The Effect of Neuraxial Versus General Anesthesia Techniques on Postoperative Quality of Recovery and Analgesia After Abdominal Hysterectomy: A Prospective, Randomized, Controlled Trial
Lucas J. Santana Catro-Alves, MD*,
Vera Lucia Fernandes De Azevedo, MD, MS*,
Tania F. De Freitas Braga, MD*,
Antonio C. Goncalves, MD*and
Gildasio S. De Oliveira Jr., MD, MSCI†
+Author Affiliations
- Address correspondence to Gildasio De Oliveira, Jr., MD, MSCI, Department of Anesthesiology, Northwestern Memorial Hospital, 251 E. Huron St., F5–704, Chicago, IL 60611. Address e-mail to G-jr@northwestern.edu.
Abstract
BACKGROUND: Patients undergoing abdominal hysterectomy often have significant postoperative pain despite the use of concurrent multimodal pain strategies. Neuraxial anesthesia has opioid-sparing effects and may provide better postoperative recovery to patients when compared with general anesthesia. Our main objective in this study was to compare the effects of neuraxial and general anesthesia on postoperative quality of recovery after abdominal hysterectomy.
METHODS: The study was a prospective, randomized, controlled clinical trial. Seventy healthy females were recruited and randomized to a general anesthesia or neuraxial technique as their primary anesthetic regimen. The primary outcome was the global quality of recovery—40 questionnaire (QoR–40) at 24 hours after the surgical procedure. Other data collected included postoperative pain scores and opioid consumption. Data were analyzed using the Mann–Whitney U test, Fisher's exact test, and linear regression. A P value <0.05 was considered statistically significant.
RESULTS: The median difference (95% confidence interval [CI]) in the global QoR–40 score at 24 hours between the neuraxial and general anesthesia groups was 17 (11 to 21.5) (P < 0.001). Patients in the neuraxial anesthesia group had better quality of recovery scores in all the QoR–40 subcomponents than did the general anesthesia group (all P < 0.005). The median difference in global QoR–40 scores at 48 hours between the neuraxial anesthesia and the general anesthesia groups was 8 (6–10) (P < 0.001). Postoperative opioid consumption and pain scores were higher in the general anesthesia group than in the neuraxial anesthesia group. There was an inverse linear relationship between opioid consumption and postoperative quality of recovery at 24 hours, r2 = 0.67 (P < 0.0001, 95% CI of 0.77 to 0.51), and at 48 hours, r2 = 0.58 (P < 0.0001, 95% CI of 0.72 to 0.42).
CONCLUSION: Neuraxial anesthesia provides better quality of recovery than does general anesthesia for patients undergoing abdominal hysterectomy. The opioid-sparing effects of neuraxial anesthesia were associated with a better quality of recovery in patients after the surgical procedure. In the absence of contraindications, neuraxial anesthesia seems to be a better anesthetic plan for those patients.
Hysterectomy is one of the most common surgical procedures with >600,000 surgeries performed each year in the United States.1 It has been estimated that the majority of hysterectomies are performed through an open incision.2 The lack of beneficial recovery outcomes comparing anesthetic techniques after abdominal hysterectomy often leaves the choice of the anesthetic plan to the preferential decision of surgeons and anesthesiologists who frequently choose general anesthesia.3,4 Even though neuraxial anesthesia may be an option to the anesthetic management of patients undergoing abdominal hysterectomy, it can also be time-consuming and has an intrinsic failure rate, even in experienced hands. It is also unknown whether the possible reduction in the use of systemic opioids provided by neuraxial opioids is associated with a better quality of recovery to patients.
The primary objective of this study was to compare the effect of 2 anesthetic techniques (general and neuraxial anesthesia) on the postoperative quality of recovery and analgesia for patients undergoing abdominal hysterectomy. We hypothesized that patients having abdominal hysterectomy under neuraxial anesthesia would have a better quality of recovery than patients having the same procedure under general anesthesia. Secondary outcomes for the study included postoperative pain scores and opioid consumption.
