Risk factors for cerebrospinal fluid leakage after spinal tumor surgery: a systematic review and meta-analysis

Introduction

Surgery remains the primary treatment modality for spinal tumors, with cerebrospinal fluid (CSF) leakage representing one of the most common postoperative complications (1). The reported incidence of CSF leakage following spinal surgery ranges from 1.0% to 14.0% (2), primarily caused by intraoperative dural membrane rupture (3). Clinical manifestations include headache, nausea, vomiting, and wound drainage (4). If inadequately managed, CSF leakage may result in poor wound healing, infection, meningitis, and other severe complications, ultimately prolonging hospitalization and increasing economic burden (5). Previous studies (6-9) have suggested that factors such as age, gender, smoking history, number of operations, surgical approach, and postoperative activity may influence the incidence of CSF leakage after spinal tumor surgery. Nevertheless, inconsistencies remain among studies, with some reporting non-significant findings. Differences in cultural background, lifestyle, study design, and sample size across populations have further contributed to these discrepancies.

Previous observational studies (6-9) have reported inconsistent findings regarding the predictors of postoperative CSF leakage after spinal intradural or tumor surgery. For example, some studies identified age or revision surgery as significant predictors, whereas others did not confirm these associations after adjustment for operative or disease-related variables. Moreover, surgery-related factors such as operative time, blood loss, and dural repair technique have not been evaluated consistently across studies. These discrepancies may reflect differences in study design, sample size, patient selection, tumor characteristics, and statistical adjustment. In addition, associations observed in univariate analyses may be attenuated after multivariable adjustment, suggesting that some reported predictors may reflect confounding by surgical complexity or case mix rather than independent effects. Therefore, this review aimed to synthesize the best available evidence, with priority given to adjusted risk estimates where possible.

To better clarify the risk factors for CSF leakage following spinal tumor surgery and assist clinicians in adopting effective preventive measures, we performed a meta-analysis to identify influencing factors. We present this article in accordance with the PRISMA reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2079/rc).

Methods

Search strategy

A comprehensive search of Chinese and English literature was conducted for case-control and cohort studies examining risk factors for CSF leakage after spinal tumor surgery, up to December 2022. The following databases were searched: PubMed, Web of Science, Embase, ProQuest, Cochrane Library, China National Knowledge Infrastructure (CNKI), China Biology Medicine disc (CBMdisc), Wanfang Data, and China Science and Technology Journal Database (VIP). The search strategy included keywords and free-text terms such as (“spine” OR “cervical vertebra” OR “thoracic vertebra” OR “lumbar vertebra” OR “sacrum”) AND (“tumor” OR “neoplasm”) AND (“surgery” OR “operation”) AND (“cerebrospinal fluid leakage” OR “CSF leak”) AND (“risk factors” OR “influencing factors”). Reference lists of included studies were also manually searched to minimize the risk of missing eligible studies.

The time frame of the search was from database inception to December 2022, without a lower date limit. Both English and Chinese language publications were considered. The search strategy combined MeSH terms with free-text terms, and titles/abstracts were scanned to identify relevant studies.

Inclusion and exclusion criteria

The inclusion criteria for article selection were as follows: eligibility criteria (PICOS): Population: adult patients undergoing surgery for spinal tumors or intradural spinal lesions. Exposure: candidate clinical or surgical risk factors for postoperative CSF leakage. Comparator: patients without postoperative CSF leakage or those without the exposure of interest. Outcome: postoperative CSF leakage or CSF leakage-related complications. Study design: observational studies, including cohort and case-control studies. Exclusion criteria: case reports, reviews, conference abstracts without sufficient data, duplicate publications, studies without extractable outcome data, and non-human studies.

Study design

Two independent researchers screened studies by title and abstract, followed by full-text review. Data extraction was conducted using a standardized form, which included: (I) study characteristics (title, author, year, country, study type); (II) patient information (sample size, tumor type, CSF leakage incidence); and (III) reported risk factors. Discrepancies were resolved by consensus or consultation with a third senior researcher.

