Sage Journals: Discover world-class research

Abstract

Study Design

Systematic review

Objectives

It remains unknown whether general anesthesia (GA) or local ± epidural anesthesia (LA) results in superior outcomes with percutaneous endoscopic lumbar discectomy (PELD). The present study sought to examine the impact of anesthesia type on patient-reported outcomes (PROs) and complications with PELD.

Methods

Systematic review and meta-analysis examining PELD performed under GA or LA was conducted. Patient-reported outcomes including Visual Analog Scale (VAS)-leg/back, and Oswestry Disability Index (ODI) scores were collected. Complication, recurrent disc herniation, durotomy, and reoperation rates as well as surgical data were recorded. All outcomes were compared between pooled studies examining GA or LA.

Results

Sixty-eight studies consisting of 5269 patients (724 GA, 4465 LA) were included in the meta-analysis. Overall complication rate was significantly higher in the GA group (9% vs 4%, P = .003). Durotomy rates, length of stay, recurrent disc herniation and reoperation rates were similar between groups. At the first follow-up timepoint, the LA group demonstrated significant improvements in VAS back and ODI scores (P < .05) while the GA group did not (P > .05). At the final follow-up (> 6 months), the percent of patients achieving an excellent McNab score was significantly higher in the GA vs LA group (P < .001).

Conclusions

Percutaneous endoscopic lumbar discectomy with LA may be associated with greater short-term improvement in VAS back pain and ODI scores. General anesthesia may be associated with more durable pain relief but a higher complication rate. Further systematic investigation is necessary to determine what short and long term benefits are associated with PELD performed under LA and GA.

Keywords

local anesthesia general anesthesia percutaneous endoscopic lumbar discectomy patient-reported outcomes complications

Introduction

Symptomatic lumbar disc herniation is one of the most common causes of lower back and leg pain.¹ When aggressive nonoperative management fails to provide symptomatic relief, surgical intervention is necessary. While open laminectomy and discectomy has traditionally shown acceptable outcomes for lumbar disc pathology, surgeons have developed minimally invasive techniques including percutaneous endoscopic lumbar discectomy (PELD).²

The percutaneous endoscopic approach for lumbar discectomy has gained popularity amongst some spine surgeons due to the shorter recovery times, hospital stays, and less muscular trauma encountered by the patient during the approach.^3-5 Percutaneous endoscopic lumbar discectomy can be performed under both general anesthesia (GA) or local ± epidural anesthesia (LA).⁶ Local ± epidural anesthesia is now being recommended by some spine surgeons due to its rapid analgesic onset, shorter operative times, decreased systemic complications, and adequate intraoperative and postoperative pain control.^7,8 Downsides to LA include patient discomfort potentially necessitating shorter operative times, in addition to a diminished ability to maintain strict hemodynamic and anesthetic parameters and risk of insufficient discectomy.^3,9-11 Controversy surrounds the feasibility, efficacy, and safety of LA compared to GA when it comes to PELD.¹² Therefore, the aim of this study was to conduct a systematic review and meta-analysis examining complication rates and patient-reported outcomes (PROs) after PELD performed under LA vs GA.

Methods

This study was conducted with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA)¹³ and was exempt from institutional review board approval.

Search Strategy

The PubMed, EMBASE and SCOPUS databases were used to identify relevant studies. A search was conducted on August 16, 2021, of these 3 databases and included all studies published between 1990 and the date of the search. The search strategy consisted of combinations of key words and commonly used synonyms and abbreviations. Two authors (AS, JM) performed separate literature searches. The advanced search for articles in each database included the terms (((conscious sedation [Title/Abstract]) OR (awake [Title/Abstract]) OR (without GA [Title/Abstract]) OR (local [Title/Abstract])) AND (PELD [Title/Abstract]))) to identify relevant articles for the LA group. Articles for the GA group were identified with the following terms: ((PELD [Title/Abstract])). References cited in the eligible studies were also evaluated to identify any additional studies.

Records were screened by several reviewers working independently including JM and either AS, NE, AM, BM and NL. Reviewers JM, AS, NE, AM, BM and NL independently collected the data from the final list of included studies and this data was reviewed by JM. Any discrepancies were adjudicated by a third reviewer.

Inclusion and Exclusion Criteria

Study inclusion criteria were: (1) first-time single or two-level PELD between L1-S1; (2) patient age ≥ 18; (3) > 10 patients; (4) study published in the English language; (5) patients with at least 6 months of reported follow-up with (6) documented pre and post-operative PROs in one or more of the categories including visual analog scale (VAS) leg, VAS back, Oswestry Disability Index (ODI), Japanese Orthopedic Association (JOA) score, McNab Score or 36-item short form physical component score (SF-36 PCS) (7) inclusion of standard deviation values for included PROs and any other continuous variables.

Percutaneous endoscopic lumbar discectomy was defined as any purely endoscopic approach performed percutaneously and included both transforaminal and interlaminar approaches. Studies involving cases performed with an open approach as well as those involving a tubular retractor or operating microscope were excluded.

For the LA group, studies were required to have a patient cohort with >90% of patients undergoing surgery with local ± LA with or without conscious sedation. For the GA group, studies were required to have a patient cohort with >90% of patients undergoing GA prior to surgery.

Meta-analyses, review articles, case reports, and studies involving pathologies other than disc herniation, studies involving revision surgeries, non-English language, open surgical procedures and lumbar fusion procedures were excluded. Studies involving patients <18 years old were excluded in addition to studies with a lack of adequate follow-up (6 months) or lack of PRO data categories.

