Introduction
Advanced surgical techniques have triggered a notable increase in degenerative spine surgeries over the recent decades [1–7]. Despite its infrequency, venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), remains a catastrophic complication following spine surgery [8] and ranks third in terms of prevalence among all hospital-acquired complication following elective spine surgeries, with an incidence rate of approximately 1% [9]. Incidence rates of VTE vary significantly across different populations and regions [10,11]. Although VTE following spine surgeries could be potentially life-threatening [11], the availability of safe anticoagulants has helped reduce the burden of VTE through enhanced thromboprophylaxis [12].
Anticoagulation, though generally safe for non-spine surgeries, could potential cause compressive hematomas that can lead to severe complications, such as paralysis, and require emergency spinal decompression surgeries to alleviate the risk of permanent neurological deficits [13]. Concerns over chemoprophylaxis in spine surgery primarily stem from its potential to trigger the development of spinal epidural hematomas (SEHs), an uncommon yet severe complication [14]. Early initiation of anticoagulation therapy increases the risk of heightened bleeding, particularly the formation of hematomas at surgical sites, which necessitate urgent interventions, such as hematoma evacuation [15]. Although mechanical prophylactic measures like patient ambulation and sequential compression devices are widely accepted due to their minimal disadvantages [16], considerable variability in chemoprophylaxis protocols have been observed among spine surgeons, with limited availability of high-quality studies and evidence-based guidelines [17].
The choice of pharmaceutical agents (enoxaparin, unfractionated heparin, aspirin, or none) and timing of chemoprophylaxis initiation remain points of contention among spine surgeons [18,19]. This variability underscores the need for comprehensive research to establish optimal VTE prophylaxis strategies in the context of degenerative spine surgery. Several existing reviews have been limited by their inclusion of diverse spine surgery types or their use of broad categories for elective spine surgeries [20]. As such, this systematic review aimed to analyze the effectiveness of chemical prophylaxis in preventing VTE and SEH after degenerative spine surgery. Moreover, we intend to evaluate different chemoprophylaxis strategies in this surgical setting.
Materials and Methods
This study was approved by the Ethics Committee of Tehran University of Medical Sciences (reference number: IR.TUMS.SINAHOSPITAL.REC.1403.032). This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) protocol [21] and was registered on PROSPERO with the registration number CRD42024585493.
Literature search
On February 2, 2024, an electronic search was conducted across five databases, namely Embase, Cochrane Library, Scopus, and Web of Science, to identify studies comparing chemical prophylaxis with non-chemical prophylaxis among patients undergoing degenerative spine surgery, without restrictions on dates, study designs, or languages. The inquiry terms included keywords related to chemical prophylaxis and degenerative spine surgeries. The detailed search strategy is available in Appendix 1.
Selection criteria
Studies examining thromboembolic events in patients undergoing degenerative spine surgery were considered for inclusion. The included studies had to satisfy the following criteria: (1) contained at least two groups in which one received chemical prophylaxis and the other did not; (2) all patients should have undergone degenerative spine surgeries; and (3) at least one of the complications, either VTE (DVT and PE) or she, should have been reported. The control group in our study was defined as patients who had not received any pharmacological VTE prophylaxis, including no single dose for VTE chemoprophylaxis. The absence or presence of mechanical prophylaxis was not relevant to our inclusion criteria. The key requirement was that patients did not receive any pharmacological VTE prophylaxis. In contrast, the chemoprophylaxis group included patients who had received any pharmacological VTE prophylaxis, including both single-dose perioperative and continuous postoperative regimens, in order to compare the outcomes between the single-dose and continuous-dosing strategies. The exclusion criteria were as follows: (1) patients younger than 18 years; (2) inadequate details on surgery or outcomes; (3) study population containing trauma, tumors, infections, congenital deformities, and adolescent idiopathic scoliosis either totally or pooled with degenerative ones; (4) editorials, letters, conference abstracts or posters, technical reports, systematic reviews, meta-analyses, nonclinical papers (in vitro, in vivo, ex vivo, and in silico), case reports, and case series; and (5) studies that reported solely on mechanical prophylaxis. The use of mechanical prophylaxis utilized in each group (with or without chemical prophylaxis) was not considered as an exclusion criterion. No restrictions on operative spinal region was set given that the anticipated variability in surgical levels required an unrestricted approach.
