Sage Journals: Discover world-class research

Abstract

Study Design

Retrospective, propensity-matched analysis

Objectives

Cervical disc arthroplasty (CDA) is being increasingly utilized for cervical disc generation. Surgeon specialty has been shown to influence the risk for postoperative complications in spine surgery, but this has not yet been explored for CDA. Thus, the purpose of this study is to determine whether there is any difference in 30-day complications between patients undergoing single-level CDA by neurosurgeons vs by orthopaedic surgeons.

Methods

A retrospective, 1:1 propensity score matched analysis was performed using the NSQIP database from 2015 to 2020. Patient demographics, operative characteristics, and postoperative complications were recorded. Independent multivariate logistic regression models were constructed using the propensity-matched dataset to assess surgical specialty influence on any complication, any site complication, any operative infection, and any medical complications.

Results

3179 single-level CDAs (28.8% orthopaedic surgery patients, 71.2% neurosurgery patients) were identified that met the inclusion criteria. Well-matched cohorts of 916 patients each were generated. After controlling for all possible confounders, orthopedic surgery specialty was not associated with a higher odds for any complication (OR: .87, 95% CI: .35 – 2.20, P = .7696), any site complication (OR: .32, 95% CI: .08 – 1.32, P = .1359), any operative infection (OR: .31, 95% CI: .07 – 1.34), P = .1172), nor any medical complication (OR: 2.11, 95% CI: .62 – 7.20, P = .2311) vs neurosurgery.

Conclusion

This is the first propensity-matched analysis to show that spine surgeon specialty does not influence the risk for any complication, any site complication, any operative infection, nor any medical complication following single-level CDA within the first 30 days after surgery.

Keywords

cervical disc arthroplasty CDA complications surgeon specialty spine

Introduction

Cervical disc arthroplasty (CDA) is being increasingly utilized, with an estimated 22,282 CDA procedures performed between 2004 and 2013 in the United States.^1,2 Although anterior cervical discectomy and fusion (ACDF) has generally been considered the gold standard for cervical disc generation, CDA is being utilized as an alternative due to several advantages. Most importantly, CDA preserves more physiologic motion of the spine and may decrease the risk for adjacent segment disease.^3,4 Additionally, prior research suggests that reoperation rates at the index level and adjacent levels are significantly less following CDA compared to ACDF.^5,6 Likewise, a national database study showed that CDA has a lower rate of complications, a lower rate of 90-day readmissions, and a shorter length of stay.⁷ Although CDA may have these benefits, it is not without complications.

Virk et al analyzed complications of 1347 entries of 7 different FDA-approved CDA implants and found that 25.2% had migration of the implant, 23.3% had insertion problem/failure, and 8.1% had heterotopic ossification.⁸ Likewise, Shah et al conducted a retrospective study of 16,510 patients from a state database and found that those who had undergone 1- or 2-level CDA had an increased risk for 2-year surgical complications, overall complications, and revisions compared to ACDF patients.⁹ Furthermore, a systematic review by Xu et al found that rates for dysphagia/dysphonia ranged from 1.3% to 27.2%, rates for cervical wound infection ranged from 1.2% to 22.5%, rates for vascular compromise ranged from 1.1% to 2.4%, and rates for cerebrospinal fluid leak ranged from .8% to 7.1%.¹⁰

There is a need to better identify risk factors for complications and worse outcomes in order to improve patient outcomes. One area of interest has been the influence of surgeon specialty on complication rates. For example, Bronheim et al conducted a retrospective study using a national database of 3182 patients undergoing anterior lumbar fusion and found that those patients undergoing surgery by a neurosurgeon had a higher risk of reoperation and urinary tract infection within 30-days compared to those undergoing surgery by an orthopaedic surgeon.¹¹ Any differences in complication rates between neurosurgery and orthopaedic surgery patients may also have long-term impacts on surgical residency and fellowship training models, reimbursements, and other aspects of the healthcare system. Whether or not such an influence of spine surgeon specialty exists on 30-day complications following single-level CDA has not yet been explored. Thus, this is the first study to conduct a propensity-matched analysis for examining whether there is any difference in 30-day complication rates between patients undergoing single-level CDA by neurosurgeons vs by orthopaedic surgeons.

