Intrathecal fentanyl added to bupivacaine and morphine for cesarean delivery may induce a subtle acute opioid tolerance

International Journal of Obstetric Anesthesia (2012) 21, 29 34 0959-289X/$ - see front matter c 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijoa.2011.09.002 ORIGINAL ARTICLE Intrathecal fentanyl added to bupivacaine and morphine for cesarean delivery may induce a subtle acute opioid tolerance B. Carvalho, a D.R. Drover, a Y. Ginosar, b S.E. Cohen, a E.T. Riley a a Department of Anesthesiology, Stanford University, Stanford, CA, USA b Department of Anesthesiology and Critical Care Medicine, Hadassah Hebrew University Medical School, Jerusalem, Israel www.obstetanesthesia.com ABSTRACT Background: Previous studies have demonstrated that the addition of intrathecal fentanyl to a spinal anesthetic for cesarean delivery improves intraoperative analgesia. However, intrathecal fentanyl may induce acute tolerance to opioids. The objective of this study was to investigate whether the addition of intrathecal fentanyl to spinal anesthesia with intrathecal morphine increases postoperative analgesic requirements and pain scores. Methods: In this randomized, double-blinded study, 40 women having elective cesarean delivery were enrolled. Patients received spinal anesthesia with hyperbaric bupivacaine 12 mg, morphine 200 lg, and fentanyl 0, 5, 10 or 25 lg. Each patient received intravenous patient-controlled analgesia morphine for 24 h postoperatively. Outcome measures included postoperative morphine usage and pain scores, as well as intraoperative pain, nausea, hypotension and vasopressor use. Results: Total morphine use over the 24-h post-spinal study period was similar among the study groups (P = 0.129). Postoperative pain scores were higher in patients receiving fentanyl 5, 10 and 25 lg compared to fentanyl 0 lg control group (P = 0.003). Conclusions: The study results suggest that intrathecal fentanyl may induce acute tolerance to intrathecal morphine. However, because there was no difference in postoperative analgesia requirement and the difference in pain scores was small, the clinical significance of this finding is uncertain. c 2011 Elsevier Ltd. All rights reserved. Keywords: Analgesic-opioids; Anesthesia-obstetric; Anesthesia-spinal; Tolerance; Spinal; Anesthesia; Intrathecal; Fentanyl; Acute; Tolerance Introduction Opioids are commonly added to the local anesthetic solution when administering spinal anesthesia for cesarean delivery. Intrathecal (IT) fentanyl improves intraoperative analgesia, especially during uterine exteriorization, and prolongs the sensory component of the block associated with IT local anesthetics. 1 However, IT fentanyl provides limited postoperative analgesia with median time to first analgesia of 4 h (range 2 13 h). 2 Neuraxial morphine has a long duration of action and provides prolonged postoperative analgesia compared to other routes of administration. 3,4 Using a lipophilic opioid such as fentanyl, is helpful because Accepted September 2011 Correspondence to: B. Carvalho, Department of Anesthesia, Stanford University Medical Center, 300 Pasteur Avenue, Stanford, CA 94305, USA. E-mail address: bcarvalho@stanford.edu the onset time for neuraxial morphine may be too slow for it to contribute to intraoperative analgesia. 5 A potential problem is that patients may experience more postoperative pain if intraoperative opioids are administered due to opioid-induced hyperalgesia. 6 Animal and human volunteer data suggest that administration of high doses of intravenous (i.v.) opioids induces acute tolerance to subsequently administered opioids, 7,8 whereas clinical studies have yielded inconsistent and inconclusive results. 5 To our knowledge, only one previous study has investigated whether administration of IT fentanyl with a spinal anesthetic for cesarean delivery induces acute tolerance to opioids. 9 Cooper et al. showed that when patients were given a large dose of IT fentanyl (25 lg) with their spinal anesthetic, they had an increased requirement for i.v. morphine postoperatively. However, in contrast to common clinical practice, IT morphine was not given in their study. 2 The primary aim of this study was to determine if the addition of IT fentanyl to a spinal anesthetic solution

