The effect of addition of low dose fentanyl to epidural bupivacaine (0.5%) in patients undergoing elective caesarean section: A randomized, parallel group, double blind, placebo controlled studyLH Parate1, SP Manjrekar2, TC Anandaswamy1, B Manjunath3
1 Department of Anaesthesia, MS Ramaiah Medical College, Bangalore, Karnataka, India
2 Department of Anaesthesia, Indira Gandhi Medical College, Nagpur, Maharashtra, India
3 Department of Community Medicine, MS Ramaiah Medical College, Bangalore, Karnataka, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.147032 Clinical trial registration CTRI/2014/06/004698
Source of Support: None, Conflict of Interest: None
Clinical trial registration CTRI/2014/06/004698
Background: Opioids have synergistic action with local anesthetics which may alter characteristics of epidural block. Giving opioids to mother before delivery of baby is still fully not accepted with some fearing risk of neonatal depression. Aims: Our primary aim was to evaluate the analgesic effect of addition of 50 μg fentanyl to epidural 0.5% bupivacaine in patients undergoing elective caesarean section using visual analog scale. The secondary aim was to assess onset of analgesia, volume of drug required to achieve T6 level, grade and duration of motor block and Apgar score. Materials and Methods: In this prospective, randomized, double blind, placebo controlled study 64 patients scheduled for elective caesarean section under epidural anesthesia were randomly divided into two groups of 32 each. The fentanyl group received 1ml of 50 μg fentanyl and the saline group received 1ml of normal saline mixed with 10ml of 0.5% bupivacaine for epidural anesthesia. VAS score, time to achieve T6 level, dose of bupivacaine, intraoperative analgesic consumption and duration of analgesia, grade and duration of motor block and any adverse maternal and neonatal effects were noted. Statistical Analysis: Data was analyzed using Students t test, chi-square test and Mann-Whitney U-test. The values of P < 0.05 were considered statistically significant. Results: Fentanyl improved the VAS score significantly (1.6 ± 1.32) compared to the saline group (3.77 ± 1.0, P < 0.0001). It also reduced the intraoperaitve analgesic supplementation compared to the saline group. (P = 0.031). The postoperative duration of analgesia was prolonged in the fentanyl group (275.80 ± 13.61 min) compared to the saline group (191.47 ± 12.16 min, P < 0.0001). The other characteristics of epidural block were unaltered. Conclusion: Addition of 50 μg fentanyl to epidural 0.5% bupivacaine significantly reduces the VAS score. It also reduces intra-operative analgesia supplementation and prolongs the duration of postoperative analgesia without altering the other characteristics of block. The neonatal outcome is not affected with addition of fentanyl before delivery of baby.
Keywords: Elective caesarean section, epidural 0.5% bupivacaine, fentanyl
Epidural anesthesia as a technique for caesarean section has gained popularity in recent years because of increase in demand for labor analgesia.  In spinal and epidural anesthesia, despite adequate sensory block, the incidence of visceral pain among caesarean section can be as high as 50%.  At times, this pain can be so stressful that it requires conversion to general anesthesia. Epidural anesthesia is less preferred compared to spinal anesthesia as it provides less dense block, has longer duration of onset and requires large volume of local anesthetics. Various adjuvants have been suggested to augment the effect of local anesthetics and these include sodium bicarbonate, phenylephrine, midazolam, ketamine and opioids.  Though opioids remain the adjuvants of choice, they are deferred until umbilical cord clamping in obstetrics. Additionally, giving opioid to mothers before delivery carries risk of maternal and neonatal respiratory depression. Various doses of fentanyl and local anesthetic combination have been studied for epidural anesthesia in obstetric and non-obstetric patients. These studies though have used different combinations of fentanyl and local anesthetic, different parameters to assess the quality of epidural anesthesia, and had small sample sizes with varying results. ,,,,,
A randomized, parallel group, double blind study was thus conducted by us to evaluate the analgesic effect of addition of 50 μg fentanyl to epidural 0.5% bupivacaine on characteristics of epidural block in patients undergoing elective caesarean section. In addition, we assessed the effect of fentanyl on other parameters of epidural block in term of onset of sensory block, degree and duration of motor block and volume required to achieve T6 level. Neonatal outcome was assessed using Apgar score.
Ethics: The study was approved by the institutional ethics committee and written, informed consent was obtained from all participants.
Selection criteria: ASA I and II full-term patients posted for elective caesarean section were included in the study. Patients in labor, with fetal distress, spine problems or requiring general anesthesia due to technical failure were excluded.