Methods

Database

A retrospective analysis was conducted using the National Surgical Quality Improvement Program (NSQIP). The NSQIP is a prospective, multi-institutional database utilized to assess surgical quality improvement within the Unites States. The methodologies for data collection for NSQIP have been previously described elsewhere.¹² This study was deemed exempt from requiring informed consent as all data used was deidentified and publicly available. This study did not receive any grant or other external funding.

Population Selection

All patients who underwent a single-level CDA from the years 2015 to 2020 were identified using Current Procedure Terminology (CPT) Code 22856. Patients undergoing any concomitant procedures were excluded from the analysis. Patients with “null” and missing preoperative variables were also excluded from the analysis.

Data Collection

Demographic variables (age, sex, BMI, race, patient status (inpatient or outpatient), functional status), comorbidities (diabetes, smoking status, history of chronic obstructive pulmonary disease, hypertension, ascites, cancer, chronic health failure, steroid use, bleeding disorders, and American Society of Anesthesiologists (ASA) classification), operative characteristics (operative (OR) time, length of stay, discharge location), and complications were collected and compared between the 2 surgical specialties.

Postoperative complications were then stratified as any surgical site complication, any medical complication, and any complication overall in which all the above was included. Surgical site complications included any operative infection such as superficial surgical site infections (SSI), deep incisional SSI, organ/space SSI, and wound disruption. Medical complications consisted of urinary tract infections (UTI), sepsis, septic shock, pneumonia, unplanned intubation, pulmonary embolisms (PE), being on a ventilator for greater than 48hrs, renal insufficiency, acute renal failure, stoke, cardiac arrest, myocardial infraction, blood transfusions, deep vein thrombosis (DVT), and any unplanned reoperations within the 30-day period. All the variables used in this analysis were used as defined by the NSQIP user guide and outcomes were followed for 30 days after the initial surgery.¹³

Statistical Analysis

Univariate analysis was conducted to compare the aforementioned variables for any statistical differences between the 2 surgical specialties. Both fisher’s exact test and chi-squared tests were used to compare categorical variables while independent t-tests were used to compare continuous variables. In all cases, a P-value of .05 or less was considered to be statistically significant. Propensity score matching analysis was then conducted to account for any differences in the aforementioned variables between the 2 cohorts. Propensity score matching was used to minimize any potential bias that may have risen due to the nonrandomization design of this study. This type of analysis allows for a more accurate estimate of the effect of surgical specialty by simultaneously controlling for any possible confounders between the 2 cohorts. The matching was completed using a 1:1 nearest neighbor algorithm signifying that for each randomly selection case from the orthopedic cohort, a case from the neurosurgery cohort with the closest propensity score was paired based on the matched preoperative and demographical characteristics. This process was repeated until every orthopedic case was matched with a neurosurgery case. Independent multivariate logistic regression models were then constructed using the propensity-matched dataset for single-level CDA to assess the association of surgical specialty with the following outcomes: any complication, any site complication, any operative infection, and any medical complications. Preoperative variables with a P-value < .2 were identified and included in each regression model. Neurosurgery served as the reference specialty. Hosmer-Lemeshow and C statistics were calculated to assess the calibration and goodness-of-fit of the model, respectively. All descriptive and comparative statistics were performed using SAS software, version 9.2 (SAS Institute, Cary, NC).

Results

A total of 3179 single-level CDAs were identified that met the inclusion criteria (Figure 1). A total of 377 cases were excluded from the original dataset containing 3556 single-level CDAs due to missing preoperative variables that we would then incorporate into our analysis. There were 916 patients (28.8%) and 2263 patients (71.2%) in the orthopedic surgery and neurosurgery cohorts, respectively. Demographically, the neurosurgery cohort was older (46.9±11.3yrs vs 43.7±9.6yrs, P<.0001), and consisted of less patients that identified as white (79.5 vs 90%, P < .0001). The same cohort consisted of more diabetics (9.3 vs 6.9%, P = .026), patients with hypertension (28 vs 19.5%, P < .0001), patients that use steroids for a chronic condition (3 vs 1.4%, P = .0102), patients classified as having a class 1 wound (99.9 vs 99.1, P = .0012), and patients that have an ASA class greater than 2 (24.4% vs 17.8%, P < .0001). All other demographic variables and comorbidities can be found in Table 1. Univariate analyses demonstrated no statistical differences in 30-day postoperative complications nor among the discharge location between the 2 cohorts. However, the neurosurgery cohort did experience longer OR times (106.7±51.1min vs 97.4±47.1min, P < .0001) and a longer length of stay (1.1 ± 3 days vs .7 ± 3.5 days, P = .0055) (Table 2).