Selection process
Initially, the search results were imported into EndNote (X9; Clarivate, Philadelphia, PA, USA) where duplicates were removed and then transferred to Rayyan QCRI (web application; Qatar Computing Research Institute [Data Analytics], Doha, Qatar) [22]. Two reviewers then independently screened the titles and abstracts, with discrepancies being resolved by the senior author. Irrelevant titles were excluded, whereas full-text articles were obtained for potentially relevant or uncertain titles. These full texts were assessed against the inclusion and exclusion criteria by two reviewers, with any disagreements resolved through discussions with the senior author.
Data extraction
Data were extracted using a Google Form collection tool, covering study identifiers (authors, year, and title), study and patient characteristics (design, country, inclusion and exclusion criteria, sample sizes, demographics, body mass index [BMI], VTE risk factors [smoking, obesity, comorbidities, and neurological deficits]), clinical characteristics (pathology, surgical approach, and operative spinal region), and anticoagulant details (type, dosage, timing of chemical prophylaxis, and mechanical prophylaxis). The primary outcomes of interest included VTE (including DVT and/or PE) and SEH incidences. Non-English studies were first translated into English using the Yandex Translate web page (launched on March 22, 2011, Russia; https://translate.yandex.com/en/) before undergoing data extraction. Three investigators independently extracted and verified the data, with discrepancies being resolved by consensus and senior author intervention when necessary.
Quality assessment
The risk of bias in randomized clinical trials (RCTs) was evaluated using version 2 of the Cochrane risk-of-bias tool (RoB 2; Cochrane, London, UK) following the Cochrane Handbook guidelines for Systematic Reviews of Interventions [23]. The risk of bias in cohort studies was assessed using the Newcastle-Ottawa scale (NOS; Ottawa Hospital Research Institute, Ottawa, ON, Canada) [24]. Two independent authors conducted these evaluations, with any disagreements being resolved by the senior author. The results then were visualized using risk-of-bias visualization (robvis; https://www.riskofbias.info/) [25].
Statistical analysis
All statistical analyses were conducted using STATA ver. 17.0 software (2023; Stata Corp., College Station, TX, USA). To compare baseline characteristics between the two groups, we first pooled the relevant quantitative data reported across the included studies. Afterward, continuous variables were expressed as mean±standard deviation and compared using independent samples t-tests, assuming equal variances. Categorical variables were presented as numbers and percentages and compared between groups using the chi-square test. A p-value of <0.05 indicated statistical significance.
The meta-analysis compared patients who did and did not receive chemical prophylaxis, focusing on VTE. Risk ratios (RRs) and 95% confidence intervals (CIs) were calculated for dichotomous outcomes across all studies. Heterogeneity among studies was assessed using the I2 statistic, with an I2 value of 0.00% indicating no observed heterogeneity. A random-effects model was applied to the overall analysis to account for potential variability across studies, using the Knapp-Hartung standard errors to enhance robustness. Subgroup analyses were conducted based on the type of medication used for the chemical prophylaxis, study design, dosage regimen, and study quality.
Results
In total, 5,535 records from five databases (Medline, Embase, Cochrane Library, Scopus, and Web of Science) were found based on our selection criteria. After removing duplicate records, 2,566 titles/abstracts were screened. Thereafter, 86 full texts were assessed for eligibility, and 17 studies were ultimately included. The PRISMA flowchart for this study is presented in Fig. 1.