Randomization, allocation concealment and blinding: This was done using a computer-generated random numbers table and allocation was concealed using sealed numbered opaque envelopes. Both the patient and anesthetist were blinded to the group allocation.
Study medications: One group received 10ml of 0.5% bupivacaine and 1 ml of normal saline (Total volume = 11 ml) while the other group received 10 ml of 0.5% bupivacaine plus 1 ml of 50 μg fentanyl (Total volume = 11 ml). All medications were prepared by an anesthetist who was not involved in the study.
Study procedure: On arrival of patient in the operating room, baseline pulse, blood pressure, respiratory rate, and oxygen saturation were noted. All patients received premedication with intravenous Pantoprazole (40 mg) and metoclopramide (10 mg) as also preloading with 500 ml of Ringer's lactate. All epidural blocks were performed in left lateral position. After providing local infiltration with 1% lignocaine, 18G Tuohy needle was inserted at L2-L3 or L3-L4 interspace. Epidural space was identified by the loss of resistance. An 18G epidural catheter was inserted in cephalic direction and 4 cm of catheter was kept in the epidural space. A test dose of 3 ml of 2% lignocaine with adrenaline 1:200000 was given to rule out its accidental intravascular or intrathecal placement.
After the initial bolus, the level of analgesia was tested by loss of pinprick sensation at 5 min interval till 15 min. Additional increments of 3 ml of 0.5% bupivacaine were given every 5 min if bilateral sensory block at T6 level was not achieved at 15 minutes. The total volume injected and the time taken to attain T6 level were noted. The duration of first epidural injection to the achievement of T6 sensory anesthesia was recorded as onset of analgesia. After attaining T6 level, motor block was assessed using a Bromage scale:  0 = No paralysis, 1 = Inability to raise extended legs. 2 = Inability to flex the knee, 3 = Inability to flex the ankle. The duration of motor block was defined as the time from establishment of motor block till complete recovery of motor block, i.e. ability to raise both legs independently. Intraoperative pain was managed by nitrous oxide with oxygen by mask and IV midazolam 1mg. If there was no relief then analgesia was supplemented with IV ketamine 0.5 mg/kg in addition to midazolam and nitrous oxide. If pain persisted despite analgesics, then general anesthesia with endotracheal intubation was done. Patients requiring general anesthesia were excluded from the study. Intraoperatively continuous ECG, S p O 2 monitoring was done. Blood pressure, respiratory rate was monitored every 5 min. All patients received supplementary oxygen by face mask. Hypotension was defined as fall in systolic blood pressure below 90 mmHg or >20% fall from baseline. Hypotension was treated with IV mephentermine 6 mg and was repeated as necessary. Time of start of surgery and baby out time were noted. Oxytocin drip of 10 units in 500 ml normal saline was started after extraction of baby.
Safety evaluation: Neonatal assessment was done using APGAR scores at 1 min and 5 min by attending neonatologist who was unaware of study group. The side effects evaluated were pruritis, sedation, nausea, vomiting, hypotension and respiratory depression. Respiratory depression was defined as respiratory rate <10 per min or fall in S p O 2 < 90%. Injection naloxone was kept ready to treat respiratory depression. If nausea, vomiting occurred, IV ondansetron 4 mg was given as rescue antiemetic. Pruritis if severe was treated with IV pheniraminemaleate. Urinary retention as a side effect was not possible to evaluate as all patients were catheterized for twenty four hrs in the postoperative period.
Efficacy evaluation: Immediately after surgery, the quality of intraoperative analgesia and patients' satisfaction during surgery was assessed using visual analog scale (VAS) (0 = no pain, 10 = worst imaginable pain).  Postoperative pain assessment was done at every 1hour interval using VAS till the time to first analgesic request, i.e. VAS > 4. Once score was above 4, intravenous diclofenac 50mg twice a day and intravenous paracetamol 1gm three times a day were given as postoperative analgesia. The time for first analgesia demand was noted and termed as duration of postoperative analgesia. The primary outcome measure was the Visual Analog Score (VAS). The secondary outcome measures were the intraoperative analgesic supplementation, and duration of postoperative analgesia, fentanyl-related side effects, intraoperative hemodynamic changes, onset of analgesia, volume of drug required to achieve T6 level, grade and duration of motor block and adverse events in newborn.
