Patients undergoing thoracic surgery may experience severe postoperative pain. Indeed, thoracotomy is recognised as one of the most painful surgical procedures.1,2 Therefore, when video-assisted thoracoscopic surgery (VATS) was introduced, it soon became very popular and widely regarded as the gold standard for thoracic surgery.3–6 Despite smaller incisions, absence of rib retraction and less tissue damage; however, acute and chronic pain control remains a major clinical problem with VATS.7–8 In this scenario, ultrasound-guided serratus anterior plane block (SAPb) is a promising interfascial plane block, having the potential to provide adequate analgesia. SAPb consists of the injection of local anaesthetic under ultrasound guidance into the interfacial plane, either above or below the serratus anterior muscle.3 This regional block provides analgesia of the hemithorax, usually extending between the 2nd and 9th thoracic dermatomes, by blocking the lateral cutaneous branches of the intercostal nerves.9
Although SAPb is widely used in addition to general anaesthesia for pain control in thoracic surgery, the scientific literature on its efficacy is limited, and systematic reviews and meta-analysis of randomised controlled trials (RCTs) are missing.
The primary aim of our meta-analysis was to assess the analgesic efficacy of adding SAPb to general anaesthesia, as opposed to general anaesthesia alone, for peri-operative pain control in adult patients undergoing VATS. The secondary aim was to evaluate differences postoperative opioid use, in intra-operative hypotension and postoperative side-effects, such as nausea and vomiting, respiratory complications, dizziness, time for chest tube removal and length of hospital stay (LOS).
Materials and methods
PRISMA Statement Guidelines (Preferred Reporting Items for Systematic Reviews and Meta-analyses) were followed to prepare this article.10
A review protocol was written before conducting this study on 11 November 2019 (Supplemental Digital Content 1, https://links.lww.com/EJA/A340).
We performed a systematic research of the medical literature for the identification, screening and inclusion of articles; the search was performed by two study members (MZ and SZ) in close collaboration with the rest of the research team. It was then reviewed by another member of the team (ADC) following the PRESS checklist.11
The search was performed on 6 December 2019 in the following databases: PubMed, Web of Science, Google Scholar and the Cochrane Central Register of Controlled Trials. We also checked the reference lists of included studies. To reduce publication bias, the ongoing trials at ClinicalTrials.gov and the proceedings from the American Society of Anesthesiologists annual meetings over the last 5 years were also retrieved. For specific information regarding our search strategy (see Supplemental Digital Content 2, https://links.lww.com/EJA/A341). We did not apply any restrictions on publication type, language, or year of publication.
Two researchers independently screened titles and abstracts of the identified articles to identify relevant and non-relevant articles. Each citation was reviewed in duplicate by two of the reviewers, with full-text retrieval of any citation considered potentially relevant. The following PICOS criteria were used for study inclusion: patients of 18 years or older undergoing VATS (P); patients received single-shot SAPb (I); general anaesthesia, with or without wounds infiltration (C); analgesic efficacy measured as postoperative pain, use of intra and postoperative opioids, intra-operative hypotension, postoperative side-effects, such as nausea and vomiting, and respiratory complications, dizziness, time for chest tube removal and LOS (O); RCTs (S).
Data extraction and data retrieval
After identifying those studies meeting the inclusion criteria, two members of our team (AB and ADC) independently reviewed and assessed each of the included studies. Any disagreement on studies was planned to be solved by a third author (FZ) or by contacting the corresponding author. In addition, the following information was collected: first author, year of the study, total number of patients per group, opioid dosage both in the intra-operative setting and in the first 24 postoperative hours, postoperative pain at 6, 12 and 24 hours, occurrence of complications (intra-operative hypotension, postoperative nausea and vomiting, postoperative respiratory complications), time for chest-tube removal (days) and LOS (days) (Supplemental Digital Content 3 contains the review data extraction form, https://links.lww.com/EJA/A342). The different opioids were converted to intravenous morphine equivalents using the GlobalRPh morphine equivalent calculator considering 0% cross-tolerance modifier (http://www.globalrph.com/narcotic).
If data were missing, a request was sent by mail to the corresponding author of the study. If no response was received after our initial request, a second request was sent 1 week later. A third and final request was sent 1 week after the second request.
Quality assessment and quality of evidence assessment
The two researchers who performed the searching and data extraction subsequently read all included randomised comparative studies and independently evaluated their quality using two tools: the Jadad Scale,12 and the Risk of Bias (RoB) 2 Tool.13 Disagreements were resolved by discussion.
