Introduction
Obesity has emerged as a critical global health crisis that severely affects surgical outcomes and poses extraordinary challenges for patients and healthcare providers [1,2]. As obesity rates continue to increase worldwide, they are increasingly associated with complications, such as dural tears, wound dehiscence, and surgical site infections [3,4]. These complications increase healthcare costs and contribute to higher patient morbidity and prolonged recovery time [5].
Traditional open surgical techniques, which are characterized by substantial tissue damage and prolonged recovery period, can exacerbate the abovementioned problems, particularly in obese patients [6]. This makes the management of lumbar spinal pathologies challenging. To address these issues, minimally invasive surgical techniques have become useful alternatives that effectively reduce surgical morbidity while maintaining high level of effectiveness [7]. Among these techniques, endoscopic spine surgery has emerged as one of the most innovative approaches, offering the potential for smaller incisions, less muscle dissection, and faster recovery time [8]. By minimizing soft tissue damage, endoscopic procedures can enhance recovery and reduce the risk of postoperative complications, which is especially important in obese patients [8,9]. Despite the growing interest in endoscopic techniques, there remains a notable gap in literature regarding their specific application and outcomes in obese patient populations. To date, no systematic review or meta-analysis has focused exclusively on the safety and efficacy of endoscopic spine surgery in obese patients and has compared outcomes between obese and nonobese patients.
The impact of obesity on surgical outcomes remains controversial. Some studies suggested that obesity significantly increases the risk of complications and worsens outcomes. Kalanithi et al. [10] in 2012 found that morbid obesity was associated with relatively high rates of surgical site infections, dural tears, and prolonged hospital stay among patients undergoing spinal fusion surgery (spine). Conversely, other studies have reported that obesity does not significantly affect clinical results. In the Spine Patient Outcomes Research Trial, Rihn et al. [11] in 2012 found that compared with nonobese patients, obese patients had higher complication rates but similar overall surgical outcomes, such as pain relief and functional improvement. This discrepancy underscores the need for further research on the impact of obesity on the outcomes of surgery, especially minimally invasive techniques, such as endoscopic spine surgery [12]. This review and meta-analysis aimed to evaluate and compare the safety and efficacy of endoscopic spine surgery for low back pain (LBP) between obese and nonobese patients.
Methods
Eligibility criteria
This review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines [13]. The PICOS search strategy was used as follows: P, patients undergoing endoscopic spinal surgery; I, the intervention group comprised patients with obesity; C, the comparator group comprised nonobese patients; O, the primary outcomes assessed were efficacy and safety; and S: the study types included cohort comparative studies.
The inclusion and exclusion criteria were established to select pertinent studies. There were no restrictions based on sex, country, or ethnicity, and only comparative studies were included. Nonhuman (animal) and in vitro studies were excluded, as were certain publication types, such as book chapters, editorials, author responses, conference papers, posters, reviews, letters, patents, and papers available only as abstracts. Studies that lacked extractable data, those with overlapping or unreliable data, and those with duplicates were excluded.
Information sources and search methods for identification of studies
A comprehensive search was performed across multiple databases, including PubMed, Scopus, and the Virtual Health Library. No filters based on publication date or language were applied. The specific search terms used were “obesity,” “body mass index,” “endoscopic spine surgery,” “endoscopic discectomy,” “minimally invasive spinal surgery,” “minimally invasive spine surgery,” and “lumbar.” Two authors selected the studies, and if they disagreed, a third author joined the discussion. A manual search of references from the included studies was also conducted.
Data extraction
Two researchers independently evaluated the data collected from the studies. In situations in which consensus could not be reached, a third researcher stepped in to make the final decision. The characteristics extracted from the studies included study type, geographical region, sex, age, and patient count. The primary outcomes were the duration of operation in minutes; Visual Analog Scale (VAS) scores for back and leg pain and assessments of Oswestry Low Back Disability Index (ODI) at 3, 6, and 12 months after the operation; and incidence of reherniation.