30 Intrathecal fentanyl and acute tolerance containing IT morphine induces acute opioid tolerance, quantified by increased postoperative morphine requirements and pain scores. The secondary aim was to determine if there is a dose response associated with this phenomenon. Our null hypothesis was that the addition of IT fentanyl would not change postoperative analgesic use or pain scores. Methods A total of 40 patients were enrolled (10 per group) in this randomized, double-blinded, controlled, dose-ranging study after obtaining Institutional Review Board approval and written informed consent. We enrolled healthy women (American Society of Anesthesiologists physical status 1 or 2) with term singleton pregnancies (>37 weeks of gestation) undergoing elective cesarean delivery under spinal anesthesia before noon. Patients were excluded from study participation if they met any of the following criteria: morbid obesity (body mass index >40 kg/m 2 ), planned postpartum tubal ligation, previous adverse reaction to opioid medications or history of chronic opioid use. The spinal anesthetic was administered with a 25- gauge Whitacre needle (B. Braun Medical, Bethlehem, PA, USA) at the L2 3 or L3 4 intervertebral space with the patient in the sitting position. Each participant received IT hyperbaric bupivacaine 12 mg and preservative-free morphine sulfate 200 lg (Astramorph PF, APP Pharmaceuticals LLC, Schaumburg, IL, USA) with an added randomized dose of 0, 5, 10 or 25 lg IT fentanyl (Baxter Healthcare Corporation, Deerfield, IL, USA). Randomization was by computer-generated random-number allocation. The mixture was diluted to a total of 2.5 ml with sterile saline. The investigator, patient, anesthesiologist treating the patient, and all study staff remained blinded to the assigned treatment. An anesthesiologist not involved in the study or patient care prepared the fentanyl doses. Patients were immediately placed supine with left uterine displacement after the spinal anesthetic was administered. The height of the spinal block was assessed by pinprick and touch every 2 min until the block reached the T4 dermatome bilaterally. Intraoperative vital-sign measurements included: mean arterial pressure measured with a non-invasive blood pressure monitor, heart rate, respiratory rate and oxygen saturation. Ephedrine was given in 5-mg increments to treat systolic arterial pressure <80% of the preanesthetic value. Patient complaints of intraoperative pain or nausea were treated as needed at the discretion of the anesthesiologist. Pain was assessed with a 100 mm visual analog pain scale (VAPS, 0 = no pain, 100 = worst pain imaginable). Intraoperative pain scores were assessed at the time of spinal placement, skin incision, uterine incision, delivery, and skin closure. The quality of intraoperative anesthesia was assessed by the presence or absence of any patient request for supplemental anesthesia. Postoperative pain scores, oxygen saturation, and respiratory rate were assessed on arrival to the post-anesthesia care unit (PACU), and at 30 min and 60 min, 4, 8, 12, and 24-h following the end of surgery. In the PACU, each patient received i.v. morphine patient-controlled analgesia (PCA) for the first 24-h. The settings were as follows: bolus 1 mg, lockout interval 8 min, hourly maximum 8 mg. No background infusion was utilized. The interval dose and cumulative dose of morphine were recorded at 4, 8, 12 and 24-h following the administration of spinal anesthesia. No other forms of analgesia were administered. Statistical analysis Based on a pilot study, a priori power analysis predicted that we required 10 patients in the control group and 30 patients in the treatment group to detect a mean difference of 30% in postoperative morphine use in the first 24 h (expected mean 17.7 mg, standard deviation 5.5 mg, power 0.8, alpha 0.05). Descriptive statistics were used to summarize demographic and outcome data. Normal distribution was determined using QQ plots and the Kolmogorov Smirnov test. The area under the VAPS time curve (AUC) was calculated using the trapezoid rule to assess the pain burden over the 24-h study period. The Kendal Correlation test was used to assess the effect of increasing fentanyl doses on postoperative pain AUC and morphine (mg) consumption. One-way ANOVA was used to analyze normally distributed continuous variables with the post-hoc Tukey test if an overall difference was found. The Kruskal Wallis test was utilized for ordinal and non-parametric continuous data with post-hoc comparisons using the Mann Whitney U test. Associations among categorical variables (requirement for supplemental intraoperative anesthesia or treatment