The Jadad Scale is a three-point questionnaire investigating randomisation (2 points), blinding (2 points) and withdrawals/dropouts (1 point). Studies scoring 2 points or less are considered as studies of low quality. The recently introduces RoB 2 Tool evaluates study quality and risk of bias by exploring six domains (bias arising from the randomisation process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in measurement of the outcome, bias in the selection of the reported result) and each domain is judged on a three-grade scale (low, high or some concerns). Overall risk of bias is then defined as ‘Low risk of bias’ if the study is judged to be at low risk of bias for all domains; ‘some concerns’ if the study is judged to raise some concerns in at least one domain, but not to be at high risk of bias for any domain; and at ‘high risk of bias’ if the study is judged to be at high risk of bias in at least one domain, or the study is judged to have some concerns for multiple domains.
We used the GRADE approach to assess the quality of evidence related to each of the key outcomes.14 At the start of the assessment, a ‘high quality’ of evidence was assigned to all outcomes, because only RCTs have been investigated. Then, the quality was downgraded by one level of evidence for serious study limitations, or by two, for very serious study limitations. These included indirectness of evidence, inconsistency, imprecision of effect estimates or potential publication bias.
We assessed ‘inconsistency of the outcome’ as a large variation in effect; or not overlapping confidence intervals (CIs); or a P value more than 0.05 or I2 more than 50% considering heterogeneity; or when important differences between studies or subgroups remained without explanation.
‘Indirectness of evidence’ was considered when patients, intervention or outcome were different from those of primary interest for the meta-analysis.
‘Imprecision of effect’ occurred in cases of small sample size, number of events and uncertainty about magnitude of effect given by large intervals of confidence.
A potential ‘publication bias’ was recorded when insufficient studies were available for investigating an outcome and the potential bias was detected in the funnel plot.
Meta-analysis of data was performed using RevMan version 5.3 (Foundation for Statistical Computing, Vienna, Austria).
Published data with mean and SD were entered as continuous variables. Continuous variables expressed as median and interquartile range or median and first and third quartiles were converted to mean and SD using Hozo's method.15 Data with a binary outcome were entered as dichotomous variables. For meta-analysis investigating intra-operative and postoperative opioid use, postoperative pain scores, chest tube indwelling time and LOS, continuous outcomes were summarised using the mean difference. For meta-analysis of complications (nausea and vomiting, intra-operative hypotension) data were summarised using the risk ratio. A random effects model was preferred. The confidence interval (CI) was set at 95%.
For assessment of study heterogeneity, the χ2 distribution and I2-statistic were used (considering I2 values as follows: low, <25%; moderate, 25 to 50%; high, >50%).16 If a moderate or high level of heterogeneity was noted, subgroup analysis was performed to explore the cause of the heterogeneity. Considering the strict inclusion criteria adopted, we identified in the review protocol low quality studies and different local anaesthetic concentration as possible causes of heterogeneity. Meta-analysis was performed using RevMan version 5.3 (Foundation for Statistical Computing, Vienna, Austria). Values of P less than 0.05 were considered to be statistically significant in all cases.
A post-hoc trial sequential analysis (TSA) was performed on the primary outcome peri-operative pain at 6, 12 and 24 h on the basis of the least likely anticipated interventions effect. This post-hoc conservative approach allows us to assess if the data are convincing enough to prove that there is such a small effect (Supplemental Digital Content 4 contains reported methods and results of the TSA, https://links.lww.com/EJA/A343).17
Study selection and data retrieval
The results of bibliographic search are summarised in the PRISMA diagram (Fig. 1) indicating the number of articles included and excluded at each stage. Briefly, we identified 102 articles. After removing duplicates, 71 articles were assessed for further examination after inspecting their titles and abstracts. Of these, 34 full-text articles were examined for eligibility, 27 of which were excluded (eight reviews, three retrospective studies, eight case reports, three studies evaluating continuous SAPb, two studies comparing SAPb with epidural anaesthesia, two studies comparing SAPb with paravertebral block and one study comparing SAPb with erector spinae plane block). Seven studies, comprising total of 489 patients, entered the qualitative and quantitative analysis.9,18–23 The inter-reviewer agreement was high (100%) and the third reviewer was never necessary to solve controversies.
We asked all seven of the corresponding authors for missing data: three responded, but only two18,19,22 provided some part of the missing data required.18,22
From the seven RCTs, our meta-analysis included 489 patients undergoing thoracoscopic surgery. Half of the subjects (48.9%) received SAPb and the other half (51.1%) received standard care. Table 1 illustrates the characteristics of the seven studies included. All studies briefly described one, two and/or three port surgical techniques.