Assessment of the risk of bias
The quality of the studies included in this review was independently assessed by two authors using the Methodological Index for Non-randomized Studies (MINORS) criteria [14]. For comparative studies, the maximum possible score was 24, and quality was assessed as very low (score of 0–6), low (score of 7–10), fair (score of 11–15), or high (score of ≥16).
Statistical analysis
Continuous variables measured on the same scale were examined for mean difference (MD) and 95% confidence interval (CI). For outcomes that were evaluated using different scales or the same scale with differing units, standard mean difference and 95% CI were calculated, as well as odds ratio (OR) for binary variables. The meta-analysis was performed using Review Manager ver. 5.4 (RevMan; Cochrane, London, UK). Heterogeneity was assessed using chi-square test and I2 statistics, and I2 values of 25%, 50%, and 75% indicated low, moderate, and high levels of heterogeneity, respectively. A fixed effects model was applied, unless there was statistical evidence of heterogeneity. Missing data were addressed according to the Cochrane Handbook guidelines.
Additional analysis
The Review manager software was used to evaluate publication bias based on the widely used visual method of funnel plots. Sensitivity analysis was conducted by excluding the study with the highest weight after comparison of all outcomes. Grading of Recommendations Assessment, Development, and Evaluation (GRADE) was performed using the GRADEpro tool [15]. This method evaluated the quality of evidence and strength of recommendations by considering bias risk, heterogeneity, result precision, and consistency among studies.
Results
Study selection
The first electronic search of the three databases identified 1,553 studies. After eliminating duplicates by title and abstract screening, 20 studies remained for full-text assessment. Of these, only seven studies fulfilled our inclusion criteria [16–22]. No further studies were found during manual search (Fig. 1).
Study characteristics
A total of seven studies involving 659 patients were included, and their baseline characteristics are summarized in Table 2. The average age ranged from 15.7 to 65.2 years, the percentage of women was between 16.6% and 52.2%, and the average body mass index (BMI) ranged from 20.2 to 37.2 kg/m2. Only one study reported the American Society of Anesthesiologists score [4], which had an average score of 2.3 for the obese group and 1.8 for the nonobese group.
Operative time
The analysis of operative time included three studies with 339 patients (146 in the obese group and 193 in the nonobese group). Using a random-effects model, the pooled MD was 9.86 (95% CI, −4.93 to 24.65), suggesting no significant difference in operative time between the two groups (Z=1.31, p=0.19). However, a high degree of heterogeneity was noted (I2=94%, p<0.00001) (Fig. 2A). Sensitivity analysis revealed significant differences between the groups (MD, 17.31; 95% CI, 11.83 to 22.79; p<0.00001; participants=165; studies=2; I2=0%). To evaluate the impact of BMI and age on operative time, a meta-regression analysis was performed. The regression coefficients were 0.018 (standard error [SE], 0.012; 95% CI, −0.005 to 0.04; p=0.13) for BMI and 0.006 (SE, 0.01; 95% CI, −0.01 to 0.02; p=0.56) for age, indicating that high BMI or old age were not significantly associated with prolonged operative time.
Length of hospital stay
The analysis of the length of hospital stay in days included three studies with 210 patients (69 in the obese group and 141 in the nonobese group). Using random-effects model, the pooled MD was −0.32 (95% CI, −1.04 to 0.39), suggesting no significant difference in length of hospital stay between the two groups (Z=0.89, p=0.38). However, a high degree of heterogeneity was noted (I2=74%, p=0.02) (Fig. 2B). Sensitivity analysis revealed significant differences between the groups (MD, −0.66; 95% CI, −1.23 to −0.09; p=0.02; participants=163; studies=2; I2=0%). Subgroup analysis according to obesity severity (lower obesity: BMI <35 kg/m2 vs. higher obesity: BMI ≥35 kg/m2) revealed that heterogeneity in the lower obesity group decreased to 0% and significantly decreased the length of hospital stay in the obese group (p=0.02). This result suggested that obesity severity may influence the length of hospital stay (Supplement 1).