of nausea) were investigated using the v 2 test and Fisher s exact test as appropriate. The association between pain (VAPS time AUC) and analgesic consumption (total 24-h IV morphine use) was performed using Spearman s rho. A P-value <0.05 was considered statistically significant for the primary outcome measure of analgesic use, as well as for pain scores, in the control group compared to the fentanyl groups combined. For the betweengroups secondary analysis, to account for multiple comparisons, a P-value <0.017 was used to reject the null hypothesis. Data were analyzed using NCSS version 17 for Windows (NCSS, LLC, Kaysville, UT, USA). Results Forty patients were recruited and enrolled. One patient in the control group dropped out secondary to pruritus

B. Carvalho et al. 31 after 12 h. Two patients dropped out because of postoperative pain: one in the fentanyl 5 lg group who had complete data until 8 h, and another in the fentanyl 25 lg group who had complete data until 12 h. Therefore, partial data for three patients and complete data for 37 patients were analyzed. There were no statistically significant differences in the demographics among groups with respect to age, height, weight, gestational age and parity (Table 1). Total postoperative PCA morphine consumption did not differ among the groups (Fig. 1). Postoperative pain scores during the study period are outlined in Fig. 2. There was a significant difference between the postoperative pain AUC with increasing fentanyl doses (Tau 0.38, P = 0.003, Kendal correlation test). The median [interquartile range] pain AUC in the fentanyl 0 lg control group was 267 [120 360] mm.h, compared with 487 [367 760] mm.h in the fentanyl 5 lg group (P = 0.018), 520 [381 647] mm.h in the fentanyl 10 lg group (P = 0.007), and 608 [483 757] mm.h in the fentanyl 25 lg group (P = 0.023). There was a poor correlation between pain, AUC and total morphine consumption (q = 0.194; P = 0.265). There were no significant differences in postoperative oxygen saturation or respiratory rate between the groups. At no time did any patient have an oxygen saturation <96%. One patient in the fentanyl 25 lg group had a respiratory rate of 8 10 breaths/min from 30 to 50 min after the spinal anesthetic. Otherwise, respiratory rates in all patients varied from 14 24 breaths/min during the intraoperative and postoperative periods. No patient reported pain at spinal placement, skin incision, uterine incision, or skin closure. However, some patients spontaneously reported pain at other times during the surgery and were treated at the discretion of the anesthesiologist providing clinical care. Four patients in the control group versus two patients in all the fentanyl groups combined required treatment for intraoperative nausea (P = 0.031, Table 2). Interventions to treat intraoperative pain were similar in the control group compared to the fentanyl groups combined (P = 0.164, Table 2). There were no differences in blood pressure (minimum or average) or ephedrine use among the study groups (Table 3). Fig. 1 Box plots showing cumulative morphine consumption by patients for each fentanyl dose group. The plot depicts morphine consumption from 0 to 24 h after block placement. The bottom and top of the box are the 25th and 75th percentiles, the horizontal line in the box is the 50th percentile, the ends of the lines are the 5th and 95th percentiles and dots outside the ends represent outliers. There were no significant differences among groups (P = 0.129, Kendal correlation test) and no statistical differences (P = 0.246, Mann Whitney U test) were found comparing the control group to all the fentanyl groups combined. Fig. 2 Visual analog pain scores (VAPS, 0 = no pain, 100 mm = worst pain imaginable) from the time the patient arrived in the PACU to 24 h post-spinal anesthesia in the four study groups. Pain scores as measured by the area under the VAPS time curve (AUC) over the 24-h study period were significantly different among the groups (P = 0.003, Kendal correlation test). D/C = Discharged from PACU, Postop = Post-operatively. Note that the x-axis is not drawn to scale. Table 1 Patient demographics and obstetric characteristics 0 lg 5 lg Fentanyl dose 10 lg 25 lg Age (years) 34 ± 6 36 ± 4 30 ± 7 31 ± 6 Height (cm) 163 ± 8 163 ± 5 163 ± 5 158 ± 8 Weight (kg) 88 ± 7 93 ± 14 82 ± 23 92 ± 18 Gestation (weeks) 38 ± 1 38 ± 1 40 ± 1 39 ± 1 Data are mean ± SD. No differences among groups except for gestational age (P = 0.046).