Data Collection and Analysis

Studies were grouped as having patient cohorts undergoing either GA or LA. If a study was comparative between GA and LA, the cohorts were analyzed separately in their respective groups for the purpose of meta-analysis. Information on patient demographics and follow-up time in addition to PROs including VAS general, VAS back/leg, JOA score, McNab Score (Percent Excellent), ODI and SF-36 PCS were extracted from the studies. Additional information including mean surgery time, blood loss, complication rate, durotomy rate, length of stay, disc recurrence rate and reoperation rates were extracted from each study when available, pooled and compared between the LA and GA groups. Only studies that reported data with standard deviation values for a given category were included in the final analyses. Additionally, subgroup analysis was performed comparing outcomes with LA and GA for PELD using data from only the prospective and randomized control trials (RCTs).

Outcomes

Primary outcomes were complication rate, durotomy rate, recurrent disc herniation rate, reoperation rate, and length of surgery, as these were the most consistently reported values across all studies and found to be the most objective. Complications included transient paresthesias, nerve root injuries, infection, wound dehiscence, sexual dysfunction, urinary retention, durotomy, discitis and post-operative hematoma. Complications attributed specifically to anesthesia and not related to the surgical intervention were not included.

Secondary outcome data including the PROs VAS general, VAS back/leg, JOA score, McNab Score (Percent Excellent), ODI score and SF-36 PCS score were recorded when available pre-operatively, at the first post-operative follow-up (within 1 month) and at the 3, 6, 12 and 24 months follow-up timepoints. Given the range in follow-up between studies, only the first post-operative follow-up and last follow-up time-points were pooled for comparison. Changes in PROs from pre-op to the first and last post-op follow-up appointment were also calculated and compared between LA and GA groups.

Statistical Analysis

Descriptive statistics were calculated for baseline, surgical, and outcomes data using R (R Foundation for Statistical Computing, Vienna, Austria). For categorical variables, absolute counts were extracted to calculate proportions. For continuous variables, means and associated standard deviations were extracted. After Freeman-Tukey double arcsine transformation for count variables, weighted pooled means and associated 95% confidence intervals (CI) were computed using the DerSimonian-Laird method. A random effects model was used due to the assumptions of clinical diversity and differences in methodology among the included studies. The LA and GA cohorts were compared using a subgroup meta-analysis. Forest plots and the I2 statistic were used to investigate heterogeneity. Heterogenity between different studies were evaluated by χ² and I², and P < .05 was considered statistically significant. I² values of <25%, 25% to 75%, and > 75% represent mild, moderate and severe heterogeneity respectively. All statistical tests were two-tailed, and P < .05 was considered statistically significant.

Risk of Bias of Individual Studies

The Newcastle-Ottowa Scale⁶ was used to assess the quality of the included studies. Several reviewers reviewed each study working independently and discrepancies were adjudicated by a third author.

Results

Search Results

After the initial search of the electronic databases, a total of 806 articles were identified. Articles were further screened by titles, abstracts and full texts according to the specific inclusion and exclusion criteria. Ultimately, 687 articles were excluded and 119 were considered eligible for inclusion in this meta-analysis. The final articles included consisted of 68 articles (12 in the GA group, 56 in the LA group) (Figure 1). These 68 articles included a total of 5189 patients (724 GA, 4465 LA). Descriptive data related to the included articles is presented in Table 1. The risk of bias as assessed with the Newcastle-Ottawa Scale is summarized in Table 2.

Figure 1.

Flow diagram depicting the literature review, search strategy and selection process.

Table 1.

Study Characteristics – Primary Surgical Outcomes and Operative Data.

