Materials and Methods
After approval by the Institutional Review Board at White Plains Hospital Center, written informed consent was obtained preoperatively from women 18 years old or older undergoing gynecological laparoscopic surgery using a 4-port symmetrical technique. Patients with a known allergy to local anesthetic were not enrolled. Cases requiring a conversion to a laparotomy or the use of more than 4 laparoscopic ports were also excluded from the analysis. After informed consent, patients were randomized according to a computer-generated randomization list in sealed white envelopes to either right- or left-sided TAP block (experimental arm), and in each case the contralateral side of the abdomen was treated with a trocar site infiltration of anesthetic (control arm). Patients and postoperative assessors were blinded to the treatment assignment.
All patients received general anesthesia with endotracheal intubation. A standard multimodal intraoperative intravenous analgesic regimen was used including 1 g of acetaminophen and 30 mg of ketorolac. All the surgeries were performed by a single surgeon trained in advanced laparoscopy and a fellow in minimally invasive gynecology.
The surgical approach consisted of 4 laparoscopic ports inserted at or below the umbilicus: 1 periumbilical balloon trocar of 12 mm (T10 dermatome), 2 accessory ports of 5 mm inserted into the right lower quadrant and the left lower quadrant, and 1 accessory port of 5 or 12 mm in the suprapubic region (T12-L1 dermatomes). In every case, peritoneal access was obtained using an open port placement technique in the periumbilical area.
The control arm consisted of a total of 30 mL of 0.25% bupivacaine with epinephrine injected in divided doses in the trocar sites on 1 side of the abdomen: the lateral port received 40% of the dose, the midline umbilical port was injected with 40% on the control side only, and the suprapubic site was injected with 20% on the control side only.
The experimental arm consisted of the same total amount of 30 mL of 0.25% bupivacaine with epinephrine injected in the midaxillary line between the costal margin and the iliac crest.
In cohort 1, the TAP block was performed using a posterior approach under ultrasound guidance by 1 of 2 anesthesiologists, both with significant experience with this technique. A high-frequency linear ultrasound probe was positioned in a transverse plane on the anterolateral abdominal wall in the midaxillary line, between the lower costal margin and the iliac crest.
A 20-gauge 10 cm needle was inserted in the plane of the ultrasound beam and followed visually until it reached the plane between the internal oblique and the transversus abdominis muscles. Two milliliters of the local anesthetic solution were injected to visualize the spread of the solution and confirm correct needle position after which the remainder of the 30 mL was administered ( Figure 1 ).
In cohort 2, the surgeon performed the TAP block injection under laparoscopic guidance. A 20-gauge 10 cm needle attached to a syringe of 30 mL of 0.25% bupivacaine with epinephrine was introduced through the abdominal wall using the same anatomic landmarks. After skin puncture, the needle was advanced to the level of the parietal peritoneum under direct laparoscopic visualization. Loss of resistance was felt while the needle passed through the external oblique and the internal oblique muscles. As the needle approached the peritoneum, slow injection was performed to tent the peritoneum, and the needle was then slightly withdrawn to infiltrate the correct plane. A diffuse bulge could be visualized expanding anterior to the peritoneum and transversus abdominis ( Figure 2 ). The needle was withdrawn and redirected in the event of incidental peritoneal perforation.
All treatments were performed at the end of the surgical procedure and immediately prior to extubation. In all patients, the total amount of 0.25% bupivacaine with epinephrine used did not exceed the recommended safe dose of 2.5 mg/kg. Patients weighing less than 60 kg received a reduced amount of local anesthetic; otherwise, patients received a total volume of 60 mL. Careful aspiration was performed prior to all injections to minimize the risks of intravascular entry and subsequent local anesthetic toxicity.
At the completion of the surgery, patients were transferred to the postanesthetic care unit (PACU). Throughout the postoperative period, analgesic medications including acetaminophen, nonsteroidal antiinflammatory drugs, and systemic narcotics were administered upon patient request. Postoperative abdominal pain was assessed at 1, 2, 4, 6, 8, 12, 18, 24, and 48 hours (0 hours being at admission to the PACU) on the TAP and contralateral sides. During the hospital stay, the nursing staff completed all timed assessments.