METHODS
To compare the effects of general and neuraxial anesthesia on postoperative quality of recovery, we performed a prospective, randomized, controlled trial. Study approval was obtained from Santo Antonio Hospital Institutional Ethics Review Board, and written informed consent was obtained from all the study participants. Eligible subjects were ASA physical status I and II females undergoing total abdominal hysterectomy for benign (noncancerous) disease. Patients with a history of coagulopathy, recent infection (<1 month), current use of an opioid analgesic or corticosteroid, allergies to drugs included in the study protocol, gross neurologic impairment or suspected difficult airway were excluded. Reasons for exclusion after randomization were protocol violations or patient request. Subjects were randomized using a computer-generated table of random numbers into 2 groups: general anesthesia and neuraxial anesthesia. Group assignments were sealed in sequentially numbered opaque envelopes that were opened after patient inclusion in the study.
All subjects were premedicated with 0.04 mg/kg IV midazolam. After arrival in the operating room, standard ASA monitors were applied. In the general anesthesia group, induction was performed using 2 to 3 mcg/kg of IV fentanyl, 1.5 to 2.0 mg/kg of propofol, and 0.6 mg/kg of rocuronium. Anesthetic maintenance was achieved with isoflurane at 1 minimum alveolar concentration (MAC), fentanyl 1 mcg/kg titrated to avoid arterial blood pressure values above 20% of baseline, and additional doses of rocuronium to keep 2 twitches using a train-of-4 monitor (TOF Watch SX, Bluestar Enterprises, Inc., Omaha, NE). At skin closure, neuromuscular blockade was antagonized with 0.01 mg/kg of atropine and 0.05 mg/kg of neostigmine. At the end of the surgical procedure before tracheal extubation, subjects received 0.1 mg/kg of IV morphine. In the neuraxial anesthesia group, subjects received 10 mL/kg of a lactated Ringer's solution IV and were placed in the sitting position. After the lumbar region was prepared and draped in a sterile fashion, topical 1% lidocaine was given for skin analgesia. The spinal block was achieved at the L3 to 4 or L4 to 5 interspace using a spinal introducer and a 27 G Pecan spinal needle (B. Braun Medical Inc., Bethlehem, PA). Hyperbaric bupivacaine 17.5 mg 0.5% and 60 mcg of preservative-free morphine sulfate were injected intrathecally. The correct placement of the needle was confirmed by aspiration of clear cerebral spinal fluid before and after the injection. After confirmation of an appropriate dermatomal level of block (T4 level), the surgeon was allowed to proceed.
Patients in both groups received ephedrine 5 mg titrated to keep arterial blood pressure within 20% of the baseline value. All subjects received 8 mg of dexamethasone after induction and 4 mg of ondansetron at the end of the surgical procedure to decrease postoperative nausea and vomiting. In the postanesthesia care unit (PACU), subjects were asked to rate their pain upon arrival and at regular intervals on a 0 to 10 pain numeric rating scale, in which 0 means no pain and 10 is the worst pain imaginable. The scores were assessed while at rest and after asking the subjects to cough. Morphine sulfate 1 to 2 mg IV was administered every 10 minutes to maintain a numeric rating scale pain score <4 (1 mg for pain <7 out of 10 and 2 mg for pain ≥7 out of 10). Nausea was assessed at the same intervals and recorded as present or absent. Number of vomiting episodes was also recorded. Discharge readiness from the PACU was assessed by using the modified Aldrete's score5 every 15 minutes until patients met discharge criteria (score ≥9). All subjects received a multimodal regimen consisting of ketoprofen 0.1 mg/kg IV every 8 hours and metamizole 30 mg/kg IV every 6 hours. Postoperative nausea and vomiting were treated with metoclopramide 10 mg IV followed by 4 mg IV ondansetron if symptoms persisted. The presence of pruritus was recorded and treated with 25 mg of diphenhydramine every 6 hours as needed.
