Patient factors rather than surgical approach are associated with readmission and reoperation for cervical spondylotic myelopathy

Introduction

The prevalence of cervical spondylotic myelopathy (CSM) is increasing annually and contributes significantly to societal burden of disability (1). Many surgical approaches exist to treat cervical myelopathy, including posterior cervical decompression and fusion (PCDF), posterior cervical decompression (PCD), anterior cervical discectomy and fusion (ACDF), or a combination of these approaches. Reoperation and readmission contribute significantly to rising healthcare costs and, in situations of relative equipoise, should be considered when selecting a surgical approach (2). In this single-center retrospective cohort study, we analyze rates of readmission and reoperation for patients undergoing PCDF, PCD, and ACDF.

Various factors influence a surgeon’s decision to perform anterior vs. posterior surgery and whether to fuse or perform decompression alone for cervical myelopathy. The anterior approach allows for decompression of neural elements and simultaneous stabilization of the anterior column of the spine (3). An anterior approach may be preferred if the compressive pathology is ventral and centrally located, or if significant loss of disc height or lordosis is present. Posterior decompression alone decompresses the spinal cord while avoiding the potential for adjacent segment disease, pseudoarthrosis, and instrumentation failure. While this approach allows for preservation of cervical mobility, the posterior approach, with or without fusion, requires dissection of the cervical musculature, which is implicated in greater post-operative pain (4-6). Without stabilization and complete access to the disc space, recurrence of symptoms is possible and may require additional surgery (7).

Although there may be clinical or radiographic indications for selecting one procedure over another, it remains unknown whether there exist differences in risk of reoperation and readmission between these surgical approaches. This single institutional retrospective cohort of ACDFs, open PCDFs, and open PCDs compares the incidence and indications for reoperation at 90-day, 1-year, and 2-year follow-up as well as readmission rates at 30 days, 90 days, and 1 year in patients with CSM. We present this article in accordance with the STROBE reporting checklist (available at https://jss.amegroups.com/article/view/10.21037/jss-24-101/rc).

Methods

Sample

A retrospective study of patients at a tertiary-care institution who underwent ACDF, PCD, or PCDF between C3 and C7 from June 2013 to April 2019 for myelopathy or myeloradiculopathy was conducted. Six surgeons’ cases were included, and all surgeons performed all types of procedures. Electronic medical records were reviewed to identify qualifying patients. Patients were excluded from the study if they were younger than 18 years old, if the indication for surgery was for anything other than myelopathy or myeloradiculopathy, and if the etiology was for anything other than degenerative disease. Additionally, minimally invasive (i.e., endoscopic and tubular approaches) and motion preserving (i.e., total disc replacement and laminoplasty) approaches were not included in this study in order to homogenize the dataset. Patients who were missing data were excluded from analysis. Patient symptoms and the procedures performed were defined by the pre-operative clinic and operative notes by the surgeon. Reoperations were recorded at 30-day, 90-day, 1-year, and 2-year intervals. Reoperations were defined as both index-level revisions as well as adjacent segment surgeries. Readmissions were recorded at 30-day, 90-day, and 1-year intervals. Both reoperations and readmissions were tracked within our single-center database, which did not include outside hospital encounters.

Patient data collected from the medical record included gender, age, race, mean body mass index (BMI), American Society of Anesthesiologists (ASA) score, and smoking status. Other pre-operative comorbidities extracted include chronic kidney disease (CKD), congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), peripheral vascular disease (PVD), diabetes mellitus (DM), prior cerebral vascular accident (CVA), cognitive decline, rheumatologic disorders, human immunodeficiency virus (HIV), and prior myocardial infarction (MI). Charleston Comorbidity Indices (CCI) were calculated. Intraoperative data, including the number of levels fused and operative time, were collected. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Institutional review board approval was obtained from the Hospital of the University of Pennsylvania (IRB #857838), and informed consent was waived due to the retrospective nature of the study.