Study	Study Design	Evidence Level^a	No. Patients	Mean Age (Years)	Mean Follow-Up (Months)	Complication Rate (%)	Durotomy Rate (%)	Disc Herniation Recurrence (%)	Reoperation Rate (%)	Length of Surgery, mins	Blood Loss (cc)
Local ± epidural anesthesia
Ye et al, 2020	RCT	II	30	42.1	12.0	0.0	0.0	0.0	0.0	50.1	NA
Feng et al, 2020	Retrospective	III	50	41.5	6.0	0.0	0.0	0.0	0.0	50.3	NA
Heo et al, 2020	Retrospective	III	28	51.1	22.5	0.0	0.0	3.6	0.0	78.0	NA
Ren et., al. 2020	Retrospective	IV	45	46.2	6.0	8.9	0.0	0.0	0.0	NA	NA
Fang et al, 2016	Retrospective	IV	121	40.5	24.0	12.4	1.7	0.0	0.0	NA	NA
Fang et al, 2016	Retrospective	IV	165	40.0	24.0	9.7	1.2	0.0	0.0	NA	NA
Ahn et al, 2009	Retrospective	IV	45	45.0	38.8	20.0	2.2	2.2	8.9	61.5	NA
Chen et al, 2011	Prospective	III	73	38.0	12.0	13.7	0.0	2.7	0.0	67.0	NA
Amoretti et al, 2021	Prospective	III	77	76.0	6.0	3.9	3.9	3.9	3.9	30.0	NA
Gibson et al, 2017	RCT	II	70	42.0	24.0	5.7	2.9	7.1	7.1	61.0	NA
Yan et al, 2020	Retrospective	III	84	45.3	24.0	14.3	0.0	NA	10.7	73.0	NA
Li et al, 2020	Prospective	III	35	47.1	24.0	0.0	0.0	8.6	8.6	69.5	7.4
Xu et al, 2019	Retrospective	III	182	38.8	29.6	6.6	NA	4.4	NA	NA	NA
Ren et al, 2017	Retrospective	IV	26	41.2	19.3	7.7	0.0	3.8	3.8	NA	NA
Abudurexiti et al, 2018	Prospective	III	82	38.2	6.0	7.3	2.4	NA	NA	64.0	15.0
Shen et al, 2019	Retrospective	IV	241	38.5	23.8	NA	NA	NA	NA	NA	NA
Choi et al, 2013	Prospective	III	89	46.6	24.0	2.2	NA	2.2	9.0	60.0	NA
Choi et al, 2017	Retrospective	III	149	47.2	12.5	0.7	0.0	2.7	1.3	NA	NA
Zhou et al, 2021	Retrospective	IV	19	42.2	18.6	10.5	5.3	0.0	0.0	85.8	NA
Choi et al, 2016	Retrospective	IV	20	33.9	27.5	0.0	0.0	5.0	5.0	67.0	NA
Li et al, 2016	Retrospective	IV	85	56.7	24.0	3.5	0.0	2.4	0.0	55.0	NA
Ahn et al, 2010	Prospective	III	79	36.9	24.0	6.3	NA	0.0	0.0	NA	NA
Lee et al, 2016	Retrospective	IV	80	40.0	13.0	0.0	0.0	2.5	0.0	NA	NA
Liu et al, 2019	Retrospective	IV	127	46.4	12.0	4.7	0.8	0.0	0.0	NA	NA
Ahn et al, 2016	Retrospective	IV	32	22.4	13.7	12.5	0.0	0.0	0.0	48.7	NA
Choi et al, 2011	Prospective	III	52	36.2	24.0	0.0	0.0	5.8	5.8	NA	NA
Wu et al, 2016	Retrospective	IV	22	41.1	18.1	4.5	0.0	4.5	4.5	88.9	NA
Jin et al, 2021	Retrospective	III	78	34.6	13.5	2.6	0.0	1.3	1.3	69.4	NA
Kong et al, 2019	RCT	II	20	34.6	24.0	5.0	0.0	0.0	0.0	59.1	25.5
Ao et al, 2019	Prospective	III	60	42.4	9.0	11.7	0.0	3.3	3.3	113.8	NA
Eun et al, 2016	Retrospective	IV	38	53.7	134.6	NA	NA	NA	NA	NA	NA
Gotecha et al, 2016	Prospective	III	120	40.5	6.0	3.0	0.0	5.0	NA	52.3	NA
Lee et al, 2008	Prospective	III	51	35.6	12.0	13.7	0.0	9.8	3.9	49.2	NA
Zheng-Mei et al, 2017	Prospective	III	50	41.3	12.0	2.0	0.0	0.0	0.0	74.4	NA
Song et al, 2017	Case control	III	30	54.8	18.0	NA	NA	NA	NA	78.7	50.6
Liu et al, 2021	Retrospective	III	60	50.7	20.0	6.7	1.7	5.0	5.0	76.2	18.0
Chen et al, 2021	Retrospective	III	71	41.8	63.0	NA	0.0	7.0	NA	90.0	NA
Jing et al, 2021	Retrospective	III	31	51.3	24.0	12.9	6.5	6.5	NA	49.9	20.9
Yu et al, 2021	Retrospective	III	632	41.2	32.3	10.0	2.1	4.3	3.5	71.2	18.6
Chen et al, 2020	RCT	II	111	40.9	24.0	4.5	0.9	3.6	9.0	97.7	NA
Ren et al, 2020	Prospective	III	54	45.6	36.0	3.7	0.0	16.7	7.4	77.0	NA
Li et al, 2020	Retrospective	III	21	49.8	12.0	4.8	9.5	0.0	0.0	94.5	18.1
Nakamae et al, 2019	Retrospective	IV	21	22.9	28.1	0.0	0.0	4.8	0.0	82.0	NA
Kong et al, 2019	Prospective	III	20	34.6	24.0	5.0	0.0	0.0	0.0	59.1	25.5
Jiang et al, 2018	Retrospective	IV	143	45.4	6.0	NA	NA	NA	NA	94.5	23.6
He et al, 2018	Retrospective	IV	52	59.0	48.0	0.0	0.0	0.0	0.0	116.0	19.0
Zhou et al, 2018	Prospective	III	72	26.3	12.0	0.0	0.0	2.8	2.8	97.5	NA
Shi et al, 2018	Prospective	III	22	40.0	12.0	0.0	0.0	0.0	0.0	78.6	0.0
Kang et al, 2017	Retrospective	IV	37	36.8	12.1	0.0	0.0	5.4	8.1	41.3	NA
Nakamura et al, 2017	Retrospective	IV	50	40.3	15.0	8.0	0.0	0.0	0.0	71.3	0.0
Choi et al, 2013	Retrospective	IV	67	46.6	12.0	4.5	0.0	0.0	4.5	NA	NA
Gadjradj et al, 2016	Prospective	III	166	43.5	12.0	2.4	0.6	7.2	7.2	51.0	NA
Sinkemani et al, 2015	Retrospective	III	36	44.2	12.0	0.0	0.0	0.0	0.0	93.9	NA
Liu et al, 2010	Retrospective	III	104	42.9	79.7	0.0	0.0	0.0	7.7	NA	NA
Liu et al, 2018	Retrospective	III	60	36.2	28.2	0.0	0.0	5.0	5.0	43.0	5.0
Chen et al, 2018	RCT	II	80	40.2	12.0	13.8	1.3	6.3	6.3	97.2	NA
GENERAL ANESTHESIA
Ye et al, 2020	RCT	II	30	43.6	12.0	3.3	0.0	0.0	0.0	78.6	NA
Ren et al, 2020	Retrospective	III	41	48.8	6.0	14.6	0.0	0.0	0.0	68.6	NA
Chen et al, 2011	Prospective	III	50	40.5	12.0	26.0	2.0	4.0	0.0	74.8	NA
Meyer et al, 2020	RCT	II	23	47.2	12.0	30.4	0.0	4.3	13.0	NA	NA
Song et al, 2021	Retrospective	III	267	47.2	36.0	9.4	0.4	6.7	NA	78.5	10.8
Yoon et al, 2012	Prospective	III	25	45.9	20.0	12.0	4.0	4.0	0.0	141.7	35.2
Yang et al, 2021	Retrospective	III	31	51.3	24.0	16.1	0.0	NA	NA	70.0	NA
Lübers et al, 2012	Retrospective	IV	22	53.8	42.0	0.0	NA	NA	NA	44.7	NA
Porto et al, 2021	Retrospective	III	71	55.7	5.7	7.0	NA	36.6	23.9	49.1	NA
Song et al, 2021	Retrospective	III	27	42.9	12.0	7.4	7.4	7.4	7.4	63.9	NA
Sasani et al, 2007	Retrospective	IV	66	52.0	12.0	6.1	0.0	7.6	7.6	NA	NA
Chen et al, 2021	Retrospective	IV	71	41.8	63.0	2.8	0.0	7.0	NA	90.0	NA