Table 1 -
||No of subjects
||Local anaesthetic used
|Chen et al.23
||0.4 ml kg−1 of 0.25% ropivacaine
||T5 to T6
|Lee and Kim21
||20 ml of 0.375% ropivacaine
||General anaesthesia care
|Kim et al.22
||0.4 ml kg−1 of 0.375% ropivacaine
||0.4 ml kg−1 of 0.45% sodium chloride
|Ökmen and Metin Ökmen20
||20 ml of 0.25% bupivacaine
||General anaesthesia care
|Park et al.9
||30 ml of 0.375% ropivacaine
||T5 and T7
||General anaesthesia care
|Semyonov et al.19
||2 mg kg−1 of 0.25% bupivacaine and dexamethasone 8 mg
||T4 to T5
||General anaesthesia care
|Viti et al.18
||30 ml of 0.3% ropivacaine
||General anaesthesia care
Five studies were considered as having ‘high/moderate quality’9,18,21–23; while two had ‘low quality’ (Jadad) (Fig. 2).19,20 Specifically, the study by Ökmen and Metin Ökmen20 received only two points because of inappropriate blinding, whereas the trial by Semyonov et al.19 had zero points related to both blinding and randomisation.
As depicted in Fig. 2, all studies were considered at low risk of bias (green circles) for ‘bias due to deviations from intended interventions’, ‘bias due to missing outcome data’ and ‘bias in the selection of the reported result’. For two studies there were ‘some concerns’ regarding the randomisation process for unclear concealment of the randomisation sequence.19,20 Finally, two studies had ‘some concerns’ in ‘measurement of the outcome’ domain because the assessors were likely unblinded to treatment.9,19
Three studies reported the intra-operative administration of opioids appropriately,21–23 one of which showed a significant reduction in the amount of opioid administered in the SAPb group. However, the overall effect is not statistically significant, with a mean difference of −164.3 μg (95% CI −423.25 to 94.64) (Fig. 3). Due to inconsistency and heterogeneity, we consider the quality of evidence for this outcome low (Table 2).
Table 2 -
Summary of findings
||Illustrative comparative risk
||Assumed risk (standard care group)
||Corresponding risk reduction in the SAPb group
||Relative effect (95% CI)
||No of subjects (no of studies)
||Quality of the evidence (GRADE)
||−423.25 to 94.64
||Range of VAS pain scores
||3.5 to 6.1
||−2.35 to −1.37
||2.8 to 6
||−1.66 to −1.25
||2.7 to 6
||−1.40 to −0.56
||24-h Postoperative opioid consumption range: 7.87 to 52 mg
||−8.41 to −1.22
||RR 0.6 [0.22 to 1.63]
||56 per 182 patients in the control group experienced PONV
||NNT 7 (4.2 to 15.8)
||RR 0.53 [0.36 to 0.79]
||5.19 ± 1.82 Days
||−0.41 to 0.12
CI, confidence interval; LOS, length of hospital stay; NNT, number needed to treat; PONV, post operative nausea ; RR, risk ratio; SAPb, serratus anterior plane block; VAS, visual analogue scaleand vomiting.
As shown in Fig. 4, patients receiving SAPb had significantly lower pain scores at 6, 12 and 24 h after surgery (Fig. 4). The quality of evidence results were considered to be moderate for pain at 6 and 12 postoperatively. At 24 hour, the quality is downgraded to low, due to remarkable inconsistency of results. The post-hoc TSA confirmed the statistical and clinical effectiveness of SAPb in reducing postoperative pain at 6, 12 and 24 h (Supplemental Digital Content 4 contains reported methods and results of the TSA, https://links.lww.com/EJA/A343).
Four studies reported opioid consumption during the postoperative course.19,20,22,23 Rescue doses were requested by patients and either administered by nurses or delivered through patient-controlled analgesia pumps.19,20,22,23
Subjects who receive SAPb block showed a significant reduction in postoperative morphine equivalent consumption (mean difference: −4.81 mg, 95% CI −8.41 to −1.22, P = 0.03) (Fig. 5a). We used a random-effects model because of high heterogeneity. Considering the inconsistency and heterogeneity of this outcome, we define the evidence for this outcome as low quality.
Three studies report arterial hypotension during surgery. Our analysis does not show any significant difference between the groups (risk ratio 0.6, 95% CI 0.22 to 1.63, P = 0.32). The quality of evidence for this outcome variable is moderate (Fig. 5b).18,20,22
Postoperative side-effects and complications
Five studies assessed postoperative nausea and vomiting (Fig. 5c): these side effects were significantly reduced in patients receiving SAPb, as opposed to controls.9,19,20,22,23 The number needed to treat is 7 (95% CI 4.2 to 15.8), that is, one in every seven patients avoids nausea and vomiting because of treatment. The quality of evidence for this outcome variable is high.
