British Journal of Anaesthesia (5): 72 35 (3) Advance Access publication 27 June 3. doi:0.093/bja/aet24 Duration of analgesic effectiveness after the posterior and lateral transversus abdominis plane block techniques for transverse lower abdominal incisions: a meta-analysis F. W. Abdallah,2 *, J. G. Laffey,2, S. H. Halpern,3 and R. Brull,4,5 Department of Anesthesia, University of Toronto, ON, Canada 2 Department of Anesthesia and Keenan Research Centre, Li Ka Shing Knowledge Institute, St Michael s Hospital, Toronto, ON, Canada 3 Department of Anesthesia, Sunnybrook Hospital, Toronto, ON, Canada 4 Department of Anesthesia, Women s College Hospital, Toronto, ON, Canada 5 Department of Anesthesia, Toronto Western Hospital, Toronto, ON, Canada * Corresponding author: Department of Anaesthesia, St Michael s Hospital, 30 Bond Street, Toronto, ON, Canada M5B W8. E-mail: abdallahf@smh.ca Editor s key points The duration of effectiveness of transversus abdominis plane (TAP) block in providing postoperative analgesia after abdominal surgery was studied in this meta-analysis. The authors conclude that TAP block using the posterior approach reduced the rest and dynamic pain as well as the of morphine for up to 48 h. The effect was not seen when a TAP block was performed using the lateral approach. The authors call for randomized controlled trials, which will compare the two approaches of performing a TAP block. Background. Both posterior and lateral transversus abdominis plane (TAP) block techniques provide effective early (0 2 h) postoperative analgesia after transverse incision surgery. However, whether either technique produces prolonged analgesia lasting beyond 2 h remains controversial. This meta-analysis examines the duration of analgesia associated with posterior and lateral TAP blocks in the first 48 h after lower abdominal transverse incision surgery. Methods. We retrieved randomized controlled trials (RCTs) investigating the analgesic effects of TAP block compared with control in patients undergoing lower abdominal transverse incision surgery. Outcomes sought included interval postoperative i.v. morphine and also rest and dynamic pain scores at 2,, 36, and 48 h postoperatively. Opioid-related side-effects and patient satisfaction at and 48 h were also assessed. The 2 h interval morphine was designated as a primary outcome. Results. Twelve RCTs including 64 patients were analysed. Four trials examined the posterior technique and eight assessed the lateral technique. Compared with control, the posterior TAP block reduced postoperative morphine during the 2 h and 48 h intervals by 9. mg (95% CI: 26.83, 2.45; P¼0.02) and 5 mg (95% CI: 29.54,.52; P¼0.03), respectively. It also reduced rest pain scores at, 36, and 48 h, and also dynamic pain scores at 2,, 36, and 48 h. Differences were not significant with the lateral TAP block. Conclusion. Based on the comparisons with control, the posterior TAP block appears to produce more prolonged analgesia than the lateral TAP block. Future RCTs comparing these two techniques are required to confirm our findings. Keywords: acute pain, novel techniques; anaesthetic blocks, regional; analgesia, postoperative; regional blockade; surgery, abdominal Accepted for publication: 7 April 3 Patients undergoing obstetric or gynaecological surgery using transverse lower abdominal incisions, such as the Pfannenstiel, Cherney, and Maylard 2 incisions, often suffer severe pain during the first 48 h postoperatively. 3 5 Not surprisingly, the incidence of persistent postoperative pain, an undesirable outcome of surgery influenced by the duration and efficacy of postoperative analgesia, 6 after 7 and total abdominal hysterectomy 8 approaches 2 and 32%, respectively. As the abdominal wall is a major contributor to acute postoperative pain after abdominal surgery, 9 field blocks like the transversus abdominis plane (TAP) block 0 can provide effective 2 analgesia for a variety of abdominal surgical procedures. However, the relative efficacy of the TAP block for transverse lower abdominal incisions may vary depending on the block technique. 2 While posterior injections in the triangle of Petit and lateral injections at the midaxillary techniques have both demonstrated efficacy in the immediate postoperative period, 2 the potential for either technique to produce more prolonged analgesic benefits ( 2 h) after lower transverse incision surgery remains controversial. A recent retrospective review 3 & The Author [3]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email: journals.permissions@oup.com
BJA Abdallah et al. suggests that prolonged post- analgesia lasting into the 48 h postsurgical period can be achieved by performing the lateral TAP block technique. In contrast, several clinical trials 4 6 and a recent qualitative systematic review 2 suggest that only the posterior TAP block technique provides prolonged analgesia. This meta-analysis examines the effect of each TAP block technique on analgesic outcomes in the first 48 h after laparotomy surgeries with a lower abdominal transverse incision. Methods The authors abided by the PRISMA guides 7 in the preparation of this review. We used a pre-determined protocol to review and evaluate the results of randomized controlled trials (RCTs) that measured the duration of analgesic effectiveness of the TAP block. Eligibility criteria We sought and retrieved full reports of RCTs that investigated the effects of TAP block (TAP group) compared with placebo or systemic analgesia (control group) on analgesic outcomes in patients undergoing abdominal surgery using a lower abdominal transverse incision. Literature search The US National Library of Medicine database, MEDLINE; the Excerpta Medica database, Embase; Cochrane Database of Systematic Reviews; and Cochrane Central Register of Controlled Trials databases were searched by two of the authors (F.W.A. and R.B.) independently. The search terms TAP/TAP block/ Transversus Abdominis/Transverse Abdominis/Transversus Abdominis Plane block/and Transverse Abdominis Plane block alone and coupled with the search keywords lateral and posterior were queried. The results of the search were combined by the Boolean operator AND with medical subject headings analgesia/pain relief/pain control/pain prevention/and pain management and with the medical subject headings abdomen/abdominal wall/abdominal muscles/abdominal surgery/and abdominal incision. We also hand searched the bibliographies of included articles for additional RCTs that met the inclusion criteria. The search was limited to RCTs on human subjects published between January 05 and December 2; no language restrictions were imposed. RCTs were excluded if analgesic outcomes were not assessed, if surgeries other than lower abdominal transverse incision were performed, if unilateral or continuous TAP blocks were performed, or if adjuvants that may prolong the duration of nerve block analgesia were used. Trials examining the subcostal TAP block technique were not included as it does not provide analgesia for lower abdominal transverse incisions. 