Introduction
Chronic kidney disease (CKD) is a progressive and incurable condition with a high morbidity and mortality rate. Individuals with diabetes and hypertension are particularly susceptible to CKD, which is characterized by a reduction in glomerular filtration rate or kidney injury that has persisted for three months or more [1]. As a major global public health problem, CKD poses a significant threat, impacting millions of people. According to the Global Burden of Diseases, Injuries, and Risk Factors Study 2017, an estimated 697.5 million individuals worldwide suffered from CKD in 2017, corresponding to a prevalence of 9.1%. This enormous burden has far-reaching implications for global health, with CKD accounting for 35.8 million disability-adjusted life years and 1.2 million deaths in a single year [2]. The increasing number of hemodialysis patients with CKD requiring spine surgery is likely driven by aging populations and improved survival rates. This trend highlights the importance of understanding the unique challenges faced by this patient population after undergoing spine surgery, underscoring the need for further research and the development of targeted treatment strategies [3].
Spinal surgery for hemodialysis patients remains a challenging procedure despite advances in surgical and dialysis methods. The distinct challenges posed by end-stage renal disease (ESRD), including impaired bone metabolism, increased susceptibility to infection, and bone fragility, underlie this complexity. To achieve optimal outcomes, meticulous consideration and tailored surgical techniques are needed [4]. Hemodialysis patients undergoing spinal surgery have a high prevalence of comorbid conditions, such as diabetes, heart disease, anemia, and malnutrition. These comorbidities can significantly impact surgical outcomes, wound healing, and bone health. Therefore, careful evaluation of these aspects is crucial throughout the planning and implementation of surgical treatments to optimize outcomes for this complex patient population [5]. Individuals with CKD undergoing spine surgery are at a higher risk of morbidity and death [6,7]. Despite the widespread use of spine fusion procedures for various spinal disorders, there is a notable knowledge gap in the literature regarding bone union, a critical factor for good surgical outcomes. This paucity of research hinders our understanding of the complex biological mechanisms underlying bone healing and limits the development of evidence-based strategies to optimize fusion rates and improve patient outcomes [7–9]. Although the biological interplay between the kidneys and the skeleton is well-established, the mechanisms by which their interactions influence bone strength and metabolism remain poorly understood. This knowledge gap hampers the development of comprehensive treatment strategies for associated bone disorders and complicates monitoring bone health in individuals with renal impairment [10]. Ho et al. [11] in 2020 examined the results of instrumented interbody fusion and posterolateral fusion (PLF) in uremic patients undergoing surgery for degenerative lumbar illnesses. Although both methods showed promising fusion and clinical outcomes, the fusion rates were lower in ESRD patients than in non-ESRD patients, indicating possible fusion difficulties in this group. However, they did not assess the fusion rate according to kidney function [11]. This systematic review and meta-analysis aims to comprehensively examine the existing literature on spinal surgery outcomes in individuals with CKD and dialysis. A systematic search and critical appraisal of relevant studies were conducted to synthesize the available evidence on the impact of CKD and dialysis on surgical outcomes. Separate meta-analyses were performed to quantitatively assess the effects of CKD and dialysis on operative and postoperative outcomes. This comprehensive approach may yield valuable insights into the specific challenges and potential complications associated with spinal surgery in patients with CKD and dialysis, ultimately informing the development of optimal surgical strategies for this complex patient population.
Methods
Eligibility criteria
A PICOS approach was employed to guide the study selection process. The population (P) included adult patients undergoing spinal surgery, specifically comparing those with CKD to those without CKD and those on dialysis to those not on dialysis. The intervention groups (I) comprised patients with CKD and those undergoing dialysis, while the comparison groups (C) involved those without CKD and those not on dialysis, respectively. The primary outcomes (O) evaluated were operative time (minutes), estimated blood loss (mL), number of patients requiring blood transfusion, blood transfusion units needed, length of stay (days), and various complications such as deep vein thrombosis (DVT), pulmonary embolism (PE), sepsis, surgical site infection (SSI), reoperation, and in-hospital mortality. In terms of study design (S), eligible studies were comparative observational studies, such as prospective or retrospective cohort and case-control investigations.