Quality assessment

Study quality was assessed using the Newcastle-Ottawa Scale (NOS), evaluating study population selection, comparability, and exposure/outcome assessment. The maximum score was 9 points (2 points for comparability; 1 point each for other items). Higher scores indicated better study quality. Variables were defined according to thresholds used in previous studies, including operative time >4 hours and intraoperative blood loss >1,000 mL. Because the included observational studies comprised both case-control and cohort designs, we performed subgroup and/or sensitivity analyses by study design where sufficient data were available. Given the relatively low incidence of postoperative CSF leakage in most included studies, ORs were used as the common summary measure; however, this issue is acknowledged as a limitation where outcome frequency or effect measure reporting differed across studies. Methodological quality was assessed using the Newcastle-Ottawa Scale. No study was excluded solely on the basis of NOS score; instead, study quality was considered in the interpretation of findings and sensitivity analyses.

Statistical analysis

Meta-analysis was performed using Stata 12.0. Weighted mean differences (WMD) were calculated for continuous variables, and odds ratios (OR) with 95% confidence intervals (CI) for categorical variables. Heterogeneity was assessed by Q test and I² statistic. A fixed-effects model was applied if P>0.05 or I²≤50%; otherwise, a random-effects model was used. Sensitivity analysis was performed to explore heterogeneity sources, and robustness was confirmed if effect estimates remained stable. Publication bias was evaluated using Begg’s test, with P>0.05 indicating no bias. For each risk factor, multivariable-adjusted effect estimates were preferentially extracted. If adjusted estimates were not reported, the most completely reported unadjusted estimates were extracted. The level of adjustment for each study was recorded and considered during interpretation of the pooled results.

Results

Search results

The study selection process is shown in Figure 1. A total of 643 publications were identified. After removing 122 duplicates using NoteExpress, 521 studies remained. Of these, 499 were excluded after title and abstract screening. Full-text review of the remaining 22 studies resulted in the inclusion of seven studies (four Chinese, three English).

Figure 1 Flow diagram depicting the study selection process. CNKI, China National Knowledge Infrastructure.

Quality of included studies

Quality assessment revealed that one study (6) scored 5 points (moderate quality) due to unclear exposure description, while the remaining six scored 7 points (high quality). Several studies reported incomplete follow-up, and some excluded patients with missing data, potentially introducing bias.

Study description

As shown in Table 1, five case-control and two retrospective cohort studies published between 2010 and 2022 were included, conducted in China, South Korea, Switzerland, and Germany. Participants were aged 19–94 years, with surgical sites primarily in the thoracic and lumbar vertebrae. Sample sizes ranged from 302 to 4,955 cases. Patients were divided into CSF leakage and non-leakage groups. Reported risk factors included age >60 years, smoking history, male gender, multisegment surgery, revision surgery, delayed postoperative mobilization (>3 days), long operative time, and substantial intraoperative blood loss.

Results of meta-analysis

The pooled results (Table 2) demonstrated that advanced age (P=0.01), smoking history (P=0.03), revision surgery (P<0.001), prolonged operative time (P=0.007), and significant intraoperative bleeding (P<0.001) were associated with increased CSF leakage risk. Conversely, no significant associations were found for gender (P=0.95), multi-segment surgery (P=0.48), or delayed mobilization (P=0.25).

Sensitivity analysis and publication bias

Significant heterogeneity was observed in the analyses of “revision surgery” and “operative time” (Figures 2,3). Sensitivity analysis indicated that heterogeneity mainly originated from Lei et al. (8), whose study had a substantially larger sample size than others. However, exclusion of individual studies did not materially change the results, suggesting robustness of the findings. Publication bias was assessed using Begg’s test, which revealed no significant bias.

Figure 2 Sensitivity analysis of the influence of revision surgery on cerebrospinal fluid leakage. CI, confidence interval.

Figure 3 Sensitivity analysis of the influence of operation time on cerebrospinal fluid leakage. CI, confidence interval.

Discussion

The occurrence of CSF leakage after spinal tumor surgery is not only associated with iatrogenic intraoperative injuries—such as mechanical trauma from surgical instruments, excessive traction of nerve roots, and rough operative techniques (13)—but also with several high-risk factors that contribute to increased incidence. However, limited research has focused specifically on these influencing factors, and the findings have been inconsistent. Therefore, this study employed a meta-analysis to identify risk factors more comprehensively.