Visual Analog Scale score for back pain
To assess the VAS score at 3 months, five studies including 437 patients (152 in the obese group and 285 in the nonobese group) were analyzed. Using a fixed effects model, the pooled MD was 0.26 (95% CI, −0.11 to 0.63), indicating no significant difference in VAS scores between the two groups (Z=1.38, p=0.17). Heterogeneity was low (I2=32%, p=0.21) (Fig. 3A). Sensitivity analysis yielded similar results (MD, 0.23; 95% CI, −0.29 to 0.75; p=0.39; participants=329; studies=4; I2=48%). Subgroup analysis according to obesity severity revealed that heterogeneity in the higher obesity group decreased to 0% and significantly decreased the VAS at 3 months in the nonobese group (p=0.04). This result suggested that obesity severity may influence VAS outcomes at 3 months (Supplement 2).
For the VAS for back pain at 6 and 12 months, three studies including 211 patients (71 in the obese group and 140 in the nonobese group) were analyzed. Using a fixed effects model, the pooled MD for VAS Back at 6 months was 0.26 (95% CI, −0.05 to 0.56), which did not show a significant difference between the groups (Z=1.63, p=0.10) and indicated no heterogeneity (I2=0%, p=0.87) (Fig. 3B). Sensitivity analysis confirmed these findings (MD, 0.11; 95% CI, −0.54 to 0.75; p=0.74; participants=164; studies=2; I2=0%). The pooled MD for VAS Back at 12 months was −0.54 (95% CI, −1.70 to 0.62), indicating no significant difference (Z=0.91, p=0.36) with high heterogeneity (I2=77%, p=0.01) (Fig. 3C). Sensitivity analysis showed consistent results (MD, −1.16; 95% CI, −3.60 to 1.28; p=0.35; participants=164; studies=2; I2=84%).
Visual Analog Scale score for leg pain
Five studies that included 437 patients (285 in the nonobese group and 152 in the obese group) were analyzed for leg VAS at 3 months. According to the fixed effects model, the pooled MD was 0.17 (95% CI, −0.06 to 0.41), indicating no significant difference in the VAS Leg scores at 3 months between the two groups (Z=1.47, p=0.14). No heterogeneity was observed (I2=0%, p=0.83) (Fig. 4A). Sensitivity analysis produced similar outcomes (MD, 0.28; 95% CI, −0.01 to 0.58; p=0.06; participants=388; studies=4; I2=0%).
For the leg VAS at 6 and 12 months, three studies including 211 patients (71 in the obese group and 140 in the nonobese group) were analyzed. The fixed effects model showed a pooled MD for VAS Leg at 6 months of 0.23 (95% CI, −0.13 to 0.59), which showed no significant difference between groups (Z=1.27, p=0.20) and low heterogeneity (I2=37%, p=0.21) (Fig. 4B). Sensitivity analysis confirmed these findings (MD, −0.40; 95% CI, −1.19 to 0.39; p=0.32; participants=164; studies=2; I2=0%). Subgroup analysis according to obesity severity revealed that heterogeneity in the lower obesity group decreased to 0%, suggesting a more consistent effect within this subgroup (Supplement 3). For the VAS Leg at 12 months, the pooled MD was 0.18 (95% CI, −0.10 to 0.45), again showing no significant difference between groups (Z=1.27, p=0.21) and low heterogeneity (I2=44%, p=0.17) (Fig. 4C). Sensitivity analysis indicated significant differences between the groups (MD, 0.85; 95% CI, −0.01 to 1.70; p=0.05; participants=164; studies=2; I2=0%). Subgroup analysis based on obesity severity revealed that heterogeneity in the lower obesity group decreased to 0% (Supplement 4).
Oswestry Low Back Disability Index
In the analysis of ODI at 3 months, five studies including 437 patients (152 in the obese group and 285 in the nonobese group) were analyzed. Using a fixed effects model, the pooled MD was 1.02 (95% CI, −0.14 to 2.18). The overall effect at 3 months was not significantly different between the groups (Z=1.72, p=0.09). Analysis indicated no heterogeneity (I2=0%, p=0.57) (Fig. 5A). Sensitivity analysis confirmed these findings (MD, 0.85; 95% CI, −0.63 to 2.32; p=0.26; participants=390; studies=4; I2=0%).
