Key PointsQuestion
Is the use of intercostal nerve block (ICNB) analgesia safe and beneficial for adults undergoing thoracic surgery?
Findings
In this systematic review and meta-analysis of 66 studies including 5184 adult patients undergoing thoracic surgery, the use of ICNB was associated with a clinically and statistically relevant analgesic benefit during the first 24 hours after thoracic surgery, with outcomes that were superior to systemic analgesia and noninferior to other techniques. Although ICNB was associated with a reduction in postoperative opioid consumption, the extent of this reduction was inferior to that of thoracic epidural and paravertebral block analgesia.
Meaning
This study found that ICNB was safe and beneficial for adults undergoing thoracic surgery, providing a reduction in pain during the first 24 hours after thoracic surgery; ICNB may be most beneficial for cases in which thoracic epidural or paravertebral block analgesia are not indicated.
Importance
The use of intercostal nerve block (ICNB) analgesia with local anesthesia is common in thoracic surgery. However, the benefits and safety of ICNB among adult patients undergoing surgery is unknown.
Objective
To evaluate the analgesic benefits and safety of ICNB among adults undergoing thoracic surgery.
Data Sources
A systematic search was performed in Ovid MEDLINE, Ovid Embase, Scopus, and the Cochrane Library databases using terms for ICNB and thoracic surgery (including thoracic surgery, thoracoscopy, thoracotomy, nerve block, intercostal nerves). The search and results were not limited by date, with the last search conducted on July 24, 2020.
Study Selection
Selected studies were experimental or observational and included adult patients undergoing cardiothoracic surgery in which ICNB was administered with local anesthesia via single injection, continuous infusion, or a combination of both techniques in at least 1 group of patients. For comparison with ICNB, studies that examined systemic analgesia and different forms of regional analgesia (such as thoracic epidural analgesia [TEA], paravertebral block [PVB], and other techniques) were included. These criteria were applied independently by 2 authors, and discrepancies were resolved by consensus. A total of 694 records were selected for screening.
Data Extraction and Synthesis
This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline. Data including patient characteristics, type of surgery, intervention analgesia, comparison analgesia, and primary and secondary outcomes were extracted independently by 3 authors. Synthesis was performed using a fixed-effects model.
Main Outcomes and Measures
The coprimary outcomes were postoperative pain intensity (measured as the worst static or dynamic pain using a validated 10-point scale, with 0 indicating no pain and 10 indicating severe pain) and opioid consumption (measured in morphine milligram equivalents [MMEs]) at prespecified intervals (0-6 hours, 7-24 hours, 25-48 hours, 49-72 hours, and >72 hours). Clinically relevant analgesia was defined as a 1-point or greater difference in pain intensity score at any interval. Secondary outcomes included 30-day postoperative complications and pulmonary function.
Results
Of 694 records screened, 608 were excluded based on prespecified exclusion criteria. The remaining 86 full-text articles were assessed for eligibility, and 20 of those articles were excluded. All of the 66 remaining studies (5184 patients; mean [SD] age, 53.9 [10.2] years; approximately 59% men and 41% women) were included in the qualitative analysis, and 59 studies (3325 patients) that provided data for at least 1 outcome were included in the quantitative meta-analysis. Experimental studies had a high risk of bias in multiple domains, including allocation concealment, blinding of participants and personnel, and blinding of outcome assessors. Marked differences (eg, crossover studies, timing of the intervention [intraoperative vs postoperative], blinding, and type of control group) were observed in the design and implementation of studies. The use of ICNB vs systemic analgesia was associated with lower static pain (0-6 hours after surgery: mean score difference, −1.40 points [95% CI, −1.46 to −1.33 points]; 7-24 hours after surgery: mean score difference, −1.27 points [95% CI, −1.40 to −1.13 points]) and lower dynamic pain (0-6 hours after surgery: mean score difference, −1.66 points [95% CI, −1.90 to −1.41 points]; 7-24 hours after surgery: mean score difference, −1.43 points [95% CI, −1.70 to −1.17 points]). Intercostal nerve block analgesia was noninferior to TEA (mean score difference in worst dynamic panic at 7-24 hours after surgery: 0.79 points; 95% CI, 0.28-1.29 points) and marginally inferior to PVB (mean score difference in worst dynamic pain at 7-24 hours after surgery: 1.29 points; 95% CI, 1.16 to 1.41 points). The largest opioid-sparing effect of ICNB vs systemic analgesia occurred at 48 hours after surgery (mean difference, −10.97 MMEs; 95% CI, −12.92 to −9.02 MMEs). The use of ICNB was associated with higher MME values compared with TEA (eg, 48 hours after surgery: mean difference, 48.31 MMEs; 95% CI, 36.11-60.52 MMEs) and PVB (eg, 48 hours after surgery: mean difference, 3.87 MMEs; 95% CI, 2.59-5.15 MMEs).
Conclusions and Relevance
In this study, single-injection ICNB was associated with a reduction in pain during the first 24 hours after thoracic surgery and was clinically noninferior to TEA or PVB. Intercostal nerve block analgesia had opioid-sparing effects; however, TEA and PVB were associated with larger decreases in postoperative MMEs, suggesting that ICNB may be most beneficial for cases in which TEA and PVB are not indicated.