RCT – randomized control trial.

^aPer Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria.

Table 2.

Assessment of the Quality of Included Studies According to the Newcastle-Ottawa Quality Assessment Scale.

Study	Representativeness of Cohort	Selection of the Non-exposed Cohort	Ascertainment of Exposure	Outcome of Interest	Comparability of Cohorts	Assessment of Outcome	Adequate Duration of Follow-Up	Adequate Follow-up of Cohort
Local ± epidural anesthesia
Ye et al, 2020	1	1	1	1	1	1	1	1
Feng et al, 2020	1	—	1	1	—	1	1	1
Heo et al, 2020	1	—	1	1	—	1	1	1
Ren et., al. 2020	1	1	1	1	1	1	1	1
Fang et al, 2016	1	1	1	1	1	1	1	1
Fang et al, 2016	1	1	1	1	1	1	1	1
Ahn et al, 2009	1	—	1	1	—	1	1	1
Chen et al, 2011	1	1	1	1	1	1	1	1
Amoretti et al, 2021	1	—	1	1	—	1	1	1
Gibson et al, 2017	1	1	1	1	1	1	1	1
Yan et al, 2020	1	—	1	1	—	1	1	1
Li et al, 2020	1	1	1	1	1	1	1	1
Xu et al, 2019	1	—	1	1	—	1	1	1
Ren et al, 2017	1	—	1	1	—	1	1	1
Abudurexiti et al, 2018	1	—	1	1	—	1	1	1
Shen et al, 2019	1	—	1	1	—	1	1	1
Choi et al, 2013	1	—	1	1	—	1	1	1
Choi et al, 2017	1	—	1	1	—	1	1	1
Zhou et al, 2021	1	—	1	1	—	1	1	1
Choi et al, 2016	1	—	1	1	—	1	1	1
Li et al, 2016	1	—	1	1	—	1	1	1
Ahn et al, 2010	1	—	1	1	—	1	1	1
Lee et al, 2016	1	—	1	1	—	1	1	1
Liu et al, 2019	1	—	1	1	—	1	1	1
Ahn et al, 2016	1	1	1	1	2	1	1	1
Choi et al, 2011	1	—	1	1	—	1	1	1
Wu et al, 2016	1	—	1	1	—	1	1	1
Jin et al, 2021	1	1	1	1	1	1	1	1
Kong et al, 2019	1	1	1	1	2	1	1	1
Ao et al, 2019	1	1	1	1	1	1	1	1
Eun et al, 2016	1	—	1	1	—	1	1	1
Gotecha et al, 2016	1	—	1	1	—	1	1	1
Lee et al, 2008	1	1	1	1	1	1	1	1
Zheng-Mei et al, 2017	1	1	1	1	1	1	1	1
Song et al, 2017	1	—	1	1	-—	1	1	1
Liu et al, 2021	1	1	1	1	1	1	1	1
Chen et al, 2021	1	1	1	1	1	1	1	1
Jing et al, 2021	1	1	1	1	1	1	1	1
Yu et al, 2021	1	1	1	1	1	1	1	1
Chen et al, 2020	1	1	1	1	2	1	1	1
Ren et al, 2020	1	1	1	1	1	1	1	1
Li et al, 2020	1	1	1	1	1	1	1	1
Nakamae et al, 2019	0	—	1	1	—	1	1	1
Kong et al, 2019	1	1	1	1	2	1	1	1
Jiang et al, 2018	1	—	1	1	—	1	1	1
He et al, 2018	1	1	1	1	1	1	1	1
Zhou et al, 2018	1	—	1	1	—	1	1	1
Shi et al, 2018	1	—	1	1	—	1	1	1
Kang et al, 2017	1	—	1	1	—	1	1	1
Nakamura et al, 2017	1	—	1	1	—	1	1	1
Choi et al, 2013	1	—	1	1	—	1	1	1
Gadjradj et al, 2016	1	—	1	1	—	1	1	1
Sinkemani et al, 2015	1	1	1	1	1	1	1	1
Liu et al, 2010	1	1	1	1	1	1	1	1
Liu et al, 2018	1	1	1	1	1	1	1	1
Chen et al, 2018	1	1	1	1	1	1	1	1
General anesthesia
Ye et al, 2020	1	1	1	1	1	1	1	1
Ren et al, 2020	1	1	1	1	1	1	1	1
Chen et al, 2011	1	1	1	1	1	1	1	1
Meyer et al, 2020	1	1	1	1	2	1	1	1
Song et al, 2021	1	1	1	1	0	1	1	1
Yoon et al, 2012	1	1	1	1	0	1	1	1
Yang et al, 2021	1	1	1	1	1	1	1	1
Lübers et al, 2012	1	—	1	1	1	1	1	—
Porto et al, 2021	1	1	1	1	0	1	1	1
Song et al, 2021	1	1	1	1	1	1	1	1
Sasani et al, 2007	1	—	1	1	—	1	1	1
Chen et al, 2021	1	—	1	1	—	1	1	1

