Share this article on:

Paravertebral Block for Inguinal Herniorrhaphy: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Law, Lawrence Siu-Chun MD*; Tan, Mingjuan MD*; Bai, Yaowu MD*; Miller, Timothy E. MB ChB, FRCA*; Li, Yi-Ju PhD; Gan, Tong-Joo MD, MHS, FRCA

doi: 10.1213/ANE.0000000000000835
Regional Anesthesia: Research Report

BACKGROUND: Paravertebral block (PVB) is a safe and effective anesthetic technique for thoracotomy and mastectomy. However, no systematic review or meta-analysis has focused on PVB for inguinal herniorrhaphy. Our study compares PVB with general anesthesia/systemic analgesia, neuraxial blocks, and other peripheral nerve blocks.

METHODS: We analyzed 14 randomized controlled trials from PubMed, MEDLINE, CENTRAL, EMBASE, and CINAHL up to February 2015, without language restriction, comparing PVB under sedation with general anesthesia/systematic analgesia (135 vs 133 patients), neuraxial blocks (191 vs 186 patients), and other peripheral nerve blocks (119 vs 117 patients). We investigated pain scores, consumption of postoperative analgesia, incidence of postoperative nausea and vomiting (PONV), length of hospital stay, postanesthesia care unit bypassing rate, time to perform blocks, intraoperative hemodynamics, and incidence of urinary retention. Joint hypothesis testing was adopted for pain and analgesics, PONV, and hemodynamic variables. All analyses were performed with RevMan 5.2.11 (Cochrane Collaboration, Copenhagen). Hartung-Knapp-Sidik-Jonkman method was used for post hoc testing.

RESULTS: PVB reduced PONV (nausea: risk ratio [RR] = 0.22; 95% confidence interval [CI], 0.05–0.93; numbers needed to treat [NNT] = 4.5; I2 = 15% and vomiting: RR = 0.15; 95% CI, 0.03–0.76; NNT = 8.3; I2 = 0%) compared with general anesthesia/systematic analgesia (quality of evidence [QoE]: high). Compared with neuraxial blocks, PVB resulted in less postoperative nausea (RR = 0.34 [95% CI, 0.13–0.91], NNT = 8.3, I2 = 0%) and urinary retention (RR = 0.14 [95% CI, 0.05–0.42], NNT = 7.4, I2 = 0%) than neuraxial blocks (QoE: high). More time was needed to perform PVB than neuraxial blocks (standardized mean difference = 1.90 [95% CI, 0.02–3.77], I2 = 92%; mean difference = 5.33 minutes; QoE: moderate). However, the available data could not reject the null hypothesis of noninferiority on all pain scores and analgesic requirements for both PVB versus general anesthesia/systematic analgesia and PVB versus neuraxial blocks (QoE: low), as well as on hemodynamic outcomes for PVB versus neuraxial blocks (QoE: moderate). Our systematic review showed that PVB decreased postoperative pain scores and analgesic requirement as compared with ilioinguinal block and transversus abdominis plane block.

CONCLUSIONS: This meta-analysis shows that PVB provides an anesthesia with fewer undesirable effects for inguinal herniorrhaphy. The choice between general anesthesia/systematic analgesia, neuraxial blocks, PVB, and other peripheral nerve blocks should be based on time available to perform the block and a complete coverage over the relevant structures by the blocks.

Supplemental Digital Content is available in the text.Published ahead of print June 17, 2015

From the *Department of Anesthesiology, Duke University, Durham, North Carolina; Department of Biostatistics and Bioinformatics, Duke University, Durham, North Carolina; and Department of Anesthesiology, Stony Brook University, Stony Brook, New York.

Lawrence Siu-Chun Law, MD, is currently affiliated with the Centre for Quantitative Medicine, Duke-NUS Graduate Medical School, Singapore, Singapore.

Mingjuan Tan, MD, is currently affiliated with the Duke-NUS Graduate Medical School, Singapore, Singapore.

Yaowu Bai, MD, currently affiliated with the Department of Anesthesiology, Tangshan Maternal and Child Healthcare Hospital, Hebei United University, Tangshan, China.

Accepted for publication March 27, 2015.

Published ahead of print June 17, 2015

Funding: None.

The authors declare no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Drs. Siu-Chun Law and Mingjuan Tan contributed equally to this study and are co-first authors.

Reprints will not be available from the authors.

Address correspondence to Tong-Joo Gan, MD, MHS, FRCA, Department of Anesthesiology, Stony Brook University, HSC Level 4, Rm 060, Stony Brook, NY 11794. Address e-mail to tong.gan@stonybrookmedicine.edu.

General anesthesia with systemic analgesia is considered undesirable because opioids are invariably needed for adequate pain control. Opioids may cause respiratory depression, pruritus, urinary retention, nausea/vomiting, drowsiness, confusion, and constipation.1 Multimodal analgesia reduces pain and minimizes usage of opioids.2,3 Such techniques include non-opioid systemic analgesics,4,5 local anesthetics,6–8 and regional anesthesia.9–16 Although epidural and spinal blocks provide adequate analgesia,17,18 neuraxial techniques are associated with hemodynamic instability and urinary retention.18 IV fluids, vasopressors, and urinary catheterization may be required although their effects on the length of hospital stay (LOS) remain inconclusive.19–22

Paravertebral block (PVB) involves local anesthetic injection around the spinal nerve roots exiting the vertebral column. A number of meta-analyses have shown PVB to be a safe, effective anesthetic technique for thoracotomy and mastectomy.18,23–26 PVB provides superior acute pain control in these surgeries over general anesthesia/systematic analgesia23,26 or equivalent pain control to neuraxial blocks.17,18 The other advantages of PVB include fewer incidents of postoperative nausea and vomiting (PONV), shorter LOS, and greater patient satisfaction.23,26,27