Baseline characteristics
Study characteristics
Among the 17 studies analyzed, 10 (58.8%), 6 (35.3%), and 1 (5.9%) were RCTs, retrospective cohort studies, and prospective cohort studies, respectively. Publication dates ranged from May 1984 to June 2023. Countries of origin included China (8 [47.1%]), the United States (2 [11.7%]), the United Kingdom (2 [11.7%]), Iran (1 [5.9%]), Switzerland (1 [5.9%]), South Korea (1 [5.9%]), Denmark (1 [5.9%]), and Hong Kong (1 [5.9%]). Overall, 5,383 patients were evaluated in the included studies, among which 2,863 (53.2%) and 2,520 (46.8%) did and did not receive chemical prophylaxis.
Patient characteristics
Baseline characteristics, including age, sex, BMI, and operative duration, were compared between the chemoprophylaxis and control (non-chemoprophylaxis) groups. No significant differences in age, BMI, and operative duration were noted between the two groups, indicating similar baseline characteristics. However, the proportion of female patients was significantly greater in the chemoprophylaxis group than in the control group (p=0.017) (Table 1).
Details regarding the baseline characteristics of all included studies are presented in Table 2 and Appendix 2. And studies excluded following email contact with the authors are presented in Appendix 3. The most common pathology among the patients was degenerative lumbar spinal stenosis in six studies (35.3%). Based on the 14 studies (82.4%) that separately reported the average age for each group, the mean age ranged from 41.5 to 66.7 years and 44.5 to 69.8 years for the patients who did and did not receive chemoprophylaxis, respectively. Sex distribution was reported in all studies except for two [26,27]. The other 15 studies (88.2%) included 1,819 men (44.7%) and 2,246 women (55.3%). Degenerative spinal disease was the indication for spine surgery in all patients. The average patient BMI ranged from 22.75±2.39 to 27.56±5.86 and from 22.24±2.78 to 26.51±4.11 for the control and treatment groups, respectively, although only 10 studies (58.85%) separate reported on BMI for each studied group. Regarding VTE risk factors, only two studies (11.77%) mentioned the smoking status of patients [28,29]. Obesity was reported in one study (5.88%) [30]. Comorbidities were addressed in three studies (17.65%): one study (5.88%) discussed them generally [29], whereas two studies (11.77%) provided a detailed breakdown of chronic diseases [30,31]. Moreover, only one study mentioned the number of patients with preoperative neurological deficits [31].
Clinical characteristics
Chemical prophylaxis
A total of 2,863 (53.2%) and 2,520 (46.8%) patients did and did not receive chemical prophylaxis, respectively. The most commonly used medications for chemical prophylaxis were tranexamic acid (TXA) in 11 studies (64.7%), low-molecular-weight heparin in two studies, aspirin in two studies, heparin in one study, and enoxaparin in one study. The most prevalent route of administration was the intravenous (IV) route in 10 studies. VTE chemical prophylaxis agents were administered only perioperatively (single-dose regimen) in 8 (47.06%) [28,30,32–37] and postoperatively (continuous-dose regimen) in eight the other studies (47.06%) [26,27,29,38–42]. Moreover, seven studies (41.18%) mentioned that they used mechanical prophylaxis [26,27,29,34,39,40,42], although 3 (50%) of these studies failed to specify the type of mechanical prophylaxis (Table 3).
Surgical procedures
Among the included studies, 7 (41.2%) reported on decompression/discectomy. Moreover, the lumbar region emerged as the predominant focus of surgical interventions, with 10 studies (58.8%) specifically investigating degenerative conditions within the lumbar spine. In nine studies (52.9%), the average operative time was reported separately for both groups, with mean durations ranging from 49.5 to 295.3 minutes and 54.7 to 320.2 minutes in the chemoprophylaxis and control groups, respectively (Table 3).