Perioperative data collected included subject's age, height, weight, ASA physical class, surgical duration, total IV fluids, pain scores at rest and after coughing, and total amount of morphine given in the PACU, at 24 and 48 hours. Opioid use did not include that given in the operating room. Perioperative data collection was performed by one of the investigators not involved with patient care. The perioperative data collected were cross-checked with assessments of the nursing staff. In cases in which a discrepancy was noted, the subject was reassessed by the nurse provider and the investigator responsible for the data collection. A quality of recovery questionnaire (QoR–40) was completed by the subjects at 24 and 48 hours after the surgical procedure.6 The QoR–40 scoring system was explained to all subjects and reviewed to provide accurate understanding of all questions. The questionnaire evaluates 5 components of patient recovery: physical comfort (12 questions), physical independence (5 questions), emotional state (9 questions), psychological support (7 questions), and pain (7 questions). The sum of the individual components generates an aggregate score. Global QoR–40 scores range from 40 to 200, representing respectively very poor to outstanding quality of recovery. The instrument was translated to the Portuguese language by one of the investigators fluent in both English and Portuguese (GSD). The simple language content of the instrument makes changes in semantics during the translation process extremely unlikely to alter the instrument validity, and this fact has justified the use of the instrument in languages other than English despite lack of formal validation.7,8 The quality of recovery instrument has had its validity confirmed in patients undergoing procedures using regional anesthesia.9 It has also been used to evaluate outcomes after neuraxial anesthesia and analgesia.8,10
The primary outcome was the QoR–40 aggregate score at 24 hours. A sample size of 31 subjects per group was estimated to achieve 90% power to detect a 10-point difference in the aggregated QoR–40 scores for the 2 study groups to be compared. A 10-point difference represents a 15% improvement on quality of recovery on the basis of previously reported values on the mean and range of the QoR–40 in patients after anesthesia and surgery.6 The responsiveness of this instrument has been assessed in patients evaluated before and after surgery.11 In a study of outcomes after cardiac surgery, a poorer quality of life at 3 months was found in subjects who had median QoR–40 global values 10 points less than those with higher QoR–40 values 3 days after cardiac surgery.12 To account for dropouts, we randomized 70 subjects.
The Shapiro–Wilk and Kolmogorov–Smirnov tests were used to test the hypothesis of normal distribution. Normally distributed continuous data (age and body mass index) are reported as mean (SD) and were evaluated with a 2-sided independent ttest for equal variances. Nonnormally distributed interval data and ordinal data are reported as median (interquartile range [IRQ]) and were evaluated using the Mann–Whiney U test. Median differences and 95% confidence intervals (CI) of the differences in global QoR–40 scores and pain scores were calculated using a 10,000 sample bootstrapping method.13 Bootstrapping is recognized as a robust method to obtain confidence intervals because it does not rely on parametric assumptions to perform calculations. Categorical variables are presented as count (percentages) and were evaluated using Fisher's exact test. A bivariate analysis with simple linear regression was used to detect an association between opioid consumption and global quality of recovery with the change in r2 evaluated for significance. To avoid the chance of a type I error, the criterion for rejection of the null hypothesis was a 2-tailed P < 0.01 for comparisons involving the primary outcome variable. A value of P < 0.05 was used for all other comparisons.
Statistical analysis was performed using STATA version 11 (College Station, TX).
RESULTS
The details of the conduct of the study are shown in Figure 1. Seventy subjects were randomized, and 68 completed the study. Patients were enrolled consecutively from September 2010 through March 2011. Patients' baseline characteristics and surgical factors were not different between groups (Table 1).
Figure 1.
Consort flow study diagram.
View this table:
Table 1.
Subjects' Baseline Characteristics and Operative Data
Subjects in the neuraxial group had better postoperative quality of recovery than did the general anesthesia group. The median difference (95% CI) in global recovery scores (QoR–40) at 24 hours after surgery between the neuraxial anesthesia and the general anesthesia was 17 (11 to 21.5) (P < 0.001). The differences between individual subcomponents in the QoR–40 are presented inFigure 2. The median difference in global QoR–40 scores at 48 hours between the neuraxial anesthesia and the general anesthesia groups was 8 (6–10) (P < 0.001). The individual subcomponents in the QoR–40 at 48 hours are presented in Table 2. The neuraxial anesthesia group had higher scores than did the general anesthesia group in all individual subcomponents at 24 and at 48 hours postoperatively (all Ps < 0.01).
Figure 2.
Box plot of dimensions of quality of recovery—40 questionnaire (QoR–40) questionnaires completed 24 hours after abdominal hysterectomy surgery. Median values shown as solid line within box of 25th and 75th percentile values. Whiskers represent 5th and 95th percentile values. †Different from general anesthesia, P = 0.05. Data were compared using the Mann–Whitney U test.