Inguinal herniorrhaphy is a common type of ambulatory surgery. Successful anesthesia and analgesia for inguinal herniorrhaphy include general anesthesia/systematic analgesia, neuraxial blocks or PVB with sedation, and general anesthesia with peripheral nerve blocks. However, a survey showed that most anesthesiologists favor general anesthesia over regional anesthesia.28,29 Many of the postoperative side effects are associated with anesthesia (i.e., postoperative ileus, urinary retention, postoperative hypotension, and PONV) rather than the herniorrhaphy itself (i.e., wound hemorrhage or infection), especially using general anesthesia/systematic analgesia and neuraxial blocks.28–30 This meta-analysis compares PVB with general anesthesia/systematic analgesia, neuraxial blocks, and other peripheral nerve blocks. From the results of previous meta-analyses, we generated a joint hypothesis31 that PVB for inguinal herniorrhaphy was noninferior to general anesthesia/systematic analgesia and neuraxial blocks on all pain outcomes (postoperative pain at rest [0–6, 6–24, and 24–72 hours] and analgesic requirements) and superiority on ≥1 outcome. Our hypotheses for secondary outcomes included that patients having PVB had a shorter LOS, greater postanesthesia care unit (PACU) bypass rate, lower incidence of PONV (joint hypothesis testing of nausea and vomiting: superiority on all outcomes), less urinary retention, and less intraoperative hemodynamic instability (joint hypothesis testing of mean arterial blood pressure [MAP], heart rate [HR], and ephedrine requirement: noninferiority on all outcomes and superiority on ≥1). We also descriptively reported the failure rate and complications of PVB.

Back to Top | Article Outline

METHODS

Search Strategy

Our meta-analysis complies with the Transparent Reporting of Systematic Reviews and Meta-Analyses (PRISMA) statement (Supplemental Digital Content, http://links.lww.com/AA/B150).32 We searched PubMed, MEDLINE, CENTRAL, EMBASE, and CINAHL for randomized controlled trials (RCTs) on PVB among patients who underwent herniorrhaphy. We searched “paravertebral block AND inguinal AND (herniorrhaphy OR hernia)” in “All Fields” without restrictions on language or year of publication (Appendix 1). The last database search was in February 2015.

Back to Top | Article Outline

Selection of Included Studies

Two authors simultaneously (not independently) retrieved studies from the initial search and determined eligibility by reading the titles and abstracts. Any discrepancy was discussed and resolved immediately during the process of the literature search. References of retained studies were manually searched. All types of PVB were included (unilateral and bilateral; single, multiple, and continuous). All studies other than RCT and conference abstracts >3 years old were excluded. We included all parallel RCT comparing PVB with any other modality.

Back to Top | Article Outline

Data Extraction

We identified outcome variables that had sufficient data (≥2 studies for the same comparison) for meaningful analyses from all studies searched. We attempted to contact the authors of studies with insufficient data. We measured the graphs if authors failed to provide the required numerical data.

For comparing PVB with general anesthesia/systematic analgesia, we had the following outcome variables on ≥2 studies: (1) pain scores at rest in 0 to 72 hours after surgery and postoperative analgesic requirements, (2) PONV, (3) LOS, and (4) PACU bypassing rate (fast-tracking rate).

For comparing PVB with neuraxial blocks, we had ≥2 studies on (1) pain scores at rest in 0 to 24 hours, (2) postoperative nausea, (3) LOS, (4) time required to perform block, (5) surgery duration, (6) intraoperative hemodynamics, and (7) urinary retention. We did not have ≥2 studies to compare PVB with other regional anesthesias.

We also found some RCT comparing PVB with other peripheral nerve blocks. Because of heterogeneous types of peripheral nerve blocks such as patient groups (adult versus pediatric) and outcome variables reported, meta-analysis was not conducted. We qualitatively reported those RCTs.

Back to Top | Article Outline

Primary Outcome

Postoperative pain scores at rest (visual analog scale or numeric rating scale) were converted into a scale of 0 to 10. We categorized the pain scores into intervals of 0 to 6 hours, 6 to 24 hours, and 24 to 72 hours postoperatively. The worst pain scores were used if there was >1 score provided for the same interval. We compared postoperative oral tramadol consumption between PVB and neuraxial blocks. Between PVB and general anesthesia/systematic analgesia, the data only allowed us to compare the number of patients requiring postoperative analgesics.

Back to Top | Article Outline

Secondary Outcomes

For PONV, 1 study16 that combined both into 1 variable, we regarded those as vomiting because all patients who vomit feel nausea, but not the reverse. For intraoperative hemodynamic indices, we analyzed MAP, HR, and ephedrine requirements (incidence required a 6-mg bolus). We considered incidence of urinary retention and incidence requiring postoperative urinary catheterization as the same variable.

Back to Top | Article Outline

Statistical Analyses

SEM and 95% confidence intervals (95% CIs) were converted into SD by: SD = SEM√N and SEM = (95% CI) / 1.96. If categorical data were reported for a continuous variable, we estimated the mean and SD by assuming a normal distribution. If SD was still unavailable, we substituted the missing SD with the pooled SD of the other studies within the same comparison by: √(∑(N SD2) / ∑N).