Advanced age emerged as a significant risk factor. With increasing age, spinal degeneration becomes more severe, dural thickness decreases, and epidural fat diminishes, thereby elevating the likelihood of dural rupture during surgery (14). Herren (15) reported that the risk of dural injury increased by 2% with each additional year of age. Our analysis confirmed that patients over 60 years had a significantly higher risk of postoperative CSF leakage compared with younger individuals. Elderly patients are also more prone to postoperative complications such as impaired sputum clearance and constipation (16). Episodes of severe coughing or straining during defecation can abruptly increase intra-abdominal pressure, potentially precipitating dural rupture at vulnerable sites, leading to leakage. Consequently, preoperative planning should be individualized, intraoperative dissection of adhesions performed meticulously, and postoperative nursing care intensified. Interventions include instructing patients on effective coughing techniques, promoting dietary habits that prevent constipation, administering laxatives when necessary, and securing drainage tubes to prevent tension-related wound complications (17).

Smoking was also identified as an independent risk factor. Prolonged exposure to nicotine and carbon monoxide compromises dural elasticity and vertebral blood flow, while also reducing fibrinolytic function of dural surface vessels, thereby promoting adhesion between the dura and surrounding tissues (18,19). This pathological state increases operative difficulty and the likelihood of CSF leakage. Additionally, smokers are more prone to postoperative coughing, which may exacerbate leakage if coughing techniques are inadequate (20). Our findings are consistent with Goyal (21), who confirmed smoking as a risk factor for CSF leakage. Thus, preoperative smoking cessation counseling is essential. Postoperatively, nebulized glucocorticoids may facilitate sputum clearance and reduce coughing-related strain on the dura.

Revision surgery further increases the risk of CSF leakage. Revision procedures often involve altered local anatomy, scar tissue formation, and dural adhesions, making the dura more susceptible to rupture (22-24). Literature has shown that revision surgery significantly raises the incidence of postoperative CSF leakage (23). For reoperations, the surgical approach should ideally progress from non-adherent to scarred tissue areas (2). Common indications for revision include infection, inadequate drainage, and implant loosening (25). To mitigate risks, strict aseptic technique, maintenance of unobstructed drainage, careful postoperative wound management, and timely intervention in cases of abnormal drainage are critical.

Prolonged operative time and excessive intraoperative blood loss were also identified as significant contributors. Patients with extensive tumor invasion often present with blurred anatomical planes, leading to longer operations, greater bleeding, and destruction of paraspinal tissue structures. These changes alter pressure gradients across the dura and create potential cavities for CSF accumulation, predisposing to leakage (13). For such cases, minimizing operative time and blood loss, stabilizing hemodynamics with intraoperative fluid management, and maintaining appropriate postoperative positioning are essential strategies to reduce complications (26). In addition to tumor-specific risk factors, general predictors of CSF leakage after spinal surgery have been reported, such as increased body mass index (BMI) and diabetes mellitus (27). These comorbidities have been shown to impair wound healing and dural integrity. Interestingly, our analysis did not validate these factors in the spinal tumor population, suggesting that the spectrum of risk determinants may differ between tumor-related and degenerative spinal procedures.

In contrast, no significant associations were found for gender, surgical segment involvement, or timing of postoperative mobilization. Some studies have reported higher leakage rates in men, likely attributable to higher smoking prevalence and associated dural adhesions rather than gender per se (9). Although multisegment procedures are technically challenging and theoretically increase the risk of dural sac injury, our results did not confirm a significant correlation. Similar findings were reported by Zhong et al. (18), where more laminae were removed in leakage cases, but regression analysis showed no statistical difference. Regarding postoperative activity, some scholars advocate prone positioning to reduce pain and complications (28), whereas others emphasize the benefits of early mobilization (29). Our findings suggest that neither early (<3 days) nor delayed (>3 days) mobilization demonstrated a clear causal relationship with CSF leakage, indicating the need for further prospective studies to clarify this association. Based on these findings, it may be feasible to develop a risk prediction model incorporating the most significant risk factors such as advanced age (>60 years), smoking, and revision surgery. A scoring system derived from these variables could stratify patients into different risk categories, thereby assisting surgeons in preoperative counseling and tailoring perioperative preventive strategies.