Studies focusing exclusively on non-adult populations were excluded, as the risk factors and considerations differ significantly in pediatric patients. Additionally, studies with incomplete or inadequate data were excluded to prevent difficulties in extracting essential information required for the meta-analysis. Studies utilizing duplicate or overlapping samples were also excluded to avoid non-independent observations, which could violate statistical assumptions in the meta-analysis. Duplicate papers, non-original studies, abstract-only papers, animal studies, narrative or systematic reviews, meta-analyses, video reports, case reports, and case series were excluded. Lastly, studies with a high risk of bias were excluded to limit the potential skewing of results from low-quality evidence.
Data sources
PubMed, EMBASE, Scopus, and the Cochrane Library were systematically searched without language or publication status restrictions. In addition, the reference lists of the included studies and relevant reviews were manually screened to identify other potentially eligible papers.
Literature search strategy
The search strategy combined controlled vocabulary and keywords as follows: chronic kidney disease AND dialysis AND spinal surgery AND spine fusion AND complications AND outcomes AND operative time AND blood transfusion AND hospital stay AND deep vein thrombosis AND pulmonary embolism AND surgical site infection AND reoperation AND mortality AND perioperative care. Similar search strategies were adapted for EMBASE, Scopus, and Cochrane Library databases with appropriate database-specific terminology. Two reviewers independently screened the titles and abstracts of all retrieved articles. The full texts of potentially relevant reports were evaluated according to the prespecified eligibility criteria. Any discrepancies between reviewers during the initial screening or full-text evaluation stages were resolved through discussion until a consensus was reached.
Data extraction and data items
Two reviewers independently extracted data from all included studies. If consensus could not be achieved, a third reviewer assessed the disputed data items. The following baseline study characteristics were collected: study period, region, study type, type of surgery, duration of follow-up, comparison group, sample size, age, male-to-female numbers, conflict of interest, and funding information. The primary outcomes were operative time (minutes), estimated blood loss (mL), number of patients requiring blood transfusion, blood transfusion units required, length of stay (days), DVT, PE, sepsis, SSI, reoperation, and in-hospital mortality. In studies that included a CKD group, a control group without CKD was included for the CKD versus non-CKD comparison. For the dialysis versus non-dialysis comparison, we extracted ESRD patients as the dialysis group, as all studies had specified that ESRD patients were on dialysis; the control group for this comparison consisted of patients who were not on dialysis or did not have ESRD.
Quality assessment
Assessment of risk of bias in included studies
The methodological quality of the cohort studies was assessed using the Newcastle-Ottawa Scale (NOS). The NOS is a standardized tool to evaluate the risk of bias in nonrandomized studies, focusing on three domains: selection, comparability, and outcome. Studies were awarded stars based on how well they met specific criteria within these domains, with a maximum possible score of nine stars. Two reviewers independently conducted the quality assessment, and any discrepancies were resolved through discussion. Based on their final scores, studies were categorized as high quality (7–9 stars), moderate quality (4–6 stars), or low quality (0–3 stars). This assessment ensured that the conclusions of the meta-analysis were derived from studies with thoroughly evaluated methodologies [12].
Assessment of quality of evidence
The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system was employed to evaluate the quality of evidence for each outcome included in our meta-analysis [13]. The assessment considered factors such as study design, consistency of results, precision of estimates, potential biases, and the clinical relevance of the findings.