We conducted the meta-analyses with Review Manager 5.2.11 (Cochrane Collaboration, Copenhagen). We compared the standardized mean differences (SMD)33,34 by the inverse-variance method for continuous variables and computed risk ratios (RRs) by the Mantel-Haenszel method for dichotomous variables. Because of the limited number of studies, random effect was used for all comparisons regardless of the heterogeneity between studies. A post hoc procedure to compute the t statistic for Hartung-Knapp-Sidik-Jonkman (HKSJ) method34–37 was performed as described by IntHout et al.34 (using the provided program34; degrees of freedom = number of studies − 1). Mean differences (MDs) were reported for increasing interpretability33 if HKSJ revealed a significant result.

A 2-stage joint hypothesis testing (noninferiority on all outcomes and superiority on ≥1) was adopted for the primary outcome (pain scores and opioid consumption) and the hemodynamic outcome.31 In the first stage, we tested the noninferiority hypothesis on all outcomes with a noninferiority delta of RR <1.5 or SMD <0.5 (less than a moderate effect). Correction for multiple testing was not required at this stage because the alternative hypothesis was an intersection (“AND”) of multiple tests. In the second stage, we tested for superiority of ≥1 outcome with Bonferroni correction for multiple testing because the alternative hypothesis was a union (“OR”) of multiple tests. RR >1.5 and SMD >0.5 were considered as clinically meaningful.

For PONV, a single-stage joint hypothesis testing (superiority on all outcomes) was used. The null hypotheses of both nausea and vomiting should be rejected to claim a significant effect on PONV (postoperative nausea “AND” vomiting). Correction for multiple testing was not required because the alternative hypothesis was an intersection (“AND”) of multiple tests.

For all significant dichotomous variables, we also report numbers needed to treat (NNT). For publication biases, we reported fail-safe number (FSN) using the Rosenberg method38 only if a significant outcome with I2 <25%. We also conducted the intercept regressions by Egger et al.39 for comparisons with >3 included studies and I2 >25%. Sensitivity and subgroup analyses were conducted to explore the heterogeneity among studies. Statistical significance was set at P value <0.05 (2 sided) or 0.025 (1 sided) when applicable. Finally, all results of the meta-analysis were graded the quality of evidence (QoE) using the GRADE Working Group system (Canada).

Back to Top | Article Outline

RESULTS

Study Selection

Fourteen of 51 nonduplicated search items met the inclusion criteria and were retained for analysis (Fig. 1; Appendix 2). Characteristics of included studies and risk of bias are shown in Table 1 and Figure 2, A–C, respectively.

Table 1

Table 1

Figure 1

Figure 1

Figure 2

Figure 2

Four studies10,13,15,16 compared PVB under sedation (135 patients) with general anesthesia/systematic analgesia (133 patients). No patient in the PVB group received general anesthesia. We subsequently referred to PVB versus general anesthesia/systematic analgesia, for simplification.

Six studies9,11,12,14,15,40 compared PVB under sedation (191 patients) with neuraxial blocks under sedation (186 patients). Five9,11,12,14,15 of the studies compared PVB with spinal anesthesia among adults. One study40 compared PVB with caudal block among pediatric patients. Three studies9,12,14 gave continuous propofol infusion titrated to light sleep with easy arousability. One study11 only used propofol sedation in the PVB group but not in the control group. The pediatric study40 used sevoflurane with N2O at a minimal alveolar concentration of 0.6 to 1.541 (assuming 1-year-old child). Because of the heterogeneity of populations (adult versus child) and techniques (spinal anesthesia versus general anesthesia with caudal block), we only performed meta-analysis on the studies9,11,12,14,15 comparing PVB with spinal anesthesia among adults and descriptively reported the pediatric study40 comparing PVB with caudal block.

Five studies42–46 compared PVB (119 patients) with other peripheral nerve blocks (117 patients). One study44 involved pediatric patients. The types of peripheral nerve blocks included ilioinguinal nerve block,42,44,46 transversus abdo minis plane (TAP) block,45 and field block.43 One study46 used a cryoanalgesic technique. All patients received supplementation with general anesthesia, spinal anesthesia, or sedation.

Back to Top | Article Outline

Joint Hypothesis Testing of Postoperative Pain at Rest and Analgesic Requirements

Comparing PVB with general anesthesia/systematic analgesia, pain scores at rest during 0 to 6 and 24 to 72 hours were analyzed in 2 studies (PVB 55 versus general anesthesia/systematic analgesia 55 patients)10,16 and during 6 to 24 hours in 3 studies (PVB 80 versus general anesthesia/systematic analgesia 80 patients).10,13,16 We conducted a 2-stage joint hypothesis testing for the following outcomes: postoperative pain at 0 to 6 hours, 6 to 24 hours, and 24 to 72 hours as well as requirements for postoperative analgesics. No significant difference was found for pain scores at all time points (0–6 hours: SMD = −1.25 [HKSJ 95% CI, −2.98 to 0.48], I2 = 0%; 6–24 hours: SMD = −0.63 [HKSJ 95% CI, −2.09 to 0.83], I2 = 75%; and 24–72 hours: SMD = −0.72 [HKSJ 95% CI, −6.75 to 5.31], I2 = 82%). Three studies10,13,16 showed no significant difference in the number of patients who required postoperative analgesics (RR = 0.49 [HKSJ 95% CI, 0.08–2.84], I2 = 95%) between PVB (35 of 77 patients, 45.5%) and general anesthesia/systematic analgesia (63 of 75 patients, 84.0%). The insignificant results and the wide 95% CI could not reject the null hypothesis of noninferiority on all outcomes (QoE: low; i.e., publication bias and imprecision) (Fig. 3, A and B).

