded after NSAIDs were administered, in violation of protocol. Data from these patients were not included in the analysis.
Patients receiving IT morphine were compared as two groups: patients who met our criteria for success of IT morphine analgesia and patients who met our criteria for failure of IT morphine analgesia. Overall, 25 patients (62.5%) met our criteria for success and 15 patients (37.5%) met our criteria for failure after receiving IT morphine. We encountered four patients who met the criteria for success at an IT morphine dose of 0, and the results of the up-down dose allocation did not seem to stabilize around an ED50 (Figure 1 ).
Figure 1: Up-down allocation of the intrathecal morphine dose as determined by sequential patient response. Above 150 [micro sign]g, the dose was increased by 50 [micro sign]g after a patient met criteria for "failure" ([black square]) and decreased by 25 [micro sign]g after a patient met criteria for "success" ([black circle]). Below 150 [micro sign]g, the dose was adjusted by 25 [micro sign]g after both success and failure. After a patient met criteria for success at a dose of 0 [micro sign]g, this dose was repeated.
Because we were unable to lower the IT morphine dose below 0 once we achieved success at this dose, we could not perform a true up-down allocation or analyze our data in a manner based on up-down allocation. Instead, we used an ad hoc probit analysis to arrive at an ED50 . Performing such an analysis based on results generated by a patient population with relatively few patients at a relatively large number of doses increased the uncertainty of our ED50 . We were unable to arrive at an ED50 with precision using either untransformed or log(x + 1)-transformed dose values. Our best estimate of an ED50 of IT morphine in the clinical setting we studied is 22 +/- 53 [micro sign]g.
Demographic data did not differ between patients who met criteria for IT morphine analgesia success and failure (Table 1 ). Respiratory rate, sedation score, consumption of hydrocodone/acetaminophen, and need to treat side effects were not significantly different among IT morphine patients (Table 2 ). Four patients who met criteria for success at 12 h required morphine treatment during the 24-h study period. These four patients received either one or two doses of morphine over the second 12 h of the study. No patient in any group received IV morphine past 24 h after spinal anesthesia. Patients who met the criteria for success were significantly more likely to leave the PACU after their cesarean section without requiring IV fentanyl for pain. There were no episodes of respiratory depression.
Table 1: Patient Demographics:
Table 2: Measured Variables and Treatments for the First 24 Hours
Patient survey results were not statistically different between patients who met criteria for IT morphine analgesia success and failure. Patients' expectations of postoperative pain, complaints of pain, and complaints of side effects were not different between the groups. Intrathecal morphine was well accepted by both groups: 80% would choose IT morphine treatment again, and 88% rated their analgesia as "good" or "excellent."
We also compared the two groups of patients who received IT morphine with the historical group who received PCA. Patient demographic data were not different among these three groups (Table 1 ). Verbal pain scores did not differ among these groups. The group of patients who met predefined criteria for success 12 h after IT morphine consumed significantly less IV morphine over their entire recovery compared with the group who met criteria for failure at 12 h and with the group who received PCA morphine.
Drug acquisition costs for our IT morphine protocol ($15.13 +/- $4.40) were significantly lower than those historically incurred at our institution using PCA ($34.64 +/- $15.55). Time spent managing patients based on nursing personnel estimates did not differ between IT morphine analgesia (150 +/- 57 min) and PCA (148 +/- 61 min).
Discussion
Although up-down sequential allocation of doses is a powerful methodology for determining the ED50 , the clinical utility of any ED50 obtained by up-down sequential allocation depends on the appropriateness of the criteria for defining success versus failure, the clinical setting in which the dose is assessed, and the appropriateness of applying the up-down approach to the clinical question being addressed.
We believe that our criteria were clinically appropriate because they identified patients likely to achieve adequate analgesia without significant morphine consumption. We defined success or failure as a request for IV morphine within 12 h of spinal anesthesia. Those patients who met the criteria for success consumed less IV morphine to a statistically significant degree over their entire postoperative course (Table 2 ). Conversely, the morphine requirements of patients who met the criteria for failure were not statistically different from the morphine requirements of patients treated with PCA. We believe that the difference between the groups defined by our criteria has clinical utility because only 4 of the 25 patients who met criteria for success requested any IV morphine postoperatively. Furthermore, these four patients requested either only one or two doses during their entire postoperative course.