Statistical analysis
This meta-analysis was conducted using the Review Manager software ver. 5.4 (The Cochrane Collaboration, London, UK) [14]. Continuous outcomes were pooled using mean difference (MD) with a 95% confidence interval (CI), while dichotomous outcomes were pooled using odds ratios with corresponding 95% CI. The fixed effect model was employed for pooled analysis. However, if substantial heterogeneity was detected (p<0.1, I2>50%), the random effects model (DerSimonian-Liard method) was employed. This model considers variability within and between studies by allocating a larger standard error to the pooled estimate, thereby accommodating disparate effect sizes and giving more weight to studies with smaller sample sizes. Consequently, we must account for these potential discrepancies in our estimates. Heterogeneity was assessed using the Cochrane Q test (chi-square test) and quantified using the I2 statistic by Higgins et al. [14], which measures the proportion of variability in effect estimates attributed to heterogeneity rather than chance. A significant chi-square test was defined as an alpha level below 0.1, and substantial heterogeneity was considered present if the I2 value exceeded 50%. To explore the source of the heterogeneity, sensitivity analyses were performed by sequentially excluding one study at a time and rerunning the analysis for each scenario.
This ensured that the overall effect estimates were not unduly influenced by individual studies. Finally, to assess publication bias, we employed funnel plot analysis using Review Manager ver. 5.4 software. The funnel plot visually displays each study’s standard error on the y axis against its effect estimate on the x axis, allowing for inspection of asymmetry that may indicate publication bias.
Results
Study selection
The database searches yielded a total of 187 records. After removing 44 duplicates, the titles and abstracts of 143 remaining articles were screened. Following this initial screening, 108 studies were excluded based on predefined criteria, leaving 35 articles for full-text assessment. After a detailed full-text evaluation, 23 studies were excluded due to insufficient data, inappropriate study design, or irrelevant outcomes. This resulted in 12 studies that met all inclusion criteria. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) flow chart of the selection process is shown in Fig. 1.
Study characteristics
CKD versus non-CKD
The following four studies were included in the analysis comparing CKD patients with non-CKD individuals: Adogwa et al. [6] (2018), Martin et al. [15] (2015), De la Garza Ramos et al. [8] (2016), and Yu et al. [16] (2011). These studies involved 2,760 CKD patients (1,284 males and 1,474 females) compared to 164,750 non-CKD patients (76,644 males and 88,104 females). The mean age of patients in the CKD and non-CKD groups was 66.83 years and 62.01 years, respectively. The studies comparing CKD versus non-CKD groups were conducted in the United States and Taiwan between 2000 and 2016, with follow-up times ranging from 1 month to 32 months. More detailed information on patient characteristics and study specifics are presented in Tables 1 and 2.
Dialysis versus non-dialysis
In the comparison of dialysis versus non-dialysis patients, the following ten studies were analyzed: Alvi et al. [17] (2019), Chikuda et al. [18] (2012), Hori et al. [19] (2018), Inoue et al. [20] (2018), Mitchell et al. [21] (2020), Nyam et al. [9] (2019), Puvanesarajah et al. [22] (2016), De la Garza Ramos et al. [8] (2016), Yoshihara and Yoneka [23] (2018), and Yu et al. [16] (2011). These studies included 11,266 dialysis patients (6,010 males and 4,677 females; mean age, 64.15 years) and 5,136,073 non-dialysis patients (2,414,109 males and 2,718,514 females; mean age, 61.36 years).
The follow-up times in studies comparing dialysis versus non-dialysis patients ranged from 30 days to over 2 years, with some studies focusing only on short-term outcomes such as hospital stays. The studies were conducted mainly in the United States, Japan, and Taiwan, highlighting a broad geographic distribution. More detailed information on patient characteristics and study specifics are presented in Tables 1 and 2.
Risk of bias assessment
The quality assessment of the included studies using the NOS revealed high methodological rigor in most studies. Martin et al. [15] (2015), Mitchell et al. [21] (2020), and Puvanesarajah et al. [22] (2016) each received the highest score of 9 stars, indicating high quality across all domains. Studies by Alvi et al. [17] (2019), Hori et al. [19] (2018), Eric Nyam et al. [9] (2019), De la Garza Ramos et al. [8] (2016), Yoshihara and Yoneka [23] (2018), and Yu et al. [16] (2011) scored eight stars, showing strong performance, though some lacked follow-up adequacy. Chikuda et al. [18] (2012) received seven stars, reflecting good quality with minor limitations. Adogwa et al. [6] (2018) and Inoue et al. [20] (2018) scored 6 and 5 stars, respectively, indicating moderate quality with potential biases in selection and outcome domains. Overall, the studies included in the meta-analysis were of high to moderate quality, supporting the robustness of the analysis (Supplement 1).