Figure 3

Figure 3

For PVB versus spinal anesthesia, we conducted a 2-stage joint hypothesis testing for the following outcomes: postoperative pain at 0 to 6 hours, 6 to 24 hours, and 24 to 72 hours as well as requirements for postoperative analgesics. Pain scores at rest for 0 to 24 hours were analyzed in 3 studies (PVB 79 versus spinal anesthesia 80 patients)9,11,12 and showed no significant difference (0–6 hours: SMD = −0.75 [HKSJ 95% CI, −2.99 to 1.49], I2 = 88%; 6–24 hours: SMD = −0.76 [HKSJ 95% CI, −3.66 to 2.14], I2 = 91%). Only 1 study (PVB 29 versus spinal anesthesia 30 patients)9 reported pain scores for 24 to 72 hours, with no significant difference (SMD = −0.27 [95% CI, −0.78 to 0.24]). Four studies9,11,12,14 (PVB 101 versus spinal anesthesia 104 patients) reported that patients receiving PVB required less postoperative oral tramadol than those receiving spinal anesthesia (SMD = −1.34 [HKSJ 95% CI, −2.58 to −0.10], I2 = 83%; MD = −76.27 mg). However, the insignificant results and the wide 95% CI (HKSJ) of pain scores at all time points implied that the null hypothesis of noninferiority on all outcomes could not be rejected (QoE: low; i.e., publication bias and imprecision) (Fig. 4, A and B).

Figure 4

Figure 4

For PVB versus caudal block, behavioral face, legs, activity, cry and consolability scores were not significantly different between PVB (35 patients) and caudal block groups (35 patients).40 Patients who underwent PVB consumed less analgesics than patients who underwent caudal block (35 vs 35 patients).40

For PVB versus other peripheral nerve blocks, 1 study44 reported that PVB (40 patients) significantly provided better pain control than iliohypogastric nerve block (INB) (40 patients) at all time points in 0 to 48 hours. Another study45 showed that patients who underwent PVB (30 patients) had lower pain scores at 0.5, 3, 6, and 12 hours but not at 1 and 24 hours than patients who underwent TAP block (30 patients). Furthermore, patients who underwent PVB consumed less analgesics than patients who underwent INB (64 vs 62 patients)42,44 and TAP block (30 vs 30 patients).45 However, 1 study (10 vs 10 patients)46 showed that patients who underwent cryoanalgesia required less analgesics than patients who underwent PVB.

Back to Top | Article Outline

Postoperative Nausea and Vomiting

Four studies10,13,15,16 compared PVB (nausea: 5 of 110 patients, 4.5%; vomiting: 0 of 135 patients, 0.0%) with general anesthesia/systematic analgesia (nausea: 29 of 108 patients, 26.9%; vomiting: 16 of 133 patients, 12.0%). We conducted a single-stage joint hypothesis testing for the incidences of nausea and vomiting. The RR for nausea was 0.22 (HKSJ 95% CI, 0.05–0.93; I2 = 15%; NNT = 4.5; 95% CI, 3.17–7.64; FSN = 5.4). The RR for vomiting was 0.15 (HKSJ 95% CI, 0.03–0.76; I2 = 0%; NNT = 8.3; 95% CI, 5.70–15.38; FSN = 2.5). Both null hypotheses of nausea and vomiting were rejected, so the joint null hypothesis was rejected (QoE: high; i.e., publication bias and large effect) (Fig. 4, A and B).

For PVB versus spinal anesthesia, 5 studies9,11,12,14,15 showed that patients undergoing PVB had a significantly lower rate of nausea (6 of 156 patients, 3.8%) than those undergoing spinal anesthesia (24 of 151 patients, 15.9%). The RR was 0.34 (HKSJ 95% CI, 0.13–0.91; I2 = 0%; NNT = 8.3; 5.37–18.24; FSN = 2.7). Null hypothesis was rejected. However, joint hypothesis testing cannot be conducted because no study reported the incidence of vomiting (QoE: high; i.e., publication bias and large effect) (Fig. 5C).

Figure 5

Figure 5

Back to Top | Article Outline

Length of Hospital Stay

Four studies10,13,15,16 comparing PVB (135 patients) with general anesthesia/systematic analgesia (133 patients; SMD = −0.94; HKSJ 95% CI, −2.23 to 0.34) and 3 studies9,11,15 comparing PVB (79 patients) with spinal anesthesia (80 patients; SMD = −3.93; HKSJ 95% CI, −15.26 to 7.40). Although no significant difference was found, PVB was noninferior to general anesthesia/systematic analgesia (QoE: moderate; i.e., publication bias). In terms of the criteria for hospital discharge, 2 studies10,11 used a postanesthetic discharge scoring system. One study16 was based on clinical decisions of surgeons. Other studies9,13,15 did not mention any discharge criterion.

Back to Top | Article Outline

PACU Bypassing Rate

Two studies10,13 comparing PVB with general anesthesia/systematic analgesia reported the PACU bypass rate (fast-tracking rate). More patients bypassed the PACU in the PVB group (43 of 54 patients, 79.6%) than in the general anesthesia/systematic analgesia group (7 of 54 patients, 13.0%). The RR was 5.92 (HKSJ 95% CI, 0.38–93.13; I2 = 0%; NNT = 1.5; 95% CI, 1.24–1.90) (QoE: high; i.e., publication bias and large effect).

One study10 reported the time for phase 1 PACU stay. Patients in the PVB group (30 patients, mean 26.0, SD 2.8) had a significantly shorter time for phase 1 PACU stay than patients in the general anesthesia/systematic analgesia group (30 patients, mean 41.2, SD 7.6) (Cohen d = 2.7; Hedges g = 2.6).