Figure 1.

Flow chart demonstrating patient population selection.

Table 1.

Characteristics of Patients Undergoing Single-Level Cervical Disc Arthroplasty

Variables	Neurosurgery		Orthopedic Surgery		P-value
Variables	(N = 2263)		(N = 916)		P-value
Age, mean (std)	46.9	11.3	43.7	9.6	<.0001
Gender, n (%)	—				.6723
Male	1170	51.7	466	50.9	—
Female	1093	48.3	450	49.1	—
BMI (kg/m^ 2), mean (std)	29.6	6.4	29.8	6.6	.5043
Race, n (%)	—				<.0001
White	1800	79.5	824	90	—
Black/AA	158	7	67	7.3
Asian	263	11.6	22	2.4
Other	42	1.9	3	0.3
Functional health status, n (%)	—				.2607¥
Independent	2255	99.6	913	99.7	—
Partially Dependent	8	0.4	3	0.3	—
Anesthesia, n (%)	—				.0123¥
General	2259	99.8	910	99.3	—
Local	1	.04	0	0
Other	3	.13	6	.66
Comorbidities	—
Diabetes, n (%)	—				.026
No	2052	90.7	853	93.1	—
Yes	211	9.3	63	6.9	—
Current Smoker, n (%)	—				.3492
No	1829	80.8	727	79.4	—
Yes	434	19.2	189	20.6	—
Hx of COPD, n (%)	—				.3288
No	2231	98.6	907	99	—
Yes	32	1.4	9	1	—
Hypertension, n (%)	—				<.0001
No	1629	72	737	80.5	—
Yes	634	28	179	19.5	—
Ascites, n (%)	—				NA
No	2263	100	916	100	—
Yes	0	0	0	0	—
Dialysis, n (%)	—				.3606¥
No	2260	99.9	916	0	—
Yes	3	0.1	0	0	—
Disseminated Cancer, n (%)	—				.2566¥
No	2259	99.8	916	100	—
Yes	4	0.2	0	0	—
CHF in 30 days before SX, n (%)	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Steroid Use for Chronic Condition, n(%)					.0102
No	2195	97	903	98.6	—
Yes	68	3	13	1.4	—
Greater than 10% weight loss in the past 6 months, n (%)	—				.5067¥
No	2261	99.9	916	100	—
Yes	2	0.1	0	0	—
Bleeding Disorders, n (%)	—				.0698¥
No	2255	99.7	909	99.2	—
Yes	8	0.3	7	0.8	—
Transfusion ≥ 1u PRBC 72hrs prior SX, n(%)	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Wound Classification, n (%)	—				.0012¥
1	2260	99.87	908	99.1	—
2	2	.09	7	0.8
3	1	.04	1	0.1
ASA Classification, n (%)	—				<.0001
≤ 2	1710	75.6	753	82.2	—
> 2	553	24.4	163	17.8	—

Table 2.

Complications of Patient Population Undergoing Single-Level Cervical Disc Arthroplasty.