8 The decision on inclusion of qualifying studies in the review was obtained by consensus between two of the authors (F.W.A. and R.B.). Data collection and presentation Two of the authors (F.W.A. and R.B.) independently assessed the quality of the reviewed RCTs using the Jadad score, 9 and a final score was designated by consensus. An RCT was considered to be of good quality if the methodological score was between 3 and 5. As an additional indicator of quality, only trials with a sample size.0 per group and that maintained a concealed assignment were considered. For the purpose of this review, we evaluated interval opioid analgesic (converted to i.v. morphine equivalent) and also pain severity at rest and with movement [visual analogue scale (VAS), a 00 mm scalewhere 0 represents no pain and 00 represents maximum pain] at 2,, 36, and 48 h postoperatively. We also assessed the incidence of opioid-related adverse effects (postoperative nausea and vomiting, pruritus, and excessive sedation) and patient satisfaction at and 48 h. The 2 h interval postoperative cumulative morphine was designated as a primary outcome; and other outcomes were classified as secondary. The authors independently used a standardized data collection form to extract data; any discrepancies were resolved by reexamination of the source data. Meta-analysis Two of the authors (F.W.A. and R.B.) entered and cross-checked the data into the statistical programme. Meta-analytic techniques (Revman 5., Cochrane Library, Oxford, UK) were used to combine the data where possible. The random effect modelling was utilized in analysing both dichotomous and continuous data. Data from trials with more than two intervention groups receiving TAP block were combined into a single group as recommended by the Cochrane Handbook. 2 We calculated the odds ratio (OR) and 95% confidence interval (CI) for the dichotomous outcomes, and the standardized mean difference and 95% CI for the continuous outcomes. Differences were considered statistically significant if the 95% CI of OR excluded, or if the 95% CI excluded 0 for the standardized mean difference. We verified the extent of heterogeneity using the I 2 statistic. 22 As prolonged analgesia has been attributed to the posterior technique, 4 6 we hypothesized a priori that combining the results of trials using the posterior and lateral techniques would generate significant heterogeneity among the pooled trials. We, therefore, performed subgroup analysis according to the TAP block technique. Additional confounding factors that were identified pre hoc as potential sources of heterogeneity included differences in the population studied (pregnant or non-pregnant), and the use of intrathecal morphine in some trials. When data relating to the primary outcome (i.e. 2 h interval postoperative cumulative morphine ) were heterogeneous, we explored whether alternative subgrouping based on these pre-determined factors influenced the level of heterogeneity and significance of the treatment effects. Results Our search retrieved 29 articles, 2 of which met the inclusion criteria. 4 5 23 32 Figure summarizes the retrieved, excluded, and reviewed RCTs. The median and range of the methodological quality score 9 of the 2 trials were 5 (3 5); and they included a total of 64 patients for analysis: 329 patients in 722
Posterior and lateral TAP block BJA TAP OR TAP block OR Transverse Abdominis OR Transversus Abdominis OR Transverse Abdominis Plane block OR Transversus Abdominis Plane block Abdomen OR Abdominal wall OR Abdominal muscles OR Abdominal surgery OR Abdominal incision AND Analgesia OR Pain relief OR Pain control OR Pain prevention OR Pain management records identified through other sources 83 records 07 records 95 records after removing duplicates 95 records screened 66 excluded records 29 articles assessed 7 articles did not fulfil inclusion criteria 2 trials included Fig Flow chart summarizing retrieved, included, and excluded RCTs. the TAPgroup and 32 in the control group. Table presents the trial characteristics and outcomes assessed for each trial. Four trials had TAP blocks performed using the posterior technique, 4 5 27 3 while another eight used the lateral technique. 23 26 28 30 32 None of the trials compared the posterior and the lateral techniques; and the separation of results into two subgroups based on the TAP block technique was found to marginally reduce the heterogeneity of results of the primary outcome from 99 to 97% and 74% for the two subgroups. Alternative subgrouping did not further reduce the heterogeneity. Table 2 describes the TAP block technique and the analgesic regimens used in the trials reviewed. One RCT 26 723