Sample size was calculated based on the study of Shapiro et al. who reported mean difference in VAS scale of 2.6 cm using n master 2.0 software.  We considered mean difference of 0.2 cm, standard deviation [SD] of 0.3 for the saline group and 0.08 for the fentanyl group with an alpha error of 5% and power of 95%. We estimated that at least 30 patients would be required in each group and made the sample 32/group to account for dropouts. SPSS version 16.0 was used for analysis. Descriptive data were expressed as mean ± SD or percentages. Quantitative variables were tested for normality at the outset and subsequently analyzed using the Students t test or the Mann Whitney U test. Categorical variables were compared using the chi square test. All analyses were carried out at 5% significance.
Demographics: Total 64 patients were randomized for the study. Two patients in each group were excluded from the study due to failed epidural anesthesia, requiring conversion to general anesthesia. Thirty cases each in the bupivacaine alone and bupivacaine + fentanyl group were eligible for the per protocol analysis [Figure 1]. Both groups were comparable for age, weight, height, parity. The mean duration of surgery, administration - baby out interval and mean Apgar score in both the groups did not showed any statistical difference [Table 1].
Efficacy outcomes: A comparison of characteristic of epidural block showed no difference in time of onset of analgesia and total dose of bupivacaine in both groups [Table 2]. The mean duration of onset of analgesia in the control group was 21 ± 2.21 min and in the combination group 21.07 ± 2.39 min (P = 0.911). The mean total volume of bupivacaine required to achieve T6 level in the control group was 17.60 ± 1.77 ml while in the combination group it was 16.80 ± 1.79 ml (P = 0.087). Both differences were not statistically significant. The mean duration of the motor block in the control group was 192.60 ± 9.46 min and in the combination group 197.73 ± 11.85 min (P = 0.069). The mean duration of postoperative analgesia was 191.47 ± 12.16 min in the bupivacaine alone group and 275.80 ± 13.61 min in the bupivacaine + fentanyl group (P < 0.0001).
A comparison of intraoperative analgesic supplementation, grade of motor block and VAS is shown in [Table 3]. In the bupivacaine + fentanyl group, 22/30 patients had no pain as compared to 12/30 in the bupivacaine only group (P value = 0.031). Nitrous oxide and midazolam were given in 12 and 6 patients respectively in the control and combination groups. Six patients in the control group and two patients in the fentanyl group required ketamine supplementation (P value = 0.031).
A comparison of the grade of motor block in both the groups showed that two patients in each group had no apparent motor block. Twelve patients in the control group and 10 in the combination group had grade 1 motor block. Fourteen patients from each group had grade 2 motor block. Two patients in the saline group and four patients in the fentanyl group had grade 3 block. However, the grade of motor block did not differ significantly in both groups (P = 0.837). The mean intraoperative VAS score in the saline group was 3.77 ± 1.0 and in the fentanyl group was 1.6 ± 1.32 and this was significantly different (P = 0.0001).
Safety: The incidence of maternal complication is shown in [Table 4]. Though the proportion of patients suffering from nausea and vomiting in the combination group was less, the difference was not statistically significant. There were also no significant differences between the groups with regards to pruritus, hypotension, and number of patients needing sedation. No patient in either group developed respiratory depression.
This study demonstrated that addition of 50 μg fentanyl to epidural 0.5% bupivacaine significantly reduces the VAS score, reduces the intraoperative analgesia supplementation and increases the duration of postoperative analgesia with other characteristics of epidural block remaining unaffected.