Acute pain after thoracic surgery is common and severe, and can lead to increased morbidity.1,2 In the thorax, nociception travels primarily via the intercostal nerves.3 Therefore, blockade of the intercostal nerve is used to provide analgesia after thoracic surgery.4 Intercostal nerve blocks (ICNBs) are a common component of multimodal analgesia for thoracic surgery.5 Current guidelines suggest that continuous intercostal analgesia is similar to thoracic epidural analgesia (TEA).6 Aside from consideration of technical aspects and costs,7 the selection of any analgesic approach is typically based on its clinical benefits and disadvantages.8,9
In recent years, the use of minimally invasive techniques in thoracic surgery has substantially increased.10 This increase coincided with a decrease in the use of TEA and the emergence of fascial plane blocks.4 Previous reviews have found that ICNB was superior to systemic analgesia and was associated with reductions in opioid consumption.11,12 However, although Joshi et al12 recommended TEA and paravertebral block (PVB) as first-line options, Detterbeck et al11 questioned the superiority of TEA vs continuous extrapleural techniques. Considering the findings of these previous studies,4,10-12 we conducted a systematic review and meta-analysis to synthesize the evidence on the benefits and safety of ICNB among adult patients undergoing thoracic surgery.
We hypothesized that the use of ICNB would be associated with superior analgesia and reductions in opioid consumption compared with systemic analgesia alone, while being inferior to PVB and TEA. We also expected ICNB to be superior to systemic analgesia but inferior to PVB or TEA with regard to postoperative complications.
This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.13 The protocol was registered in the PROSPERO database (registration number: CRD42021224783).
Selected sources included observational and experimental studies of adults 18 years and older undergoing any cardiothoracic surgery in which ICNB with local anesthesia was administered via single injection, continuous infusion, or a combination of both techniques in at least 1 group of patients. The use of ICNB was separately compared with both systemic analgesia and different forms of regional analgesia, including TEA and PVB. The coprimary outcomes were acute postoperative pain intensity (dynamic and static) before hospital discharge and opioid consumption. The secondary outcomes were pulmonary function and 30-day postoperative complications.
A systematic literature search was constructed by a medical librarian (R.S.H.). Ovid MEDLINE, Ovid Embase, Scopus, and the Cochrane Library databases were queried using the following natural language and controlled vocabulary terms for ICNB and thoracic surgery: thoracic surgery, thoracic surgical procedures (including cardiac surgical procedures, mediastinoscopy, pulmonary surgical procedures, sternotomy, thoracoplasty, thoracoscopy, thoracostomy, thoracotomy, thymectomy, tracheostomy, tracheotomy, cardiac, heart, pulmonary, and lung), and ICNB-related terms (including nerve block, intercostal nerves, ICNB, nerve, and block). Records included were limited to human studies published in the English language. A sample electronic search is available on the PROSPERO website.14 Case reports, conference abstracts, editorial letters, and pediatric-only studies were excluded. Identification of other unpublished studies was not attempted. The search and results were not limited by date. After deduplication, 694 unique records were identified. The last date of search was July 24, 2020.
Only articles with available full text were included. Records were screened independently by 2 authors (C.E.G.-L. and J.P.C.) based on title and abstract. Discrepancies were resolved by consensus. The 2 authors then obtained full text of the remaining articles to assess eligibility.
Three authors (C.E.G.-L., A.P., and J.P.C.) extracted the data independently using Excel spreadsheets (Microsoft Corp) that were subsequently merged after the data were collected. Attempts were made to contact authors for missing data. The following data were extracted: demographic characteristics of participants (age, sex, and body mass index), type of surgery (sternotomy, thoracoscopy, thoracotomy, and not specified), type of intervention (single-injection, continuous administration, dose administered, and type of local anesthetic used), and comparator groups (eg, placebo, systemic analgesia, TEA, PVB, and erector spinae plane block). Pain was extracted as the worst static or dynamic pain on a validated 10-point scale (verbal, numerical, or visual, with 0 indicating no pain and 10 indicating severe pain) within the following postoperative periods: 0 to 6 hours, 7 to 24 hours, 25 to 48 hours, 49 to 72 hours, and more than 72 hours. Clinically relevant analgesia was defined as a 1-point or greater difference in pain intensity score at any interval.15 Opioid consumption was extracted for the same intervals and converted to intravenous morphine milligram equivalents (MMEs).
Nominal data were summarized using proportions, whereas continuous data were reported as means with SDs. Data provided as medians and ranges were converted to means and SDs according to the methods described in Hozo et al.16 Data from figures were digitized using WebPlotDigitizer software, version 4.4 (Ankit Rohatgi).17 Risk of bias in individual studies was assessed according to the criteria described by the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) Working Group18 and was considered for each outcome.
To provide estimates of intervention outcomes, a quantitative fixed-effects meta-analysis was performed using the meta package in R software, version 3.6.3 (R Foundation for Statistical Computing), when data from at least 3 studies were available for each outcome. Summary estimates, including odds ratios (ORs), mean differences, and 95% CIs, were calculated for each outcome. Heterogeneity was assessed using the I2 statistic and, when serious heterogeneity was detected, subgroups were investigated to identify potential differences. Qualitative analysis was followed by quantitative analysis using the GRADE criteria to rate the quality of evidence.19 Results were interpreted in the context of pooled effect estimates; risk of bias, heterogeneity (measured using the I2 statistic), imprecision, and indirectness were assessed for each outcome across the respective informing studies. Publication bias was determined by inspection of funnel plots for each outcome. Synthesis results, including information on the quality of evidence, are shown in Table 1.