When comparing ODI at 6 months between the obese (71 patients) and nonobese (140 patients) groups among the three studies, the fixed effects model showed an MD of 1.40 (95% CI, 0.14 to 2.67), indicating a small but significantly higher ODI score at 6 months in the obese group than in the nonobese group (Z=2.17, p=0.03). There was no heterogeneity among the studies (I2=0%, p=0.42) (Fig. 5B). Sensitivity analysis indicated no significant differences between the groups (MD, 2.52; 95% CI, −0.22 to 5.27; p=0.07; participants=164; studies=2; I2=0%).
For the ODI score at 12 months, three studies including 211 patients (71 in the obese group and 140 in the nonobese group) were analyzed. Using a fixed effects model, the pooled MD was 0.10 (95% CI, −1.14 to 1.33), indicating no significant difference between groups (Z=0.15, p=0.88) and moderate heterogeneity (I2=52%, p=0.12) (Fig. 5C). Sensitivity analysis showed similar results (MD, −2.61; 95% CI, −5.50 to 0.28; p=0.08; participants=164; studies=2; I2=0%). Subgroup analysis according to obesity severity revealed that heterogeneity in the lower obesity group decreased to 0%, suggesting a more consistent effect within this subgroup (Supplement 5).
Reherniation
Six studies including 541 patients (194 in the obese group and 347 in the nonobese group) were analyzed for reherniation. Based on a fixed effects model, the OR was 1.35 (95% CI, 0.73 to 2.49), suggesting no significant difference in reherniation rates between obese and nonobese patients (Z=0.96, p=0.33). There was no heterogeneity among the studies (I2=0%, p=0.62) (Fig. 2C). Sensitivity analysis revealed similar results (OR, 1.43; 95% CI, 0.66 to 3.10; p=0.36; participants=367; studies=5; I2=0%).
Publication bias
Egger regression analysis revealed no evidence of publication bias for any of the outcomes (p>0.05 for all). Furthermore, the funnel plots showed symmetrical distribution.
GRADE
Table 3 presents the evaluations based on GRADE. Certainty was low for the VAS back and leg measurements at 3 and 6 months, ODI at 12 months, and the remaining variables.
Discussion
This meta-analysis examined the relationship between obesity and endoscopic spine surgery outcomes in patients undergoing LBP treatment for LBP. The results point to a fairly limited impact of body weight on most of the measured parameters, with similar operative time between patients who were obese and nonobese. Moreover, both groups had similar VAS scores for back and leg pain at 3, 6, and 12 months after surgery and similar rates of reherniation. However, further studies are warranted, especially because heterogeneity was high in some comparisons. In addition, the relatively high ODI at 6 months postoperatively among patients who were obese could have reflected some delay in functional recovery, although this finding was not sustained in the sensitivity analyses, suggesting limited quality of evidence. These conclusions implied that under some conditions, obesity may be a poor determinant of outcome after endoscopic spine operation for LBP, except when transient functional recovery is a concern.
The similar operative time between obese and nonobese patients was expected because of the high variability in surgical techniques, anesthesia protocols, surgeon experience, and patient type among the included studies. For example, variability may have been affected by differences between minimally invasive and open techniques among patients with different comorbidities, as highlighted by Hartmann et al. [23] for minimally invasive discectomy and Weinstein et al. [24] for open procedures. Standardized reporting of these variables should not be neglected in future research [23]. Similarly, psychological and behavioral aspects were more likely to have affected pain perception and recovery [25,26]. Gatchel et al. [25] emphasized the role of psychosocial factors, such as depression, anxiety, and coping mechanisms, in influencing pain perception and recovery outcomes after spine surgery. Similarly, Hansson et al. [26] demonstrated that psychological distress and maladaptive behaviors rather than body weight were significant predictors of poor surgical outcomes and prolonged recovery. These findings suggested that the small differences attributable to obesity may be masked by the stronger influence of psychological and behavioral factors, underscoring the importance of addressing these aspects in patient care. Future studies should incorporate these measures for deeper understanding.
