Posterior cervical decompression and fusion (PCDF) is an effective methodology used to treat a variety of cervical spine pathologies.1,2 In recent years, the age and medical complexity of the surgical population undergoing this procedure has been rising.3,4 With this comes an increased risk of in-hospital complications for a procedure that, in comparison with other cervical fusion procedures, is already at a higher risk for short-term perioperative complications.5,6 An investigation into the factors driving perioperative complications in PCDF is critical.
Complications rates are a common focus of analysis; not only do they function as indicators of care quality,7 but they also drive resource utilization, cost, and perioperative outcomes.8 Previous research has analyzed complication rates9 and risk factors for short-term perioperative complications in cervical spine surgery, including teaching-hospital status,10 age,11 and smoking status.12 PCDF is regularly performed by two surgical specialties: orthopedic surgery and neurological surgery. Although recent literature has cited differences in surgical opinion13 and practice patterns14 between these two specialties, results evaluating the impact of surgeon specialty on outcomes have been contradictory; some studies show no difference in complication rates,15 while others report significant differences between the specialties.16,17 As PCDF represents a surgical population especially at risk for complications relative to other fusion procedures,5,6 it warrants an individual analysis to determine if surgical specialty is driving differences in outcomes.
This study uses 10 years of data from a single institution and nine from within the American College of Surgeons National Surgical Quality Improvement Project Participant User Files (ACS-NSQIP PUFs), a prospectively collected clinical database with demonstrated validity and reproducibility,16,18 to characterize in-hospital complications for patients undergoing PCDF by both neurological spine surgeons and orthopedic spine surgeons. Differences in rates of in-hospital complications between the two surgical services are analyzed to compare performance and encourage collaboration between the two specialties.
The Data Sets
After Institutional Review Board approval, records of PCDF cases occurring between January 1, 2006 and November 30, 2016 were obtained from a single institution. This database contains demographic and perioperative data for all patients undergoing surgical procedures at a single, urban, academic medical center. Records of PCDF cases occurring between 2007 and 2015 were also obtained from the ACS-NSQIP PUFs database, which prospectively collects demographic data, patient factors, and postoperative outcomes from >600 institutions.19 Informed consent was waived for this study.
Defining the Cohort
Patients undergoing PCDF were queried from a single institution and the ACS-NSQIP PUFs database by Current Procedural Terminology (CPT) code. Cases were preliminarily identified using the CPT codes 63045, 63001, 63015, 22110, 22210, and 22600, and confirmed using the International Classification of Disease (ICD) Ninth Revision code 81.03 or ICD Tenth Revision code M43.22. Patients undergoing anterior cervical fusion during the same hospital stay were identified using the CPT codes 63075, 22551, and 22554, and were excluded from the analysis. Pediatric patients, patients not operated on by a spine surgeon, and patients with missing data were also excluded from the analysis.
Demographic variables collected from both databases included age, sex, and American Society of Anesthesiology Physical Status Classification (ASA status). ICD-9 and ICD-10 codes (see Table 1) were utilized to query the following complications: perioperative or postoperative bleeding requiring transfusion, cardiac arrest, cerebrovascular accident, deep vein thrombosis, myocardial infarction, pneumonia, pulmonary embolism, renal failure, superficial surgical site infection, urinary tract infection, wound dehiscence, sepsis, septic shock, and death. Case descriptions were utilized to identify airway complications requiring reintubation or tracheostomy, as well as additional cases of renal failure. Complications were limited to in-hospital complications using the index hospital stay. Descriptions of each complication can be found in the ACS-NSQIP PUFs Data User Guide.20
Statistical Analysis Software version 9.4 (SAS v9.4—SAS Institute, Cary, NC) was utilized to perform all statistical analyses. Cohorts were created based on if patients underwent PCDF by an orthopedic surgeon or a neurological surgeon. Categorical variables were compared using chi-square or Fischer exact tests, depending on the complication frequency. Continuous variables were compared using the Student t test. Univariate logistic regression was utilized with the outcome of any in-hospital complication, the exposure variable of surgical specialty, and the reference group as the neurological surgeon cohort. A multivariable model was also created controlling for age, sex, and ASA status.