To synthesize data, the Cochrane criteria86 were used for randomized clinical trials, and the GRADE criteria19 were used for nonrandomized and observational studies. To evaluate the quality of the body of evidence for each individual outcome according to GRADE criteria, the seriousness of risk was assessed across all informing studies to grade the certainty of evidence for each effect estimate.18
Of 694 records screened, 608 were excluded based on the prespecified exclusion criteria discussed in Methods (Figure 1). The remaining 86 full-text articles were assessed for eligibility, and 20 of those articles were excluded (5 were not written in the English language, 4 included interventions that were mislabeled as ICNB, 3 did not include a group of patients who received ICNB with local anesthesia, 3 were systematic reviews, 2 were editorials or short reports, 1 examined outcomes outside the scope of the present study, 1 had insufficient reporting, and 1 examined nonthoracic surgery). All of the remaining 66 studies20-85 (5184 patients; mean [SD] age, 53.9 [10.2] years; approximately 59% men and 41% women) were included in the qualitative analysis; of those, 59 studies20,22,25,26,28-34,36-42,44-59,65-69,71-74,77-81,83,84 (3325 patients) that provided data for at least 1 outcome were included in the quantitative meta-analysis.
Among 66 studies20-85 included in the qualitative analysis, 55 studies20-49,52-59,61,64-71,73,74,76-78,80-82 (3024 patients) were experimental, and 11 studies50,51,60,62,63,72,75,79,83-85 (2160 patients) were observational. Because analgesic techniques vary based on the type of surgery performed, we divided the populations accordingly. Thirty-nine studies20-25,27-30,32-34,36,38-40,44-49,60,64-68,70,73,74,76-82 (1805 patients) examined thoracotomy, 20 studies26,41,50-59,61-63,69,71,75,84,85 (2512 patients) examined thoracoscopy, 4 studies35,37,42,43 (235 patients) examined sternotomy, 2 studies72,83 (602 patients) examined both thoracotomy and thoracoscopy, and 1 study31 (30 patients) did not specify the type of thoracic surgery examined (Table 1).
The intervention used was single-injection ICNB in 51 studies20-26,29-34,36,39,41-48,50-53,55-61,64,65,67-72,74-76,78,79,81,83-85 (4690 patients) and continuous ICNB in 15 studies27,28,35,37,38,40,49,54,62,63,66,73,77,80,82 (494 patients). The comparison groups received TEA, interpleural analgesia, intercostal cryoanalgesia, transcutaneous electrical stimulation, PVB, erector spinae plane block, or serratus anterior plane block. The systemic analgesia group comprised patients in treatment arms who received any form of systemic analgesia (with or without placebo) and did not receive any form of regional analgesia (eg, TEA, PVB, or ICNB).
The risk of bias assessments across studies and for each of the coprimary outcomes are provided in eFigure 1 and eFigure 2 in the Supplement. Most experimental studies had a high risk of bias for allocation concealment,20-22,24-26,28,30,32-40,42-49,52,54,55,57-59,64-71,74,78,82 blinding of outcome assessors,29,32,44,46,58,59,63,67,71,74 blinding of participants and personnel,29,32,44,46,58,59,63,67,71,74 and other sources of bias,29,46,59,67,71 such as study design (eg, crossover studies and studies that used unvalidated pain measures). In 22 experimental studies21,22,24,25,29,33,38,40,45,49,52,55,61,67-71,73,74,80,81 (40.0%), a high risk of incomplete data for all outcomes was detected, primarily because of the lack of an intention-to-treat analysis. Most observational studies had a high risk of bias for inadequate control of confounding51,60,62,75,79,84,85 and a low risk of bias for incomplete follow-up50,51,62,63,72,75,79,83,84 and flawed measurement of exposure.50,51,62,63,72,75,79,83-85
Among 59 studies20,22,25,26,28-34,36-42,44-59,65-69,71-74,77-81,83,84 included in the quantitative meta-analysis, 54 studies20,22,25,26,28-34,36-42,44-49,52-59,65-69,71,73,74,77-81 (2615 patients) were experimental, and 5 studies50,51,72,83,84 (710 patients) were observational. Forest plots summarizing the coprimary outcomes are provided in Figure 2 and eFigure 3 in the Supplement. A summary of findings in Table 2 shows the evidence profile and the specific grading of the level of certainty for each outcome. The certainty of evidence for most outcomes was downgraded for reasons including risk of bias, heterogeneity, and imprecision.
Static and Dynamic Pain Intensity
Overall, ICNB was superior to systemic analgesia with regard to static and dynamic pain during the first 24 hours after surgery (eFigure 3A in the Supplement). The largest pain reduction occurred at 0 to 6 hours after surgery for both static pain (mean score difference, −1.40 points; 95% CI, −1.46 to −1.33 points) and dynamic pain (mean score difference, −1.66; 95% CI, −1.90 to −1.41). The benefit of ICNB analgesia decreased progressively over time. Static pain scores were lower in the ICNB group at 25 to 48 hours after surgery (mean score difference, −0.37 points; 95% CI, −0.60 to −0.14 points) and shifted in favor of systemic analgesia after 48 hours (mean score difference, 1.51 points; 95% CI, 0.94-2.08 points). Dynamic pain scores changed in favor of systemic analgesia at 25 to 48 hours after surgery (mean score difference, 0.51 points; 95% CI, 0.03-0.98 points).
In the thoracotomy subgroup, ICNB was superior to systemic analgesia with respect to static pain at 0 to 6 hours (mean score difference, −1.88 points; 95% CI, −2.07 to −1.69 points), 7 to 24 hours (mean score difference, −1.55 points; 95% CI, −1.81 to −1.29 points), and 25 to 48 hours (mean score difference, −0.38 points; 95% CI, −0.62 to −0.15 points) after surgery (eFigure 4 in the Supplement). In the thoracoscopy subgroup, ICNB was superior to systemic analgesia for static pain during the first 6 hours after surgery (mean score difference, −1.33 points; 95% CI, −1.40 to −1.27 points).