Variables	Neurosurgery		Orthopedic Surgery		P-value
	(N = 2263)		(N = 916)
	n	%	n	%
Any Complications	—				.3861
No	2241	99	910	99.3	—
Yes	22	1	6	0.7	—
Any Surgical Site Complication	—				.1350¥
No	2254	99.6	915	99.9	—
Yes	9	0.4	1	0.1	—
Any operative Infection	—				.1056¥
No	2253	99.6	915	99.9	—
Yes	10	0.4	1	0.1	—
Superificial SSI Infections	—				.2134¥
No	2256	99.7	915	99.9	—
Yes	7	0.3	1	0.1	—
Deep Incisional SSI Inf.	—				.5067¥
No	2261	99.9	916	100	—
Yes	2	0.1	0	0	—
Organ/Space SSI Inf.	—				.7119¥
No	2262	99.96	916	100
Yes	1	.04	0	0
Wound Disruption	—				NA
No	2263	100	916	100	—
Yes	0	0	0	0	—
Any Medical Complication	—				.6712
No	2251	99.5	910	99.3	—
Yes	12	0.5	6	0.7	—
UTI	—				.1458¥
No	2256	99.7	911	99.5	—
Yes	7	0.3	5	0.5	—
Sepsis	—				.2881¥
No	2263	100	915	99.9	—
Yes	0	0	1	0.1	—
Septic Shock	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Pneumonia	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Unplanned Intubation	—				NA
No	2263	100	916	100	—
Yes	0	0	0	0	—
Pulmonary Embolism	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
On Ventilator > 48hrs	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Renal Insufficiency	—				NA
No	2263	100	916	100	—
Yes	0	0	0	0	—
Acute Renal Failure	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Stroke/CVA	—				.7119¥
No	2262	99.96	916	100	—
Yes	1	.04	0	0	—
Cardiac Arrest	—				NA
No	2263	100	916	100	—
Yes	0	0	0	0	—
Mycocardial Infraction	—				NA
No	2263	100	916	100	—
Yes	0	0	0	0	—
Bleeding Requiring Transfusions	—				.5067¥
No	2261	99.9	916	100
Yes	2	0.1	0	0	—
DVT	—				.5067¥
No	2261	99.9	916	100	—
Yes	2	0.1	0	0	—
Unplanned Reoperation	—				.1712
No	2246	99.2	913	99.7	—
Yes	17	0.8	3	0.3	—
Discharge Location	—				.0963
Home	2247	99.3	914	99.8	—
Rehab or Acute Care	16	0.7	2	0.2	—
OR Time (min), mean (std)	106.7	51.1	97.4	47.1	<.0001
Length of Stay (days), mean (std)	1.1	3	0.7	3.5	.0055

A 1:1 propensity match of single level CDA’s was performed generating well-matched cohorts consisting of 916 patients in each cohort. After matching, the neurosurgery cohort consisted of more patients that identified as white (91.9% vs 90.0%, P=.0006), patients that use steroids for chronic conditions (2.9% vs 1.4%, P = .0252), and patients classified as having a class 1 wound (99.9% vs 99.1%, P = .0039) (Table 3). Similar to the unmatched analysis, univariate analyses demonstrated no statistical differences in 30-day postoperative complications nor among the discharge location between the 2 cohorts. However, a similar difference was seen among OR times and total length of stay with the neurosurgery cohort experiencing longer times (102.3±41.4min vs 97.4±47.1min, P = .0167) and longer length of stay (1±1.5days vs .7±3.5days, P = .0201) (Table 4).

Table 3.

Characteristics of Propensity Matched Population.

Variables	Neurosurgery		Orthopedic Surgery		P-value
Variables	(N = 916)		(N = 916)		P-value
Age, mean (std)	43.8	9.5	43.7	9.6	.7971
Gender, n (%)	—				1.0000
Male	450	49.1	450	49.1	—
Female	466	50.9	466	50.9	—
BMI (kg/m^ 2), mean (std)	29.4	6.2	29.8	6.6	.1894
Race, n (%)	—				.0006¥
White	842	91.9	824	90.0	—
Black/AA	49	5.3	67	7.3
Asian	23	2.5	22	2.4
Other	2	0.2	3	0.3
Functional health status, n (%)	—				.3128¥
Independent	914	99.8	913	99.7	—
Partially Dependent	2	0.2	3	0.3	—
Anesthesia, n (%)	—				.0547¥
General	913	99.7	910	99.3	—
Local	1	0.1	0	0
Other	2	0.2	6	0.7
Comorbidities	—				—
Diabetes, n (%)	—				.9262
No	854	93.2	853	93.1	—
Yes	62	6.8	63	6.9	—
Current Smoker, n (%)	—				.8621
No	730	79.7	727	79.4	—
Yes	186	20.3	189	20.6	—
Hx of COPD, n (%)	—				.5103
No	904	98.7	907	99.0	—
Yes	12	1.3	9	1.0	—
Hypertension, n (%)	—				.4162
No	723	78.9	737	80.5	—
Yes	193	21.1	179	19.5	—
Ascites, n (%)	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Dialysis, n (%)	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Disseminated Cancer, n (%)	—				.5000¥
No	915	99.9	916	100	—
Yes	1	0.1	0	0	—
CHF in 30 days before SX, n (%)	—				.5000¥
No	915	99.9	916	100	—
Yes	1	0.1	0	0	—
Steroid Use for Chronic Condition, n(%)	—				.0252
No	889	97.1	903	98.6	—
Yes	27	2.9	13	1.4	—
Greater than 10% weight loss in the past 6 months, n (%)	—				.5000¥
No	915	99.9	916	100	—
Yes	1	0.1	0	0	—
Bleeding Disorders, n (%)	—				.0700¥
No	914	99.8	909	99.2	—
Yes	2	0.2	7	0.8	—
Transfusion ≥ 1u PRBC 72hrs prior SX, n(%)	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Wound Classification, n (%)	—				.0039¥
1	915	99.9	908	99.1	—
2	0	0	7	0.8
3	1	0.1	1	0.1
ASA Classification, n (%)	—				.0765
≤ 2	723	78.9	753	82.2	—
> 2	193	21.1	163	17.8	—