32 Intrathecal fentanyl and acute tolerance Table 2 Incidence of intraoperative pain and nausea requiring treatment 0 lg 5 lg Fentanyl dose 10 lg Patients requiring treatment for pain 2 0 0 1 Patients requiring treatment for nausea 4 * 0 2 0 Data are number of patients per group. * There was a significant difference compared to the fentanyl groups combined (P = 0.031). 25 lg Table 3 Intraoperative hemodynamic data 0 lg 5 lg Fentanyl dose 10 lg 25 lg Mean BP (mmhg) 70 ± 5 71 ± 7 77 ± 26 70 ± 15 Minimum BP (mmhg) 59 ± 7 57 ± 11 55 ± 9 56 ± 20 Ephedrine dose (mg) 17 ± 10 13 ± 11 19 ± 10 19 ± 15 Data are mean ± SD. There were no significant differences among the groups. BP: blood pressure. Discussion Our primary objective was to investigate if clinically significant acute tolerance to opioids developed with the addition of IT fentanyl to a bupivacaine spinal anesthetic solution containing IT morphine. This was assessed by measuring the postoperative i.v. morphine requirement and postoperative pain for patients who had received IT fentanyl in doses 5, 10, and 25 lg compared to a no-fentanyl group. We expected that if acute tolerance occurred, the amount of morphine consumed and postoperative pain experienced would be greater for the patients who were given fentanyl versus no fentanyl, assuming subjects in each group used the PCA to treat their pain to the same level. In addition, morphine usage and pain would likely increase successively with increasing doses of IT fentanyl if there was a dose response associated with IT fentanyl dose and acute tolerance. Our study did not find any significant difference in postoperative morphine usage with increasing fentanyl dosages. However, pain scores over the first 24 h after surgery, as measured by AUC, were increased in the groups receiving IT fentanyl. Our assumption that patients would self-administer the PCA to keep their pain levels the same across groups was incorrect, which makes interpretation of the data difficult. Patients who received IT fentanyl, on average, did not treat themselves to the same pain level as the patients not receiving IT fentanyl. A number of factors may influence morphine use including side effects such as nausea, patient preferences for analgesics and concerns for exposure of medication to breast-feeding infants. 10 However, we did not measure or control for these factors and potential confounders. In addition, large inter-patient variability in PCA use (24-h morphine doses ranged from 3 to 68 mg) made opioid consumption data difficult to interpret. A post-hoc power calculation based on mean PCA use, showed that we would require 170 patients (control group 42: fentanyl group 128) to have enough power to have shown a difference in analgesic use (SD 10.75 mg, power 0.8, alpha 0.05). Although the higher postoperative pain scores in patients receiving fentanyl did not result in increased analgesic use, we believe that the differences in pain scores suggests that intrathecal fentanyl may induce acute opioid tolerance when IT morphine is included in the spinal anesthesia solution. One limitation with our study design is that it does not allow us to definitively differentiate if IT fentanyl results in tolerance to the IT morphine administered in the spinal solution or the i.v. PCA morphine used to treat postoperative pain. Differentiation between IT and i.v. morphine tolerance could have occurred if only non-opioids were administered for postoperative analgesia, however, we felt that non-opioids (non-steroidal antiinflammatory drugs, acetaminophen) would provide inadequate post-cesarean analgesia and could not be ethically justified at our institution. Four patients in the control group versus two patients in all the fentanyl groups combined required treatment for intraoperative nausea (P = 0.031). A number of other studies have shown less intraoperative pain and nausea when fentanyl is added to the local anesthetic for spinal anesthesia. 2,11 15 Intrathecal fentanyl blunts the noxious input from visceral afferents that are stimulated by pulling on the peritoneal structures and exteriorization or manipulation of the uterus. 16 These visceral afferent stimuli can manifest clinically as nausea, vomiting or cramping and do not appear to be blocked completely with the local anesthetic block alone. Even when IT morphine 200 lg was added, as in this study, visceral stimuli appeared to be inadequately blocked in the absence of IT fentanyl. Although hypotension can cause nausea and be a potential confounder, there was no evidence in our study that patients