Baseline Data

Patients undergoing GA were slightly older than those undergoing LA (47.2 vs 42.8 years, P = .03). Only 7 studies reported baseline JOA scores (1 GA – 8%, 6 LA – 11%); however, patients in the GA group had slightly higher JOA scores than the LA group (16 vs 13, P = .002). The GA and LA groups had similar baseline VAS leg, VAS back and ODI scores (Table 3).

Table 3.

Baseline Age and Patient-Reported Outcomes.

Parameter	Local ± Epidural			General
Parameter	# Studies Pooled	Pooled Value	SD	# Studies Pooled	Pooled Value	SD	P-Value
Age	43	42.8	8.9	10	47.2	6	.03
VAS back	30	5.5	1.5	7	4.8	2.5	.4
VAS leg	34	7.1	1	7	6.8	1.2	.5
JOA	6	13	2.4	1	16	.6	.002
ODI	42	53.4	15.7	5	55.5	23.8	.8

Overall Complications

Sixty-three studies reported data on complications (12 GA – 100%, 51 LA – 91%). The complication incidence was .09 (95% confidence interval [CI] .06 – .15) in the GA group and .04 (95% CI .03 – .06) in the LA group (P = .003). These results are presented in Figure 2 as forest plots.

Figure 2.

Forest plots of complication incidence in PELD with GA and LA.

Durotomy

Fifty-nine studies reported data on durotomy incidence (10 GA – 83%, 49 LA – 88%). Durotomy incidence in the GA group was .007 (95% CI .001 – .03) and .006 (95% CI .003 – .01) in the LA group (P = .86). Results by study are presented in Figure 3 as forest plots.

Figure 3.

Forest plots of durotomy incidence in PELD with GA and LA.

Recurrent Disc Herniation

Sixty studies reported data on recurrent disc herniation incidence (10 GA – 83%, 50 LA – 89%). Recurrent disc herniation incidence in the GA group was .06 (95% CI .02 – .13) and .02 (95% CI .02 – .04) in the LA group (P = .06). Results by study are presented in Figure 4 as forest plots.

Figure 4.

Forest plots of recurrent disc herniation incidence in PELD with GA and LA.

Reoperations

Fifty-five studies reported data on reoperation incidence (8 GA – 67%, 47 LA – 84%). Reoperation incidence in the GA group was .03 (95% CI .004 – .18) and .02 (95% CI .01 – .03) in the LA group (P = .74). Results by study are presented in Figure 5 as forest plots.

Figure 5.

Forest plots of reoperation incidence in PELD with GA and LA.

Operative Data

Thirty-eight studies reported data on length of surgery (7 GA – 58%, 31 LA – 55%). Length of surgery in the GA group was 65 minutes (95% CI 53 - 78) and 75 minutes (95% CI 67 - 82) in the LA group (P = .14). Results by study are presented in Figure 6 as forest plots. Eleven studies (1 GA – .5%, 10 LA – 18%) reported data on blood loss. Blood loss in the GA group was 11cc (95% CI 10 - 12) and 19cc (95% CI 11 - 28) in the LA group (P = .03).

Figure 6.

Forest plots of length of surgery in PELD with GA and LA.

Patient Reported Outcomes

The greatest number of studies presented data on VAS back (37 studies: 7 GA – 58%, 30 LA – 54%) VAS leg (41 studies: 7 GA – 58%, 34 LA – 61%) and ODI scores (47 studies: 5 GA – 42%, 42 LA – 81%). Pre-operative VAS back, leg, and ODI scores were similar between the GA and LA groups. Seven studies (1 GA – 8%, 6 LA – 11%) reported JOA scores which overall were higher in the GA group pre-operatively (16 vs 13, P = .002) (Table 3).