Table 4.

Complications of Propensity-Matched Population.

Variables	Neurosurgery		Orthopedic Surgery		P-value
	(N = 916)		(N = 916)
	n	%	n	%
Any Complications	—				.5916
No	908	99.1	910	99.3	—
Yes	8	0.9	6	0.7	—
Any Surgical Site Complication	—				.0935¥
No	911	99.5	915	99.9	—
Yes	5	0.5	1	0.1	—
Any operative Infection	—				.0935¥
No	911	99.5	915	99.9	—
Yes	5	0.5	1	0.1	—
Superificial SSI Infections	—				.2500¥
No	913	99.7	915	99.9	—
Yes	3	0.3	1	0.1	—
Deep Incisional SSI Inf.	—				.2499¥
No	914	99.8	916	100	—
Yes	2	0.2	0	0	—
Organ/Space SSI Inf.	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Wound Disruption	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Any Medical Complication	—				.1641¥
No	913	99.7	910	99.3	—
Yes	3	0.3	6	0.7	—
UTI	—				.1640¥
No	914	99.8	911	99.5	—
Yes	2	0.2	5	0.5	—
Sepsis	—				.5000¥
No	916	100	915	99.9	—
Yes	0	0	1	0.1	—
Septic Shock	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Pneumonia	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Unplanned Intubation	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Pulmonary Embolism	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
On Ventilator > 48hrs	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Renal Insufficiency	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Acute Renal Failure	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Stroke/CVA	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Cardiac Arrest	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Mycocardial Infraction	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
Bleeding Requiring Transfusions	—				NA
No	916	100	916	100	—
Yes	0	0	0	0	—
DVT	—				.5000¥
No	915	99.9	916	100	—
Yes	1	0.1	0	0	—
Unplanned Reoperation	—				.2047
No	909	99.2	913	99.7	—
Yes	6	0.8	3	0.3	—
Discharge Location	—				.1640¥
Home	911	99.5	914	99.8	—
Rehab or Acute Care	5	0.5	2	0.2	—
OR Time (min), mean (std)	102.3	41.4	97.4	47.1	.0167
Length of Stay (days), mean (std)	1	1.5	0.7	3.5	.0201

Multivariate logistic regression models using the propensity-matched dataset were then constructed to adjust for possible confounders (Table 5). Our analysis found that after controlling for all possible confounders, the orthopedic surgery specialty was not associated with a higher odds for any complication (OR: .87, 95% CI: .35 – 2.20, P = .7696), any site complication (OR: .32, 95% CI: .08 – 1.32, P = .1359), any operative infection (OR: .31, 95% CI: .07 – 1.34), P = .1172), nor any medical complication (OR: 2.11, 95% CI: .62 – 7.20, P = .2311) vs neurosurgery.

Table 5.