Back to Top | Article Outline

Time to Perform Blocks and Duration of Surgery

The 4 studies9,11,12,14 comparing PVB (101 patients) with spinal anesthesia (104 patients) showed that more time was needed to perform PVB than spinal anesthesia (SMD = 1.90 [HKSJ 95% CI, 0.02–3.77], I2 = 92%; MD = 5.33 minutes; QoE: moderate; i.e., publication bias). Three studies10,13,16 comparing PVB (77 patients) with general anesthesia/systematic analgesia (75 patients) and 2 studies44,45 comparing PVB (70 patients) with other peripheral nerve blocks (70 patients) also found no difference in duration of surgery. One study13 found that general anesthesia/systematic analgesia (24 patients) had a 5-minute shorter induction time than PVB (24 patients).

Back to Top | Article Outline

Joint Hypothesis of Intraoperative Hemodynamics

We conducted a 2-stage joint hypothesis testing for the following outcomes: MAP, HR, and ephedrine requirements. Three studies9,12,14 (PVB 81 versus spinal anesthesia 84 patients) reported intraoperative MAP and HR. Patients receiving PVB had greater MAP (SMD = 0.49 [HKSJ 95% CI, 0.12–0.86], I2 = 0%, FSN = 3.6; MD = 3.03 mm Hg) and a noninferior HR (SMD = 0.54 [HKSJ 95% CI, −0.49 to 1.57], I2 = 54%). Two studies12,14 compared intraoperative ephedrine (PVB group: 0 of 52 patients, 0%; spinal anesthesia group: 10 of 54 patients, 18.5%). The RR was 0.11 (HKSJ 95% CI, 0.00–416.77; I2 = 0%, NNT = 5.4; 95% CI, 3.46–12.26). The null hypothesis of noninferiority on all outcomes was not rejected (QoE: moderate; i.e., publication bias) (Fig. 6, A–C).

Figure 6

Figure 6

One study44 comparing PVB (40 patients) with INB (40 patients) demonstrated that patients who underwent PVB had a smaller percentage increase of intraoperative systolic blood pressure and HR from baseline than patients who underwent INB. Another study45 comparing PVB (30 patients) with TAP block (30 patients) reported no significant difference in preoperative, intraoperative, and postoperative blood pressure, HR, and SaO2.

Back to Top | Article Outline

Urinary Retention

Three studies11,12,14 reported the incidence of urinary retention and/or the need for postsurgical catheterization. PVB (0 of 72 patients, 0.0%) was related to a lower incidence of urinary retention than spinal anesthesia (10 of 74 patients, 13.5%). The RR was 0.14 (HKSJ 95% CI, 0.05–0.42; I2 = 0%; NNT = 7.4; 95% CI, 4.69–17.47; FSN = 0.9) (QoE: high; i.e., publication bias and large effect) (Fig. 7).

Figure 7

Figure 7

Back to Top | Article Outline

Sensitivity and Subgroup Analyses

PVB was still significantly associated with a lower incidence of postoperative vomiting than general anesthesia/systematic analgesia after 1 deleted study15 had nearly 50% weight of the total effect.

For postoperative pain, we analyzed PVB subgroups using a landmark approach9,10,12,13 and PVB using a nerve stimulator approach11,16 separately. PVB using a nerve stimulator approach was associated with lower postoperative pain scores than PVB using a landmark approach. The P values (DerSimonian-Laird approach) for subgroup differences ranged from <0.00001 to 0.04 for 0- to 72-hour postoperative pain at rest. However, the effects were not confirmed by the HKSJ method.

Back to Top | Article Outline

Publication Bias and Small Study Effect

All FSN were lower than 5n + 10, where n is the original number of studies. Small study effects (by Egger intercept regression) were not found in all comparisons, except postoperative analgesic requirements (for both PVB versus general anesthesia/systematic analgesia and PVB versus spinal anesthesia).

Back to Top | Article Outline

DISCUSSION

Our meta-analysis demonstrates that PVB with sedation reduced PONV compared with general anesthesia/systematic analgesia for patients undergoing inguinal herniorrhaphy. PVB also resulted in less postoperative nausea and urinary retention but required a longer time to perform compared with spinal anesthesia. However, the current data failed to reject the null hypothesis of noninferiority on all pain and analgesic outcomes for both PVB versus general anesthesia/systematic analgesia and PVB versus spinal anesthesia, as well as on hemodynamic outcomes for PVB versus spinal anesthesia.

Our systematic review found that PVB provides better pain control and decreases postoperative analgesic requirement than INB and TAP block. Finally, surgery duration did not differ between PVB and all other groups.

Back to Top | Article Outline

Implications for Clinical Practice

For PVB versus general anesthesia/systematic analgesia, systematic reviews17,26 and meta-analyses23 reveal that PVB results in better postoperative pain control, less PONV, and shorter LOS than general anesthesia/systematic analgesia for thoracotomy and mastectomy. Our meta-analysis found that PVB reduced PONV and had a noninferior LOS. However, the pain control effect of PVB could not be reproduced because of the limited number of studies. More studies with larger sample sizes are required to confirm the possible superiority of pain control using PVB over general anesthesia/systematic analgesia.

For PVB versus neuraxial blocks, systematic reviews17 and meta-analyses18 have shown that PVB provides comparable postoperative pain control to neuraxial blocks for thoracotomy while causing less hypotension, urinary retention, and PONV. Our meta-analysis confirmed that PVB caused less postoperative nausea and urinary retention over spinal anesthesia for inguinal herniorrhaphy. However, the currently available data were not strong enough to demonstrate noninferiority on all outcomes of pain control or superiority on hemodynamic outcomes.

For PVB versus other peripheral nerve blocks, PVB seems to provide better pain control than other forms of peripheral nerve blocks from our systematic review. More RCTs are required to confirm these findings.