The slight increase in ODI after six months in obese patients suggested the possibility of delayed functional recovery, although this was not strongly indicated in the sensitivity analysis. These results could have been confounded by common obesity-associated aspects, such as relatively diminished preoperative mobilization, difficulty in postoperative rehabilitation, and a relatively high likelihood of wound complications [27]. Chou et al. [27] highlighted that obese patients undergoing spine surgery were at increased risk of wound infections, delayed healing, and prolonged recovery because of reduced mobility and relatively high rates of comorbidities. Careful analysis of the studies included in this meta-analysis, particularly those that contributed to the 6-month ODI, is required to ascertain the clinical importance of these factors and their impact on functional recovery in obese patients.
The impact of obesity varies among surgical specialties. For example, in hip and knee arthropla sy, there is a clear association between obesity and increased rates of complications, such as surgical site infection, implant failure, and prolonged recovery time [28,29]. However, less invasive forms of spinal surgery, such as endoscopic procedures, do not seem to exhibit a comparable effect of obesity on outcomes. Obesity is a well-established risk factor for increased complications following gynecological, cardiovascular, and neurological surgery. For instance, Aarts et al. [30] found that obese patients undergoing gynecological surgery had relatively high rates of wound complications and infections. Similarly, Engelman et al. [31] highlighted the increased risk of postoperative complications, including sternal wound infections and prolonged hospital stay, among obese patients undergoing cardiovascular surgery. Patil et al. [32] reported that obesity was associated with increased rates of complications, such as cerebrospinal fluid leak and infection, following cranial and spinal procedures. A meta-analysis on spinal fusion surgery indicated that obesity had a significantly more negative effect on outcomes [33], probably because this more complex procedure is highly invasive, involves extensive soft tissue dissection, and requires a longer recovery period, all of which can be further complicated by obesity. In contrast, less invasive endoscopic spine surgery may mitigate several of these risks and make the impact of obesity relatively minor. However, future studies that focus on comparison of outcomes among various types of spine surgeries in obese and nonobese patients are required to verify this.
In addition to the VAS and ODI, other factors, such as quality of life and patient-reported outcomes, were not considered in this meta-analysis and require further investigation. Future meta-analyses should include these factors to provide a comprehensive understanding of surgical outcomes. Moreover, the minimal clinically important difference (MCID) for both VAS and ODI must be considered to determine the clinical importance of the observed differences [34,35]. Our analysis suggested that differences in VAS were not significant. However, determining whether the subgroups exceeded the MCID may have demonstrated clinically significant changes. Furthermore, ODI MCID assessment may shed light on the functional significance of the changes at 6 months. Future research should focus on the advantages of any preoperative intervention, such as weight reduction or even bariatric surgery, in obese patients undergoing endoscopic spine surgery. The relationship between BMI and outcome requires further clarification, and optimization of surgical outcomes needs to be distinguished between morbid and less severe obesity.
This meta-analysis had several limitations. The small number of studies and few complications analyzed, such as reherniation, may have limited the statistical power of detecting significant differences. Variations in BMI among the studies introduced heterogeneity, because the effects of obesity differed between mild and morbid obesity. Furthermore, high heterogeneity in outcomes, such as operative time, indicated uncontrollable discrepancies in surgical techniques or protocols. Lastly, the predominance of nonrandomized studies brought about bias, as shown by the very low overall quality of evidence by GRADE assessment.
Conclusions
Endoscopic spine surgery did not result in significant differences in operative times, patient-reported outcomes, or reherniation rates between obese and nonobese patients. However, the small number of studies, different BMIs, and high heterogeneity in certain outcomes in this meta-analysis pave the way for further investigation. Future research should standardize BMI within variables, surgical techniques, and protocols and use larger sample sizes for more definitive conclusions. Moreover, the role of preoperative interventions, such as weight loss programs, should be examined to maximize the outcomes of endoscopic spine surgery for obese patients.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
Supplement