Multivariate Analysis of Orthopedic Surgery Specialty on Complications.

	cervical disc arthroplasty
	Odds Ratio	95% CI	P-value	c-stat
Any complication	.87	.35–2.20	.7696	.52
Any site complication	.32	.08–1.32	.1359	.67
Any operative infections	.31	.07–1.34	.1172	.66
Any medical complications	2.11	.62–7.20	.2311	.65

Discussion

CDA provides a good alternative to ACDF for patients with cervical disc generation by allowing for more preserved physiological motion of the spine and a reduced risk for adjacent disc disease. Nonetheless, it continues to have perioperative complications and identifying risk factors for such complications or worse outcomes is necessary. Such identification can allow us to improve patient outcomes and healthcare delivery. The influence of surgeon specialty on spinal surgery outcomes in other surgeries has been reported previously. However, this is the first study to examine the influence of surgeon specialty on 30-day complications following single-level CDA. In the propensity-matched cohorts, neurosurgery patients were found to have longer OR time and length of stay on bivariate analysis. However, these differences are not clinically significant. On multivariate analysis, orthopaedic surgery specialty was not a risk factor for any complication, any site complication, any operative infections, or any medical complications.

This study’s findings are consistent with prior studies showing that spine surgeon specialty does not increase the risk for overall perioperative complications.^14–16 Kim et al conducted a propensity score matching analysis using a national database of patients undergoing single-level lumbar fusion to evaluate the influence of the surgeon’s specialty.¹⁶ The authors found no significant differences in the risk for developing any complication, medical complications, surgical complications, or reoperation within 30-days of surgery.¹⁶ Similarly, Minhas et al conducted a propensity score-matching analysis of single-level ACDF patients comparing neurosurgery and orthopaedic surgery patients.¹⁵ The authors found that spine surgery specialty did not influence the risk for overall complications, surgical site complications, or medical complications.¹⁵ Furthermore, Malik et al used a national database to assess the influence of surgeon specialty on risks following fusions, laminectomies, or osteotomy/corpectomy for spinal metastases.¹⁴ The authors did not find any significant differences in risk for 90-day complications, readmission, or mortality between both spine surgery specialties.¹⁴

However, several studies are not consistent with this study’s findings.^17–20 Alomari et al conducted a retrospective, 1:4 propensity score-matched analysis using a national database to evaluate the influence of spine surgeon specialty on outcomes following elective posterior spinal fusion for adolescent idiopathic scoliosis.¹⁸ The authors found that orthopaedic surgery patients had a shorter total hospital stay, shorter operation time, lower rate of return to operating room within same admission, and lower unplanned readmission rates, whereas neurosurgery patients had only a lower perioperative blood transfusion rate.¹⁸ Likewise, Alomari et al also conducted a retrospective, 1:1 propensity score-matched study using a national database to evaluate the influence of surgeon specialty on outcomes following elective single-level and multilevel anterior/lateral lumbar interbody fusion.¹⁹ The authors found that neurosurgery patients had longer operative times, shorter total hospital stays, lower rates of return to the operating room, lower rates of nonhome discharge, and lower rates of perioperative blood transfusion in both single and multilevel surgery.¹⁹ Furthermore, another 1:1 propensity score-matched study of patients undergoing elective ACDF aimed to evaluate the influence of spine surgeon specialty on 30-day outcomes.¹⁷ The authors found that in both single-level and multilevel ACDF, neurosurgery patients had shorter total hospital stay and lower rates of return to operating room, nonhome discharge, sepsis, and perioperative blood transfusion but longer operative times compared to orthopaedic surgery patients.¹⁷ Additionally, Seicean et al conducted a propensity matched analysis of patients undergoing elective spine fusion and laminectomy by neurosurgeons and orthopaedic surgeons.²⁰ The authors found that patients operated on by orthopaedic surgeons had an increased risk for a prolonged length of stay and perioperative transfusion.²⁰

Further investigations on the influence of spine surgeon specialty on patient outcomes following other surgeries are warranted. Both neurosurgery and orthopaedic surgery specialists operate on the spine despite having significant differences in training, but the impact this has on clinical outcomes has not been explored for all spine surgeries. For example, it was recently found that graduating neurosurgery residents have logged 6.8 times more adult spine cases when graduating compared to orthopaedic surgery residents.²¹ Moreover, spine procedures were found to have accounted for 33.5% of all total surgical cases over 10-years for neurosurgery residents compared to 6.2% for orthopaedic surgery residents.²² However, although orthopaedic residents perform less spine cases than neurosurgery residents, fellowship training likely equalizes this difference as significantly more orthopaedic spine surgeons are fellowship trained compared to neurosurgery spine surgeons.^23,24 Therefore, this study finding no clinically significant difference in complication rates between both specialties provides further evidence supporting the equalizing effect of fellowship training.