Over a 10-year period, a total of 1221 patients at a single institution and 11,116 patients within the NSQIP database underwent PCDF (Table 2). In both populations, a greater proportion of patients underwent PCDF by the neurological surgery service as opposed to the orthopedic surgery service (single institution: 55.94% vs. 44.06%; NSQIP: 78.32% vs. 21.68%) (Table 2). The mean age of patients operated on by the neurological surgeon service was slightly higher in both data sets, although the difference was only significant in the NSQIP database (61.11 vs. 60.16 yr, P = 0.002) (Table 2). The proportion of males was significantly higher in the neurological surgeon service in the NSQIP database (59.74% vs. 57.26%, P = 0.03), but did not differ significantly in the single institution data (Table 2). The mean ASA status of patients in the neurological surgeon service was significantly higher in the institutional sample (2.63 vs. 2.45, P < 0.0001) (Table 2).
In the single institution cohort, when individual in-hospital complications were compared between surgical services, patients in the orthopedic surgeon cohort had a significantly higher proportion of bleeding requiring transfusions (14.5% vs. 9.08%, P = 0.003) and PE (0.74% vs. 0.00%, P = 0.04), while patients in the neurological surgeon cohort had a significantly higher proportion of airway complications (1.17% vs. 0.00%, P = 0.01) and pneumonia (4.25% vs. 0.74%, P = 0.0002) (Table 3). There were no significant differences between the two cohorts for the other in-hospital complications analyzed (see Table 3).
Within the NSQIP database, patients in the orthopedic surgeon cohort had a significantly higher proportion of bleeding requiring transfusion (11.16% vs. 6.18%, P < 0.0001) and septic shock (0.71% vs. 0.32%, P = 0.009), while patients in the neurological surgeon cohort had a significantly higher proportion of deep vein thrombosis (0.76% vs. 0.37%, P = 0.04) (Table 4). There were no significant differences between the two cohorts for the other in-hospital complications analyzed (see Table 4).
When in-hospital complications were considered as a whole, both univariate and multivariate analysis of the institutional data revealed that patients in the orthopedic surgeon service were no more likely to have an in-hospital complication than patients in the neurological surgeon service (Table 5). Within the NSQIP database, however, univariate analysis revealed that patients in the orthopedic surgery service were 1.5 times as likely to have an in-hospital complication when compared with patients in the neurological surgery service (95% CI: 1.31–1.71, P < 0.0001) (Table 5). After adjusting for age, sex, and ASA status, patients in the orthopedic surgery service were 1.66 times as likely to have an in-hospital complication when compared with patients in the neurological surgery service (95% CI: 1.44–1.90, P < 0.0001) (Table 5).
This study used a large single-institution sample and an even larger national sample to compare in-hospital complication rates for patients undergoing PCDF by orthopedic and neurological surgeons. Although results in the literature have been anything but conclusive when examining this topic, some studies have demonstrated significant differences in complication rates between the two specialties.16,17 Understanding if surgical specialty drives differences in outcomes is key in moving toward practices that can limit in-hospital complications for patients. When complication rates were compared between specialties, discrepancies in individual in-hospital complications were identified in both the institutional sample and the national sample (Tables 3 and 4). Although overall complication rates were similar in the single institution sample, in the national sample, orthopedic surgeons were found to be more likely to encounter an in-hospital complication than neurological surgeons (Table 5).
While this study identified discrepancies in outcomes between the two surgical services, there were also significant differences in patient population. In both the institutional sample and the national sample, patients operated on by neurological surgeons had a significantly higher ASA designation (Table 2). This difference in the degree of comorbidity of patients treated by orthopedic and neurological surgeons has been shown across multiple databases in the literature.17 Higher ASA status has been linked to an increased risk for in-hospital complications,9,21 making the finding that patients operated on by the orthopedic surgeon service are more likely to have an in-hospital complication (Table 5) counterintuitive. The authors believe that the difference in ASA status between the two cohorts retains more statistical significance than clinical significance and that this marginal difference in ASA status between the two patient populations is not driving outcomes. Similarly, although age has also been linked to increased risk for in-hospital complications in cervical fusion,11,22 the small age difference between the two cohorts (Table 2) is likely insignificant. These conclusions are supported by the fact that the difference in overall in-hospital complication rates retained significance in a multivariable model controlling for age and ASA status (Table 5).