Grade summary
The GRADE assessment of the studies in this meta-analysis revealed varying levels of certainty in the evidence regarding spinal surgery outcomes in patients with CKD and those undergoing dialysis. Although the studies were generally of moderate to high quality according to the NOS, the certainty of evidence ranged from low to moderate across different outcomes. In the dialysis versus non-dialysis comparison, most outcomes had low certainty of evidence due to high heterogeneity, while reoperation, SSI, and estimated blood loss had moderate certainty. In the CKD versus non-CKD comparison, outcomes such as length of hospital stay, operative time, reoperation, and red blood cell (RBC) units required were supported by moderate certainty of evidence, with other outcomes having low or very low certainty. These findings highlight the variability and potential biases in the evidence (Tables 3, 4).
Primary outcomes
CKD versus non-CKD
Estimated blood loss: The pooled analysis of estimated blood loss during spinal surgery showed no significant difference in blood loss between patients with CKD and those without CKD. The pooled MD was 104.99 mL (95% CI, −153.43 to 363.42; p=0.43), based on 361 patients and two studies (Fig. 2A), with low certainty of evidence. Notably, there was low heterogeneity among the studies (I2=0%).
Operative time: Pooled analysis revealed that patients with CKD had a significantly shorter operative time compared to those without CKD. The pooled MD was −12.63 minutes (95% CI, −14.49 to −10.78; p<0.00001), based on 164,750 patients and four studies (Fig. 2B), with moderate certainty of the evidence. There was moderate heterogeneity among the studies (I2=37%).
RBC units required: Pooled analysis revealed that patients with CKD required significantly more RBC units than those without CKD. The pooled MD was 1.17 units (95% CI, 0.40–1.95; p=0.003, based on 361 patients and two studies (Fig. 2C), with moderate certainty of evidence. The heterogeneity was low (I2=0%).
Length of hospital stay: Patients with CKD had a significantly longer hospital stay than those without CKD. The pooled MD was 1.51 days (95% CI, 1.28–1.74; p<0.00001), based on 164,188 patients and three studies (Fig. 2D), with moderate certainty of evidence. There was low heterogeneity among the studies (I2=0%).
Deep vein thrombosis: Patients with CKD had higher, though not statistically significant, odds of developing postoperative DVT compared to controls without CKD. The pooled OR was 5.21 (95% CI, 0.74–36.93; p=0.10), based on 167,442 patients and three studies (Fig. 3A), with low certainty of evidence. However, there was high heterogeneity among the studies in this pooled analysis (I2=91%). The heterogeneity was best resolved (I2=0%) by excluding the study by Martin et al. [15], following which the overall odds of DVT became significantly higher among CKD patients (OR, 10.8; 95% CI, 5.93–19.66; p<0.00001) (Supplement 2). This indicates that the initial insignificance was driven by the study by Martin et al. [15].
Pulmonary embolism: There was no significant difference between patients with CKD and those without CKD regarding the incidence of PE. The pooled OR was 4.94 (95% CI, 0.11–214.08; p=0.41), based on 167,442 patients and three studies (Fig. 3B), with very low certainty of evidence. However, there was very high heterogeneity among the studies (I2=97%). On sensitivity analysis, the heterogeneity was best resolved (I2=0%) by excluding the study by De la Garza Ramos et al. [8]. However, the overall difference between the CKD and non-CKD groups remained insignificant after the exclusion of the study by De la Garza Ramos (OR, 1.19; 95% CI, 0.45–3.11; p=0.73) (Supplement 3).