7 Table Outcomes of reviewed trials. GA, general anaesthesia; ITM, intrathecal morphine; PACU, post-anaesthesia care unit Author/year Jadad score Surgery N Groups (n) Anaesthesia Primary outcome Rest pain scores Dynamic pain scores Opioid Time to first analgesic Opioid-related adverse effects Patient satisfaction Early Late Early Late Early Late request McDonnell and 08 4 5 Carney and 4 Total abdominal hysterectomy 08 5 Belavy and 4 09 23 Costello and 5 09 Baaj and 3 0 McMorrow and 5 26 Amr and 3 Total abdominal hysterectomy 27 Shin and 3 Gynecologic surgery 28 Atim and 5 Total abdominal hysterectomy 29 Tan and 5 2 30 Eslamain and 5 2 3 Bollag and 5 2 32 52. TAP block () 2. Sham block () Spinal Cumulative 48 h opioid 53. TAP block () GA Cumulative 48 h 2. Sham block (26) opioid 50. TAP block (23) Spinal Cumulative h 2. Sham block () opioid 00. TAP block+itm (49) Spinal Dynamic pain 2. Sham block+itm (47) scores at h 40. TAP block (9) Spinal Cumulative h 2. Sham block () opioid 80. TAP block+itm () Spinal Dynamic pain 2. Sham block+itm () scores (time point 3. TAP block () not specified) 4. Sham block () 75. Pre-incisional TAP block (22) 2. Post-incisional TAP block (22) 3. Sham block (23) GA Cumulative 48 h opioid 32. TAP block (6) GA Rest pain scores in 2. No block (6) PACU 60. TAP block (8) GA Cumulative h 2. Sham block (8) opioid 3. Infiltration (9) 40. TAP block () GA Cumulative h 2. No block () morphine 50. TAP block () GA Postoperative 2. No block () pain (time point not specified) 90. TAP block () Spinal Hyperalgesia at 2. Sham block (30) 48 h 3. TAP block+clonidine adjuvant (26) BJA Abdallah et al.
7 Table 2 Analgesic regimen. PCA, patient-controlled analgesia; PACU, post-anaesthesia care unit Author/year Surgery Pre-incisional analgesia McDonnell and 08 4 Carney and 08 5 Belavy and 09 23 Costello and 09 Baaj and 0 McMorrow and 26 Amr and 27 Shin 28 Atim and 29 Tan and 2 30 Total abdominal hysterectomy Total abdominal hysterectomy Gynecologic surgery Total abdominal hysterectomy I.V. morphine, rectal diclofenac, rectal acetaminophen, TAP block in experimental group TAP block in experimental group TAP block in experimental group Surgical analgesia Spinal+intrathecal: mg fentanyl I.V. fentanyl, i.v. morphine Spinal+intrathecal: 5 mg fentanyl Spinal+intrathecal: 0 mg fentanyl, 00 mg morphine Spinal+intrathecal: mg fentanyl Spinal+intrathecal: 0 mg fentanyl, 00 mg morphine Supplemental postoperative analgesia dose rectal diclofenac, dose rectal acetaminophen, then i.v. PCA morphine, oral acetaminophen, and rectal diclofenac I.V. PCA morphine, rectal diclofenac, and rectal acetaminophen dose rectal diclofenac, dose rectal acetaminophen, then i.v. PCA morphine, oral acetaminophen, and oral ibuprofen dose i.v. ketorolac, dose rectal acetaminophen, then i.v. morphine, oral diclofenac, and oral acetaminophen TAP block Block timing Localization Site of injection Postoperative Anatomical Triangle of Petit Preoperative Anatomical Triangle of Petit Postoperative US Midaxillary Postoperative US Midaxillary I.V. PCA morphine Postoperative US Midaxillary dose rectal paracetamol, dose rectal diclofenac, then i.v. PCA morphine, oral paracetamol, and rectal diclofenac Postoperative Anatomical Midaxillary I.V. fentanyl I.V. morphine Pre-+postoperative Anatomical Triangle of Petit I.V. remifentanil I.V. fentanyl or ketorolac, then ketorolac+sufentanil via i.v. PCA, i.v. meperidine I.V. diclofenac I.V. fentanyl I.V. tramadol, then i.v. PCA tramadol, and i.m. meperidine Preoperative US Midaxillary Postoperative US Midaxillary I.V. morphine I.V. PCA morphine Postoperative US Midaxillary Block solution.5 mg kg 2 ropivacaine 0.75% to a total dose of 50 mg.5 mg kg 2 ropivacaine 0.75% up to a total dose of 50 mg ml ropivacaine 0.5% ml ropivacaine 0.375% ml bupivacaine 0.% mgkg 2 bupivacaine 0.375% ml levobupivacaine 0.375% ml ropivacaine 0.375% ml bupivacaine 0.% ml levobupivacaine 0.% Continued Posterior and lateral TAP block BJA
BJA Abdallah et al. TAP block Block solution Surgical analgesia Supplemental postoperative analgesia Block timing Localization Site of injection 5 ml bupivacaine 0.% Postoperative Anatomical Triangle of Petit I.V. tramadol and rectal diclofenac I.V. sufentanil, i.v. morphine ml bupivacaine 0.375% Postoperative US Midaxillary dose i.v. ketorolac, i.v. morphine in PACU, then oral acetaminophen, oral diclofenac, and oral tramadol Spinal+intrathecal: mg fentanyl, 00 mg morphine included four groups and performed two comparisons between the TAP block and control groups, in the presence and absence of intrathecal morphine. For the sake of this review, each of the two comparisons was treated as a separate trial. One trial included two TAP block groups 27 ( pre-incisional and post-incisional ) that were combined into one for the purpose of this review. Two of the authors whom we contacted provided additional unpublished results pertaining to postoperative morphine 27 32. Interval morphine Results describing postoperative i.v. morphine were available from nine trials for the 0 2 h interval, 4 5 23 27 30 nine trials for 2 h, 4 5 23 27 30 two trials for the 36 h, 4 26 4 26 and two trials for the 36 48 h postoperatively. Because so few trials reported morphine at 36 h postoperatively, the 36 and 36 48 h intervals sought were combined into one interval ( 48 h) for the purpose of this review. Thus, data reflecting postoperative i.v. morphine during the 48 h interval were available from a total of eight trials. 4 5 26 28 32 Compared with the control group, performing the posterior TAP block technique was effective in reducing the interval morphine by 23.2 mg or a relative decrease of 64.5% (95% CI: 227.70, 28.78; P,0.0000) at 0 2 h; by 9. mg or 5.9% (95% CI: 26.83, 2.45; P¼0.02) at 2 h; and by 5 mg or 63% (95% CI: 29.54,.52; P¼0.03) at 48 h postoperatively (Table 3, Fig. 2). Performing a TAP block using the lateral technique reduced morphine by 8.3 mg or 49.7% (95% CI: 26.64,.05; P¼0.05) at 0 2 h postoperatively (Table 3, Fig. 2). The lateral technique did not differ from control at any other time interval up to 48 h postoperatively. Morphine data for both the posterior and lateral TAP block subgroups during the 2 h interval (primary outcome) were characterized by high heterogeneity (I 2 ¼95 and 92%, respectively; P,0.0000). The heterogeneity remained when pregnancy and intrathecal morphine use were considered (Fig. 3). Table 2 Continued Author/year Surgery Pre-incisional analgesia Eslamaian and 2 3 Bollag and 2 32 Rest pain 4 5 23 27 28 30 32 Rest pain was assessed at 2 h in 0 trials, at h in 3 trials, 4 5 23 32 at 36 h in 3 trials, 4 5 32 and 48 h in 4 5 26 28 32 8 trials. Compared with control, performing the posterior TAP block technique reduced the rest pain VAS score by 7 mm at 2 h (95% CI: 22.2, 22.; P,0.0000), by 3 mm at h (95% CI: 22.7, 23.8; P¼0.005), by 8 mm (95% CI: 222.7, 22.9; P,0.0000) at 36 h, and by 5 mm (95% CI: 2.2, 28.9; P,0.0000) at 48 h postoperatively (Table 3, Fig. 4). When a TAP block was performed using the lateral technique, rest pain VAS scores were reduced by 5 mm at 2 h (95% CI: 27.5, 22.8; P,0.000). There was no difference in rest pain scores at, 36, and 48 h postoperatively between the lateral technique and control groups (Table 3, Fig. 4). 726