At the first and last post-operative follow-ups (within 1 month and >6 months), the LA group had significant improvements in VAS back (5.5 to 2.3 to 1.5, P < .001), VAS leg (7.1 to 2.1 to 1.4, P < .001) and ODI (53.4 to 21.0 to 12.4, P < .001) scores. The GA group only experienced significant improvements in VAS leg score (6.8 to 2.6, P < .001) at the first follow-up but demonstrated significant improvements in VAS leg (6.8 to 1.3, P < .001), VAS back (4.8 to 1.7, P = .01) and ODI scores (55.5 to 18.1, P = .04) at the final follow-up (Table 4).

Table 4.

Changes in Patient-Reported Outcomes – Preoperative, Postoperative and Last Follow-Up.

PRO	Pre-Operative			First Postoperative				Last Follow-Up
PRO	# Studies Pooled	Pooled Value	SD	# Studies Pooled	Pooled Value	SD	P-Value*	# Studies Pooled	Pooled Value	SD	P-Value**
Local ± epidural
VAS back	30	5.5	1.5	13	2.3	1.0	<.001	29	1.5	.7	<.001
VAS leg	34	7.1	1	15	2.1	.8	<.001	33	1.4	.6	<.001
ODI	42	53.4	15.7	17	21	9.0	<.001	40	12.4	6.8	<.001
General
VAS back	7	4.8	2.5	4	2.5	1.4	.13	6	1.7	1.2	.01
VAS leg	7	6.8	1.2	4	2.6	1.2	< .001	6	1.3	.8	<.001
ODI	5	55.5	23.8	1	36.5	.7	.33	3	18.1	10.3	.04

Twenty-three studies reported McNab scores at follow-up > 6 months (19 LA – 34%, 4 GA – 33%). At the final post-operative follow-up, the percent of patients achieving an excellent McNab score was significantly higher in the GA vs LA group (62% vs 20%, P < .001).

Subgroup Analysis (Prospective and RCT Studies)

In the LA group, outcomes from 21 included studies (5 RCTs, 16 prospective) were compared to those using 4 studies from the GA group (2 RCTs, 2 prospective). There were no differences in baseline age, VAS back/leg or ODI scores between the GA and LA groups using these studies (Table 5). Overall complication incidence remained significantly higher in the GA group when compared to the LA group (.16 vs .04; P = .003; Figure 7). Durotomy incidence (.02 vs .006; P = .31; Figure 8), recurrent disc herniation incidence (.03 vs .04; P = .64; Figure 9), reoperation incidence (.003 vs .04; P = .37; Figure 10), length of surgery (76.1 vs 75.8 minutes; P = .97; Figure 11) and length of stay (4.2 vs 4.9 days; P = .59; Figure 12) were not significantly different between GA and LA groups respectively. There was not enough available data utilizing only the RCT and prospective studies to examine PROs in the GA and LA groups.

Table 5.

Baseline Age and Patient-Reported Outcomes (RCT and Prospective Studies).

	Local ± Epidural			General
Parameter	# Studies Pooled	Pooled Value	SD	# Studies Pooled	Pooled Value	SD	P-Value
Age	15	42.3	11.6	3	43.6	6.7	.68
VAS back	12	5.6	1.3	2	5.9	2.1	.43
VAS leg	12	6.8	1.1	2	7.8	4.6	.07
ODI	13	54	12.8	2	39	93	.16

Figure 7.

Forest plots of complication incidence in PELD with GA and LA (Prospective and RCT studies).

Figure 8.

Forest plots of durotomy incidence in PELD with GA and LA (Prospective and RCT studies).

Figure 9.

Forest plots of recurrent disc herniation incidence in PELD with GA and LA (Prospective and RCT studies).

Figure 10.

Forest plots of reoperation incidence in PELD with GA and LA (Prospective and RCT studies).

Figure 11.

Forest plots of length of surgery in PELD with GA and LA (Prospective and RCT studies).

Figure 12.

Forest plots of length of surgery in PELD with GA and LA (Prospective and RCT studies).

Discussion

Percutaneous endoscopic lumbar discectomy is commonly performed under LA which has been reported to reduce the risk of nerve injury.^14,15 Pain experienced by the patient during PELD under LA; however, can pose various challenges to successfully completing the operation. Percutaneous endoscopic lumbar discectomy under GA has thus been reported as an alternative to keep patient’s more comfortable; however, may pose an increased risk of neurologic injury as patients are unable to provide feedback during the surgery.^16,17

The present study demonstrated that the use of GA was associated with a higher percentage of patients achieving an excellent McNab score at final follow-up. Both GA and LA groups demonstrated significant improvement in VAS and ODI scores at the final follow-up. At the first post-operative follow-up, however, the GA group only demonstrated improvement in VAS leg score while the LA group demonstrated improvements in all PROs. Prior study has demonstrated that PELD under GA results in greater post-operative dysesthesias, possibly due to greater nerve root manipulation with lack of patient feedback compared to LA.¹⁸ This could partially explain the non-significant improvement in ODI scores seen at the first follow-up timepoint in the GA group. Lower post-operative cognitive function resulting from GA use as noted by Ye et al may also contribute to ODI scores.¹⁹ The higher percent of patients achieving an Excellent McNab score in the GA group may indicate that PELD performed under GA confers a more sustained benefit for the treatment of radiculopathy through providing the surgeon with increased ability to adequately mobilize nerve roots for inspection of residual disc material. The ability to achieve optimal decompression may be limited due to patient discomfort with the use of LA.