Subjective and objective methods were used to evaluate postoperative pain at each time point. Patients were first asked whether they had more pain on 1 side of the abdomen or the other (yes or no). Next, patients were asked to subjectively quantify their pain level on each side using a 10 point visual analog scale (VAS), a simple and validated measure of pain ranging from 0, representing no pain, to 10, representing the worst pain imaginable. Then the nurse palpated the incisions bilaterally and recorded the VAS scores obtained on each side. Finally, patients were asked to rate their overall abdominal pain using the 10 point VAS at each time point.
Intraoperative and postoperative analgesic medications used were abstracted from the chart. For the purpose of analysis, all narcotic use was converted to morphine sulfate equivalents using equianalgesic tables. Patients discharged prior to 48 hours postoperatively were asked to record the amount of pain medications used and complete the remaining pain assessments. The 48 hour pain diary was then collected at the 2 week postoperative visit.
Demographic and surgical data were retrieved from the patient charts. This included age, race, body mass index, American Society of Anesthesiologists physical status classification, diagnosis, type of procedure, operative skin-to-skin time, estimated blood loss, and length of stay in the PACU and in the hospital. Antiemetic medications used postoperatively on request or prophylactically were abstracted from the nursing flow sheets. The sites of injection of the TAP block were inspected for hematoma formation or signs of infection.
The primary outcome of the study is the difference in patient-reported pain scores between the abdominal side treated with TAP block and the contralateral side treated with local trocar infiltration at different points in time. Secondary outcomes included total narcotic consumption, opioid-related adverse effects (nausea and vomiting), and TAP block related complications.
We based the sample size calculation on the assumption that a 2 point difference in VAS pain scores would be clinically meaningful. A sample size of 34 patients in each cohort would be able to detect a difference of 2.0 points in pain score with a power of 80% and a type 1 error (alpha) of 0.05 using a 2-tailed test, assuming the SD of the differences to be 4. With an exclusion rate of 20%, 42 patients would need to be enrolled in each cohort.
Normally distributed continuous variables were described as means with SD, whereas continuous variables that were skewed were described as medians with ranges (minimum-maximum). Means were compared between study groups using a Student t test, whereas medians were compared using a Wilcoxon rank-sum test. Differences in mean pain scores between TAP and non-TAP sides were estimated with 95% confidence intervals, and paired t tests were used to test whether the difference in pain scores at each follow-up was equal to zero. Categorical data were expressed as percentages and then compared between groups using a Fisher exact test.
To test for difference in pain scores between the TAP and non-TAP sides over time, a mixed model analysis of variance was used with random intercepts and an unstructured covariance to account for the correlation among measures made on the same patient over time. All hypothesis testing was conducted at the 5% level of significance. Statistical analyses were performed using SAS software version 9.2 (SAS Institute, Inc, Cary, NC).
Results
Between January and August 2013, 93 patients signed a preoperative written consent form and were enrolled in the study. Five patients were excluded: 1 because of an intraoperative conversion to a laparotomy secondary to malignancy and 4 requiring an additional laparoscopic port. Eighty-eight patients were eligible for randomization, 45 and 43 patients in cohorts 1 and 2, respectively. There were no patients lost to follow-up. The statistical analysis included a total of 88 patients ( Figure 3 ).
There was no significant difference between cohorts with respect to age, race, body mass index, American Society of Anesthesiologists physical status, preoperative diagnosis, operative time, estimated blood loss, type of procedure, time spent in the recovery room, or length of hospital stay ( Table 1 ). In all patients, the TAP block was easily performed on the intended abdominal side, and there were no described complications with either approach. Upon inspection, the sites of injection revealed no bruising, hematomas, swelling, or infection. There were no reported anaphylactic reactions or episodes of toxicity.