Meta-analysis
Venous thromboembolism
Pooled for VTE incidence rates were 125/2,863 (4.4%) and 30/2,520 (1.2%) in the treatment and control groups, respectively. A meta-analysis comparing the chemoprophylaxis and control groups in terms of VTE prevention showed that chemical prophylaxis offered no significant benefits (RR, 1.09; 95% CI, 0.82 to 1.46; p=0.988) (Figs. 2–4). Subgroup analysis according to the type of medication used for chemical prophylaxis revealed that TXA showed no significant benefit (RR, 0.88; 95% CI, 0.64 to 1.22; p=0.994) based on 11 studies using TXA for chemical prophylaxis. Moreover, the six other studies that used medications other than TXA for chemical prophylaxis showed that these medications offered no significant benefit (RR, 1.40; 95% CI, 0.76 to 2.60; p=0.752) (Fig. 2). Subgroup analyses according to study design (Fig. 3), dosage regimen (patients who received only a single perioperative dose versus those who received continuous postoperative VTE chemoprophylaxis) (Fig. 4), and study quality (Appendices 4–11) showed consistent results (for subgroup analyses, all p>0.05). Table 2 presents the incidence rates of VTE specific to each study.
Spinal epidural hematoma
A total of 2,025 patients with spinal epidural hematoma were included. The pooled SEH rate for the treatment and control group was 0.2% (two out of 1,353 patients) and 0.5% (three out of 672 patients). Details regarding study-specific SEH rates can be found in Table 2.
Quality assessment
Fig. 5 and Table 4 show the risk of bias evaluations for all studies included. The NOS was applied to the seven cohort studies included in the meta-analyses (Table 4), with six rated as good quality and one as fair quality. Additionally, each of the ten RCTs was assessed using the RoB 2 tool (Fig. 5), with our results revealing that three trials had low overall bias risk, three raised some concerns, and four exhibited a high overall risk of bias.
Discussion
The extent of VTE risk in patients undergoing spine surgery has not been comprehensively elucidated and varies considerably based on the specific procedure and level of neurological impairment [43]. The consensus among most researchers is that minor spine surgery like discectomy and single-level laminectomy carries a low risk for VTE, whereas major spinal reconstruction involving multiple-level procedures and instrumented fusions increases this risk significantly [44,45]. Kepler et al. [43] recommends chemical prophylaxis for individuals anticipated to undergo extended surgeries and thoracolumbar surgeries across multiple levels. Despite the lack of a unanimous agreement on protocols for antithrombotic treatments, the vast majority of spine surgeons frequently recommend pharmacological DVT prevention after surgery [43]. However, a balance needs to be maintained between the potential for significant VTE events and the risks for postoperative bleeding and SEH following spine surgery [43].
The current review found no significant difference between the incidence of VTE and SEH following degenerative spine surgery among patients who did and did not receive chemical prophylaxis. In this study, we considered numerous VTE risk factors and aimed to narrow our inclusion criteria to ensure that all included studies focused on patients with shared characteristics. This approach was selected to ensure a comparable risk of thromboembolic events across all patients. First, we compared the baseline characteristics between the chemoprophylaxis and control groups. Although most baseline characteristics were similar between both groups, a significant sex imbalance was observed between the groups, with the proportion of women being higher than men in the chemoprophylaxis group. This difference may partly reflect clinical tendencies considering that high estrogen levels in females have been known to enhance coagulation factor production and reduce anticoagulant activity. Consequently, clinicians may be more inclined to prescribe anticoagulants after surgery in female patients [46].
Another common feature between the groups was the degenerative nature of their conditions. Through an extensive literature review, we identified some studies indicating that spinal tumors, spinal cord injuries, and their respective surgeries were associated with increased risk for bleeding and VTE [20,47–50]. Additionally, studies considered patients receiving VTE chemoprophylaxis alone after surgery or those receiving a single dose as separate groups. Although none of our results were statistically significant, the RR of 0.91 (95% CI, 0.49 to 1.68) for single perioperative VTE chemoprophylaxis surprisingly suggests its potentially better protective effects over a continuous-dose regimen after surgery. Nonetheless, the lack of a statistical significance and wide confidence intervals indicate uncertainty, emphasizing the need for further research to determine whether single-dose regimens are more effective than continuous-dose ones. Another critical risk factor for thromboembolic events that warrants attention is neurological impairment. However, we were unable to standardize this factor across studies given that many of the included studies did not explicitly report on the neurological status of patients. Instead, they only mentioned whether patients underwent pre- or postoperative neurological assessments.