B. Carvalho et al. 33 treated for nausea had more hypotension than the other patients. The intraoperative benefit of using IT fentanyl suggested in this study and a number of previous studies appears to outweigh the potential for inducing acute postoperative tolerance. An important message from the data in this study, and other studies of post-cesarean delivery pain 4,17 19 is the variability in the postoperative analgesic use and pain scores across all groups. The phenomenon of acute tolerance that we were studying accounted for very little of the variation in the amount of pain or analgesia required. For example, the range of morphine doses used by any group was three times the median dose in the groups. Large inter-patient variability in opioid use is not unique to this study. 20 Although this study was underpowered to show small differences among groups, we found no apparent benefit with larger IT fentanyl doses. These findings are consistent with those of Hunt et al., who studied IT fentanyl doses, ranging from 2.5 to 50 lg in combination with IT bupivacaine for cesarean delivery. 21 They found improved intraoperative analgesia compared to bupivacaine alone, but minimal additional benefit beyond IT fentanyl doses of 6.25 lg and interestingly also no difference in postoperative morphine use amongst the groups. Side effects from IT opioids are dose-dependent and larger doses of fentanyl may increase the risk of respiratory depression. 11 We did not observe any significant respiratory depression in the study groups. One patient who received fentanyl 25 lg had a decreased respiratory rate intraoperatively for 20 min. Although the patient s respiratory rate fell below 10 breaths/min, the oxygen saturation remained above 96% and the episode resolved spontaneously. A potential limitation of this study is the multiple small-group comparisons, which were a function of the dose response design. The strength of this design is that we obtain information over a wide range of doses. However, the multi-group design reduces the statistical power to show small differences between the groups. Due to the variability in postoperative morphine use, combined with the small sample size and multiple groups, we lacked power to show a small difference in morphine use. Although there was a statistically significant difference in postoperative pain in the fentanyl groups, the clinical effect size was small and did not lead to increased analgesic use despite a five-fold increase in fentanyl dose. However, we acknowledge that subtle differences among study groups would not have been detected due to study design limitations acknowledged above. In conclusion, our findings suggest that IT fentanyl in doses from 5 to 25 lg co-administered with IT bupivacaine and morphine for cesarean delivery under spinal anesthesia may induce subtle acute opioid tolerance as evidenced by higher overall pain scores in patients who received fentanyl compared with the control group. However, this did not result in increased analgesic use post-cesarean delivery. Therefore the clinical importance of our findings is uncertain. Future studies with larger sample sizes are required to fully elucidate this phenomenon and determine its clinical significance before clinicians consider avoid using IT fentanyl for fear of inducing acute tolerance postoperatively, especially with in light of previous studies showing that IT fentanyl results in less intraoperative pain and nausea during cesarean delivery. Disclosure The authors involved in this study and the preparation of the manuscript received no external funding. This study was financially supported by the Department of Anesthesia, Stanford University School of Medicine, Stanford, California. We thank Karen Giarrusso, MD for her valuable help with patient enrollment and data collection. We would also like to thank Alex McMillan, Human Health and Policy Biostatistics, Stanford University. For his expertise and advice on this analysis. References 1. Liu S, Chiu AA, Carpenter RL, et al. Fentanyl prolongs lidocaine spinal anesthesia without prolonging recovery. Anesth Analg 1995;80:730 4. 