Statistical analysis

Patients undergoing either ACDF, PCDF, or PCD between 2014 and 2021 were identified using Current Procedural Terminology (CPT) codes. Circumferential approaches were excluded from analysis. Charts were reviewed for baseline patient characteristics (age, gender, ethnicity, BMI, smoking status, and medical comorbidities), surgical characteristics (operative levels and procedure length), and perioperative outcomes [length of stay (LOS), reoperation rates and indications, and readmission rates]. Baseline demographic variables were compared between groups using t-tests for continuous characteristics and χ² testing for categorical characteristics. Similar analyses were used to compare outcome variables between groups. To control for significant baseline demographic differences between groups, linear regressions were performed using demographic variables, surgical characteristics (type of procedure and levels treated) with LOS as the dependent variable. Similarly, logistic regressions were performed with either readmission or reoperation as the dependent variable while controlling for BMI, age, number of levels treated, CCI, and anterior vs. posterior approach, which could act as possible confounding characteristics. The logistic regression for readmission at any point also includes covariates controlling for CHF, CKD, COPD, and DM. Continuous variables were evaluated for distributional assumptions, and non-parametric methods were used when appropriate to account for deviations from normality. All statistical analyses were performed on SPSS (IBM, Armonk, NY, USA).

Results

A total of 366 patients meeting the inclusion criteria were identified (86 PCDFs, 105 PCDs, 175 ACDFs). The groups differed in age with PCDF and PCD groups being significantly older than the ACDF group (P<0.001). There was no difference between groups with respect to mean BMI. There was a significant difference in sex between groups with patients undergoing PCD being more likely to be male compared to PCDF and ACDF (72.4% vs. 58.1% and 53.1%, P=0.009; Table 1). Patients undergoing PCDF and PCD both had significantly higher CCIs than ACDF patients (P<0.001, P<0.001). Other comorbidities, including DM, CKD, COPD, and CHF, are displayed in Figure 1. PCDs were more likely to have pre-operative DM (16.2%, P=0.03) compared to the other groups. PCDF and PCD were more likely to have pre-operative CHF and PVF than ACDF patients (18.6% and 13.3% vs. 5.7%, P=0.007; and 16.3% and 16.2% vs. 8.3%, P=0.04). Given the significant differences between groups with respect to age, gender, and CCI, subsequent regression analyses were performed to control for possible confounding effects of these variables.

Figure 1 Comorbidities by surgery type. The most commonly observed comorbidity across all groups was COPD. ACDF, anterior cervical discectomy and fusion; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; PCD, posterior cervical decompression; PCDF, posterior cervical decompression and fusion; PVD, peripheral vascular disease.

Reoperation rate was significantly higher in the PCDF group with 16.3% (14 out of 72 patients) requiring surgery within 30 days compared to 1.9% and 4.6% in the PCD and ACDF groups, respectively (P<0.001). At 1 year, the reoperation rate in the PCDF group remained significantly higher than the other groups (P=0.003) (Table 2). There was, however, no significant difference between groups at 2 years with respect to reoperation rates. Patients undergoing ACDF had a longer time between initial surgery and reoperation compared to the two posterior groups, which trended towards significance (P=0.052). To control for the effects of age and comorbidity, which significantly differed between the groups, logistic regression models controlling for age, CCI, BMI, and number of levels treated, showed that the surgical approach used (anterior vs. posterior) was no longer significantly associated with reoperation, while the number of levels treated was significantly associated with 30-day reoperation.

Reoperations were most frequently performed in the PCDF group for wound complications, 38% vs. 21% and 10% in the PCD cohort and ACDF cohorts, respectively. Reoperation for recurrent symptoms was the most common reason for reoperation in the PCD cohort, accounting for 60% of the reoperations vs. 35% in the ACDF group and 11% in the PCDF group. Adjacent segment disease was noted in two patients in the ACDF group and no patients in the PCDF group. Hardware failure or pseudoarthrosis was noted in one patient in the PCDF group and two patients in the ACDF group (Figure 2). Hardware failures occurred on average 181 days after the procedure with a range of 21–342 days. Three patients required reoperation following PCDF for post-operative hematoma, compared to one patient in the ACDF cohort and no patients in the PCD group.