727 Table 3 Secondary endpoint results. N/A, not applicable; PONV, postoperative nausea and vomiting Outcome Subgroup Studies included TAP block mean or n/n Control mean or n/n OR or weighed mean [95% CI] P-value for statistical significance P-value for heterogeneity Interval morphine for 0 2 h (mg) Posterior 4 5 27.82 33.34 223. [227.70, 28.78],0.0000 0.002 84% Lateral 23 26 30 7.7 4.26 28.34 [26.64,.05] 0.05,0.0000 99% Interval morphine for 2 h (mg) Posterior 4 5 27 7.6 4.88 29.4 [26.83, 2.45] 0.02,0.0000 97% Lateral 23 26 30 6.06 8.04 2.83 [.44, 0.77] 0.7 0.002 74% Interval morphine for 48 h (mg) Posterior 4 5 27 3.36 9.06.03 [29.54,.52] 0.03,0.0000 95% Lateral 26 32 5.80 3.49 0.94 [.84, 2.7] 0.3 0.04 64% Rest pain at 2 h (cm) Posterior 4 5 27 3.33 2.88 2.67 [22.2, 2.2],0.0000 0.003 78% Lateral 23 28 30 32.45.84.5 [.75,.28],0.000 0.9 32% Rest pain at h (cm) Posterior 4 5 27 3 0.92.98 2.28 [22.7,.38] 0.005,0.0000 97% Lateral 23 26 28 30 32.53.76.3 [.66, 0.05] 0.09,0.000 78% Rest pain at 36 h (cm) Posterior 4 5 0.49 2.26 2.78 [22.27, 2.29],0.0000 0.06 72% Lateral 32.00.00 0.00 [.74, 0.74] N/A N/A Rest pain at 48 h (cm) Posterior 4 5 27.7 2.9 2.46 [22.02,.89],0.0000,0.000 9% Lateral 26 28 32.7.06 0.08 [.49, 0.65] 0.78 0.0002 82% Dynamic pain at 2 h (cm) Posterior 4 5 27 3 3.05 4.59 2.59 [2.88, 2.30],0.0000 0.36 6% Lateral 23 28 30 32 2.45 3.04.77 [2.60, 0.05] 0.07,0.0000 94% Dynamic pain at h (cm) Posterior 4 5 27 3 2.34 4.36 22.9 [23.28, 2.0],0.000,0.0000 90% Lateral 23 26 28 30 32 2.53 3.08.75 [2.52, 0.02] 0.06,0.0000 94% Dynamic pain at 36 h (cm) Posterior 4 5 2.39 3.6 2.9 [2.87,.52] 0.0006 0.39 0% Lateral 32 2.00 2.00 0.00 [.80, 0.80].0 N/A N/A Dynamic pain at 48 h (cm) Posterior 4 5 27.92 3.83 22.22 [23.32, 2.2],0.000 0.0002 88% Lateral 26 28 32 2.6 2.29.07 [2.06, 0.92] 0.89,0.0000 89% Incidence of PONV at h (n/n) Posterior 4 5 27 27/95 36/73 0.30 [0.05,.73] 0.8 0.02 75% Lateral 23 26 30 32 2/27 30/34 0.65 [0.33,.28] 0.2 0.63 0% Incidence of PONV at 48 h (n/n) Posterior 4 5 0/49 5/5 0.07 [0.00,.40] 0.08 N/A N/A Lateral 26 32 3/65 3/70 0.95 [0.8, 4.93] 0.95 0.4 0% Incidence of pruritus at h (n/n) Posterior N/A N/A N/A N/A N/A N/A N/A Lateral 23 26 30 32 6/08 47/4 2.28 [.22, 4.] 0.0 0.42 0% Incidence of pruritus at 48 h (n/n) Posterior N/A N/A N/A N/A N/A N/A N/A Lateral 26 32 5/65 4/70.5 [0.35, 3.76] 0.82 0.22 33% Incidence of sedation at h (n/n) Posterior 4 5 27 /95 38/73 0.23 [0.0, 0.57] 0.00 0.29 9% Lateral 23 30 32 34/68 30/74.80 [0.67, 4.84] 0. 0.5 0% Incidence of sedation at 48 h (n/n) Posterior 4 5 27 0/95 9/73 0.03 [0.00, 0.63] 0.02 N/A N/A Lateral N/A N/A N/A N/A N/A N/A N/A Patient satisfaction at h (mm) Posterior N/A N/A N/A N/A N/A N/A N/A Lateral 23 26 28 82. 77.97 6.96 [.74, 9.67] 0.28 P,0.000 83% Patient satisfaction at 48 h (mm) Posterior N/A N/A N/A N/A N/A N/A N/A Lateral 26 28 87.83 82.08 3.84 [26.2, 3.89] 0.45,0.0000 96% I 2 test for heterogeneity Posterior and lateral TAP block BJA
BJA Abdallah et al. A Study or subgroup McDonnell, 08 Carney, 08 Amr, TAP block Control Mean difference Mean difference Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl 3 7.2 9.4.5 6.8 4.39 46 95 9 3.8 23.3 3.5 7.8 3.6 26 22 73 Heterogeneity: t 2 =44.37; c 2 =77.69, df=2 (P<0.0000); I 2 =97% Test for overall effect: Z=2.33 (P=0.02).7%.%.6% 34.4% 6.00 [ 7.49, 4.5] 3.40 [ 0.65, 7.45] 3.90 [ 5.73, 2.07] 9.4 [ 6.83,.45] Belavy, 09 Costello, 09 Baaj, 0 McMorrow- McMorrow-2 Tan, 2 8 0.3 2.68 5 4.62 4.7 5.6 0.7 6.7 3.94.06 3.9 23 49 9 5 5 0.3 6.05 6.5 9.23 0.6 8.9 0.7 5.48 4.9 8.46.82 Heterogeneity: t 2 =6.42; c 2 =9.44, df=5 (P=0.002); I 2 =74% Test for overall effect: Z=.38 (P=0.7) 47 5.%.7%.2%.4% 0.5% 9.7% 65.6% 7.00 [.23, 2.77] 0.00 [ 0.28, 0.28] 3.37 [ 7.22, 0.48].5 [ 3.9,.6] 5.39 [ 0.7,.49] 6.43 [ 4.43,.57].83 [ 4.44, 0.77] 6 2 00.0% Heterogeneity: t 2 =79.; c 2 =645.63, df=8 (P<0.0000); I 2 =99% Test for overall effect: Z=.46 (P=0.4) Test for subgroup differences: c 2 =3., df= (P=0.08); I 2 =67.8% 4.44 [ 0.42,.53] 0 0 0 B TAP block Control Mean difference Mean difference Study or subgroup Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl McDonnell, 08 0. 6 4.2 8.% 6.00 [ 7.70, 4.30] Carney, 08 6. 9.3 5.6 8.7 26 5.7% 9.50 [ 4.50, 4.50] Amr, 3.75 0.8 46 4.8 0.9 22.5%.05 [.49, 0.6] 95 73 48.3% 5.03 [ 9.54, 0.52] Heterogeneity: t 2 =3.90; c 2 =40.54, df=2 (P<0.0000); I 2 =95% Test for overall effect: Z=2.9 (P=0.03) Costello, 09.3 0.2 49.3 0.3 McMorrow- 3.5 6.5 7.7 3.9 McMorrow-2 4.3.3 7.3 7.2 Bollag, 2.67 3.2.56 2.87 4 Heterogeneity: t 2 =.62; c 2 =8.33, df=3 (P=0.04); I 2 =64% Test for overall effect: Z=.03 (P=0.30) 47 30 7.% 3.5% 4.4% 8.5% 5.7% 0.00 [ 0.0, 0.0] 5.80 [ 0.92, 2.52] 7.00 [.3, 2.87] 0. [.5,.73] 0.94 [ 0.84, 2.7] 9 90 00.0% Heterogeneity: t 2 =.93; c 2 =89.55, df=6 (P<0.0000); I 2 =93% Test for overall effect: Z =.94 (P = 0.05) Test for subgroup differences: c 2 =5.82, df= (P=0.02); I 2 =82.8%.35 [ 2.72, 0.0] 0 5 0 5 0 Fig 2 Forest plots of morphine for: (A) 2 h interval; (B) 48 h interval. The sample size, mean, SDs, and the pooled estimates of the mean difference are shown. The 95% CIs are shown as s for individual studies and as diamonds for pooled estimates. Dynamic pain Dynamic pain was assessed at 2 h in 0 trials, 4 5 23 27 32 at h in 3 trials, 4 5 23 32 at 36 h in 4 trials, 4 5 32 and at 48 h in 4 5 26 28 32 8 trials. Compared with control, the posterior TAP block technique reduced dynamic pain VAS scores by 6 mm at 2 h (95% CI: 28.8, 23.0; P,0.0000), by 22 mm at h (95% CI: 232.8, 2.0; P,0.000), by 2 mm at 36 h (95% CI: 28.7,.2; P¼0.0006), and by 22 mm (95% CI: 233.2, 2.2; P,0.000) at 48 h postoperatively (Table 3, Figs 5 and 6). Performing the lateral TAP block technique did not produce a difference in dynamic pain compared with control at 2,, 36, and 48 h postoperatively (Table 3, Figs 5 and 6). Opioid-related adverse effects Because of the diversity in the definitions of opioid-related adverse effects in the reviewed trials, the results of these outcomes are reported as standardized units. The incidence of postoperative nausea and vomiting was similar between the TAP block and the control group for both the posterior and 728