When individual in-hospital complications were assessed, a higher proportion of patients in the orthopedic surgeon cohort had perioperative bleeding requiring transfusion in both the institutional sample (14.5% vs. 9.08%, P = 0.003) and the national sample (11.16% vs. 6.18%, P < 0.0001) (Tables 3 and 4). Although this is consistent with what has been shown in the literature,16 this study was unable to correct for comorbidities that predispose patients to bleeding (i.e., bleeding disorders), which may have skewed results. This study also only captured complications occurring before discharge, missing patients who had a delayed complication or who were readmitted to the hospital. The higher rates of complications in the single institution sample relative to the national sample may be attributable to the single institution's status as a teaching hospital, which has been shown to experience more complications.10,23
While the factors driving these differences in outcomes are unclear, surgical opinion and clinical decision-making have been shown to differ between orthopedic surgeons and neurological surgeons when treating cervical spine pathologies.13,24,25 This could be a product of training differences between these two specialties. Case volume, which has been associated in some procedures with variation in postoperative outcomes,26 differs dramatically between orthopedic and neurological residency programs. Additional fellowship training, however, may narrow that gap.27 It is not unreasonable to suggest that differences in training may be driving variation in outcomes, as training pathway in other surgical specialties such as the general surgical subspecialties of lung, breast, and vascular surgery has been shown to significantly impact outcomes.26,28–31 If this is true in spine surgery as well, the creation of a joint spine training program between orthopedic and neurological surgery may be beneficial. More evidence is needed to determine if specialist specific practices are driving differences in outcomes.
The presence of a secondary surgeon in the operating room may also be driving variation in outcomes, as dual-attending surgery has been shown to improve outcomes in some spinal surgeries.32,33 It is noteworthy that at the single institution, where interdisciplinary collaboration between primary and secondary surgical teams is prevalent, no statistically significant difference was identified for overall in-hospital complication rates between surgical specialties. This suggests that a two-surgeon approach to PCDF may be an effective method to minimize complications and improve outcomes for patients. Unfortunately, the single institution data, as well as the data in the NSQIP database, does not include information on collaboration between the primary and secondary surgeon in the OR. Future research should investigate the impact two-surgeon collaboration has on short-term and long-term perioperative outcomes.
This study had several limitations. First and foremost, a portion of the single institution data is likely present in the NSQIP sample; the single institution is a contributor to the ACS-NSQIP PUFs, and ethical standards prohibit removing these patients from NSQIP before analysis. As the ACS-NSQIP PUFs samples data from more than 600 institutions,19 this redundancy should not invalidate any comparisons made between the two cohorts. Flaws inherent to a large national data bank like NSQIP, including a lack of granularity and sampling bias toward large academic institutions, should also be acknowledged.34 Institutional data was therefore presented alongside data from NSQIP to corroborate more granular findings such as differences in individual complication rates between surgical specialties. Second, this study was unable to sufficiently correct for comorbidities that have been shown to increase the risk for the in-hospital complications studied (e.g., anemia and transfusion requirements). ASA status was used as a surrogate measurement of patient illness, as it is reliably documented in perioperative data, but variability in ASA status designation is common between practictioners.35 This study was also unable to control for the number of levels fused during the procedure. This has been shown in the literature to increase the risk for complications,6,9 and may have confounded the results of this study. Finally, cases were separated into orthopedic and neurological surgeon cohorts based on the primary surgeon designation, making it impossible to determine if the specialty of the secondary surgeon impacted outcomes.
When examining a large, national sample, this study demonstrates that orthopedic surgeons are more likely to encounter in-hospital complications than neurological surgeons when performing PCDF. This difference was not present at the single institution, where spinal surgery is approached from an interdisciplinary standpoint, and two-surgeon spinal surgery is prevalent. The objective of this study is to encourage interdisciplinary collaboration between surgical specialists to decrease in-hospital complications and improve outcomes for patients undergoing PCDF. With a surgical population that is growing not only in age, but also in medical complexity, collaboration between orthopedic and neurosurgical specialties is paramount.
1. Anderson PA, Henley MB, Grady MS, et al. Posterior cervical arthrodesis with AO reconstruction plates and bone graft. Spine (Phila Pa 1976)
1991; 16 (3 suppl):S72–S79.