Intercostal nerve block analgesia was marginally inferior to TEA with regard to static pain during the first 24 hours after surgery only (0-6 hours: mean score difference, 0.49 points [95% CI, 0.18-0.79 points]; 7-24 hours: mean score difference, 0.41 points [95% CI, 0.21-0.61 points]) (eFigure 3B in the Supplement). For dynamic pain, ICNB was noninferior to TEA at 7 to 24 hours after surgery only (mean score difference, 0.79 points; 95% CI, 0.28-1.29 points). In the thoracotomy subgroup, no substantial differences in pain intensity between ICNB and TEA were observed (eg, dynamic pain at 0-6 hours: mean score difference, 0.13 points [95% CI, −0.27 to 0.52 points]; static pain at 0-6 hours: mean score difference, 0.64 points [95% CI, 0.27-1.02 points]) (eFigure 4 in the Supplement). Data for the thoracoscopy subgroup were available only for static pain at 7 to 24 hours after surgery, with a marginal difference in pain scores favoring TEA (mean score difference, 0.32 points; 95% CI, 0.04-0.60 points) (eFigure 4 in the Supplement).
Intercostal nerve block analgesia was inferior to PVB with regard to dynamic and static pain (eFigure 3B in the Supplement). The largest difference was noted in dynamic pain between 7 and 24 hours after surgery (mean score difference, 1.29 points; 95% CI, 1.16-1.41 points). In the thoracoscopy subgroup, patients who received ICNB had higher static pain scores between 7 and 24 hours after surgery compared with those who received PVB (mean score difference, 0.84 points; 95% CI, 0.72-0.96 points) (eFigure 4 in the Supplement).
The use of ICNB was associated with an opioid-sparing benefit compared with systemic analgesia. Overall, the reduction in opioid consumption associated with ICNB vs systemic analgesia started within the first 6 hours after surgery (mean difference, −3.04 MMEs; 95% CI, −3.17 to −2.90 MMEs) and peaked at 48 hours after surgery (mean difference, −10.97 MMEs; 95% CI, −12.92 to −9.02 MMEs) (Figure 2). In the thoracotomy subgroup, no difference was noted between ICNB and systemic analgesia during the first 24 hours after surgery (mean difference, 0.26 MMEs; 95% CI, −1.25 to 1.76 MMEs) (eFigure 5 in the Supplement). However, a reduction in opioid consumption was present at 48 hours after surgery (mean difference, −11.73 MMEs; 95% CI, −13.70 to −9.76 MMEs).
Intercostal nerve block analgesia was inferior to TEA with regard to opioid consumption at 24 hours after surgery (mean difference, 3.77 MMEs; 95% CI, 2.31-5.23 MMEs) (Figure 2). This effect was more marked at 48 hours after surgery, during which opioid consumption increased to 48.31 MMEs (95% CI, 36.11-60.52 MMEs). These findings were specific to patients undergoing thoracotomy. The high heterogeneity was explained by the Concha et al65 study, in which patients allocated to the TEA group only received opioids epidurally, which may have overestimated the opioid-sparing benefit of TEA.
Intercostal nerve block analgesia was only inferior to PVB at 48 hours after surgery (mean difference, 3.87 MMEs; 95% CI, 2.59-5.15 MMEs) (Figure 2).
Intercostal nerve block analgesia was associated with a reduction in the risk of nausea and vomiting compared with systemic analgesia (OR, 0.44; 95% CI, 0.20-0.94) (eFigure 6 in the Supplement). The opposite result was observed for ICNB vs TEA (OR, 1.60; 95% CI, 0.96-2.66) and PVB (OR, 1.66; 95% CI, 0.96-2.89). In the thoracotomy subgroup, no significant differences were observed between ICNB and TEA (eFigure 7 in the Supplement). In the thoracoscopy subgroup, no significant differences were observed between ICNB and PVB (eFigure 7 in the Supplement).
The risk of cardiovascular complications was similar between ICNB and systemic analgesia (OR, 1.07; 95% CI, 0.44-2.63) (eFigure 6 in the Supplement). However, ICNB was associated with a reduction in the risk of cardiovascular complications compared with TEA analgesia (OR, 0.66; 95% CI, 0.46-0.93). In the thoracotomy subgroup, ICNB was associated with a reduced risk of cardiovascular complications compared with TEA (OR, 0.68; 95% CI, 0.47-0.98) (eFigure 7 in the Supplement). In the thoracoscopy subgroup, ICNB was inferior to PVB (OR, 3.4; 95% CI, 1.27-9.08) (eFigure 7 in the Supplement).
The use of ICNB was associated with a reduction in the risk of arterial hypotension compared with TEA (OR, 0.20; 95% CI, 0.06-0.74) (eFigure 6 in the Supplement). In the thoracotomy subgroup, ICNB was associated with a reduction in the risk of hypotension compared with TEA (OR, 0.21; 95% CI, 0.05-0.87) (eFigure 7 in the Supplement). No data were available for thoracoscopic surgery.
Intercostal nerve block analgesia was associated with a reduction in the risk of pulmonary complications compared with systemic analgesia (OR, 0.45; 95% CI, 0.26-0.79) (eFigure 6 in the Supplement). No difference was noted between ICNB and TEA (OR, 0.86; 95% CI, 0.63-1.18) or PVB (OR, 1.07; 95% CI, 0.25-4.63). In the thoracotomy subgroup, ICNB was superior to systemic analgesia but noninferior to TEA (eFigure 7 in the Supplement). No data were available for thoracoscopic surgery.
With few adverse events reported, no significant difference in risk was noted between ICNB and TEA with regard to 30-day mortality, neurologic complications, catheter or injection site infection, hematoma, pruritus, or urinary retention (eFigure 6 in the Supplement). In addition, no substantial difference in the risk of urinary retention was found between ICNB and PVB. Results remained similar when stratified by type of surgery (eFigure 7 in the Supplement).