Five studies assessed respiratory complications, although the definitions were very heterogeneous among studies.9,18,20,21,23 Accordingly, it is impossible to undertake a meaningful meta-analysis of this variable. Worth mentioning, none of these trials showed significant differences in the rate of respiratory complications between SAPb and controls.
Finally, two studies investigating the incidence of dizziness did not report any relevant difference between groups.9,22
Time to chest tube removal
Four studies evaluated the time to chest tube removal (Supplemental Digital Content 5, https://links.lww.com/EJA/A344).9,18,22,23 The overall effect is not significantly influenced by SAPb (mean difference −0.3 days; 95% CI −0.75 to 0.15), with a moderate quality.9
Length of hospital stay
Four studies evaluated the LOS (Supplemental Digital Content 6, https://links.lww.com/EJA/A345).9,18,22,23 Compared with controls, SAPb does not produce significant effects (mean difference −0.15 days; 95% CI −0.41 to 0.12), with moderate quality of evidence.
Notwithstanding lack of clear asymmetry on visual inspection, a definite interpretation of the funnel plots is not possible due to the paucity of studies (Supplemental Digital Content 7, https://links.lww.com/EJA/A346).
Our meta-analysis suggests that SAPb decreases peri-operative pain in patients undergoing VATS. The effect is more remarkable in the early phase (6th hour) of recovery from surgery rather than later on (24th hour), which is consistent with the single-shot mode of administration and the pharmacokinetic characteristics of the local anaesthetics used. Patients undergoing VATS surgery often suffer from both acute and chronic pain.24 Peri-operative pain control is of paramount importance in reducing patient discomfort, limiting recurrence of pain and potentially reducing postoperative complications and hospital LOS.24,25
SAPb involves the lateral cutaneous branches of the intercostal nerves and, occasionally, the long thoracic nerve and thoracodorsal nerve. The extent and duration of nerve involvement is directly related to the amount of local anaesthetic administered, but this may be not sufficient to control intra-operative pain. In fact, few data have been reported about the potential role of SAPb to modulate different nociceptive triggers generated by deep visceral stimulation during surgery, such as lung manipulation and surgical resection.26 It is therefore not surprising that we found no significant difference in the use of intra-operative opioids between the groups. However, Lee and Kim21 showed an intra-operative opioid consumption twice as high in the control group than in SAPb group. This finding is different compared with Park and Kim,22,23 and a possible explanation could be related to different surgical settings and techniques.
There is a slight, though significant, reduction of the amount of postoperative opioids in the SAPb group, as opposed to the control group. This ‘slight’ difference between groups could be a consequence of the high heterogeneity and inconsistency among studies secondary to various confounding factors. First, the number and site of the ports might influence the dermatomes involved and, accordingly, the area of pain may extend outwith the bounds of the SAPb, and thus create a need for additional opioids. The combined administration of adjuvant analgesic drugs, such as acetaminophen and ibuprofen, or different approaches to drug administration, such as infusion after nurse evaluation or a patient controlled pump. Moreover, the finding that postoperative nausea and vomiting are also significantly reduced by SAPb, as opposed to controls, is likely consequent to the reduced opioids use.
Noteworthy, our results show that SAPb does not increase the risk of hypotensive events. However, more studies focusing on this topic are necessary to investigate the effect of SAPb on haemodynamics.
This meta-analysis does not provide clear results on the rate of respiratory complications occurring after VATS. This secondary outcome would provide indirect proof of a patient's ability to breath without pain and to cough effectively, but the definition of respiratory complications is too heterogeneous among the studies, and for this reason a proper meta-analysis was not feasible. Further investigations based on specific postoperative parameters, such as gas exchange and respiratory frequency are therefore necessary.
The time to chest tube removal and LOS are apparently not affected by SAPb. However, these results may be biased by various confounders, such as different surgical techniques, the number of ports, the duration of the surgical procedure, previous lung function, comorbidities and local protocols.
Our study has limitations deserving discussion. First of all, although all the studies under investigation are RCTs with similar key characteristics (methodology and main outcomes), the study heterogeneity is high for several of the variables under consideration and this clearly limits and weakens the strength of our conclusions. Second, we used a random effects analysis that is more conservative and borderline outcomes may have been overlooked. Third, these results are valid only in the absence of other locoregional techniques such as epidural analgesia. Fourth, the variable intra-operative opioid dose is a possible confunder that may have an effect on the postoperative opioid dose/side effect as well.