Overall complication rate of PELD under GA was significantly higher than PELD under LA in our analysis. This remained true when examining higher quality prospective and RCTs in subgroup analysis. Prior literature has demonstrated conflicting evidence regarding the effects of GA and LA on post-operative complications including nerve injury and dysesthesias. Exiting nerve root injury for PELD under LA are variably reported between 1-6.7%.^20-24 However, one study by Choi et al in 2013 noted a slightly higher rate of 8.9% in a retrospective study of 233 patients.²⁵ These findings highlight the value of patient cooperation/feedback during PELD with LA.

There is a paucity of literature assessing the use of GA with PELD. The first report comparing PELD under GA vs LA was published in 2011 by Chen et al nothing that both techniques achieve good outcomes and high patient satisfaction.¹⁸ However, the authors concluded that the use of LA was more favorable given the ability for patient feedback, less post-operative dysesthesia and shorter hospital stay. Ye et al in a randomized controlled trial comparing the effects of GA and LA in lumbar interlaminar endoscopic surgery also found shorter hospital stays and operative times with LA and no differences in ODI or VAS scores between groups.²⁶ These authors concluded that GA makes a positive intraoperative surgical experience more likely. Supporting our finding of a higher complication rate with PELD under GA, Ren et al demonstrated a complication rate of 14.6% for PELD patients undergoing GA vs 8.9% for those undergoing EA.¹² Similarly, Chen et al reported a 24% rate of post-operative transient paresthesias in patients undergoing PELD under GA compared to 14% of those with LA, while Ye et al report no differences in neurologic complications between GA and LA groups.^27,28

Overall, the present analysis is consistent with several other reports in that both anesthetic methods achieve good patient outcomes long term. The risk of short-term complications resulting from the use of GA must be balanced against the risk of patient discomfort with the use of LA. The decision to pursue either anesthetic technique should be made on a case by case basis taking into account patient preferences.

Limitations

The limitations of this study are primarily related to the limited body of literature directly examining the topic. Cohorts of patients undergoing LA and GA for PELD were combined in attempt to overcome this, resulting in 68 total studies meeting the inclusion criteria. Additionally, while transforaminal and interlaminar PELD have distinct differences based on their approaches, these were combined due to the failure of many studies to distinguish between the approaches in their cohorts. Next, outcomes assessed were not present uniformly across all studies leading to underpowered metrics in several categories. While several RCTs were included, the majority of included studies were retrospective or case-control designs of poor quality. Subgroup analysis using only the higher quality prospective and RCT studies was performed to attempt to overcome this limitation. Significant heterogeneity within and between groups limits the strength of our conclusions. Additionally, studies of varying design were combined to generate the greatest amount of data for analysis, potentially limiting the strength of conclusions. Included studies constituted a wide-range of international locations, resulting in significant variation in patient assessment and outcome measurements. Primary outcomes were thus chosen based on those reported in the greatest number of studies and felt to be the most objective. The present study design is also limited by the inability to control for biases inherent to the use of GA vs LA for PELD. It is possible GA patients may have had more complex conditions warranting the use of GA that may have affected their outcomes. Despite these limitations, this is the first systematic review and meta-analysis on the topic of anesthetic use in PELD and generates several novel hypotheses that are worthy of further investigation.

Conclusions

The use of GA may carry a slightly higher rate of short-term (resolving within 1 month) complications including nerve root injuries and post-operative dysesthesias but may result in more sustained benefits with a greater percentage of patients achieving an Excellent McNab score at final follow-up. Both groups demonstrated significant improvements in all PROs at the final follow-up; however, LA may be associated with greater short-term improvement in disability scores. These conclusions are limited by a high level of study bias and heterogeneity and further systematic investigation is necessary.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

James Mooney

References

Fjeld

Grøvle

Helgeland

, et al. Complications, reoperations, readmissions, and length of hospital stay in 34 639 surgical cases of lumbar disc herniation. Bone Joint J . 2019;101-b(4):470-477. doi:10.1302/0301-620x.101b4.Bjj-2018-1184.R1.

Alvi

Kerezoudis

Wahood

Goyal

Bydon

. Operative approaches for lumbar disc herniation: A systematic review and multiple treatment meta-analysis of conventional and minimally invasive surgeries. World Neurosurg. 2018;114:391-407.e2. doi:10.1016/j.wneu.2018.02.156

Pan

, et al. Percutaneous endoscopic lumbar discectomy: Indications and complications. Pain Physician. 2020;23(1):49-56.

Han

, et al. Percutaneous endoscopic lumbar discectomy for lumbar disc herniation. J Clin Neurosci . 2016;33:19-27. doi:10.1016/j.jocn.2016.01.043

Ren

Qin

Sun

Wang

. Transforaminal endoscopic lumbar discectomy for lumbar disc herniation causing bilateral symptoms. World Neurosurg. 2017;106:413-421. doi:10.1016/j.wneu.2017.06.191

Oksar

. Sedation for percutaneous endoscopic lumbar discectomy. ScientificWorldJournal. 2016;2016:8767410. doi:10.1155/2016/8767410

Zhao

Feng

Chen

Wang

. Stepwise local anesthesia for percutaneous endoscopic interlaminar discectomy: Technique strategy and clinical outcomes. World Neurosurg . 2020;134:e346-e352. doi:10.1016/j.wneu.2019.10.061

Sanusi

Davis

Nicassio

Malik

. Endoscopic lumbar discectomy under local anesthesia may be an alternative to microdiscectomy: A single centre's experience using the far lateral approach. Clin Neurol Neurosurg. 2015;139:324-327. doi:10.1016/j.clineuro.2015.11.001

Sairyo

Egawa

Matsuura

, et al. State of the art: Transforaminal approach for percutaneous endoscopic lumbar discectomy under local anesthesia. J Med Invest. 2014;61(3-4):217-225. doi:10.2152/jmi.61.217

10.