From baseline, ICNB was associated with higher forced expiratory volume in the first second compared with systemic analgesia at 7 to 24 hours (mean difference, 20.19%; 95% CI, 16.45%-23.93%), 25 to 48 hours (mean difference, 15.75%; 95% CI, 12.35%-19.14%), 49 to 72 hours (mean difference, 19.57%; 95% CI, 16.63%-22.52%), and more than 72 hours (mean difference, 19.75%; 95%CI, 16.32%-23.18%) after surgery (eFigure 8 and eFigure 9 in the Supplement). These results were specific to the thoracotomy subgroup.
Intercostal nerve block analgesia was associated with higher forced vital capacity from baseline compared with systemic analgesia at 7 to 24 hours (mean difference, 10.95%; 95% CI, 8.34%-13.57%) and 25 to 48 hours (mean difference, 8.89%; 95% CI, 6.39%-11.38%) after surgery (eFigure 8 and eFigure 9 in the Supplement). These results were specific to the thoracotomy subgroup. When compared with TEA, no difference was observed in the period of 7 to 24 hours after surgery (mean difference, 3.78%; 95% CI, −2.13% to 9.69%).
Data from 7 studies22,66,69,71,73,77,84 (478 participants) did not reveal differences in hospital length of stay with the use of ICNB vs TEA (mean difference, −3.38 hours; 95% CI, −10.75 to 4.00 hours) (eFigure 10 in the Supplement).
In the thoracotomy subgroup, ICNB was associated with an increased length of stay of approximately 14.3 hours (95% CI, 0.15-28.45 hours) vs TEA (eFigure 11 in the Supplement). In the thoracoscopy subgroup, the use of ICNB was associated with a decreased length of stay of approximately −9.97 hours (95% CI, −18.61 to −1.33 hours) vs TEA (eFigure 11 in the Supplement).
A slight increase in the length of stay occurred when ICNB was compared with PVB (mean difference, 5.27 hours; 95% CI, 1.11-9.42 hours) (eFigure 11 in the Supplement). The results remained similar when limited to thoracoscopic procedures.
In this systematic review and meta-analysis, the use of single-injection ICNB among adults undergoing thoracic surgery was associated with a small reduction in pain scores during the first 24 hours after surgery. Intercostal nerve block analgesia was superior to systemic opioid-based analgesia, noninferior to TEA, and marginally inferior to PVB. Because ICNB analgesia was also associated with better pulmonary function and a reduction in the risk of pulmonary complications, these findings were clinically relevant. Although ICNB was associated with reductions in opioid consumption compared with systemic analgesia alone, patients receiving ICNB consumed more opioids than those receiving TEA or PVB. However, caution is warranted when interpreting these findings because the quality of evidence was reduced by the limitations of the included studies.
Unlike previous reviews,11,12 our study provided estimates of the strength and duration of analgesic benefits. These estimates may allow clinicians to balance the benefits and harms of regional analgesia. Notably, both pulmonary and cardiovascular complications have been associated with postoperative mortality among patients undergoing thoracic surgery.10 Our results bring into question the superiority of TEA with regard to analgesia because the differences were minimal and inconsistent.87
The data suggested that the benefit of ICNB analgesia decreases progressively and disappears at 24 to 48 hours after surgery. Reliance on ICNB after this period may result in an abrupt lack of analgesia or rebound pain, represented by higher pain scores at 24 hours after surgery for dynamic pain and 48 hours after surgery for static pain.88 This finding is relevant because the severity of acute pain may be the main measure associated with the occurrence of chronic pain.89 Notably, the fact that ICNB was noninferior to TEA may underscore the known limitations of TEA, which has reported failure rates of up to 30%.90 Nevertheless, the success rate of TEA may be improved by the use of ultrasonography91 or the implementation of a preoperative block area.92 Comparisons between the use of ICNB with liposomal bupivacaine vs other regional analgesia techniques have only been performed in observational studies.72,83 Hussain et al93 recently reported that the use of liposomal bupivacaine in peripheral nerve block analgesia was not superior to plain local anesthetic formulations. Therefore, we do not consider the use of liposomal bupivacaine as indicated to provide sustained and beneficial analgesia after thoracic surgery.
Systemic hypotension is a known adverse event associated with TEA.94 We found that ICNB was associated with a reduction in the risk of hypotension compared with TEA. Whether TEA-associated hypotension is associated with cardiovascular events is unclear given that neuraxial blockade has not been independently associated with worse cardiovascular outcomes.95 Meta-analyses comparing TEA with PVB have also found limited high-quality evidence suggesting that PVB is associated with a lower risk of hypotension than TEA without differences in morbidity or mortality.9,96 However, most studies comparing the impact of TEA with that of other techniques have not investigated the incidence of silent events, such as myocardial injury after surgery, which may be associated with postoperative mortality.97
This study has several limitations. First, most studies20-40,42-49,51,52,54-62,64-71,73-75,77-82,84,85 included in the meta-analysis had at least 1 domain at a high risk of bias. In addition to small samples, differences in the protocol designs and types of analgesia produced high heterogeneity and imprecision. Differences in opioid consumption associated with different types of analgesia during certain intervals, such as ICNB vs TEA at 48 hours after surgery, may be overestimated. Second, we were unable to perform subgroup analyses incorporating continuous techniques or extended-release formulations. However, the benefits of extended-release medications have recently come into question,93 highlighting the need for pharmacological innovation. Third, the analysis of postoperative complications was limited by the use of observational data and the inadequate outcome definitions provided by the randomized clinical trials included in the meta-analysis. Nevertheless, observational studies can overcome the risk of sampling error associated with small samples.98 Fourth, most studies20,21,23,25,26,28,29,31,33,34,36,38,39,41,46,49,50,55,57,61-65,68-71,73,76,78,80,82 did not include complications as a primary or secondary outcome, raising concerns about detection bias.