2. Kuhns CA, Geck MJ, Wang JC, et al. An outcomes
analysis of the treatment of cervical pseudarthrosis with posterior fusion. Spine (Phila Pa 1976)
3. Marquez-Lara A, Nandyala SV, Fineberg SJ, et al. Current trends in demographics, practice, and in-hospital outcomes
in cervical spine
surgery. Spine (Phila Pa 1976)
4. Kadow BT, Tyagi P, Chermansky CJ. Neurogenic causes of detrusor underactivity. Curr Bladder Dysfunct Rep
5. Oglesby M, Fineberg SJ, Patel AA, et al. The incidence and mortality of thromboembolic events in cervical spine
surgery. Spine (Phila Pa 1976)
6. Memtsoudis SG, Hughes A, Ma Y, et al. Increased in-hospital complications
after primary posterior versus primary anterior cervical fusion. Clin Orthop Relat Res
7. Moore L, Lauzier F, Stelfox HT, et al. Derivation and validation of a quality indicator to benchmark in-hospital complications
among injury admissions. JAMA Surg
8. Marquez-Lara A, Nandyala SV, Hassanzadeh H, et al. Sentinel events in cervical spine
surgery. Spine (Phila Pa 1976)
9. Medvedev G, Wang C, Cyriac M, et al. Complications, readmissions, and reoperations in posterior cervical fusion. Spine (Phila Pa 1976)
10. Fineberg SJ, Oglesby M, Patel AA, et al. Outcomes
of cervical spine
surgery in teaching and non-teaching hospitals. Spine (Phila Pa 1976)
11. Radcliff K, Ong KL, Lovald S, et al. Cervical spine
surgery complications and risks in the elderly. Spine (Phila Pa 1976)
12. Yue JK, Upadhyayula PS, Deng H, et al. Risk factors for 30-day outcomes
in elective anterior versus posterior cervical fusion: a matched cohort analysis. J Craniovertebr Junction Spine
13. Glaser JA, Jaworski BA, Cuddy BG, et al. Variation in surgical opinion regarding management of selected cervical spine
injuries. Spine (Phila Pa 1976)
14. Hussain M, Nasir S, Moed A, et al. Variations in practice patterns among neurosurgeons and orthopaedic surgeons in the management of spinal disorders. Asian Spine J
15. Minhas SV, Chow I, Patel AA, et al. Surgeon specialty differences in single-level anterior cervical discectomy and fusion. Spine (Phila Pa 1976)
16. Seicean A, Alan N, Seicean S, et al. Surgeon specialty and outcomes
after elective spine surgery. Spine (Phila Pa 1976)
17. Mabud T, Norden J, Veeravagu A, et al. Complications, readmissions, and revisions for spine procedures performed by orthopedic surgeons versus neurosurgeons. Clin Spine Surg
18. Khuri SF, Henderson WG, Daley J, et al. The patient safety in surgery study: background, study design, and patient populations. J Am Coll Surg
20. User Guide for the 2015 ACS NSQIP Participant Use Data File.
21. Somani S, Capua JD, Kim JS, et al. ASA Classification as a risk stratification tool in adult spinal deformity surgery: a study of 5805 patients. Glob Spine J
22. Tetreault L, Ibrahim A, Cote P, et al. A systematic review of clinical and surgical predictors of complications following surgery for degenerative cervical myelopathy. J Neurosurg Spine
23. Lee NJ, Kothari P, Kim C, et al. The impact of resident involvement in elective posterior cervical fusion. Spine (Phila Pa 1976)
24. Grauer JN, Vaccaro AR, Beiner JM, et al. Similarities and differences in the treatment of spine trauma between surgical specialties and location of practice. Spine (Phila Pa 1976)
25. Irwin ZN, Hilibrand A, Gustavel M, et al. Variation in surgical decision making for degenerative spinal disorders. Part II: cervical spine
. Spine (Phila Pa 1976)
26. Pearce WH, Parker MA, Feinglass J, et al. The importance of surgeon volume and training in outcomes
for vascular surgical procedures. J Vasc Surg
27. Daniels AH, Ames CP, Smith JS, et al. Variability in spine surgery procedures performed during orthopaedic and neurological surgery residency training: an analysis of ACGME case log data. J Bone Joint Surg Am
28. Feasby TE, Quan H, Ghali WA. Hospital and surgeon determinants of carotid endarterectomy outcomes
. Arch Neurol
29. Goodney PP, Lucas FL, Stukel TA, et al. Surgeon specialty and operative mortality with lung resection. Ann Surg
30. Kingsmore D, Hole D, Gillis C. Why does specialist treatment of breast cancer improve survival? The role of surgical management. Br J Cancer
31. McCutcheon BA, Talamini MA, Chang DC, et al. The comparative effectiveness of surgeons over interventionalists in endovascular repairs of abdominal aortic aneurysm. Ann Surg
32. Ames CP, Barry JJ, Keshavarzi S, et al. Perioperative outcomes
and complications of pedicle subtraction osteotomy in cases with single versus two attending surgeons. Spine Deform
33. Sethi RK, Pong RP, Leveque JC, et al. The Seattle spine team approach to adult deformity surgery: a systems-based approach to perioperative care and subsequent reduction in perioperative complication rates. Spine Deform
34. Alluri RK, Leland H, Heckmann N. Surgical research using national databases. Ann Transl Med
35. Sankar A, Johnson SR, Beattie WS, et al. Reliability of the American Society of Anesthesiologists physical status scale in clinical practice. Br J Anaesth