The setting we studied also has clinical utility. In clinical practice, IT morphine may be administered along with other medicines in the first 24 h after cesarean delivery. Other authors have used up-down sequential allocation to determine the dose-response of inhaled anesthetics [13] [14] [15]
Using up-down allocation of IT morphine doses in this clinically relevant setting, we were not able to arrive at a precise ED50 . We believe that this lack of precision (an ED50 of 22 +/- 53 [micro sign]g) is the result of patient variability in a clinical context and the statistical limitations of our investigation. In distinction to dose-ranging studies, doses determined by sequential allocation result from patient response, not assignment. Our protocol allowed the administration of IT morphine in doses of 0-500 [micro sign]g. The first patient in our study was assigned a 200-[micro sign]g dose of IT morphine, a dose that is used very commonly. Subsequent patients required less based on patient response. Of our patients, 10% (or four patients) received 0 [micro sign]g of IT morphine but met our criteria for success postoperatively. Not only did we encounter patients who met our criteria for success without receiving IT morphine, but these four patients were also among the patients who required little treatment for pain and little treatment for side effects postoperatively. (Three of these patients requested eight hydrocodone/acetaminophen tablets without other medications in the first 24 h, and one patient requested four hydrocodone/acetaminophen tablets for pain and two nalbuphine treatments for pruritis. None of these four patients required IV fentanyl in the PACU or IV morphine on the ward at any time.) Postoperatively, two of these patients considered their pain relief to be "excellent" and two considered it "good." All four had lower than average pain scores. The next sequential patient demonstrated the individual variability characteristic of postoperative pain management. She was the fifth patient to receive 0 [micro sign]g of IT morphine but requested six hydrocodone/acetaminophen tablets followed by 40 mg of IV morphine in the first 24 h, she met our criteria for failure, and she ranked her quality of pain relief as "fair." Had we chosen a protocol in which medications other than the study drug were strictly excluded, as others have done [14-16] 50 with more certainty but perhaps less clinical relevance.
Adequate postoperative analgesia with zero and very small doses of IT morphine was an unanticipated finding of this study and warrants further discussion. One obvious explanation for the success of the study regimen is that IT morphine provides very effective analgesia after cesarean delivery even when administered in very small doses.
A second explanation for the success of the study regimen is that our patients received several analgesics in addition to IT morphine. Hydrocodone is a relatively potent oral opioid [16] [17] [18] [19]
Our study was not designed to determine the relative analgesia that results from IT morphine as opposed to other medications administered in this study. Our goal was to study the analgesia provided by these medications used in combination, as they may be used clinically. Authors of recent dose-ranging studies have concluded that IT morphine provides analgesia after cesarean delivery even at very small doses when these doses are administered along with oral analgesics [2,7]
However, the analgesia we found in women who received no IT morphine suggests that the oral analgesics and other medications administered also provide substantial analgesia after cesarean delivery. The lack of stabilization of our dose allocation and the lack of precision in the resultant ED50 imply that IT morphine dose was not the principal determinant of analgesia in this clinical context. An important implication of this study is that clinicians should not underestimate the importance of providing oral analgesics when IT morphine is given in very small doses. In this regard, our results are in agreement with those of other authors who have questioned the incremental benefit of minidose IT morphine when combined with other analgesics for cesarean delivery analgesia [10,11]
Whatever the relative contribution of IT morphine, fentanyl, oral hydroxycodone/acetaminophen, and other medications commonly prescribed after cesarean delivery, this study suggests that IT morphine in very small doses has a role in time- and cost-effective multimodal management of pain after cesarean delivery. This study confirms the work of Rosaeg et al. [20] [21]
A weakness of this time and cost comparison is the retrospective nature of our historical group receiving PCA morphine. This group was included only to place the analgesia achieved by our study patients in a historical context and to compare time and costs. We believed that it would be overly taxing (and perhaps unsafe) for nursing personnel to simultaneously manage PCA patients by our standard protocol while managing IT morphine patients by an entirely new, less familiar, and double-blinded protocol. The PCA patients did not have access to as many or as costly medications to treat side effects. Their analgesic regimen was less strictly controlled in that several of these patients received NSAIDs just before 24 h postoperatively, which probably lowered their PCA morphine consumption. These differences in treatment seem likely to minimize differences in time and costs between groups and would therefore seem unlikely to change our conclusions. Unlike the historical group, our study patients consented to be part of a study "evaluating a promising method of managing postoperative pain better with less pain medicine." The study patients received IT fentanyl in the operating room and IV fentanyl in the PACU, whereas the historical patients did not. These differences may have decreased interventions and costs in the IT morphine patients.
Our cost analysis is limited to drug acquisition costs. Our study does not account for equipment and supplies; their acquisition, storage, and maintenance costs; cost of nursing time; or less tangible costs of patient comfort or satisfaction. Our time analysis is based on estimates by nurses that may be less accurate than measurements of time spent on each intervention. However, measuring drug acquisition costs and estimating nursing time requirements was a simple, reasonable, and important evaluation of an approach new at our institution.