This systematic review and meta-analysis found that single-injection ICNB was associated with a reduction in pain during the first 24 hours after thoracic surgery. Within the limitations of the available evidence, ICNB was superior to systemic opioid-based analgesia, noninferior to TEA, and marginally inferior to PVB. Although a small opioid-sparing benefit was found for ICNB alone, TEA and PVB were more favorable when opioid reduction was a consideration. Therefore, ICNB analgesia may be most beneficial for cases in which TEA or PVB are not indicated. Randomized clinical trials with rigorous methodological approaches and a priori outcomes that include safety end points are needed.
Accepted for Publication: September 12, 2021.
Published: November 15, 2021. doi:10.1001/jamanetworkopen.2021.33394
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Guerra-Londono CE et al. JAMA Network Open.
Corresponding Author: Juan P. Cata, MD, Department of Anesthesiology and Perioperative Medicine, MD Anderson Cancer Center, University of Texas, 1515 Holcombe Blvd, Houston, TX 77030 (jcata@mdanderson.org).
Author Contributions: Dr Cata had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Guerra-Londono, Privorotskiy, Memtsoudis, Mariano, Cata.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Guerra-Londono, Hicklen, Mariano, Cata.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Cozowicz, Cata.
Administrative, technical, or material support: Guerra-Londono, Cozowicz, Hicklen, Cata.
Supervision: Memtsoudis, Mariano, Cata.
Conflict of Interest Disclosures: Dr Memtsoudis reported receiving personal fees from Teikoku Pharma, owning a patent for a multicatheter infusion system, being the owner of SGMN Consulting, and being the co-owner of Centauros Healthcare Analytics and Consulting outside the submitted work. No other disclosures were reported.
2.Bendixen
M, Jorgensen
OD, Kronborg
C, Andersen
C, Licht
PB. Postoperative pain and quality of life after lobectomy via video-assisted thoracoscopic surgery or anterolateral thoracotomy for early stage lung cancer: a randomised controlled trial.
Lancet Oncol. 2016;17(6):836-844. doi:
10.1016/S1470-2045(16)00173-X
PubMedGoogle ScholarCrossref 6.Batchelor
TJP, Rasburn
NJ, Abdelnour-Berchtold
E,
et al. Guidelines for enhanced recovery after lung surgery: recommendations of the Enhanced Recovery After Surgery (ERAS) Society and the European Society of Thoracic Surgeons (ESTS).
Eur J Cardiothorac Surg. 2019;55(1):91-115. doi:
10.1093/ejcts/ezy301
PubMedGoogle ScholarCrossref 9.Baidya
DK, Khanna
P, Maitra
S. Analgesic efficacy and safety of thoracic paravertebral and epidural analgesia for thoracic surgery: a systematic review and meta-analysis.
Interact Cardiovasc Thorac Surg. 2014;18(5):626-635. doi:
10.1093/icvts/ivt551
PubMedGoogle ScholarCrossref 15.Myles
PS, Myles
DB, Galagher
W,
et al. Measuring acute postoperative pain using the visual analog scale: the minimal clinically important difference and patient acceptable symptom state.
Br J Anaesth. 2017;118(3):424-429. doi:
10.1093/bja/aew466
PubMedGoogle ScholarCrossref 23.Fiorelli
S, Leopizzi
G, Menna
C,
et al. Ultrasound-guided erector spinae plane block versus intercostal nerve block for post-minithoracotomy acute pain management: a randomized controlled trial.
J Cardiothorac Vasc Anesth. 2020;34(9):2421-2429. doi:
10.1053/j.jvca.2020.01.026
PubMedGoogle ScholarCrossref 24.Shafei
H, Chamberlain
M, Natrajan
KN, Khan
MA, Gandhi
RG. Intrapleural bupivacaine for early post-thoracotomy analgesia—comparison with bupivacaine intercostal block and cryofreezing.
Thorac Cardiovasc Surg. 1990;38(1):38-41. doi:
10.1055/s-2007-1013989
PubMedGoogle ScholarCrossref 26.Bolotin
G, Lazarovici
H, Uretzky
G, Zlotnick
AY, Tamir
A, Saute
M. The efficacy of intraoperative internal intercostal nerve block during video-assisted thoracic surgery on postoperative pain.
Ann Thorac Surg. 2000;70(6):1872-1875. doi:
10.1016/S0003-4975(00)01757-4
PubMedGoogle ScholarCrossref 28.Chan
VW, Chung
F, Cheng
DC, Seyone
C, Chung
A, Kirby
TJ. Analgesic and pulmonary effects of continuous intercostal nerve block following thoracotomy.
Can J Anaesth. 1991;38(6):733-739. doi:
10.1007/BF03008451
PubMedGoogle ScholarCrossref 30.Liu
M, Rock
P, Grass
JA,
et al. Double-blind randomized evaluation of intercostal nerve blocks as an adjuvant to subarachnoid administered morphine for post-thoracotomy analgesia.
Reg Anesth. 1995;20(5):418-425.
PubMedGoogle Scholar 32.Zhan
Y, Chen
G, Huang
J, Hou
B, Liu
W, Chen
S. Effect of intercostal nerve block combined with general anesthesia on the stress response in patients undergoing minimally invasive mitral valve surgery.