Posterior and lateral TAP block BJA A TAP block Control Mean difference Mean difference Study or subgroup Pregnant Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl Baaj, 0 2.68 6.7 9 26.05 5.48.2% 3.37 [ 7.22, 9.52] Belavy, 09 8 5.6 23 5 8.9.% 7.00 [.23, 2.77] Costello, 09 0.3 0.7 49 0.3 0.7 47.7% 0.00 [ 0.28, 0.28] McDonnell, 08 3.5 9 3.5.6% 6.00 [ 7.49, 4.5] McMorrow- 5 3.94 6.5 4.9.4%.5 [ 3.9,.6] McMorrow-2 4.62.06 9.23 8.46 0.5% 5.39 [ 0.7,.49] Tan, 2 4.7 3.9 0.6.82 9.8% 6.43 [ 4.43,.57] 76 76 77.3% 5.59 [ 2.69,.50] Heterogeneity: t 2 =86.63; c 2 =483.06, df=6 (P<0.0000); I 2 =99% Test for overall effect: Z=.55 (P=0.2) Non-pregnant Amr, Carney, 08 9.4 7.2 4.39 6.8 46 70 23.3 9.8 3.6 7.8 22 26 48 Heterogeneity: t 2 =6.28; c 2 =.84, df= (P<0.0000); I 2 =96% Test for overall effect: Z=.49 (P=0.4).6%.% 22.7% 3.90 [ 5.73, 2.07] 2.60 [ 6.65,.45] 8.40 [ 9.47, 2.67] 6 2 00.0% Heterogeneity: t 2 =83.45; c 2 =680.60, df=8 (P<0.0000); I 2 =99% Test for overall effect: Z=.99 (P=0.05) Test for subgroup differences: c 2 =0.8, df= (P=0.68); I 2 =0% 6.23 [ 2.35, 0.0] 0 0 0 B TAP block Control Mean difference Mean difference Study or subgroup Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl Without intrathecal morphine McDonnell, 08 3.5 9 3.5.7% 6.00 [ 7.49, 4.5] Carney, 08 Belavy, 09 Baaj, 0 7.2 8 2.68 6.8 5.6 6.7 23 9 3.8 5 6.05 7.8 8.9 5.48 26.%.%.2% 3.40 [ 0.65, 7.45] 7.00 [.23, 2.77] 3.37 [ 7.22, 0.48] Amr, 9.4 4.39 46 23.3 3.6 22.6% 3.90 [ 5.73, 2.07] McMorrow-2 Tan, 2 4.62 4.7.06 3.9 9.23 0.6 8.46.82 0.5% 9.7% 5.39 [ 0.7,.49] 6.43 [ 4.43,.57] 77 57 76.8% 5.70 [.43, 0.04] Heterogeneity: t 2 =54.59; c 2 =45.00, df=6 (P<0.0000); I 2 =96% Test for overall effect: Z=.95 (P=0.05) With intrathecal morphine Costello, 09 McMorrow- 0.3 5 0.7 3.94 49 0.3 6.5 0.7 4.9 69 Heterogeneity: t 2 =0.00; c 2 =0.66, df= (P=0.42); I 2 =0% Test for overall effect: Z=0.08 (P=0.93) 47 67.7%.4% 23.2% 0.00 [ 0.28, 0.28].5 [ 3.9,.6] 0.0 [ 0.29, 0.27] 6 2 00.0% Heterogeneity: t 2 =79.; c 2 =645.63, df=8 (P<0.0000); I 2 =99% Test for overall effect: Z=.46 (P=0.4) Test for subgroup differences: c 2 =3.76, df= (P=0.05); I 2 =73.4% 4.44 [ 0.42,.53] 0 0 0 Fig3 Forest plots depicting alternative subgrouping of the primaryoutcome (2 h intervalmorphine ) according to: (A) pregnancy; (B) use of intrathecal morphine. The sample size, mean, SDs, and the pooled estimates of the mean difference are shown. The 95% CIs are shown as s for individual studies and as diamonds for pooled estimates. the lateral techniques at and 48 h (Table 3). The incidence of pruritus in the lateral TAP block technique was increased to an OR of 2.28 (95% CI:.22, 4.; P¼0.0) at h compared with control, but there was no difference at 48 h. There were no data on the incidence of pruritus for the posterior technique. The incidence of sedation wasreduced to an ORof 0.23 (95% CI: 0.0, 0.57; P¼0.00) at h, and to an OR of 0.03 (95% CI: 0.00, 0.63; P¼0.02) at 48 h compared with the control group when the posterior TAP block technique was performed; but there was no difference from control at h with the lateral technique. There were no data at 48 h postoperatively. There were no data related to patient satisfaction for the posterior TAP block technique. For the lateral technique, there were no statistically significant differences in patient 729
BJA Abdallah et al. A Study or subgroup McDonnell, 08 Carney, 08 Amr, Eslamian, 2 TAP block Control Mean difference Mean difference Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl 0.86 0 0.75 0.56 0.5 0. 46.46 3 3.5 0.37 0.5 0.38 0.75 26 22 9 97 Heterogeneity: t 2 =0.80; c 2 =96.9, df=3 (P<0.0000); I 2 =97% Test for overall effect: Z=2.79 (P=0.005) 7.7% 8.8% 8.9% 8.7% 34.%.46 [ 2.07, 0.85] 2.00 [ 2.30,.70].65 [.86,.44] 0.00 [ 0.32, 0.32].28 [ 2.7, 0.38] Belavy, 09 2.3 0.53 23 2.65 0.5 Costello, 09 2.79 49 2 0. 47 Baaj, 0 2.5 9 3.63 0.7 McMorrow-.33.43 0.83 0.62 McMorrow-2 2.57.52.93 0.97 Shin, 0.9.5 6 2.2 2.9 6 Atim, 0.29 0.9 8.03 0.26 8 Tan, 2 0.5 0.5 Bollag, 2.75.75 30 20 25 Heterogeneity: t 2 =0.9; c 2 =36.88, df=8 (P<0.0000); I 2 =78% Test for overall effect: Z=.68 (P=0.09) 8.8% 8.% 8.0% 7.4% 7.0% 4.% 9.0% 7.0% 6.4% 65.9% 0.35 [ 0.64, 0.06] 0.00 [ 0.50, 0.50].3 [.67, 0.59] 0.50 [ 0.8,.8] 0.64 [ 0.5,.43].30 [ 2.90, 0.30] 0.74 [ 0.89, 0.59] 0.50 [.29, 0.29] 0.00 [ 0.93, 0.93] 0.3 [ 0.66, 0.05] 329 32 00.0% Heterogeneity: t 2 =0.54; c 2 =0.82, df=2 (P<0.0000); I 2 =94% Test for overall effect: Z=2.84 (P=0.004) Test for subgroup differences: c 2 =3.87, df= (P<0.05); I 2 =74.2% 0.63 [.06, 0.] 4 2 0 2 4 Favours tap BLOCK Favours control B TAP block Control Mean difference Mean difference Study or subgroup Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl McDonnell, 08 0 0.5.5 34.2%.50 [.94,.06] Carney, 08 0.5 3 0.5 26 35.8% 2.00 [ 2.28,.72] 49 5 70.0%.78 [ 2.27,.29] Heterogeneity: t 2 =0.09; c 2 =3.57, df= (P=0.06); I 2 =72% Test for overall effect: Z=7.7 (P<0.0000) Bollag, 2 Heterogeneity: not applicable Test for overall effect: Z=0.00 (P=.00).75 30 30 30.0% 30.0% 0.00 [ 0.74, 0.74] 0.00 [ 0.74, 0.74] 74 8 00.0% Heterogeneity: t 2 =0.6; c 2 =.62, df=2 (P<0.0000); I 2 =92% Test for overall effect: Z=2.59 (P=0.00) Test for subgroup differences: c 2 =5.55, df= (P<0.000); I 2 =93.6%.23 [ 2.6, 0.30] 4 2 0 2 4 Favours TAP Favours control C Study or subgroup McDonnell, 08 Carney, 08 Amr, 95 Heterogeneity: t 2 =0.22; c 2 =2.70, df=2 (P<0.000); I 2 =9% Test for overall effect: Z=5.08 (P<0.0000) Costello, 09 McMorrow- McMorrow-2 Shin, Bollag, 2 TAP block Control Mean difference Mean difference Mean SD Total Mean SD Total Weight IV, random, 95% Cl IV, random, 95% Cl 0 0 2.42.7.9 0.4 0.48 0.5 0.57 0.2.6.3 0.7.5 46 49 6 2 3.77 0.8..6 0.74 0.5 0.5 0.3 0.9 0.8.2. 30 Heterogeneity: t 2 =0.32; c 2 =22.23, df=4 (P=0.0002); I 2 =82% Test for overall effect: Z=0.28 (P=0.78) 26 22 73 47 6 30 33 3.0% 3.2% 3.2% 39.4% 3.4%.5% 2.0% 2.0%.8% 60.6%.00 [.35, 0.65] 2.00 [ 2.28,.72].35 [.62,.08].46 [ 2.02, 0.89] 0.00 [ 0.0, 0.0] 0.90 [0.0,.70] 0.80 [0.3,.47]. [.88, 0.52] 0.00 [ 0.74, 0.74] 0.08 [ 0.49, 0.65] 2 6 00.0% 0.52 [.23, 0.] Heterogeneity: t 2 =.00; c 2 =274.32, df=7 (P<0.0000); I 2 =97% Test for overall effect: Z=.40 (P=0.6) Test for subgroup differences: c 2 =4.7, df= (P=0.0002); I 2 =92.9% 4 2 0 2 4 Favours TAP Favours control Fig 4 Forest plots of rest pain VAS scores at: (A)h;(B)36h;(C) 48 h. The sample size, mean, SDs, and the pooled estimates of the mean difference are shown. The 95% CIs are shown as s for individual studies and as diamonds for pooled estimates. 730