Single-shot SAPb in patients undergoing VATS improves patient recovery by decreasing postoperative pain, opioid consumption, and nausea and vomiting during the early postoperative period. Moreover, SAPb does not increase the risk of intra-operative hypotension. Because of missing data, it is impossible to obtain meaningful information about postoperative respiratory complications, time for chest tube removal and hospital LOS.
In the era of multimodal analgesia, SAPb block may be a valid adjunct to conventional general anaesthesia for VATS. However, well designed and properly powered RCTs are necessary to confirm the benefits of SAPb.
Acknowledgements relating to this article
Assistance with the study: we would like to thank Greta Castellini, PhD; Pamela Frigerio, MSc and Silvia Gianola, MSc for their invaluable help.
Financial support and sponsorship: none.
Conflicts of interest: none.
1. Gerner P. Postthoracotomy pain management problems. Anesthesiol Clin
2. Elmore B, Nguyen V, Blank R, et al. Pain management following thoracic surgery. Thorac Surg Clin
3. Blanco R, Parras T, McDonnell JG, et al. Serratus plane block: a novel ultrasound-guided thoracic wall nerve block. Anaesthesia
4. Sedrakyan A, van der Meulen J, Lewsey J, et al. Video assisted thoracic surgery for treatment of pneumothorax and lung resections: systematic review of randomised clinical trials. BMJ
5. Grogan EL, Jones DR. VATS lobectomy is better than open thoracotomy: what is the evidence for short-term outcomes? Thorac Surg Clin
6. Flores RM, Park BJ, Dycoco J, et al. Lobectomy by video-assisted thoracic surgery (VATS) versus thoracotomy for lung cancer. J Thorac Cardiovasc Surg
7. Neustein SM, McCormick PJ. Postoperative analgesia after minimally invasive thoracoscopy: what should we do? Can J Anesth
8. Bayman EO, Parekh KR, Keech J, et al. A prospective study of chronic pain after thoracic surgery. Anesthesiology
9. Park MH, Kim JA, Ahn HJ, et al. Seong BG.A randomised trial of serratus anterior plane block for analgesia after thoracoscopic surgery. Anaesthesia
10. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ
11. McGowan J, Sampson M, Salzwedel DM, et al. PRESS peer review of electronic search strategies: 2015 guideline statement. J Clin Epidemiol
12. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials
13. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ
14. Schünemann HJ, Oxman AD, Vist GE. Higgins JPT, Green S, et al. Interpreting results and drawing conclusions. Cochrane handbook for systematic reviews of interventions. Version 5.0.1
. Chichester, UK: Wiley-Blackwell; 2009; Available from www.cochrane-handbook.org
. [Updated September 2009].
15. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol
16. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ
17. Wetterslev J, Jakobsen JC, Gluud C. Trial sequential analysis in systematic reviews with meta-analysis. BMC Med Res Methodol
18. Viti A, Bertoglio P, Zamperini M, et al. Serratus plane block for video-assisted thoracoscopic surgery major lung resection: a randomized controlled trial. Interact Cardiovasc Thorac Surg
19. Semyonov M, Fedorina E, Grinshpun J, et al. Ultrasound-guided serratus anterior plane block for analgesia after thoracic surgery. J Pain Res
20. Ökmen K, Metin Ökmen B. Evaluation of the effect of serratus anterior plane block for pain treatment after video-assisted thoracoscopic surgery. Anaesth Crit Care Pain Med
21. Lee J, Kim S. The effects of ultrasound-guided serratus plane block, in combination with general anesthesia, on intraoperative opioid consumption, emergence time, and hemodynamic stability during video-assisted thoracoscopic lobectomy: a randomized prospective study. Medicine (Baltimore)
22. Kim DH, Oh YJ, Lee JG, et al. Efficacy of ultrasound-guided serratus plane block on postoperative quality of recovery and analgesia after video-assisted thoracic surgery: a randomized, triple-blind, placebo-controlled study. Anesth Analg
23. Chen G, Li Y, Zhang Y, et al. Effects of serratus anterior plane block for postoperative analgesia after thoracoscopic surgery compared with local anesthetic infiltration: a randomized clinical trial. J Pain Res
24. Kaplowitz J, Papadakos PJ. Acute pain management for video-assisted thoracoscopic surgery: an update. J Cardiothorac Vasc Anesth
25. Katz J, Jackson M, Kavanagh BP, et al. Acute pain after thoracic surgery predicts long term postthoracotomy pain. Clin J Pain
26. Charalampidis C, Youroukou A, Lazaridis G. Pleura space anatomy. J Thorac Dis