The limitations of this study should be acknowledged. First, the number of included studies was small, limiting the breadth of analyzable risk factors. Second, only Chinese and English publications were reviewed, raising the possibility of language bias. Third, substantial differences in sample size across studies may have influenced pooled results. Finally, the heterogeneity introduced by inconsistent surgical sites could not be fully resolved, as tumor location was not uniformly reported and thus could not be analyzed as an independent variable. The pooled estimate for substantial intraoperative blood loss was very large and should be interpreted with caution, as it appeared to be driven primarily by sparse-event data from a small number of studies and may have been affected by zero-cell instability.

Conclusions

CSF leakage is a frequent complication of spinal tumor surgery, occurring across all spinal levels and significantly impacting postoperative recovery and treatment efficacy (30). Its prevention is of paramount importance. The present study identified advanced age, smoking, revision surgery, prolonged operative time, and excessive intraoperative blood loss as major risk factors. Preventive measures include comprehensive preoperative evaluation, formulation of individualized surgical plans, and reinforcement of smoking cessation. Intraoperatively, meticulous technique and efforts to minimize operative time and blood loss are essential to reduce dural injury. Postoperatively, patient education on effective coughing, constipation prevention, and abdominal pressure management should be emphasized. From a translational perspective, these findings highlight the need for a preventive checklist for high-risk patients. Such a checklist can include preoperative smoking cessation, comprehensive imaging assessment and surgical planning, intraoperative strategies to minimize traction and blood loss, and postoperative measures such as effective cough training and constipation prevention. Implementation of these measures may help to substantially reduce the incidence of CSF leakage in clinical practice. Nevertheless, because only a limited number of studies were included, factors such as surgical site, BMI, duration of postoperative bed rest, and drainage tube retention could not be adequately assessed. Future multicenter studies with larger cohorts are warranted to validate these findings and explore additional risk factors.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2079/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2079/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2079/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Birkholz SE, Patil AA, Chamczuk AJ. Treatment of postoperative recurrent cerebrospinal fluid leak with pseudo-meningocele formation using temporary epidural drain. Interdisciplinary Neurosurgery 2019;16:25-8.
2. Menon SK, Onyia CU. A short review on a complication of lumbar spine surgery: CSF leak. Clin Neurol Neurosurg 2015;139:248-51. [Crossref] [PubMed]
3. Ghobrial GM, Theofanis T, Darden BV, et al. Unintended durotomy in lumbar degenerative spinal surgery: a 10-year systematic review of the literature. Neurosurg Focus 2015;39:E8. [Crossref] [PubMed]
4. Guerin P, El Fegoun AB, Obeid I, et al. Incidental durotomy during spine surgery: incidence, management and complications. A retrospective review. Injury 2012;43:397-401.
5. Buck JS, Yoon ST. The Incidence of Durotomy and its Clinical and Economic Impact in Primary, Short-segment Lumbar Fusion: An Analysis of 17,232 Cases. Spine (Phila Pa 1976) 2015;40:1444-50. [Crossref] [PubMed]
6. Jesse CM, Schermann H, Goldberg J, et al. Risk Factors for Postoperative Cerebrospinal Fluid Leakage After Intradural Spine Surgery. World Neurosurg 2022;164:e1190-9. [Crossref] [PubMed]
7. Zhu MD. Risk factors of cerebrospinal fluid leakage in lumbar surgery and the application value of hyperplastic bone knife. China Medical University; 2020.
8. Lei BP. The correlations of Surgical Site Infection in lumbar Spine and Cerebrospinal fluid leakage. Yangtze University; 2020.