In conclusion, very small doses of IT morphine, when used in combination with oral analgesics and other commonly prescribed medications, provide time- and cost-efficient postoperative analgesia after cesarean delivery. These results support the use of small doses of IT morphine with oral analgesics and other commonly prescribed medications for analgesia after cesarean delivery.
REFERENCES
1. Abouleish E, Rawal N, Rashad M. The addition of 0.2 mg subarachnoid morphine to hyperbaric bupivacaine for cesarean delivery: a prospective study of 856 cases. Reg Anesth 1991;16:137-40.
2. Cardoso MM, Carvalho JC, Amaro AR, et al. Small doses of intrathecal morphine combined with systemic diclofenac for postoperative pain control after cesarean delivery. Anesth Analg 1998;86:538-41.
3. Abboud TK, Dror A, Mosaad P. Mini-dose intrathecal morphine for the relief of post-cesarean section pain: safety, efficacy, and ventilatory responses to carbon dioxide. Anesth Analg 1988;67:137-43.
4. Milner AR, Bogod DG. Intrathecal administration of morphine for elective caesarean section. Anaesthesia 1996;51:871-3.
5. Norman PF, Tsang BK, Werhan CF. Comparison of postcesarean pain relief and side effects with two doses of intrathecal morphine [abstract]. Anesth Analg 1997;84:S399.
6. Swart M, Sewell J, Thomas D. Intrathecal morphine for caesarean section: an assessment of pain relief, satisfaction, and side-effects. Anaesthesia 1997;52:373-7.
7. Sibilla C, Albertazzi P, Zatelli R. Pain relief after caesarean section: comparison of different techniques of morphine administration. Int J Obst Anesth 1994;3:203-7.
8. Palmer CM, Voulgaropoulos D, Van Maren G. What is the optimal dose of subarachnoid morphine for post-cesarean analgesia? A dose-response study [abstract]. Anesthesiology 1994;81:A1151.
9. Huffnagle HJ, Norris MC, Leighton BL. A dose-response study of intrathecal morphine for post cesarean section analgesia [abstract]. Anesth Analg 1997;84:S388.
10. Cohen SE, Desai JB, Ratner EF. Ketorolac and spinal morphine for postcesarean analgesia. Int J Obst Anesthesia 1996;5:14-8.
11. Roy L, Crochetiere C, Arsenault MY. Combined use of intrathecal morphine and nonsteroidal anti-inflammatory drugs for post caesarean section analgesia [abstract]. Can J Anaesth 1997;44:A15.
12. Krukawski JA, Hood DD, Eisenach JC. Intrathecal neostigmine for post-cesarean section analgesia: dose response. Anesth Analg 1997;84:1269-75.
13. Inomata S, Watanabe S, Masakazu T. End tidal sevoflurane concentration for tracheal intubation and minimum alveolar concentration in pediatric patients. Anesthesiology 1994;80:93-6.
14. Powell H, Morgan M, Sear JW. Pregnanolone: a new steroid intravenous anaesthetic-dose-finding study. Anaesthesia 1992;47:287-90.
15. Columb MO, Lyons G. Determination of the minimum local analgesic concentrations of epidural bupivacaine and lidocaine in labor. Anesth Analg 1995;81:833-7.
16. Sunshine A, Olson NZ, O'Neill E. Analgesic efficacy of a hydrocodone with ibuprofen combination compared with ibuprofen alone for the treatment of acute postoperative pain. J Clin Pharmacol 1997;37:908-15.
17. Beaver WT, McMillian D. Methodological considerations in the evaluation of analgesic combinations: acetaminophen and hydrocodone in postpartum pain. Br J Clin Pharmacol 1980;10:215-235.
18. Dahlgren G, Hultstrand C, Jakobsson J. Intrathecal sufentanil, fentanyl, or placebo added to bupivacaine for cesarean section Anesth Analg 1997;85:1288-93.
19. Connelly NR, Dunn SM, Ingold V. The use of fentanyl added to morphine-lidocaine-epinephrine spinal solution in patients undergoing cesarean section. Anesth Analg 1994;78:918-20.
20. Rosaeg OP, Lui ACP, Cicutti NJ. Peri-operative multi-modal pain therapy for cesarean section: analgesia and fitness for discharge. Can J Anaesth 1997;44:803-9.
21. Cohen SE, Subak LL, Brose WG. Analgesia after cesarean delivery: patient evaluation and costs of five opioid techniques. Reg Anesth 1991;16:141-9.