Tacconi

Baldo

Merci

Serra

. Transforaminal percutaneous endoscopic lumbar discectomy: Outcome and complications in 270 cases. J Neurosurg Sci . 2020;64(6):531-536. doi:10.23736/s0390-5616.18.04395-3

11.

Kanno

Aizawa

Hahimoto

Itoi

. Minimally invasive discectomy for lumbar disc herniation: Current concepts, surgical techniques, and outcomes. Int Orthop . 2019;43(4):917-922. doi:10.1007/s00264-018-4256-5

12.

Ren

, et al. Comparison of the safety and effectiveness of percutaneous endoscopic lumbar discectomy for treating lumbar disc herniation under epidural anesthesia and general anesthesia. Neurospine . 2020;17(1):254-259. doi:10.14245/ns.1938366.183

13.

Moher

Liberati

Tetzlaff

Altman

. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097. doi:10.1371/journal.pmed.1000097

14.

Sairyo

Egawa

Matsuura

, et al. State of the art: Transforaminal approach for percutaneous endoscopic lumbar discectomy under local anesthesia. J Med Invest. 2014;61(34):217-225. doi:10.2152/jmi.61.217

15.

Kim

. Use of lidocaine patch for percutaneous endoscopic lumbar discectomy. Korean J Pain. 2011;24(2):74-80. doi:10.3344/kjp.2011.24.2.74

16.

Fang

Ding

Song

. Comparison of the effects of epidural anesthesia and local anesthesia in lumbar transforaminal endoscopic surgery. Pain Physician. 2016;19(7):E1001-E1004.

17.

Zhu

Zhao

Fan

, et al. Comparison of 3 anesthetic methods for percutaneous transforaminal endoscopic discectomy: A prospective study. Pain Physician. 2018;21(4):347-353.

18.

Chen

H-T

Tsai

C-H

Chao

S-C

, et al. Endoscopic discectomy of L5-S1 disc herniation via an interlaminar approach: Prospective controlled study under local and general anesthesia. Surg Neurol Int. 2011;2:93-93. doi:10.4103/2152-7806.82570

19.

H RQ

Zhu

, et al. Therapeutic effec of percutaneous transforaminal nucleus pulposus extraction on lumbar disc hernation under different anesthesia. Chin J Endosc. 2019;25:68-72.

20.

Tsou

Yeung

. Transforaminal endoscopic decompression for radiculopathy secondary to intracanal noncontained lumbar disc herniations: Outcome and technique. Spine J . 2002;2(1):41-48. doi:10.1016/s1529-9430(01)00153-x

21.

Ahn

Lee

Kim

Liu

. Transforaminal percutaneous endoscopic lumbar discectomy for upper lumbar disc herniation: Clinical outcome, prognostic factors, and technical consideration. Acta Neurochir (Wien). 2009;151(3):199-206. doi:10.1007/s00701-009-0204-x

22.

Ruetten

Komp

Merk

Godolias

. Full-endoscopic interlaminar and transforaminal lumbar discectomy versus conventional microsurgical technique: A prospective, randomized, controlled study. Spine. 2008;33(9):931-939. doi:10.1097/BRS.0b013e31816c8af7

23.

Yeung

Tsou

. Posterolateral endoscopic excision for lumbar disc herniation: Surgical technique, outcome, and complications in 307 consecutive cases. Spine. 2002;27(7):722-731.

24.

Lee

Kang

Ahn

, et al. Operative failure of percutaneous endoscopic lumbar discectomy: A radiologic analysis of 55 cases. Spine (Phila Pa 1976). 2006;31(10):E285-E290.

25.

Choi

Ahn

J-O

W-S

Lee

S-J

Choi

I-J

Kim

. Exiting root injury in transforaminal endoscopic discectomy: Preoperative image considerations for safety. Eur Spine J. 2013;22(11):2481-2487. doi:10.1007/s00586-013-2849-7

26.

X-F

Wang

A-M

, et al. Comparison of the effects of general and local anesthesia in lumbar interlaminar endoscopic surgery. Ann Palliat Med. 2020;9(3):1103-1108. doi:10.21037/apm-20-623

27.

Chen

Tsai

Chao

28.

Wang

, et al. Comparison of the effects of general and local anesthesia in lumbar interlaminar endoscopic surgery. Ann Palliat Med . 2020;9(3):1103-1108. doi:10.21037/apm-20-623

General vs Local Anesthesia for Percutaneous Endoscopic Lumbar Discectomy (PELD): A Systematic Review and Meta-Analysis

Abstract

Study Design

Objectives

Methods

Results

Conclusions

Keywords

Introduction

Methods

Search Strategy

Inclusion and Exclusion Criteria

Data Collection and Analysis

Outcomes

Statistical Analysis

Risk of Bias of Individual Studies

Results

Search Results

Baseline Data

Overall Complications

Durotomy

Recurrent Disc Herniation

Reoperations

Operative Data

Patient Reported Outcomes

Subgroup Analysis (Prospective and RCT Studies)

Discussion

Limitations

Conclusions

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

References