View this table:
Table 2.
Dimensions of the Quality of Recovery—40 (QoR–40) Questionnaire by Study Groups at 48 Hours After Surgery
The neuraxial anesthesia group had lower pain scores at rest and after coughing than did the general anesthesia group up to 48 hours after the surgical procedure (Table 3). Opioid consumption in the PACU at 24 and at 48 hours was lower in the neuraxial anesthesia group than in the general anesthesia group (Table 4). The neuraxial anesthesia group had less nausea and vomiting in the PACU and less nausea in the first 24 hours after the procedure (Table 4). Simple linear regression demonstrated an inverse relationship between cumulative opioid consumption and global quality of recovery at 24 hours, r2 = 0.67 (P < 0.0001, 95% CI of 0.77 to 0.51), and at 48 hours, r2 = 0.58 (P < 0.0001, 95% CI of 0.72 to 0.42) (Fig. 3).
View this table:
Table 3.
Pain Scores at Rest and Coughing
View this table:
Table 4.
Opioid Consumption and Postoperative Nausea and Vomiting
Figure 3.
A, Scatter plot and regression analysis demonstrating inverse relationship between global quality of recovery—40 questionnaire (QoR–40) scores and 24-hour cumulative opioid consumption after surgery. Slope of regression line −0.45 ± 0.03, goodness of fit r2 = 0.67, and slope significantly different from 0 (P < 0.0001). B, Scatter plot and regression analysis demonstrating inverse relationship between global QoR–40 scores at 48 hours and 48 hours cumulative opioid consumption after surgery. Slope of regression line −0.84 ± 0.08, goodness of fit r2 = 0.58, and slope significantly different from 0 (P < 0.0001).
The mean (95% CI) difference in time to meet discharge criteria from the PACU between the neuraxial anesthesia and general anesthesia groups was 4 minutes (−1 to 9) (P = 0.11).
Four subjects in the neuraxial group and 3 in the general anesthesia group developed postoperative pruritus. No cases of postoperative respiratory depression were documented in any of the study groups.
CONCLUSIONS
The important finding of the current investigation is the better postoperative quality of recovery promoted by neuraxial anesthesia in comparison with general anesthesia in patients undergoing abdominal hysterectomy. Improvement in quality of recovery using different anesthesia techniques after abdominal hysterectomy has not been demonstrated. Patients in the neuraxial anesthesia group had better scores in all subcomponents of the QoR–40 questionnaire than did patients in the general anesthesia group. The effects of neuraxial anesthesia in improving patients' recovery were extended up to 48 hours after the surgical procedure. These findings have important implications regarding the choice of the anesthetic plan for patients undergoing abdominal hysterectomy.
Another important finding of this study was the inverse relationship between opioid-sparing effects and the quality of recovery. Patients who required fewer opioids had better quality of recovery scores. The opioid-sparing effects of the neuraxial technique were associated with a better quality of recovery for that group when compared with the general anesthesia group. This finding is significant because despite the known opioid-sparing effects of regional anesthesia techniques, it is not determined whether the reduction in opioid consumption is associated with an improvement in postoperative recovery for patients.14
The lack of differences in major morbidity or mortality when anesthesia techniques are compared often leaves the decision of the anesthetic plan to the personal preferences of patients, surgeons, and anesthesia providers. The increased anesthetic induction time of neuraxial anesthesia in comparison with general anesthesia can influence surgeons and anesthesiologists in favor of general anesthesia techniques.15 This choice is commonly done in a context in which patient-oriented outcomes comparing the 2 techniques are lacking.16 The better postsurgical physical comfort, pain control, physical independence, psychological support, and emotional state experienced by patients receiving neuraxial anesthesia in comparison with general anesthesia for abdominal hysterectomy can change surgeons' and anesthesiologists' choice in favor of neuraxial anesthesia to perform those cases.