Exp Ther Med. 2017;14(4):3259-3264. doi:
10.3892/etm.2017.4868
PubMedGoogle ScholarCrossref 33.Joucken
K, Michel
L, Schoevaerdts
JC, Mayne
A, Randour
P. Cryoanalgesia for post-thoracotomy pain relief.
Acta Anaesthesiol Belg. 1987;38(2):179-183.
PubMedGoogle Scholar 38.Carretta
A, Zannini
P, Chiesa
G, Altese
R, Melloni
G, Grossi
A. Efficacy of ketorolac tromethamine and extrapleural intercostal nerve block on post-thoracotomy pain. a prospective, randomized study.
Int Surg. 1996;81(3):224-228.
PubMedGoogle Scholar 39.Ghafouri
A, Movafegh
A, Nasr-Esfahani
M, Gholamrezanezhad
A, Madhkhan
S. Post-thoracotomy pain and pulmonary function—comparison of intermittent intercostal bupivacaine vs intravenous pethidine.
Middle East J Anaesthesiol. 2008;19(4):767-780.
PubMedGoogle Scholar 41.Ahmed
Z, Samad
K, Ullah
H. Role of intercostal nerve block in reducing postoperative pain following video-assisted thoracoscopy: a randomized controlled trial.
Saudi J Anaesth. 2017;11(1):54-57. doi:
10.4103/1658-354X.197342
PubMedGoogle Scholar 44.Zhu
M, Gu
Y, Sun
X, Liu
X, Chen
W, Miao
C. Ultrasound-guided intercostal nerve block following esophagectomy for acute postoperative pain relief in the postanesthesia care unit.
Pain Pract. 2018;18(7):879-883. doi:
10.1111/papr.12689
PubMedGoogle ScholarCrossref 46.Wang
Y, Cheng
J, Yang
L, Wang
J, Liu
H, Lv
Z. Ropivacaine for intercostal nerve block improves early postoperative cognitive dysfunction in patients following thoracotomy for esophageal cancer.
Med Sci Monit. 2019;25:460-465. doi:
10.12659/MSM.912328
PubMedGoogle ScholarCrossref 50.Yang
H, Dong
Q, Liang
L,
et al. The comparison of ultrasound-guided thoracic paravertebral blockade and internal intercostal nerve block for non-intubated video-assisted thoracic surgery.
J Thorac Dis. 2019;11(8):3476-3481. doi:
10.21037/jtd.2019.07.77
PubMedGoogle ScholarCrossref 52.Hutchins
J, Sanchez
J, Andrade
R, Podgaetz
E, Wang
Q, Sikka
R. Ultrasound-guided paravertebral catheter versus intercostal blocks for postoperative pain control in video-assisted thoracoscopic surgery: a prospective randomized trial.
J Cardiothorac Vasc Anesth. 2017;31(2):458-463. doi:
10.1053/j.jvca.2016.08.010
PubMedGoogle ScholarCrossref 53.Chen
N, Qiao
Q, Chen
R, Xu
Q, Zhang
Y, Tian
Y. The effect of ultrasound-guided intercostal nerve block, single-injection erector spinae plane block and multiple-injection paravertebral block on postoperative analgesia in thoracoscopic surgery: a randomized, double-blinded, clinical trial.
J Clin Anesth. 2020;59:106-111. doi:
10.1016/j.jclinane.2019.07.002
PubMedGoogle ScholarCrossref 54.Kadomatsu
Y, Mori
S, Ueno
H, Uchiyama
M, Wakai
K. Comparison of the analgesic effects of modified continuous intercostal block and paravertebral block under surgeon’s direct vision after video-assisted thoracic surgery: a randomized clinical trial.
Gen Thorac Cardiovasc Surg. 2018;66(7):425-431. doi:
10.1007/s11748-018-0936-8
PubMedGoogle ScholarCrossref 56.Wu
C, Ma
W, Cen
Q, Cai
Q, Wang
J, Cao
Y. A comparison of the incidence of supraventricular arrhythmias between thoracic paravertebral and intercostal nerve blocks in patients undergoing thoracoscopic surgery: a randomised trial.
Eur J Anaesthesiol. 2018;35(10):792-798. doi:
10.1097/EJA.0000000000000837
PubMedGoogle ScholarCrossref 57.Mogahed
MM, Elkahwagy
MS. Paravertebral block versus intercostal nerve block in non-intubated uniportal video-assisted thoracoscopic surgery: a randomised controlled trial.
Heart Lung Circ. 2020;29(5):800-807. doi:
10.1016/j.hlc.2019.04.013PubMedGoogle Scholar 58.Xiang
X, Zhou
H, Wu
Y, Fang
J, Lian
Y. Impact of supraglottic device with assist ventilation under general anesthesia combined with nerve block in uniportal video-assisted thoracoscopic surgery.
Medicine (Baltimore). 2020;99(10):e19240. doi:
10.1097/MD.0000000000019240
PubMedGoogle Scholar 59.Zheng
Y, Wang
H, Ma
X, Cheng
Z, Cao
W, Shao
D. Comparison of the effect of ultrasound-guided thoracic paravertebral nerve block and intercostal nerve block for video-assisted thoracic surgery under spontaneous-ventilating anesthesia.
Rev Assoc Med Bras (1992). 2020;66(4):452-457. doi:
10.1590/1806-9282.66.4.452 60.Oksuz
G, Sayan
M, Arslan
M,
et al. The comparison of serratus anterior plane block versus intercostal block for postoperative analgesia following thoracotomy surgery.
Anestezi Dergisi. 2018;26(4):223-228.