9. Lenschow M, Perrech M, Telentschak S, et al. Cerebrospinal fluid leaks following intradural spinal surgery-Risk factors and clinical management. Front Surg 2022;9:959533. [Crossref] [PubMed]
10. Qu HY, Guo W, Yang RL. Risk factors analysis and countermeasures of wound related complications of sacral tumors. Orthopedic Journal of China 2010;18:212-5.
11. Chen T. Risk factors of postoperative complications after anterior cervical spinal surgery. Nanchang University; 2020.
12. Lee S, Cho DC, Kim KT, et al. Reliability of Early Ambulation after Intradural Spine Surgery: Risk Factors and a Preventive Method for Cerebrospinal Fluid Leak Related Complications. J Korean Neurosurg Soc 2021;64:799-807. [Crossref] [PubMed]
13. Zhu Q, Ye C, Xie D, et al. Management process of incidental durotomy and cerebrospinal fluid leakage following lumbar surgery: a literature review. Chinese Journal of Bone and Joint 2023;12:57-3.
14. Du JY, Aichmair A, Kueper J, et al. Incidental durotomy during spinal surgery: a multivariate analysis for risk factors. Spine (Phila Pa 1976) 2014;39:E1339-45. [Crossref] [PubMed]
15. Herren C, Sobottke R, Mannion AF, et al. Incidental durotomy in decompression for lumbar spinal stenosis: incidence, risk factors and effect on outcomes in the Spine Tango registry. Eur Spine J 2017;26:2483-95. [Crossref] [PubMed]
16. Chen B, Yan X, Wang X, et al. Effectiveness of precise and quantitative rapid pulmonary rehabilitation nursing program for elderly patients with lung cancer during the perioperative period: A randomized controlled trial. Pak J Med Sci 2023;39:572-7. [Crossref] [PubMed]
17. Krpenská E, Jandová K, Gürlich R. Perioperative care of elderly patients. Cas Lek Cesk 2020;159:13-6.
18. Zhong J, Wen BT, Chen ZQ, et al. Risk factors of the cerebrospinal fluid leakage after posterior circumferential decompression for thoracic ossification of posterior longitudinal ligament. Chinese Journal of Spine and Spinal Cord 2021;31:705-1.
19. Wang H, Xue R, Wu LM, et al. Prevalence and risk factors of postoperative symptomatic myocardial ischemia in surgically treated patients with degenerative spinal disorder. Chin J Anat Clin 2018;23:406-1.
20. Wang L, Yang LL, Chu PY. Risk factors analysis of postoperative cerebrospinal fluid leakage after lumbar posterior decompression fusion. Nursing Practice and Research 2021;18:1901-4.
21. Goyal DKC, Divi SN, Bowles DR, et al. How Does Smoking Influence Patient-reported Outcomes in Patients After Lumbar Fusion? Clin Spine Surg 2021;34:E45-50. [Crossref] [PubMed]
22. Chinese Association of Orthopaedic Surgeons. Evidence-based clinical practice guidelines for the treatment of dural tears and the consequent cerebrospinal fluid leak during spine surgery. Chin J Surg 2017;55:86-9.
23. Smorgick Y, Baker KC, Herkowitz H, et al. Predisposing factors for dural tear in patients undergoing lumbar spine surgery. J Neurosurg Spine 2015;22:483-6. [Crossref] [PubMed]
24. Kang YW, Yang XF, Yu BS. Research progress of treatment for cerebrospinal fluid leak after lumbar surgery. J Spinal Surg 2020;18:278-1.
25. Smith JS, Shaffrey CI, Sansur CA, et al. Rates of infection after spine surgery based on 108,419 procedures: a report from the Scoliosis Research Society Morbidity and Mortality Committee. Spine (Phila Pa 1976) 2011;36:556-63. [Crossref] [PubMed]
26. Tosun B, Ilbay K, Kim MS, et al. Management of Persistent Cerebrospinal Fluid Leakage Following Thoraco-lumbar Surgery. Asian Spine J 2012;6:157-62. [Crossref] [PubMed]
27. Wong AP, Shih P, Smith TR, et al. Comparison of symptomatic cerebral spinal fluid leak between patients undergoing minimally invasive versus open lumbar foraminotomy, discectomy, or laminectomy. World Neurosurg 2014;81:634-40. [Crossref] [PubMed]
28. Menon SK, Onyia CU. A short review on a complication of lumbar spine surgery: CSF leak. Clin Neurol Neurosurg 2015;139:248-51. [Crossref] [PubMed]
29. Dafford EE, Anderson PA. Comparison of dural repair techniques. Spine J 2015;15:1099-105. [Crossref] [PubMed]
30. Zou YB, Lu L, Xie JX, et al. Research progress of spinal surgery complicated with dural rupture and cerebrospinal fluid leakage. Zhejiang Practical Medicine 2021;26:176-9.