Similar findings have been reported in previous reviews within this field. Colomina et al. [50] evaluated the efficacy of mechanical prophylaxis, chemical prophylaxis, or both for VTE prevention in patients undergoing elective spine surgery for spinal deformity or degenerative disease. Notably, they indicated a potential preference for employing mechanical over chemical prophylaxis in uncomplicated, non-traumatic, non-tumoral surgeries, mainly when the surgery involves opening the spinal canal [50]. Consistent with our findings, Solaru et al. [47] reported that the effectiveness of VTE prevention following elective spine surgery remains contentious, with no definitive guidelines established for its implementation. Furthermore, Rahmani et al. [20] failed to detect a significant difference in VTE occurrence between patients who did and did not receive chemoprophylaxis.
According to a survey by Adeeb et al. [18] on 370 neurosurgeons, 26.5% utilize the sequential compression device as their main thromboprophylaxis approach following elective spine surgery. However, the majority of respondents expressed their willingness to begin chemical prophylaxis on the first day following surgery [18]. Moreover, Khan et al. [51] showed evidence in support of chemoprophylaxis. In spine surgeries worldwide, anticoagulants have been commonly prescribed to not only reduce the risk of VTE but also mitigate potential medicolegal risks for surgeons [52,53]. However, many degenerative spine diseases, such as discopathy and stenosis, which require discectomy and laminectomy, respectively, can be surgically treated in a relatively short period of time, especially in the hands of spine experts [54]. Our findings help to hinder this preventive approach, which seems to be impractical, at least in degenerative spine surgeries. Despite the relatively low risk of DVT following routine elective spine surgery and the North American Spine Society guidelines’ acceptance of mechanical prophylaxis as a sufficient approach for preventing morbidity and mortality associated with VTE just before elective spine surgeries, numerous neurosurgeons have preferred to prescribe postoperative chemical prophylaxis on day one following spine surgeries [10,11,18,55]. Nevertheless, the risk of hemorrhagic complications, particularly SEH, remains controversial in this regard and should be a considerable concern for neurosurgeons [47,51]. Consequently, an individualized approach to pharmacological VTE prevention based on the procedure and each patient’s needs is recommended [10,53].
Some limitations of this review include heterogeneity in the spinal region undergoing surgical treatment, the timing and dosage of chemoprophylaxis, and the methods of VTE detection. Additionally, some potentially relevant patient-level data, such as smoking status, comorbidities, and neurological deficits, were inconsistently reported across studies, limiting our ability to perform more robust subgroup analyses. Other critical limitations include the moderate to low quality of most included RCTs and the limited number of studies reporting on hemorrhagic complications, such as SEH. Therefore, future high-quality RCTs should ensure balanced patient demographics, comprehensive reporting of baseline characteristics, and standardization in chemoprophylaxis protocols to support stronger recommendations.
Conclusions
The current review provides a comprehensive overview of studies investigating the role of chemical prophylaxis in preventing thromboembolic events in the context of degenerative spine surgeries. Despite the controversial nature of chemoprophylaxis in spine surgeries and the fairly low incidence of VTE in patients undergoing degenerative spine surgeries, our findings demonstrated that the use of chemical prophylaxis before, during, and after surgery may not significantly alter the rates of VTE and SEH in this particular group of patients. Nonetheless, further high-quality studies with more homogeneous characteristics are needed to recommend a more robust approach for chemical prophylaxis in these groups of patients.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