Figure 2 Reasons for reoperation by surgery type. Reoperation for recurrent symptoms was not statistically different between groups. There was significantly more reoperation for recurrent symptoms, ASD, pseudoarthrosis, and need for stabilization in the ACDF group. There were more reoperations for hardware malposition, hematoma, and wound complications in the PCDF group. ACDF, anterior cervical discectomy and fusion; ASD, adjacent segment disease; PCD, posterior cervical decompression; PCDF, posterior cervical decompression and fusion.

Overall readmission rate was 30.3% over 1 year, and there was no significant difference between groups (ACDF 29.1%, PCDF 34.9%, and PCD 29.5%). At 30 days, there were significantly more readmissions in the PCDF group (17.4% compared to ACDF 6.3% and PCD 7.6%). At 90 days and 1 year, there was no significant difference in readmission rates between groups (Table 2). Additionally, BMI, number of levels, and anterior vs. posterior approach were not significant predictors of 90-day readmission rates. Worse scores on the CCI were associated with readmission at all time points. BMI, number of levels treated, and the procedure type did not have a significant association with the overall rate of any-time readmission on regression analysis. Results of multivariable logistic regression analyses are summarized in Table 3. Additional regression analyses were performed to examine the effects of specific comorbidities. In these models, CHF and DM were independent predictors of readmission at 30 days, 90 days, and 1 year (P=0.003, P<0.001, P<0.001 for CHF; P=0.002, P=0.01, P<0.001 for DM). CKD and COPD were also independent predictors of readmission at 90 days and 1 year (P=0.004, P=0.001 for CKD; P=0.001, P<0.001 for COPD).

LOS was significantly shorter in the ACDF group compared to the PCDF group (5 days after PCDF, 4 days after PCD, and 3 days after ACDF, P=0.02). With a linear regression model, predictors of LOS were higher CCI and increased age when comparing both fusion to non-fusion patients and comparing anterior to posterior approaches. In these models, the surgical approach was not associated with LOS.

Discussion

The present study is one of the largest retrospective reviews from a single center to examine reoperation and readmission rates in patients undergoing surgical intervention for CSM and one of the only studies to report 90-day readmission rates after ACDF and PCDF from the same healthcare system. Our findings suggest that underlying comorbidities, more so than surgical approach, contribute to readmission and reoperation rates.

In our cohort, reoperation rates were significantly higher in the PCDF group at 30 days (16.3%) and 1 year (17.4%) but not at 2 years. Other studies, including meta-analyses, have reported a wide range of reoperation rates after ACDF ranging from 1.8% to 16.8% (8-14) and from 4% to 7% after PCDF (13,15,16). One database study of CSM showed that reoperation rates were significantly higher in posterior approaches at 90 days but not at 6 months, 1 year, or 2 years (17). Reoperations in the PCDF group were most commonly performed for wound complications. This finding is consistent with prior literature that cites rates of post-operative wound complications in PCDF as high as 19.4% (18).

Similar to prior studies, our cohort of patients undergoing posterior approaches to the cervical spine tended to be older and had more comorbidities. Consequently, in univariable analyses, there were significant associations between posterior approaches and reoperation rates. However, upon controlling for these important confounders, there was no significant effect of surgical approach on reoperation rates. Rather, patient age and comorbidities, notably DM, CHF, CKD, and COPD, appear to be the drivers of reoperation in the CSM population.

Delving further into the effects of comorbidities on surgical outcomes in the CSM population, prior studies have shown that ischemic heart disease and higher number of levels treated in PCDFs were associated with increased LOS (18). In our cohort, the mean number of operative levels was significantly higher in both posterior approaches than in the ACDF group. LOS was significantly shorter in the ACDF group (3 days) than in the PCDF group (5 days). However, in multivariable linear regression models, age and CCI were significantly associated with increased LOS, whereas surgical approach was not, again suggesting that the observed differences in LOS among surgical groups are secondary to differences in patient populations rather than surgical approach or number of levels treated.