Posterior and lateral TAP block BJA A TAP block Control Mean difference Study or subgroup Mean SD Total Mean SD Total Weight IV, random, 95% Cl McDonnell, 08 3.5 4.38 9.9%.00 [.80, 0.] Carney, 08 2.95 2.05 4.63.79 26 8.4%.68 [ 2.75, 0.6] Amr, 3.6 0.73 46 4.77 0.56 22 2.3%.6 [.93,.29] Eslamian, 2 3.5 5. 0.0% 2.00 [ 2.78,.22] 9 97 40.7%.59 [.88,.30] Heterogeneity: t 2 =0.0; c 2 =3., df=3 (P=0.36); I 2 =6% Test for overall effect: Z=0.63 (P<0.0000) Mean difference IV, random, 95% Cl Belavy, 09 Costello, 09 Shin, Atim, Tan, 2 Bollag, 2 2.3 3 2.8.42 0 4 0.53.07 2.5 0.32 2.5 23 49 6 8 5 2.65 3 4.3 3.23 4 0.5.75 2.3 0.32. 2.5 47 6 8 30 55 Heterogeneity: t 2 =0.87; c 2 =84.47, df=5 (P=0.0000); I 2 =94% Test for overall effect: Z=.84 (P=0.07) 2.4%.% 5.6% 2.7% 0.5% 7.% 59.3% 0.35 [ 0.64, 0.06] 0.00 [ 0.58, 0.58].50 [ 3.6, 0.6].8 [ 2.02,.60].00 [.70, 0.30] 0.00 [.33,.33] 0.77 [.60, 0.05] 270 2 00.0%.0 [.62, 0.58] Heterogeneity: t 2 =0.55; c 2 =94.0, df=9 (P<0.0000); I 2 =90% Test for overall effect: Z=4.2 (P<0.000) 4 2 0 2 4 Test for subgroup differences: c 2 =3.35, df= (P=0.07); I 2 =70.% B Study or subgroup McDonnell, 08 Carney, 08 Amr, Eslamian, 2 TAP block Control Mean difference Mean SD Total Mean SD Total Weight IV, random, 95% Cl 2.93 3. 2.05 0.6.8 0.00. 46 4 4.47 5.00 4.73.89 0.4.5 26 22 9 97 Heterogeneity: t 2 =.07; c 2 =28.59, df=3 (P<0.0000); I 2 =90% Test for overall effect: Z=3.93 (P<0.000) 7.6% 7.4% 0.7% 7.9% 3.6%.07 [.99, 0.5].37 [ 2.39, 0.35] 2.0 [ 2.49,.07] 4.00 [ 4.78, 3.22] 2.9 [ 3.28,.0] Mean difference IV, random, 95% Cl Belavy, 09 Costello, 09 Baaj, 0 McMorrow- McMorrow-2 Shin, Atim, Tan, 2 Bollag, 2 2.3 0.53 3 3.57 3.44 2.47.9 4.9.9.5.6 0.3 0.9 2.5 2 2.5 23 49 9 6 8 20 2.65 3 7.44 2.27 3.63 3..8 3 2 0.5 2.8.23.37 2 0.32 2.75 47 6 8 30 25 Heterogeneity: t 2 =.6; c 2 =33.7, df=8 (P<0.0000); I 2 =94% Test for overall effect: Z=.92 (P=0.06) 8.7% 0.35 [ 0.64, 0.06] 6.7% 0.00 [.28,.28] 8.2% 4.00 [ 4.63, 3.37] 7.4% 0. [ 0.79,.9] 7.4%.27 [0., 2.30] 6.8%.60 [ 2.85, 0.35] 8.8% 0.88 [.05, 0.7] 7.2%.00 [ 2.0, 0.0] 7.0% 0.00 [.6,.6] 68.4% 0.75 [.52, 0.02] 329 32 00.0%. [.87, 0.53] Heterogeneity: t 2 =.33; c 2 =7.45, df=2 (P<0.0000); I 2 =95% Test for overall effect: Z=3.49 (P=0.0005) 4 2 0 2 4 Test for subgroup differences: c 2 =4.44, df= (P=0.04); I 2 =77.5% Fig 5 Forest plots of dynamic pain VAS scores at: (A)2h;(B) h. The sample size, mean, SDs, and the pooled estimates of the mean difference are shown. The 95% CIs are shown as s for individual studies and as diamonds for pooled estimates. 73
BJA Abdallah et al. A TAP block Control Mean difference Study or subgroup Mean SD Total Mean SD Total Weight IV, random, 95% Cl McDonnell, 08 2.6 2.9.8 32.3% 0.90 [.84, 0.04] Carney, 08 2.8.7 4.3.8 26 3.6%.50 [ 2.47, 0.53] 49 5 63.9%.9 [.87, 0.53] Heterogeneity: t 2 =0.00; c 2 =0.75, df= (P=0.39); I 2 =0% Test for overall effect: Z=3.45 (P=0.0006) Mean difference IV, random, 95% Cl Bollag, 2 2.5 2.5 30 30 Heterogeneity: not applicable Test for overall effect: Z=0.00 (P=.00) 36.% 36.% 0.00 [ 0.80, 0.80] 0.00 [ 0.80, 0.80] 74 8 00.0% 0.76 [.65, 0.2] Heterogeneity: t 2 =0.40; c 2 =5.75, df=2 (P=0.06); I 2 =65% Test for overall effect: Z=.70 (P=0.09) 4 2 0 2 4 Test for subgroup differences: c 2 =5.00. df= (P=0.03). I 2 =80.0% B TAP block Control Mean difference Study or subgroup Mean SD Total Mean SD Total Weight IV, random, 95% Cl McDonnell, 08 0 3.4.5 3.2% Carney, 08 2.3.8 3.5 2. 26.3% Amr, 2.76 0.82 46 4.7 0.5 22 4.3% 95 73 Heterogeneity: t 2 =0.8; c 2 =6.6, df=2 (P=0.0002); I 2 =88% Test for overall effect: Z=3.95 (P<0.000) 39.% 3.40 [ 4., 2.69]. [ 2.28, 0.2].94 [ 2.26,.62] 2.22 [ 3.32,.2] Mean difference IV, random, 95% Cl Costello, 09 2 0.3 49 3 0.3 47 4.3%.00 [.2, 0.88] McMorrow- McMorrow-2 2.9 2.8.5.5 2.3 2.3.2 2.3% 2.5% 0.60 [ 0.9,.39] 0.50 [ 0.34,.34] Shin,.2.8 6 2.6 2. 6 0.0%.40 [ 2.76, 0.04] Bollag, 2 2 2.5 2 30 0.7%.00 [ 0.2, 2.2] 30 33 60.9% 0.07 [.06, 0.92] Heterogeneity: t 2 =.05; c 2 =36.95, df=4 (P<0.0000); I 2 =89% Test for overall effect: Z=0.3 (P=0.89) 2 6 00.0% 0.9 [.65, 0.7] Heterogeneity: t 2 =0.95; c 2 =3.43, df=7 (P<0.0000); I 2 =94% Test for overall effect: Z=2.42 (P=0.02) 4 2 0 2 4 Test for subgroup differences: c 2 =8.2, df= (P=0.004); I 2 =87.7% Fig 6 Forest plots of dynamic pain VAS scores at: (A)36h;(B) 48 h. The sample size, mean, SDs, and the pooled estimates of the mean difference are shown. The 95% CIs are shown as s for individual studies and as diamonds for pooled estimates. satisfaction among the trials that reported this outcome at h 23 26 28 and 48 h 26 28 when compared with control (Table 3). None of the trials reported any block-related complications. Other outcomes It is noteworthy that the effect of TAP block on the incidence of 27 32 postsurgical chronic pain was evaluated by three trials. Two trials evaluated the rate of chronic pain after TAP block performed with the lateral technique, which was not different from the control group at 6 weeks, and 3, 6, and 2 months. 32 However, the third trial, 27 which compared pre-incisional with post-incisional TAP block using the posterior technique or no block (control), reported that none of the patients who received a pre-incisional posterior TAP block required analgesics to treat chronic pain at 3 and 6 months, compared with 3.6 and 7.4% for the post-incisional TAP block and control groups, respectively. Discussion In the absence of a direct comparison between the two TAP block techniques, and based on the comparison with control, our results suggest that performing the posterior TAP block may be a better technique for prolonged analgesia after 732