Other investigators have compared neuraxial anesthesia with general anesthesia techniques in patients undergoing abdominal hysterectomy. Despite the reduction in postoperative pain and opioid consumption by the neuraxial anesthesia technique, they did not demonstrate better postoperative recovery. Sprung et al. showed less postoperative nausea in the PACU but no improvement in other outcomes for patients receiving intrathecal anesthesia versus general anesthesia.17 Wodlin et al. showed a faster recovery of bowel function in patients receiving intrathecal anesthesia in comparison with general anesthesia.18 In the aforementioned studies, despite better postoperative pain control in the neuraxial groups, patients did not achieve faster hospital discharge. We did not examine time to hospital discharge in the current study; the lack of standardized and validated criteria make hospital discharge an unreliable outcome to compare different anesthetic techniques.15
Neuraxial anesthesia techniques improved postoperative recovery in patients undergoing surgical procedures other than abdominal hysterectomy. Caputo et al. concluded that the use of a thoracic epidural improves overall recovery in patients undergoing coronary artery bypass grafting.19 Jellish et al. demonstrated better postoperative pain control and less postoperative nausea in patients receiving spinal anesthesia in comparison with general anesthesia for spinal laminectomy.20 In contrast to our study, those investigators did not establish an association between less opioid requirement and improved quality of recovery.
Neuraxial opioids can have systemic opioid-sparing effects up to 48 hours after surgery but can also result in increased undesirable side effects such as respiratory depression and pruritus.21 In the current study, we used a small 60-mcg dose of intrathecal morphine in an attempt to minimize those side effects. This small dose was comparable to the lowest effective dose examined by Palmer et al. in patients undergoing cesarean surgery.22 Our study design was not powered to detect differences between groups in secondary outcomes such as postoperative respiratory depression and pruritus. It would require 1565 patients per group to achieve 80% power to detect a difference in the incidence of pruritus on the basis of the frequency distribution we observed in the 2 study groups.
It is important to note that the use of systemic medications can have important effects on postoperative pain and can affect a patient's quality of recovery. Lunn et al. demonstrated that a single dose of 125 mg IV methylprednisolone could prolong the duration of analgesic effects of patients undergoing total knee arthroplasty under spinal anesthesia.23 Our group has investigated the effect of a single dose of systemic dexamethasone in subjects undergoing general anesthesia for outpatient laparoscopy, and we demonstrated that dexamethasone provides an improvement in postoperative quality of recovery when compared with saline.24 All subjects in the current investigation received the same perioperative dose of systemic dexamethasone to control the effects of systemic steroids on postoperative pain and quality of recovery.
Although not evaluated in the current investigation, social culture in different regions around the world can affect a patient's input regarding the choice of the anesthetic technique. A patient's fear about the surgical procedure can influence his or her choice of general anesthesia instead of neuraxial anesthesia. We used a simple random sampling method that has probably adjusted the groups to avoid a disproportionate sample with respect to preference of anesthesia type. However, a proportionate sampling controlling for anesthesia preference would have provided greater assurance that an equal number of subjects with an anesthetic preference would have been allocated to each group. Nonetheless, because 30% of the subjects declined to participate, it is likely that subjects with a high preference for a specific technique did not elect to participate to avoid randomization. Future studies examining whether the better quality of postoperative recovery provided by neuraxial technique can change patients' preferences between the 2 anesthetic techniques are desirable.
Our study had several limitations. First, because of the nature of the question proposed by the study, we were unable to blind subjects and investigators to group allocation. We attempted to reduce measurement bias by cross-checking perioperative data collected with nurses' assessments and by having the patients directly complete the quality of recovery questionnaires. Second, we did not investigate the potential mechanistic effect of surgical stress response reduction in the postoperative quality of recovery. Lastly, we did not examine whether the better postoperative analgesia and quality of recovery in the neuraxial group was associated with better long-term outcomes such as a reduction of the incidence of chronic pain in that group.
In summary, we demonstrated that neuraxial anesthesia provides a better postoperative quality of recovery and analgesia than does general anesthesia in patients undergoing abdominal hysterectomy. There is also an inverse relationship between postoperative opioid consumption and quality of recovery. In the absence of contraindications, neuraxial anesthesia seems to be a better anesthetic plan for those patients. Clinicians should consider the improvement in postoperative quality of recovery provided by neuraxial anesthesia in comparison with general anesthesia when formulating an anesthetic plan for patients undergoing abdominal hysterectomy. Future studies to confirm our findings as well as to address the same question in different surgical procedures are needed.
Commentaires
Enregistrer un commentaire