This study is not without limitations. A large national database is prone to erroneous data entry or omission, although this likely occurs at too low of a rate to significantly influence this study’s results due to quality control measures. Additionally, the database does not have surgeon experience available for analysis, as that may influence risks from each specialty. Moreover, the database is limited to only 30-day outcomes, and it is possible that there may be differences in risks seen with spine surgery specialties with 90-day or longer outcomes. For example, complications associated with surgeon technique such as heterotopic ossification, poor range of motion secondary to misplacement, and implant subsidence or migration may not be recognized in the first 30 days after surgery. Despite these limitations, this is the first well-powered, propensity score matched analysis to evaluate the influence of spine surgeon specialty on complications following single-level CDA.

Conclusion

Single-level CDA is performed by both neurosurgeons and orthopaedic surgeons. This is the first propensity-matched analysis to show that spine surgeon specialty does not influence the risk for any complication, any site complication, any operative infection, nor any medical complication following single-level CDA within the first 30 days after surgery.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Puneet Gupta

Fthimnir M Hassan

George M Thomas

References

Kerezoudis

Alvi

Goyal

, et al. Commentary: Utilization Trends of Cervical Disk Replacement in the United States. Oper Neurosurg (Hagerstown, Md). 2018;15(4):40-43. doi:10.1093/ons/opy181

Jain

Nguyen

Formanek

, et al. Cervical Disc Replacement: Trends, Costs, and Complications. Asian Spine J. 2020;14(5):647-654. doi:10.31616/asj.2019.0246

Nunley

Kerr

3rd Cavanaugh

, et al. Adjacent Segment Pathology After Treatment With Cervical Disc Arthroplasty or Anterior Cervical Discectomy and Fusion, Part 2: Clinical Results at 7-Year Follow-Up. Int J spine Surg. 2020;14(3):278-285. doi:10.14444/7037

Laratta

Shillingford

Saifi

Riew

. Cervical Disc Arthroplasty: A Comprehensive Review of Single-Level, Multilevel, and Hybrid Procedures. Glob spine J. 2018;8(1):78-83. doi:10.1177/2192568217701095

Badhiwala

Platt

Witiw

Traynelis

. Cervical disc arthroplasty versus anterior cervical discectomy and fusion: a meta-analysis of rates of adjacent-level surgery to 7-year follow-up. J spine Surg (Hong Kong). 2020;6(1):217-232. doi:10.21037/jss.2019.12.09

Zhong

Z-M

Zhu

S-Y

Zhuang

J-S

Chen

J-T

. Reoperation After Cervical Disc Arthroplasty Versus Anterior Cervical Discectomy and Fusion: A Meta-analysis. Clin Orthop Relat Res. 2016;474(5):1307-1316. doi:10.1007/s11999-016-4707-5

Rumalla

Smith

Arnold

. Cervical Total Disc Replacement and Anterior Cervical Discectomy and Fusion: Reoperation Rates, Complications, and Hospital Resource Utilization in 72 688 Patients in the United States. Neurosurgery. 2018;82(4):441-453. doi:10.1093/neuros/nyx289

Virk

Phillips

Khan

Qureshi

. A cross-sectional analysis of 1347 complications for cervical disc replacements from medical device reports maintained by the United States Food and Drug Administration. Spine J. 2021;21(2):265-272. doi:10.1016/j.spinee.2020.09.005

Shah

N V

Jain

Moattari

, et al. Comparing Predictors of Complications After Anterior Cervical Diskectomy and Fusion, Total Disk Arthroplasty, and Combined Anterior Cervical Diskectomy and Fusion-Total Disk Arthroplasty With a Minimum 2-Year Follow-Up. J Am Acad Orthop Surg. 2020;28(17):e759-e765. doi:10.5435/JAAOS-D-19-00666

10.

Goel

Shriver

, et al. Adverse Events Following Cervical Disc Arthroplasty: A Systematic Review. Glob Spine J. 2017;8(2):178-189. doi:10.1177/2192568217720681

11.