PVB requires a longer time to perform and has a clinically comparable failure (conversion to general anesthesia) rate (our meta-analysis10,12–14,40,43: 5.1% using landmark approach; other studies18,47,48: 4.4%–6.1%) to neuraxial blocks (other studies18,49–52: 3.2%–15.0%, spinal anesthesia or epidural block). The subgroup analysis demonstrated weak evidence that PVB using nerve stimulator guidance might provide better pain control than PVB using a landmark approach.

PVB also has its own specific complications47,48: pneumothorax (0.2%–1.0%), pleural puncture (0.8%–2.0%), epidural/intrathecal spread (0.5%–1.1%), vascular puncture (3.8%–8.7%), hypotension (4.6%), and hematoma (1.9%–3.1%). Among the studies we analyzed, 4 patients (2.0%) had epidural spread, 1 (0.5%) had hypotension, and 1 (0.5%) had a hematoma. In our meta-analysis, no studies reported any cases of pleural puncture (>60 patients16,40) or vascular puncture (>25 patients16).

Back to Top | Article Outline

Implications for Clinical Research

PVB may be a potential alternative to neuraxial blocks for other kinds of abdominal surgery, including colorectal, renal,53 hepatobiliary, gynecologic,54 and urologic surgery. In addition, continuous PVB may provide longer pain control for major surgery than a single injection.55 However, only a limited number of studies have investigated these applications of PVB. In addition, no RCT compared PVB with local anesthesia by surgeons for inguinal herniorrhaphy. Future studies may investigate this important topic.

Back to Top | Article Outline

Limitations

Clinical characteristics of the studies were heterogeneous, including techniques of PVB (unilateral versus bilateral, single versus multiple versus continuous, preoperative versus postoperative, types and doses of anesthetics) and postoperative analgesics. We had already separated the analyses of PVB versus general anesthesia/systematic analgesia, spinal anesthesia, and other peripheral nerve blocks to minimize clinical heterogeneity and did subgroup analyses if the number of studies allowed analysis. The numbers of included studies and the sample sizes were relatively small. There were publication bias and small study effects for some comparisons. Statistical exploration of the sources of heterogeneities was limited, but we have adopted the HKSJ method with Student t test to control for a type I error. Moreover, some clinically important outcomes (e.g., intraoperative hemodynamic changes from baseline and opioid side effects) were not reported by all studies. These results have to be interpreted with caution. Moreover, despite multiple attempts by our team, we were rarely successful in contacting the authors for more detailed information.

Back to Top | Article Outline

CONCLUSIONS

PVB provides anesthesia with fewer undesirable effects in patients undergoing inguinal herniorrhaphy. The choice that the anesthesiologist will make between PVB with sedation, general anesthesia/systematic analgesia, neuraxial block (i.e., spinal anesthesia), and other peripheral nerve blocks must be based on the time available to perform the block before surgery and adequate blockade of the surgical site. E

Back to Top | Article Outline

APPENDIX 1

Searching Strategies

PubMed (26 Results)

Paravertebral AND inguinal AND (hernia OR herniorrhaphy).

MEDLINE (26 Results)

(TOPIC:(paravertebral) AND (TOPIC: (inguinal)OR MeSH HEADING:exp: (Groin)))AND ((TOPIC: (hernia) OR MeSH HEADING:exp: (Hernia) OR MeSH HEADING:exp: (Hernia)) OR (TOPIC:(herniorrhaphy) OR MeSH HEADING:exp: (Herniorrhaphy)))

Timespan: All years.Indexes: MEDLINE.

EMBASE (45 Results)

#4: #1 AND #2 AND #3

#3: ‘hernia’/exp OR hernia OR ‘herniorrhaphy’/exp OR herniorrhaphy

#2: inguinal

#1: paravertebral.

CENTRAL (14 Results)

Paravertebral AND inguinal AND (hernia OR herniorrhaphy) in Title, Abstract, Keywords in Trials.

CINAHL (11 Results)

Paravertebral AND inguinal AND (hernia OR herniorrhaphy).

Back to Top | Article Outline

Appendix 2. Search Results

Table

Table

Back to Top | Article Outline

DISCLOSURES

Name: Lawrence Siu-Chun Law, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Lawrence Siu-Chun Law has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Mingjuan Tan, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Mingjuan Tan has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Yaowu Bai, MD.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Yaowu Bai reviewed the analysis of the data and approved the final manuscript.

Name: Timothy E. Miller, MB ChB, FRCA.

Contribution: This author helped write the manuscript.

Attestation: Timothy E. Miller approved the final manuscript.

Name: Yi-Ju Li, PhD.

Contribution: This author helped analyze the data.

Attestation: Yi-Ju Li has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Tong-Joo Gan, MD, MHS, FRCA.

Contribution: This author helped design the study, conduct the study, and write the manuscript.

Attestation: Tong-Joo Gan has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Back to Top | Article Outline

RECUSE NOTE

Dr. Tong-Joo Gan is the Section Editor for Ambulatory Anesthesiology and Perioperative Management for the journal. This manuscript was handled by Terese T. Horlocker, Section Editor for Regional Anesthesia for the Journal, and Dr. Gan was not involved in any way with the editorial process or decision.