In the present literature, 30-day readmission rates range from 1.9% to 5.1% after ACDF and from 5.4% to 11.2% after PCDF, and 90-day rates range from 0.6% to 7.7% after ACDF and from 3.13% to 16.9% after PCDF (18-20). Our readmission rates were generally higher than those previously reported, particularly at 30-day (17.4% after PCDF and 6.3% after ACDF). The higher readmission rates may reflect our tertiary referral population with greater comorbidity burden and more complete capture of readmissions within an integrated healthcare system. In our cohort, CHF and DM were predictors of readmission at all time points, and CKD and COPD were predictive at 90 days and 1 year. Unlike our findings on reoperation rates, the number of levels treated did not impact readmission rates nor did procedure type or BMI. In fact, CCI and related comorbidities were the only variables that predicted readmission.

This study has inherent limitations to retrospective design. As the data is from a single center, it may not be generalizable to other clinical settings. Finally, we did not analyze radiographic features of patients, which may have provided additional insight into the indications for reoperation. Additionally, potential coding inaccuracies in CPT selection and the absence of patient-reported outcome measures represent further limitations of this retrospective study. Future studies should focus on prospective reduction of readmission rates and reoperation within certain at-risk patient populations.

Conclusions

The present study is one of the largest single-center retrospective analyses of reoperation rates up to 2 years and readmission rates at 30 days, 90 days, and 1 year postoperatively after surgery for CSM. Contrary to some of the previous literature on this topic, which suggests that posterior approaches may have increased rates of readmission and reoperation, our data suggest that patient characteristics and morbidities, such as age and CCI, may be more important determining factors rather than surgical approach.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jss.amegroups.com/article/view/10.21037/jss-24-101/rc

Data Sharing Statement: Available at https://jss.amegroups.com/article/view/10.21037/jss-24-101/dss

Peer Review File: Available at https://jss.amegroups.com/article/view/10.21037/jss-24-101/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jss.amegroups.com/article/view/10.21037/jss-24-101/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Institutional review board approval was obtained from the Hospital of the University of Pennsylvania (IRB #857838), and informed consent was waived due to the retrospective nature of the study.