Sepsis: Patients with CKD had significantly higher odds of developing sepsis compared to those without CKD. The pooled OR was 4.44 (95% CI, 0.23–86.29; p=0.33), based on 167,442 patients (Fig. 3C), with very low certainty of evidence. There was very high heterogeneity among the studies (I2=95%). This heterogeneity was best resolved (I2=0%) by excluding the study by De la Garza Ramos et al. [8]. However, the overall difference in sepsis occurrence remained insignificant (OR, 1.31; 95% CI, 0.54–3.19; p=0.55) (Supplement 4).
Surgical site infection: Patients with CKD had higher, though not statistically significant, odds of developing SSI compared to those without CKD. The pooled OR was 1.51 (95% CI, 0.99–2.31; p=0.06), based on 3,683 patients and three studies (Fig. 3D), with low certainty of evidence. There was low heterogeneity among the included studies (I2=0%)
Reoperation: Patients with CKD had higher, though not statistically significant, odds of undergoing reoperation compared to those without CKD. The pooled OR was 1.84 (95% CI, 0.98–3.44; p=0.06), based on 167,149 patients and two studies (Fig. 3E), with moderate certainty of the evidence. The heterogeneity was moderate (I2=70%).
In-hospital mortality: Patients with CKD had higher odds of death compared to those without CKD, although statistically significant. The pooled OR was 1.13 (95% CI, 0.42–3.04; p=0.80) based on 3,390 patients and two studies (Fig. 3F), with low certainty of evidence and low heterogeneity (I2=0%).
Dialysis versus non-dialysis
Estimated blood loss: There was no significant difference in the estimated blood loss during spinal surgery between dialysis patients and those not on dialysis. The pooled MD in blood loss was 8.42 mL (95% CI, −16.59 to 33.43; p=0.5), based on 1,107 patients and three studies (Fig. 4A), with moderate certainty of evidence. There was low heterogeneity among the studies (I2=0%).
Operative time: Patients on dialysis had a marginally significant longer operative time compared to those not on dialysis. The pooled MD was 10.65 minutes (95% CI, 0.05–21.26; p=0.05), based on 217,709 patients and six studies (Fig. 4B), with low certainty of evidence. There was moderate to high heterogeneity among the studies (I2=77%). On sensitivity analysis, the heterogeneity was reduced but not resolved (I2=65%) after excluding the study by Chikuda et al. [18]; however, the overall MD became statistically insignificant (MD, 6.79, 95% CI, −3.11 to 16.70; p=0.18) (Supplement 5).
Number of patients requiring RBC transfusion: Dialysis patients had significantly higher odds of RBC transfusion requirement than those not on dialysis. The pooled OR was 2.65 (95% CI, 1.57 to 4.46; p=0.0002), based on 68,742 patients and three studies (Fig. 4C), with low certainty of evidence. This indicates that patients on dialysis were more than twice as likely to require RBC transfusion. However, there was substantial heterogeneity among the included studies (I2=94%). On sensitivity analysis, the heterogeneity was reduced but not resolved (I2=87%) after excluding the study by Inoue et al. [20]. The overall higher transfusion rate among CKD patients remained significant (OR, 1.74; 95% CI, 1.31–2.31; p=0.0001). However, the results became nonsignificant after excluding either the study by Chikuda et al. [18] (OR, 5.49; 95% CI, 0.40–75.51; p=0.20) (Supplement 6A) or Puvanesarajah et al. [22] (OR, 6.31; 95% CI, 0.61–65.37; p=0.12), indicating that the overall results depended on either of these two studies (Supplement 6B).
Length of hospital stay: Patients on dialysis had a significantly longer hospital stay than those not on dialysis. The pooled MD was 5.89 days (95% CI, 3.58–8.20; p<0.0000), based on 521,659 patients and seven studies (Fig. 4D), with low certainty of evidence. There was very high heterogeneity among the studies (I2=98%). On sensitivity analysis, excluding the study by Yoshihara and Yoneka [23] slightly reduced the heterogeneity (I2=92%). These findings indicated no undue influence of any single study on the results.