2. Dahl JB, Jeppesen IS, Jorgensen H, Wetterslev J, Moiniche S. Intraoperative and postoperative analgesic efficacy and adverse effects of intrathecal opioids in patients undergoing cesarean section with spinal anesthesia: a qualitative and quantitative systematic review of randomized controlled trials. Anesthesiology 1999;91:1919 27. 3. Cohen SE, Subak LL, Brose WG, Halpern J. Analgesia after cesarean delivery: patient evaluations and costs of five opioid techniques. Reg Anesth 1991;16:141 9. 4. Palmer CM, Emerson S, Volgoropolous D, Alves D. Doseresponse relationship of intrathecal morphine for postcesarean analgesia. Anesthesiology 1999;90:437 44. 5. Fournier R, Van Gessel E, Macksay M, Gamulin Z. Onset and offset of intrathecal morphine versus nalbuphine for postoperative pain relief after total hip replacement. Acta Anaesthesiol Scand 2000;44:940 5. 6. Fishbain DA, Cole B, Lewis JE, Gao J, Rosomoff RS. Do opioids induce hyperalgesia in humans? An evidence-based structured review. Pain Med 2009;10:829 39. 7. Gustorff B, Hoechtl K, Sycha T, Felouzis E, Lehr S, Kress HG. The effects of remifentanil and gabapentin on hyperalgesia in a new extended inflammatory skin pain model in healthy volunteers. Anesth Analg 2004;98:401 7. 8. Angst MS, Koppert W, Pahl I, Clark DJ, Schmelz M. Short-term infusion of the mu-opioid agonist remifentanil in humans causes hyperalgesia during withdrawal. Pain 2003;106:49 57. 9. Cooper DW, Lindsay SL, Ryall DM, Kokri MS, Eldabe SS, Lear G. A. Does intrathecal fentanyl produce acute cross-tolerance to i.v. morphine? Br J Anaesth 1997;78:311 3. 10. Carvalho B, Cohen SE, Lipman SS, Fuller A, Mathusamy AD, Macario A. Patient preferences for anesthesia outcomes associated with cesarean delivery. Anesth Analg 2005;101:1182 7.

34 Intrathecal fentanyl and acute tolerance 11. Dahlgren G, Hultstrand C, Jakobsson J, Norman M, Eriksson H, Martin H. Intrathecal sufentanil, fentanyl, or placebo added to bupivacaine for cesarean section. Anesth Analg 1997;85:1288 93. 12. Manullang TR, Viscomi CM, Pace NL. Intrathecal fentanyl is superior to intravenous ondansetron for the prevention of perioperative nausea during cesarean delivery with spinal anesthesia. Anesth Analg 2000;90:1162 6. 13. Ben-David B, Miller G, Gavriel R, Gurevitch A. Low-dose bupivacaine-fentanyl spinal anesthesia for cesarean delivery. Reg Anesth Pain Med 2000;25:235 9. 14. Palmer CM, Voulgaropoulos D, Alves D. Subarachnoid fentanyl augments lidocaine spinal anesthesia for cesarean delivery. Reg Anesth 1995;20:389 94. 15. Connelly NR, Dunn SM, Ingold V, Villa EA. The use of fentanyl added to morphine-lidocaine-epinephrine spinal solution in patients undergoing cesarean section. Anesth Analg 1994;78:918 20. 16. Wang C, Chakrabarti MK, Whitwam JG. Specific enhancement by fentanyl of the effects of intrathecal bupivacaine on nociceptive afferent but not on sympathetic efferent pathways in dogs. Anesthesiology 1993;79:766 73. 17. Girgin NK, Gurbet A, Turker G, Aksu H, Gulhan N. Intrathecal morphine in anesthesia for cesarean delivery: dose-response relationship for combinations of low-dose intrathecal morphine and spinal bupivacaine. J Clin Anesth 2008;20:180 5. 18. Carvalho B, Riley E, Cohen SE, et al. Single-dose, sustainedrelease epidural morphine in the management of postoperative pain after elective cesarean delivery: results of a multicenter randomized controlled study. Anesth Analg 2005;100:1150 8. 19. Carvalho B, Roland LM, Chu LF, Campitelli 3rd VA, Riley ET. Single-dose, extended-release epidural morphine (DepoDur) compared to conventional epidural morphine for post-cesarean pain. Anesth Analg 2007;105:176 83. 20. Sia AT, Lim Y, Lim EC, et al. A118G single nucleotide polymorphism of human mu-opioid receptor gene influences pain perception and patient-controlled intravenous morphine consumption after intrathecal morphine for postcesarean analgesia. Anesthesiology 2008;109:520 6. 21. Hunt CO, Naulty JS, Bader AM, et al. Perioperative analgesia with subarachnoid fentanyl-bupivacaine for cesarean delivery. Anesthesiology 1989;71:535 40.