Posterior and lateral TAP block BJA lower abdominal transverse incision surgeries. The posterior technique can reduce opioid, rest and dynamic pain scores, and the incidence of sedation up to 48 h postoperatively. Several possible explanations may account for these findings. First, a more posterior injection point may allow the TAP block to capture lateral cutaneous branches of thoracolumbar nerves before entering into the TAP where they undergo extensive branching and anastomoses. 33 36 Secondly, the posterior but not the lateral technique results in a retrograde local anaesthetic spread that reaches the paravertebral space 37 and extends between the T4-L levels within 4 h of injection, 38 potentially producing some degree of block along the thoracolumbar sympathetic chain. Evidence suggestive of a role of the sympathetic nervous system in acute postoperative pain continues to emerge, 39 and sympathetic block may account for the prolonged analgesic effect associated with the posterior technique. Finally, while the formation of a local anaesthetic depot in the neuro-fascial TAP plane 40 might explain the prolonged effects of the TAP block, this possibility has been challenged, 40 4 and in any case would also occur with the lateral technique. The recent surge of interest in TAP blocks is likely attributable to the advent of ultrasound (US) guidance, as the benefits of US are believed to be related to enhanced accuracy of local anaesthetic deposition. 42 Interestingly, when the present data set was analysed post hoc to compare US-guidance to anatomical landmark techniques, we found that anatomically guided TAP blocks, irrespective of location (i.e. posterior or lateral), provided prolonged postoperative analgesia whereas US-guided TAP blocks did not. The latter reflects the fact that all trials that examined the TAP block performed using the posterior technique relied on anatomical localization; and all but one trial 26 where the TAP block was performed using the lateral technique used US guidance. It has been hypothesized that the double pop endpoint 4 used in the anatomical technique is more effective in depositing local anaesthetics deep to the fascia between the internal oblique and the TAP where the thoracolumbar nerves lie 34 compared with the distension of the fascial planes endpoint 43 used in the US technique. 44 However, studies on the local anaesthetic spread cast doubt on this possible explanation, and instead attribute the difference in efficacy to material anatomical differences at the sites of injection resulting in a different spread pattern of injected local anaesthetics rather than the localization 37 45 46 technique. Our meta-analysis has several limitations. First, none of the trials performed a direct comparison between the posterior and lateral techniques; hence, the observed differences between the two techniques are based on their indirect comparison. A direct comparison of these approaches is needed to provide a definitive answer to this issue. Secondly, the reviewed trials were small and characterized by high heterogeneity. Considerable differences existed in the doses and types of local anaesthetics used in performing the TAP block, in the control groups, and in the patient population. Alternative subgrouping according to, for example, the use of intrathecal morphine or pregnant population subgroup did not reduce the heterogeneity of results. This heterogeneity may limit the clinical combinability of the results. It is also noteworthy that none of the reviewed trials performed a sensory assessment of their patients to confirm block onset or offset; thus, failure of the TAP block cannot be ruled out as an explanation for some of the observed lack of analgesic efficacy. Finally, except for two trials, 27 28 invasive placebo injections were used in the control arms of the reviewed trials. These invasive approaches towards blinding are becoming less acceptable. 47 In conclusion, while much remains to be learned regarding the TAP block, it does appear that more posterior block approaches may provide more prolonged analgesia to patients having surgery using a lower abdominal transverse incision. Further research is needed, directly comparing the lateral with posterior approaches of the TAP block, to clarify whether or not the posterior approach provides superior prolonged analgesia. Authors contributions F.W.A. helped design the study, conduct the study, analyse the data, and write the manuscript. F.W.A. is also the senior author and is responsible for archiving the study files. J.G.L. helped conduct the study, analyse the data, and write the manuscript. S.H.H. helped conduct the study, analyse the data, and write the manuscript. R.B. helped design the study, conduct the study, analyse the data, and write the manuscript. Declaration of interest None declared. Funding This work was supported by departmental funding. J.G.L. and R.B. were also supported by the Merit Award Program, Department of Anesthesia, University of Toronto. References Nygaard IE. Abdominal incisions from creation to closure. Obstet Gynecol Surv 996; 5: 429 36 2 Abu-Rustum NR. Radical abdominal hysterectomy. Surg Clin North Am 0; 8: 85 28 3 Angle P, Halpern S, Leighton B, Szalai JP, Gnanendran K, Kronberg J. A randomized controlled trial examining the effect of naproxen on analgesia during the second day after cesarean. Anesth Analg 02; 95: 74 5 4 Gilron I, Orr E, Tu D, O Neill JP, Zamora J, Bell A. A placebo-controlled randomized clinical trial of perioperative administration of gabapentin, rofecoxib and their combination for spontaneous and movement-evoked pain after abdominal hysterectomy. Pain 05; 3: 9 0 5 Fassoulaki A, Stamatakis E, Petropoulos G, Siafaka I, Hassiakos D, Sarantopoulos C. Gabapentin attenuates late but not acute pain afterabdominalhysterectomy. Eur J Anaesthesiol06; 23: 36 4 6 Lavand homme P. From preemptive to preventive analgesia: time to reconsider the role of perioperative peripheral nerve blocks? Reg Anesth Pain Med ; 36:4 6 733