DVT: Patients on dialysis had significantly higher odds of developing DVT compared to those not on dialysis. The pooled OR was 6.45 (95% CI, 1.72–24.20; p=0.006), based on 206,258 patients and three studies (Fig. 5A), with low certainty of evidence. However, there was very high heterogeneity among the studies (I2=98%). The heterogeneity was not resolved or remarkably reduced by excluding any study in the sensitivity analysis. However, the results became statistically insignificant after excluding the study by Mitchell et al. [21] (OR, 7.23; 95% CI, 0.32–162.31; p=0.21) (Supplement 7).
Pulmonary embolism: Patients on dialysis had significantly higher odds of developing PE compared to those not on dialysis. The pooled OR was 6.48 (95% CI, 1.13–37.14; p=0.04), based on 519,932 patients and five studies (Fig. 5B), with low certainty of the evidence. The heterogeneity among the studies was very high (I2=98%). On sensitivity analysis, the heterogeneity was not resolved or remarkably reduced after the exclusion of any study. However, the results became statistically insignificant after the exclusion of the studies by Yoshihara and Yoneka [23], De la Garza Ramos et al. [8], or Mitchell et al. [21].
Sepsis: Patients on dialysis had significantly higher odds of developing sepsis compared to those not on dialysis. The pooled OR was 21.56 (95% CI, 2.35–197.96; p=0.007), based on 5,116,937 patients and four studies, with very low certainty of evidence. However, the heterogeneity among the studies was extremely high (I2=100%) (Fig. 5C). On sensitivity analysis, there was no remarkable reduction in heterogeneity after the exclusion of any individual study.
Surgical site infection: Patients on dialysis had significantly higher odds of developing SSI compared to those not on dialysis. The pooled OR was 2.14 (95% CI, 1.72–2.66; p<0.0000), based on 69,949 patients and five studies (Fig. 5D), with moderate certainty of evidence. There was low heterogeneity among the studies (I2=18%).
Re-operation: Patients on dialysis had significantly higher odds of undergoing reoperation than those not on dialysis. The pooled OR was 7.04 (95% CI, 2.49–19.86; p=0.0002), based on 1,039 patients and two studies (Fig. 5E), with moderate certainty of evidence. There was low heterogeneity among the studies (I2=0%).
In-hospital mortality: Patients on dialysis had significantly higher odds of in-hospital death compared to those not on dialysis. The pooled OR was 16.71 (95% CI, 6.23–44.85; p<0.00001), based on 525,359 patients and seven studies (Fig. 5F), with low certainty of evidence. However, the heterogeneity among the studies was very high (I2=99%). On sensitivity analysis, the exclusion of any single study was not found to resolve the heterogeneity.
Publication bias
Publication bias was assessed using funnel plots for outcomes with three or more studies. Visual inspection of the funnel plots revealed asymmetrical distributions for both CKD and dialysis-dependent patients for several outcomes. In the CKD group, funnel plots for operative time, length of stay, and complications (DVT, PE, and SSI) showed moderate asymmetry (Supplement 8A, B, and C, respectively), with studies scattered asymmetrically around the mean effect size. Similarly, in the dialysis group, asymmetrical distributions were noted mainly for mortality and sepsis outcomes (Supplement 8D and E, respectively), suggesting potential publication bias. However, the interpretation of these funnel plots should be approached with caution due to the high heterogeneity observed in most outcomes (I2 ranging from 70% to 100%).