Google Scholar 61.Kim
S, Bae
CM, Do
YW, Moon
S, Baek
SI, Lee
DH. Serratus anterior plane block and intercostal nerve block after thoracoscopic surgery.
Thorac Cardiovasc Surg. 2021;69(6):564-569.
PubMedGoogle Scholar 62.Hsieh
MJ, Wang
KC, Liu
HP,
et al. Management of acute postoperative pain with continuous intercostal nerve block after single port video-assisted thoracoscopic anatomic resection.
J Thorac Dis. 2016;8(12):3563-3571. doi:
10.21037/jtd.2016.12.30
PubMedGoogle ScholarCrossref 63.Wu
CF, Hsieh
MJ, Liu
HP,
et al. Management of post-operative pain by placement of an intraoperative intercostal catheter after single port video-assisted thoracoscopic surgery: a propensity-score matched study.
J Thorac Dis. 2016;8(6):1087-1093. doi:
10.21037/jtd.2016.04.01
PubMedGoogle ScholarCrossref 64.Bachmann-Mennenga
B, Biscoping
J, Kuhn
DF,
et al. Intercostal nerve block, interpleural analgesia, thoracic epidural block or systemic opioid application for pain relief after thoracotomy?
Eur J Cardiothorac Surg. 1993;7(1):12-18. doi:
10.1016/1010-7940(93)90141-W
PubMedGoogle ScholarCrossref 65.Concha
M, Dagnino
J, Cariaga
M, Aguilera
J, Aparicio
R, Guerrero
M. Analgesia after thoracotomy: epidural fentanyl/bupivacaine compared with intercostal nerve block plus intravenous morphine.
J Cardiothorac Vasc Anesth. 2004;18(3):322-326. doi:
10.1053/j.jvca.2004.03.013
PubMedGoogle ScholarCrossref 68.Scheinin
B, Scheinin
M, Asantila
R, Lindberg
R, Viinamaki
O. Sympatho-adrenal and pituitary hormone responses during and immediately after thoracic surgery—modulation by four different pain treatments.
Acta Anaesthesiol Scand. 1987;31(8):762-767. doi:
10.1111/j.1399-6576.1987.tb02660.x
PubMedGoogle ScholarCrossref 74.Vilvanathan
S, Kuppuswamy
B, Sahajanandan
R. A randomized control trial to compare thoracic epidural with intercostal block plus intravenous morphine infusion for postoperative analgesia in patients undergoing elective thoracotomy.
Ann Card Anaesth. 2020;23(2):127-133. doi:
10.4103/aca.ACA_167_18
PubMedGoogle Scholar 76.Ranganathan
P, Tadvi
A, Jiwnani
S, Karimundackal
G, Pramesh
CS. A randomised evaluation of intercostal block as an adjunct to epidural analgesia for post-thoracotomy pain.
Indian J Anaesth. 2020;64(4):280-285. doi:
10.4103/ija.IJA_714_19
PubMedGoogle ScholarCrossref 77.Sagiroglu
G, Baysal
A, Kiraz
OG, Meydan
B, Tasçi
AE, Iskender
I. Comparison of continuous use of thoracic epidural analgesia and intercostal block for pain management after thoracotomy.
Kardiochir Torakochirurgia Pol. 2013;10(3):244-250.
Google Scholar 80.Dauphin
A, Lubanska-Hubert
E, Young
JE, Miller
JD, Bennett
WF, Fuller
HD. Comparative study of continuous extrapleural intercostal nerve block and lumbar epidural morphine in post-thoracotomy pain.
Can J Surg. 1997;40(6):431-436.
PubMedGoogle Scholar 81.Meierhenrich
R, Hock
D, Kuhn
S,
et al. Analgesia and pulmonary function after lung surgery: is a single intercostal nerve block plus patient-controlled intravenous morphine as effective as patient-controlled epidural anaesthesia? a randomized non-inferiority clinical trial.
Br J Anaesth. 2011;106(4):580-589. doi:
10.1093/bja/aeq418
PubMedGoogle ScholarCrossref 84.Hung
MH, Chan
KC, Liu
YJ,
et al. Nonintubated thoracoscopic lobectomy for lung cancer using epidural anesthesia and intercostal blockade: a retrospective cohort study of 238 cases.
Medicine (Baltimore). 2015;94(13):e727. doi:
10.1097/MD.0000000000000727 85.Ambrogi
V, Mineo
TC. VATS biopsy for undetermined interstitial lung disease under non-general anesthesia: comparison between uniportal approach under intercostal block vs. three-ports in epidural anesthesia.
J Thorac Dis. 2014;6(7):888-895.
PubMedGoogle Scholar 86.Higgins JPT, Thomas J, Chandler J, et al, eds.
Cochrane Handbook for Systematic Reviews of Interventions. Version 6.2. Cochrane; 2021. Updated February 2021. Accessed August 2020.
https://training.cochrane.org/handbook 94.Leslie K, McIlroy D, Kasza J, et al. Neuraxial block and postoperative epidural analgesia: effects on outcomes in the POISE-2 trial.
Br J Anaesth. 2016;116(1):100-112. doi:
10.1093/bja/aev255 97.Botto
F, Alonso-Coello
P, Chan
MTV,
et al; Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Writing Group on behalf of the Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Investigators; Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Study Investigators Writing Group; Vascular Events in Noncardiac Surgery Patients Cohort Evaluation Operations Committee; Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Study Investigators. Myocardial injury after noncardiac surgery: a large, international, prospective cohort study establishing diagnostic criteria, characteristics, predictors, and 30-day outcomes.
Anesthesiology. 2014;120(3):564-578. doi:
10.1097/ALN.0000000000000113
PubMedGoogle Scholar