BJA Abdallah et al. 7 Nikolajsen LL. Chronic pain following caesarean section chronic pain after caesarean section. Acta Anaesthesiol Scand 04; 48: 6 8 Brandsborg B, Nikolajsen L, Kehlet H, Jensen TS. Chronic pain after hysterectomy. Acta Anaesthesiol Scand 08; 52: 327 3 9 Joris J, Cigarini I, Legrand M, et al. Metabolic and respiratory changes after cholecystectomy performed via laparotomy or laparoscopy. Br J Anaesth 992; 69: 34 5 0 Rafi AN. Abdominal field block: a new approach via the lumbar triangle. Anaesthesia 0; 56: 0 6 Johns NN, O Neill S, Ventham NT, Barron F, Brady RR, Daniel T. Clinical effectiveness of transversus abdominis plane (TAP) block in abdominalsurgery: a systematic review and meta-analysis. Colorectal Dis 2; 4: 635 42 2 Abdallah FW, Chan VW, Brull R. Transversus abdominis plane block: the effects of surgery, dosing, technique, and timing on analgesic outcomes. A systematic review. Reg Anesth Pain Med 2; 37: 93 9 3 Patel SS. Transversus abdominis plane block for postoperative analgesia after cesarean. Matern Fetal Neonatal Med 2; : 2270 3 4 McDonnell JG, Curley G, Carney J, et al. The analgesic efficacy of transversus abdominis plane blockaftercesarean : a randomized controlled trial. Anesth Analg 08; 06: 86 9 5 Carney J, McDonnell JG, Ochana A, Bhinder R, Laffey JG. The transversus abdominis plane block provides effective postoperative analgesia in patients undergoing total abdominal hysterectomy. Anesth Analg 08; 07: 56 60 6 Carney J, Finnerty O, Rauf J, Curley G, McDonnell JG, Laffey JG. Ipsilateral transversus abdominis plane block provides effective analgesia after appendectomy in children: a randomized controlled trial. Anesth Analg 0; : 998 003 7 Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg 0; 8: 336 4 8 Lee TH, Barrington MJ, Tran TM, Wong D, Hebbard PD. Comparison of extent of sensory block following posterior and subcostal approaches to ultrasound-guided transversus abdominis plane block. Anaesth Intensive Care 0; 38: 452 60 9 Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 996; 7: 2 Canadian Pharmacists Association. Compendium of Pharmaceuticals and Specialties: The Canadian Drug Reference for Health Professionals, 45th Edn. Ottawa: Canadian Pharmacists Association, 0 2 Higgins JP, Deeks JJ, Altman DG. Chapter 6: special topics in statistics. In: Higgins JP, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. Baltimore, MD: The Cochrane Collaboration, 22 Higgins JP. Quantifying heterogeneity in a meta-analysis. Stat Med 02; 2: 539 58 23 Belavy D, Cowlishaw PJ, Howes M, Phillips F. Ultrasound-guided transversus abdominis plane block for analgesia after. Br J Anaesth 09; 03: 726 30 Costello JF, Moore AR, Wieczorek PM, Macarthur AJ, Balki M, Carvalho JC. The transversus abdominis plane block, when used as part of a multimodal regimen inclusive of intrathecal morphine, does not improve analgesia after cesarean. Reg Anesth Pain Med 09; 34: 586 9 Baaj JM, Alsatli RA, Majaj HA, Babay ZA, Thallaj AK. Efficacy of ultrasound-guided transversus abdominis plane (TAP) block for postcesarean section analgesia a double-blind, placebocontrolled, randomized study. Middle East J Anaesthesiol 0; : 82 6 26 McMorrow RC, Ni Mhuircheartaigh RJ, Ahmed KA, et al. Comparison of transversus abdominis plane block vs spinal morphine for pain relief after caesarean section. Br J Anaesth ; 06: 706 2 27 Amr YM, Amin SM. Comparative study between effect of pre- versus post-incisional transversus abdominis plane block on acute and chronic post-abdominal hysterectomy pain. Anesth Essays Res ; 5: 77 82 28 Shin HJ. Preemptive analgesic efficacy of ultrasound-guided transversus abdominis plane block in patients undergoing gynecologic surgery via a transverse lower abdominal skin incision. Korean J Anesthesiol ; 6: 43 8 29 Atim A, Bilgin F, Kilickaya O, et al. The efficacy of ultrasound-guided transversus abdominis plane block in patients undergoing hysterectomy. Anaesth Intensive Care ; 39: 630 4 30 Tan TT. A randomised trial of the analgesic efficacy of ultrasoundguided transversus abdominis plane block after caesarean under general anaesthesia. Eur J Anaesthesiol 2; 29: 88 94 3 Eslamian LL. Transversus abdominis plane block reduces postoperative pain intensity and analgesic in elective cesareanundergeneral anesthesia. J Anesth 2; 26: 334 8 32 Bollag L, Richebe P, Siaulys M, Ortner CM, Gofeld M, Landau R. Effect of transversus abdominis plane block with and without clonidine on post-cesarean wound hyperalgesia and pain. Reg Anesth Pain Med 2; 37: 508 4 33 Jankovic ZB, du Feu FM, McConnell P. An anatomical study of the transversus abdominis plane block: location of the lumbar triangle of Petit and adjacent nerves. Anesth Analg 09; 09: 98 5 34 Rozen WM, Tran TM, Ashton MW, Barrington MJ, Ivanusic JJ, Taylor GI. Refining the course of the thoracolumbar nerves: a new understanding of the innervation of the anterior abdominal wall. Clin Anat 08; 2: 3 33 35 Hebbard PD. Transversalis fascia plane block, a novel ultrasoundguided abdominal wall nerve block. Can J Anaesth 09; 56: 68 36 Hebbard P, Barrington M, Vasey C. Ultrasound-guided continuous oblique subcostal transversus abdominis plane blockade: description of anatomy and clinical technique. Reg Anesth Pain Med 0; 35: 436 4 37 Carney J, Finnerty O, Rauf J, Bergin D, Laffey J, Mc Donnell J. Studies on the spread of local anaesthetic solution in transversus abdominis plane blocks. Anaesthesia ; 66: 023 30 38 Finnerty O, McDonnell JG. Transversus abdominis plane block. Curr Opin Anaesthesiol 2; : 60 4 39 McDonnell JG, Finnerty O, Laffey JG. Stellate ganglion blockade for analgesia following upper limb surgery. Anaesthesia ; 66: 6 4 40 Latzke D, Marhofer P, Kettner SC, et al. Pharmacokinetics of the local anesthetic ropivacaine after transversus abdominis plane block in healthy volunteers. Eur J Clin Pharmacol 2; 68: 49 4 Griffiths JD, Barron FA, Grant S, Bjorksten AR, Hebbard P, Royse CF. Plasma ropivacaine concentrations after ultrasound-guided transversus abdominis plane block. Br J Anaesth 0; 05: 853 6 42 Abrahams MS, Horn JL, Noles LM, Aziz MF. Evidence-based medicine: ultrasound guidance for truncal blocks. Reg Anesth Pain Med 0; 35: S36 42 43 Hebbard P, Fujiwara Y, Shibata Y, Royse C. Ultrasound-guided transversus abdominis plane (TAP) block. Anaesth Intensive Care 07; 35: 66 7 734