Back to Top | Article Outline

REFERENCES

1. Wheeler M, Oderda GM, Ashburn MA, Lipman AG. Adverse events associated with postoperative opioid analgesia: a systematic review. J Pain. 2002;3:159–80
2. Pyati S, Gan TJ. Perioperative pain management. CNS Drugs. 2007;21:185–211
3. Tan M, Law LS, Gan TJ. Optimizing pain management to facilitate enhanced recovery after Surgery pathways. Can J Anaesth. 2015;62:203–18
4. Zukowski M, Kotfis K. The use of opioid adjuvants in perioperative multimodal analgesia. Anaesthesiol Intensive Ther. 2012;44:42–6
5. Young A, Buvanendran A. Recent advances in multimodal analgesia. Anesthesiol Clin. 2012;30:91–100
6. Loizides S, Gurusamy KS, Nagendran M, Rossi M, Guerrini GP, Davidson BR. Wound infiltration with local anaesthetic agents for laparoscopic cholecystectomy. Cochrane Database Syst Rev. 2014;3:CD007049
7. Byager N, Hansen MS, Mathiesen O, Dahl JB. The analgesic effect of wound infiltration with local anaesthetics after breast surgery: a qualitative systematic review. Acta Anaesthesiol Scand. 2014;58:402–10
8. Bai Y, Miller T, Tan M, Law LS, Gan TJ. Lidocaine patch for acute pain management: a meta-analysis of prospective controlled trials. Curr Med Res Opin. 2015;31:575–81
9. Akcaboy EY, Akcaboy ZN, Gogus N. Ambulatory inguinal herniorrhaphy: paravertebral block versus spinal anesthesia. Minerva Anestesiol. 2009;75:684–91
10. Akcaboy EY, Akcaboy ZN, Gogus N. Comparison of paravertebral block versus fast-track general anesthesia via laryngeal mask airway in outpatient inguinal herniorrhaphy. J Anesth. 2010;24:687–93
11. Bahar C, Utebey G, Ozlu O. Comparison of spinal anaesthesia and paravertebral block for inguinal hernia operations [Turkish] Anestezi Dergisi. 2012;20:217–22
12. Bhattacharya P, Mandal MC, Mukhopadhyay S, Das S, Pal PP, Basu SR. Unilateral paravertebral block: an alternative to conventional spinal anaesthesia for inguinal hernia repair. Acta Anaesthesiol Scand. 2010;54:246–51
13. Hadzic A, Kerimoglu B, Loreio D, Karaca PE, Claudio RE, Yufa M, Wedderburn R, Santos AC, Thys DM. Paravertebral blocks provide superior same-day recovery over general anesthesia for patients undergoing inguinal hernia repair. Anesth Analg. 2006;102:1076–81
14. Mandal MC, Das S, Gupta S, Ghosh TR, Basu SR. Paravertebral block can be an alternative to unilateral spinal anaesthesia for inguinal hernia repair. Indian J Anaesth. 2011;55:584–9
15. Naja MZ, el Hassan MJ, Oweidat M, Zbibo R, Ziade MF, Lönnqvist PA. Paravertebral blockade vs general anesthesia or spinal anesthesia for inguinal hernia repair. Middle East J Anaesthesiol. 2001;16:201–10
16. Naja ZM, Raf M, El Rajab M, Ziade FM, Al Tannir MA, Lönnqvist PA. Nerve stimulator-guided paravertebral blockade combined with sevoflurane sedation versus general anesthesia with systemic analgesia for postherniorrhaphy pain relief in children: a prospective randomized trial. Anesthesiology. 2005;103:600–5
17. Joshi GP, Bonnet F, Shah R, Wilkinson RC, Camu F, Fischer B, Neugebauer EA, Rawal N, Schug SA, Simanski C, Kehlet H. A systematic review of randomized trials evaluating regional techniques for postthoracotomy analgesia. Anesth Analg. 2008;107:1026–40
18. Davies RG, Myles PS, Graham JM. A comparison of the analgesic efficacy and side-effects of paravertebral vs epidural blockade for thoracotomy—a systematic review and meta-analysis of randomized trials. Br J Anaesth. 2006;96:418–26
19. Fleischer M, Marini CP, Statman R, Capella J, Shevde K. Local anesthesia is superior to spinal anesthesia for anorectal surgical procedures. Am Surg. 1994;60:812–5
20. Jellish WS, Thalji Z, Stevenson K, Shea J. A prospective randomized study comparing short- and intermediate-term perioperative outcome variables after spinal or general anesthesia for lumbar disk and laminectomy surgery. Anesth Analg. 1996;83:559–64
21. Miller AG, McKenzie J, Greenky M, Shaw E, Gandhi K, Hozack WJ, Parvizi J. Spinal anesthesia: should everyone receive a urinary catheter?: a randomized, prospective study of patients undergoing total hip arthroplasty. J Bone Joint Surg Am. 2013;95:1498–503
22. Moppett IK, Rowlands M, Mannings A, Moran CG, Wiles MDNOTTS Investigators. . LiDCO-based fluid management in patients undergoing hip fracture surgery under spinal anaesthesia: a randomized trial and systematic review. Br J Anaesth. 2015;114:444–59
23. Tahiri Y, Tran DQ, Bouteaud J, Xu L, Lalonde D, Luc M, Nikolis A. General anaesthesia versus thoracic paravertebral block for breast surgery: a meta-analysis. J Plast Reconstr Aesthet Surg. 2011;64:1261–9
24. Kotzé A, Scally A, Howell S. Efficacy and safety of different techniques of paravertebral block for analgesia after thoracotomy: a systematic review and metaregression. Br J Anaesth. 2009;103:626–36
25. Richardson J, Lönnqvist PA, Naja Z. Bilateral thoracic paravertebral block: potential and practice. Br J Anaesth. 2011;106:164–71
26. Thavaneswaran P, Rudkin GE, Cooter RD, Moyes DG, Perera CL, Maddern GJ. Brief reports: paravertebral block for anesthesia: a systematic review. Anesth Analg. 2010;110:1740–4