| Variable | Cohort 1, AAT (n = 45) | Cohort 2, SAT (n = 43) | P value |
|---|---|---|---|
| Age (y), mean (SD) | 53.6 (9.8) | 53.5 (11.6) | .985 a |
| BMI (kg/m 2 ), mean (SD) | 28.2 (7.8) | 28.4 (6.0) | .882 a |
| Race, n (%) | .512 b | ||
| White | 38 (84) | 37 (86) | |
| Other | 7 (16) | 6 (14) | |
| ASA physical status, n (%) | .882 b | ||
| 1 | 12 (27) | 14 (33) | |
| 2 | 30 (66) | 26 (60) | |
| 3 | 3 (7) | 3 (7) | |
| Preoperative diagnosis, n (%) | .084 b | ||
| Fibroid uterus | 7 (16) | 10 (23) | |
| Adnexal mass | 16 (35) | 14 (33) | |
| Malignancy | 15 (33) | 17 (19) | |
| Abnormal uterine bleeding | 0 (0) | 2 (5) | |
| Chronic pelvic pain | 3 (7) | 0 (0) | |
| Other | 4 (9) | 0 (0) | |
| Operative time, min, mean (SD) | 88.9 (40.7) | 80.8 (30.5) | .293 a |
| Estimated blood loss, mL, mean (SD) | 91.6 (53.1) | 106.7 (77.1) | .283 a |
| Procedures, n (%) | .378 b | ||
| Adnexal only | 4 (9) | 1 (2) | |
| LSH with or without BSO | 13 (29) | 13 (30) | |
| TLH with or without BSO | 18 (40) | 16 (38) | |
| TLH with or without staging | 10 (22) | 13 (30) | |
| Time in the recovery room (min), mean (SD) | 125 (50.7) | 116 (39.4) | .329 a |
| Length of stay (d), median (range) | 1 (0–4) | 1 (0–9) | .051 c |
a Student t test used to compare means
b Fisher exact test used to compare proportions
When patients were asked whether 1 side of the abdomen hurt more than the other side, the majority of patients in both cohorts answered no. In cohort 1, 53% of patients reported equal pain on both sides across all time points, whereas 29% and 18% reported the TAP side as more painful and the local side as more painful, respectively. In cohort 2, 61% of patients reported equal pain on both sides across all time points, whereas 20% and 19% of patients reported the TAP side as more painful and the local side as more painful, respectively.
Patient self-assessed pain scores were reported on the 10 point VAS and analyzed using paired t tests at each time point. When the TAP was performed under ultrasound-guidance, there was a statistically significant lower mean pain score on the TAP side compared with the local side at 2 hours postoperatively. However, the 95% confidence interval (CI) excludes the clinically meaningful difference of 2 (95% CI, 0.27–1.33). At all other documented time points, no statistical significance was found ( Table 2 ).
| Hour | Mean local anesthetic pain score | Mean TAP pain score | Mean difference (SD) | 95% CI | P value |
|---|---|---|---|---|---|
| 1 | 1.97 | 1.59 | 0.38 (2.16) | −0.32 to 1.08 | .273 |
| 2 | 2.00 a | 1.20 a | 0.80 (1.65) a | 0.27–1.33 a | .004 a |
| 4 | 1.56 | 1.13 | 0.44 (1.43) | −0.03 to 0.90 | .064 |
| 6 | 1.79 | 1.34 | 0.45 (1.48) | −0.04 to 0.93 | .071 |
| 8 | 2.26 | 1.66 | 0.61 (2.46) | −0.20 to 1.41 | .137 |
| 12 | 2.46 | 2.08 | 0.38 (1.88) | −0.25 to 1.00 | .228 |
| 18 | 3.23 | 2.79 | 0.44 (2.36) | −0.33 to 1.20 | .256 |
| 24 | 3.33 | 3.08 | 0.26 (1.93) | −0.37 to 0.88 | .412 |
| 48 | 2.55 | 2.63 | −0.08 (1.73) | −0.65 to 0.49 | .780 |
| Overall | 0.41 a | 0.20–0.61 a | .001 a,b |
When analyzed across time using the mixed regression model, there was a statistically significant difference in mean pain scores between TAP and contralateral side, with the local side having a 0.41 point higher average pain score than the TAP side. Similarly, the 95% CI excludes the clinically meaningful difference of 2 (95% CI, 0.20–0.61) ( Figure 4 ).
In the second cohort, when the TAP was performed under laparoscopic guidance, there was no statistically significant difference between TAP and contralateral side mean pain scores at any time point when using the paired t test ( Table 3 ) or overall across time points when using the mixed regression model ( Figure 5 ).