Discussion
This systematic review and meta-analysis, synthesizing data from 12 studies and over 5 million patients, provides robust evidence of the significant impact of CKD and dialysis on spinal surgery outcomes. The results demonstrate a consistent pattern of increased risk and complications in patients with CKD or those undergoing dialysis, compared to their counterparts without CKD or those not on dialysis, respectively. Patients with CKD pose distinct challenges for spine surgeons. While they tend to undergo shorter surgical procedures [5,24], CKD patients face a higher risk of bleeding, necessitating more RBC transfusions [9,15,25]. Furthermore, they experience extended hospital stays, likely due to delayed wound healing, increased postoperative complications, and the need for more intensive postoperative management [3,4]. Patients undergoing dialysis exhibit an even more pronounced risk profile. They experience longer operative times, higher RBC transfusion requirements, and substantially prolonged hospital stays [18,26]. Our analysis revealed significantly higher odds of several serious complications in dialysis patients compared to those not on dialysis, including DVT, PE, SSI, reoperation, and in-hospital mortality [17,27]. These findings underscore the profound impact of dialysis on spinal surgery outcomes and underline the need for heightened awareness and individualized care. This meta-analysis boasts several strengths, including its broad scope, large sample size, and rigorous methodology. The use of NOS and GRADE systems ensured a robust assessment of methodological quality and certainty of evidence [12,28]. However, several limitations must be acknowledged. The reliance on administrative databases introduces the inherent risk of miscoding [29], and the accuracy of diagnostic codes for specific complications may be inconsistent [30]. Furthermore, the limited number of studies explicitly comparing CKD with non-CKD patients for spinal fusion surgeries, the paucity of long-term follow-up data, and the potential for publication bias limit the generalizability of the findings [31]. This study offers crucial insights for surgeons managing CKD and dialysis patients undergoing spinal surgery. Preoperative optimization, including correction of anemia and meticulous blood management strategies, is vital for CKD patients [15,25]. Dialysis patients require even more intensive perioperative care, and the decision to proceed with surgery should be made with caution, considering the high risk of life-threatening complications [18,26]. Shared decision-making, involving a thorough discussion of risks and benefits with patients and their families, is fundamental to ethical and patient-centered care. Future research should prioritize prospective studies with standardized protocols and extended follow-up durations to overcome the limitations of this meta-analysis. Investigating the impact of various dialysis modalities, the optimal timing of dialysis sessions relative to surgery, and the role of multidisciplinary preoperative optimization are crucial aspects for further research [9]. Moreover, research examining the long-term outcomes of spinal surgery in this population is critically needed, including assessments of functional status, quality of life, and long-term morbidity and mortality. This meta-analysis offers compelling evidence that both CKD and dialysis are significant risk factors for adverse outcomes following spinal surgery. These findings underscore the need for heightened awareness among surgeons and individualized care for patients with renal impairment. Future research efforts should focus on refining risk stratification, optimizing perioperative care, and investigating the long-term consequences of spinal surgery in this vulnerable population.
Conclusions
This meta-analysis highlights the profound impact of CKD and dialysis on spinal surgery outcomes, with increased risks of complications, prolonged operative times, and mortality. Heightened awareness, meticulous preoperative optimization, and tailored perioperative care are crucial to minimize risk in this highly complex patient population. Shared decision-making based on risk/benefit discussions with patients and families is imperative. Future research may emphasize prospective studies with more extended follow-up periods, focusing on refining perioperative care protocols, optimizing the timing of dialysis relative to surgery, and assessing long-term outcomes to advance the quality of care for CKD and dialysis patients undergoing spinal surgery. With more research and more collaboration, such patients’ outcomes can be improved, and more concrete and effective surgical strategies can be developed for this vulnerable population.
Key Points
Chronic kidney disease (CKD) and dialysis significantly increase surgical risks in spinal surgery patients compared to those with normal kidney function.
Dialysis patients demonstrate markedly higher odds of serious complications including deep vein thrombosis (odds ratio [OR], 6.45), pulmonary embolism (OR, 6.48), and in-hospital mortality (OR, 16.71).
CKD patients require more blood transfusions and experience longer hospital stays, while dialysis patients have substantially prolonged operative times and hospitalization.
Enhanced perioperative care protocols and careful patient selection are essential for managing CKD and dialysis patients undergoing spinal surgery.
Future research should focus on optimizing dialysis timing relative to surgery and developing standardized perioperative care protocols for this high-risk population.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
Supplement