27. Klein SM, Bergh A, Steele SM, Georgiade GS, Greengrass RA. Thoracic paravertebral block for breast surgery. Anesth Analg. 2000;90:1402–5
28. Kehlet H, White PF. Optimizing anesthesia for inguinal herniorrhaphy: general, regional, or local anesthesia? Anesth Analg. 2001;93:1367–9
29. Callesen T, Bech K, Kehlet H. One-thousand consecutive inguinal hernia repairs under unmonitored local anesthesia. Anesth Analg. 2001;93:1373–6
30. Ozgün H, Kurt MN, Kurt I, Cevikel MH. Comparison of local, spinal, and general anaesthesia for inguinal herniorrhaphy. Eur J Surg. 2002;168:455–9
31. Mascha EJ, Turan A. Joint hypothesis testing and gatekeeping procedures for studies with multiple endpoints. Anesth Analg. 2012;114:1304–17
32. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6:e1000100
33. Takeshima N, Sozu T, Tajika A, Ogawa Y, Hayasaka Y, Furukawa TA. Which is more generalizable, powerful and interpretable in meta-analyses, mean difference or standardized mean difference? BMC Med Res Methodol. 2014;14:30
34. IntHout J, Ioannidis JP, Borm GF. The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis is straightforward and considerably outperforms the standard DerSimonian-Laird method. BMC Med Res Methodol. 2014;14:25
35. Hartung J, Knapp G. On tests of the overall treatment effect in meta-analysis with normally distributed responses. Stat Med. 2001;20:1771–82
36. Hartung J, Knapp G. A refined method for the meta-analysis of controlled clinical trials with binary outcome. Stat Med. 2001;20:3875–89
37. Hartung J. An alternative method for meta-analysis. Biom J. 1999;41:901–6
38. Rosenberg MS. The file-drawer problem revisited: a general weighted method for calculating fail-safe numbers in meta-analysis. Evolution. 2005;59:464–8
39. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629–34
40. Tug R, Ozcengiz D, Güneş Y. Single level paravertebral versus caudal block in paediatric inguinal surgery. Anaesth Intensive Care. 2011;39:909–13
41. Nickalls RW, Mapleson WW. Age-related iso-MAC charts for isoflurane, sevoflurane and desflurane in man. Br J Anaesth. 2003;91:170–4
42. Klein SM, Pietrobon R, Nielsen KC, Steele SM, Warner DS, Moylan JA, Eubanks WS, Greengrass RA. Paravertebral somatic nerve block compared with peripheral nerve blocks for outpatient inguinal herniorrhaphy. Reg Anesth Pain Med. 2002;27:476–80
43. Wassef MR, Randazzo T, Ward W. The paravertebral nerve root block for inguinal herniorrhaphy—a comparison with the field block approach. Reg Anesth Pain Med. 1998;23:451–6
44. Naja ZM, Raf M, El-Rajab M, Daoud N, Ziade FM, Al-Tannir MA, Lönnqvist PA. A comparison of nerve stimulator guided paravertebral block and ilio-inguinal nerve block for analgesia after inguinal herniorrhaphy in children. Anaesthesia. 2006;61:1064–8
45. Kaya C, Köksal E, Üstün YB, Yilmaz MZ, Torun AÇ, Özkan F. Comparative effects of transversus abdominis plane and paravertebral block on postoperative pain in inguinal hernia surgery. Acta Medica Mediterranea. 2014;30:341–6
46. Wood GJ, Lloyd JW, Bullingham RE, Britton BJ, Finch DR. Postoperative analgesia for day-case herniorrhaphy patients. A comparison of cryoanalgesia, paravertebral blockade and oral analgesia. Anaesthesia. 1981;36:603–10
47. Naja Z, Lönnqvist PA. Somatic paravertebral nerve blockade. Incidence of failed block and complications. Anaesthesia. 2001;56:1184–8
48. Lönnqvist PA, MacKenzie J, Soni AK, Conacher ID. Paravertebral blockade. Failure rate and complications. Anaesthesia. 1995;50:813–5
49. Heesen M, Van de Velde M, Klöhr S, Lehberger J, Rossaint R, Straube S. Meta-analysis of the success of block following combined spinal-epidural vs epidural analgesia during labour. Anaesthesia. 2014;69:64–71
50. Miro M, Guasch E, Gilsanz F. Comparison of epidural analgesia with combined spinal-epidural analgesia for labor: a retrospective study of 6497 cases. Int J Obstet Anesth. 2008;17:15–9
51. Pan PH, Bogard TD, Owen MD. Incidence and characteristics of failures in obstetric neuraxial analgesia and anesthesia: a retrospective analysis of 19,259 deliveries. Int J Obstet Anesth. 2004;13:227–33
52. Eappen S, Blinn A, Segal S. Incidence of epidural catheter replacement in parturients: a retrospective chart review. Int J Obstet Anesth. 1998;7:220–5
53. Moawad HE, Mousa SA, El-Hefnawy AS. Single-dose paravertebral blockade versus epidural blockade for pain relief after open renal surgery: a prospective randomized study. Saudi J Anaesth. 2013;7:61–7
54. Melnikov AL, Bjoergo S, Kongsgaard UE. Thoracic paravertebral block versus transversus abdominis plane block in major gynecological surgery: a prospective, randomized, controlled, observer-blinded study. Local Reg Anesth. 2012;5:55–61
55. Júnior Ade P, Erdmann TR, Santos TV, Brunharo GM, Filho CT, Losso MJ, Filho GR. Comparison between continuous thoracic epidural and paravertebral blocks for postoperative analgesia in patients undergoing thoracotomy: systematic review. Braz J Anesthesiol. 2013;63:433–42

Supplemental Digital Content

Back to Top | Article Outline
© 2015 International Anesthesia Research Society