Twenty-two patients in the fentanyl group had excellent quality of analgesia while higher number of patients in the saline group required rescue ketamine and nitrous oxide with midazolam supplementation. Our study confirms the results of previous reports where fentanyl has improved the quality of analgesia. ,, In previous studies intraoperative pain assessment was done using various methods most commonly being patient generated VAS. In our study in addition to VAS, this assessment and further supplementation of analgesics was done by anesthetist on the basis of patients' response to surgical stimulation. We used ketamine as rescue analgesic in contrast to intravenous opioids used in most of the studies. The reason being the concern regarding using high doses of opioids before delivery of baby and this has helped us to assess the action of epidural fentanyl on characteristic of epidural block. A comparison of epidural and intravenous route has shown that epidural given opioid has more potent analgesic and anesthetic effect due to its spinal action. , Epidurally given opioid helps in early extubation and provides longer duration of analgesia. , Incision, uterine manipulation, peritoneal retraction and delivery of baby are the events with maximum surgical pain. The practice of uterus exteriorization is common in our hospital. As per National Institute for Clinical Excellence Guidelines for caesarean section exteriorization was not recommended because it is associated with more pain.  This may explain high incidence of visceral pain in our study. Visceral pain is transmitted by unmyelinated "C" fibers. Inspite of adequate sensory block, these fibers may not adequately block. Electrophysiological studies revealed that opioids selectively depress "C" fibers at the level of dorsal horn. This may justify less incidence of visceral pain in the fentanyl group. In this study the duration of postoperative analgesia was significantly improved by addition of fentanyl. The mean duration of analgesia in the fentanyl group was 275 min and the saline group 191 min. This finding complements with 8 h analgesia with 100 μg fentanyl.  Interaction with each other's bioavailability at spinal cord is suggested for this synergy. , The volume required to achieve T4-T6 sensory level in epidural anesthesia varies from 10 to 25 ml. ,,, Addition of 50 μg fentanyl has shown to reduce the volume of 0.5% bupivacaine to achieve the T6 level as minimum as 10 ml.  When used intrathecally fentanyl is able to reduce the dose of bupivacaine in spinal anesthesia.  In our study bupivacaine dose remained the same in both groups. The minimum volume required to achieve T6 level in the fentanyl group was 14 ml, average volume being 16 ml. Onset of analgesia is unaffected in our study. Though various combinations of fentanyl and local anesthetic have shown to improve onset of analgesia, the target level of sensory anesthesia and doses of fentanyl with local anesthetics were different. ,, In our study the grade and duration of motor block was unaffected. Duration of motor block was prolonged upto 8h by 100 μg fentanyl.  Prolongation of motor block is not desirable in caesarean patients.
In this study less number of patients (3.33%) in the fentanyl group complained of nausea and vomiting compared to the saline group (13.33%). It is unlikely that only hypotension has resulted in nausea and vomiting since hypotension occurred equally in both groups [40%]. Visceral pain is a known factor to cause nausea and vomiting.  Lack of visceral pain may explain lower incidence of nausea in the fentanyl group. Other explanation could be more use of nitrous oxide in the saline group compared to the fentanyl group. Three patients in the fentanyl group and one patient from the control group complained of pruritis which was mild and did not require any treatment. Incidence of urinary retention could not be assessed as all patients were catheterized. Incidence of hypotension did not differ in both groups and it was transient, responded quickly to intravenous mephentermine. Four patients in the saline group and two patients in the fentanyl group were sedated which was mild and patients were easily arousable. Sedation can be a result of ketamine and midazolam itself rather than fentanyl. None of our patient had respiratory rate below 10 or fall in SpO 2 below 90%.
All Apgar score in both groups were 9 at 5 min except for one baby in the saline group whose score was 8. Mothers who receive fentanyl exclusively by epidural route are known to deliver neonates with respiratory depression. Even at lower doses, neonates are prone to respiratory depression because of the immaturity of the respiratory system. Kumar et al., reported two cases of neonatal respiratory depression requiring naloxone treatment after receiving over 200 μg epidural fentanyl to mother.  In our study although no case of respiratory depression was observed, it is recommended that neonatologist should be alerted when opioids are used before delivery of baby.
This study is limited by the selection of T6 dermatome for highest sensory blockade. In caesarean sections, ideally T4-S5 dermatome sensory block is required. But there is no uniformity with the type of stimulus to be used to assess the level and there is wide, unpredictable correlation with the loss of touch and pinprick sensation. Loss of touch sensation at T6 level has been recommended for pain free caesarean section.  Another limitation is the choice of ketamine whch is known to cause psychiatric disturbances. Low, subanesthetic dose of ketamine gives analgesia without any emergence delirium, hallucination and hemodynamic changes in parturients.  The intensity of anesthetic block in epidural anesthesia is inferior to the dense anesthetic block produced by spinal anesthesia. With epidural 0.5% bupivacaine, satisfactory sensory and motor block score reported is 62% and 71%, respectively.  In our study 20% patients in the saline group and 7% patients in the fentanyl group required ketamine as supplementary analgesia. Hence, analgesic intervention done in this study may itself reflect the limitation of epidural anesthesia.We conclude that addition of low dose fentanyl to epidural 0.5% bupivacaine reduced the need of supplementary intraoperative analgesicsand prolonged duration of postoperative analgesia. The onset of analgesia, volume required to achieve T6 level and duration of motor block remain unaffected. Epidural fentanyl did not affect neonatal outcome but further evaluation is needed while using this technique in mothers with compromised fetuses.
[Table 1], [Table 2], [Table 3], [Table 4]