Bronheim

Cheung

Phan

White

SJW

Kim

Cho

. Anterior Lumbar Fusion: Differences in Patient Selection and Surgical Outcomes Between Neurosurgeons and Orthopaedic Surgeons. World Neurosurg. 2018;120:e221-e226. doi:10.1016/j.wneu.2018.08.034

12.

Birkmeyer

Shahian

Dimick

, et al. Blueprint for a new American College of Surgeons: National Surgical Quality Improvement Program. J Am Coll Surg. 2008;207(5):777-782. doi:10.1016/j.jamcollsurg.2008.07.018

13.

Khuri

Henderson

Daley

, et al. The patient safety in surgery study: background, study design, and patient populations. J Am Coll Surg. 2007;204(6):1089-1102. doi:10.1016/j.jamcollsurg.2007.03.028

14.

Malik

Baek

Alexander

Khan

Scharschmidt

. Orthopaedic vs. Neurosurgery - Does a surgeon’s specialty have an influence on 90-day complications following surgical intervention of spinal metastases? Clin Neurol Neurosurg. 2020;192:105735. doi:10.1016/j.clineuro.2020.105735

15.

Minhas

S V

Chow

Patel

Kim

JYS

. Surgeon specialty differences in single-level anterior cervical discectomy and fusion. Spine (Phila Pa 1976). 2014;39(20):1648-1655. doi:10.1097/BRS.0000000000000499

16.

Kim

Edelstein

Hsu

Lim

Kim

JYS

. Spine surgeon specialty is not a risk factor for 30-day complication rates in single-level lumbar fusion: a propensity score-matched study of 2528 patients. Spine (Phila Pa 1976). 2014;39(15):19-27. doi:10.1097/BRS.0000000000000394

17.

Alomari

Judy

Weingart

, et al. Early Outcomes of Elective Anterior Cervical Diskectomy and Fusion for Degenerative Spine Disease Correlate With the Specialty of the Surgeon Performing the Procedure. Neurosurgery. 2022;90(1):99-105. doi:10.1227/NEU.0000000000001748

18.

Alomari

Lubelski

S-FL

, et al. Surgeon specialty effect on early outcomes of elective posterior spinal fusion for adolescent idiopathic scoliosis: a propensity-matched analysis of 965 patients. European Spine Journal. 2022;31(9):2355-2361. doi:10.1007/s00586-022-07248-1

19.

Alomari

Porras

S-FL

, et al.

Does the Specialty of the Surgeon Performing Elective Anterior/Lateral Lumbar Interbody Fusion for Degenerative Spine Disease Correlate with Early Perioperative Outcomes?

World Neurosurg. 2021;155:e111-e118. doi:10.1016/j.wneu.2021.08.010

20.

Seicean

Alan

Seicean

Neuhauser

Benzel

Weil

. Surgeon specialty and outcomes after elective spine surgery. Spine (Phila Pa 1976). 2014;39(19):1605-1613. doi:10.1097/BRS.0000000000000489

21.

Lad

Gupta

Para

, et al. An ACGME-based comparison of neurosurgical and orthopedic resident training in adult spine surgery via a case volume and hours-based analysis. J Neurosurg Spine. 2021;35(5):553-563. doi:10.3171/2020.10.SPINE201066

22.

Pham

Jakoi

Wali

Lenke

. Trends in Spine Surgery Training During Neurological and Orthopaedic Surgery Residency: A 10-Year Analysis of ACGME Case Log Data. J Bone Joint Surg Am. 2019;101(22):e122. doi:10.2106/JBJS.19.00466

23.

Post

Dai

, et al. Workforce Analysis of Spine Surgeons Involved with Neurological and Orthopedic Surgery Residency Training. World Neurosurg. 2019;122:e147-e155. doi:10.1016/j.wneu.2018.09.152

24.

Konczalik

Elsayed

Boszczyk

. Experience of a fellowship in spinal surgery: a quantitative analysis. Eur spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc. 2014;23(suppl 1):S40-54. doi:10.1007/s00586-014-3209-y

Influence of Surgeon Specialty on 30-day Outcomes Following Single-Level Cervical Disc Arthroplasty: A Propensity-Matched Analysis

Abstract

Study Design

Objectives

Methods

Results

Conclusion

Keywords

Introduction

Methods

Database

Population Selection

Data Collection

Statistical Analysis

Results

Discussion

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iDs

References