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. Youssef JA, Heiner AD, Montgomery JR, et al. Outcomes of posterior cervical fusion and decompression: a systematic review and meta-analysis. Spine J 2019;19:1714-29. [Crossref] [PubMed]
2. King JT Jr, Abbed KM, Gould GC, et al. Cervical spine reoperation rates and hospital resource utilization after initial surgery for degenerative cervical spine disease in 12,338 patients in Washington State. Neurosurgery 2009;65:1011-22; discussion 1022-3. [Crossref] [PubMed]
3. Katz AD, Mancini N, Karukonda T, et al. Comparative and Predictor Analysis of 30-day Readmission, Reoperation, and Morbidity in Patients Undergoing Multilevel ACDF Versus Single and Multilevel ACCF Using the ACS-NSQIP Dataset. Spine (Phila Pa 1976) 2019;44:E1379-87. [Crossref] [PubMed]
4. Ruetten S, Komp M, Merk H, et al. Full-endoscopic cervical posterior foraminotomy for the operation of lateral disc herniations using 5.9-mm endoscopes: a prospective, randomized, controlled study. Spine (Phila Pa 1976) 2008;33:940-8. [Crossref] [PubMed]
5. Skovrlj B, Gologorsky Y, Haque R, et al. Complications, outcomes, and need for fusion after minimally invasive posterior cervical foraminotomy and microdiscectomy. Spine J 2014;14:2405-11. [Crossref] [PubMed]
6. Rossi V, Coric D. Minimally Invasive Posterior Cervical Fusion Strategies. Neurosurgery 2025;96:S42-50. [Crossref] [PubMed]
7. Wang TY, Lubelski D, Abdullah KG, et al. Rates of anterior cervical discectomy and fusion after initial posterior cervical foraminotomy. Spine J 2015;15:971-6. [Crossref] [PubMed]
8. Maharaj MM, Mobbs RJ, Hogan J, et al. Anterior cervical disc arthroplasty (ACDA) versus anterior cervical discectomy and fusion (ACDF): a systematic review and meta-analysis. J Spine Surg 2015;1:72-85. [Crossref] [PubMed]
9. Xie L, Liu M, Ding F, et al. Cervical disc arthroplasty (CDA) versus anterior cervical discectomy and fusion (ACDF) in symptomatic cervical degenerative disc diseases (CDDDs): an updated meta-analysis of prospective randomized controlled trials (RCTs). Springerplus 2016;5:1188. [Crossref] [PubMed]
10. Zhong ZM, Zhu SY, Zhuang JS, et al. Reoperation After Cervical Disc Arthroplasty Versus Anterior Cervical Discectomy and Fusion: A Meta-analysis. Clin Orthop Relat Res 2016;474:1307-16. [Crossref] [PubMed]
11. Wu TK, Liu H, Wang BY, et al. Minimum four-year subsequent surgery rates of cervical disc replacement versus fusion: A meta-analysis of prospective randomized clinical trials. Orthop Traumatol Surg Res 2017;103:45-51. [Crossref] [PubMed]
12. Puvanesarajah V, Jain A, Cancienne JM, et al. Complication and Reoperation Rates Following Surgical Management of Cervical Spondylotic Myelopathy in Medicare Beneficiaries. Spine (Phila Pa 1976) 2017;42:1-7. [Crossref] [PubMed]
13. Park MS, Ju YS, Moon SH, et al. Reoperation Rates After Anterior Cervical Discectomy and Fusion for Cervical Spondylotic Radiculopathy and Myelopathy: A National Population-based Study. Spine (Phila Pa 1976) 2016;41:1593-9. [Crossref] [PubMed]
14. Gordon AM, Elali FR, Saleh A. Revision Rates After Single-Level Cervical Disc Arthroplasty Versus Anterior Cervical Discectomy and Fusion: An Observational Study With 5-Year Minimum Follow-Up. Spine (Phila Pa 1976) 2025;50:19-25. [Crossref] [PubMed]
15. Dewan MC, Rattani A, Mekary R, et al. Global hydrocephalus epidemiology and incidence: systematic review and meta-analysis. J Neurosurg 2019;130:1065-79. [Crossref] [PubMed]
16. Lau D, Winkler EA, Than KD, et al. Laminoplasty versus laminectomy with posterior spinal fusion for multilevel cervical spondylotic myelopathy: influence of cervical alignment on outcomes. J Neurosurg Spine 2017;27:508-17. [Crossref] [PubMed]
17. Wadhwa H, Sharma J, Varshneya K, et al. Anterior Cervical Discectomy and Fusion Versus Laminoplasty for Multilevel Cervical Spondylotic Myelopathy: A National Administrative Database Analysis. World Neurosurg 2021;152:e738-44. [Crossref] [PubMed]
18. Zaki O, Jain N, Yu EM, et al. 30- and 90-Day Unplanned Readmission Rates, Causes, and Risk Factors After Cervical Fusion: A Single-Institution Analysis. Spine (Phila Pa 1976) 2019;44:762-9. [Crossref] [PubMed]
19. Akamnonu C, Cheriyan T, Goldstein JA, et al. Ninety-day readmissions after degenerative cervical spine surgery: A single-center administrative database study. Int J Spine Surg 2015;9:19. [Crossref] [PubMed]
20. Rumalla K, Smith KA, Arnold PM. National Rates, Causes, Risk Factors, and Outcomes Associated With 30-Day and 90-Day Readmissions Following Degenerative Posterior Cervical Spine Surgery Utilizing the Nationwide Readmissions Database. Neurosurgery 2017;81:740-51. [Crossref] [PubMed]