Single-level anterior cervical discectomy and fusion is associated with higher nonroutine discharge rates compared to cervical disc arthroplasty in obese patients

ABSTRACT

Context:

Obesity is a recognized risk factor for adverse outcomes in cervical spine surgery. While cervical disc arthroplasty (CDA) has emerged as an alternative to anterior cervical discectomy and fusion (ACDF), comparative outcomes among obese patients remain underexplored.

Aims:

The aim of this study is to compare nonroutine discharge rates and other postoperative outcomes between obese patients undergoing single-level ACDF and CDA.

Settings: and Design:

Retrospective cohort study using a national database.

Subjects and Methods:

The National Inpatient Sample was queried to identify obese patients who underwent single-level ACDF or CDA between 2016 and 2020. Inclusion and exclusion criteria were applied. Propensity score matching (1:3) was performed based on age, sex, race, Elixhauser Comorbidity Index, and primary diagnosis to create comparable cohorts. Outcomes included nonroutine discharge, length of stay (LOS), total cost, and postoperative complications.

Statistical Analysis Used:

Chi-square tests and Student’s t-tests were used for categorical and continuous outcomes, respectively, with significance set at P < 0.05.

Results:

After matching, 1455 ACDF and 485 CDA cases were analyzed. Obese patients undergoing ACDF had significantly higher nonroutine discharge rates (11.3% vs. 4.1%, P = 0.037). ACDF patients had lower total costs ($16,400 vs. $19,400, P = 0.003), with similar LOS (1.4 ± 0.1 days, P = 0.931) and adverse event rates (7.6% vs. 8.2%, P = 0.827).

Conclusions:

Obese patients undergoing ACDF experience higher nonroutine discharge rates compared to those receiving CDA. While CDA is associated with higher costs, it may offer superior discharge outcomes in this high-risk population.

INTRODUCTION

Obesity, defined as a body mass index (BMI) ≥30 kg/m², is one of the most prevalent health conditions globally.[1] While obesity is a well-established risk factor for lumbar spine diseases, its role in the development of cervical spine pathologies remains unclear due to conflicting and limited evidence.[2,3,4,5] Despite this uncertainty, assessing surgical outcomes in obese patients undergoing cervical spine procedures is crucial, as obesity has been consistently implicated as a perioperative and postoperative risk factor in these procedures.[6,7,8] Among cervical spine surgeries, cervical disc arthroplasty (CDA) has emerged as a preferred option for treating conditions such as disc herniation and degeneration. From 2011 to 2019, the use of CDA increased dramatically by 6.5-fold, compared to the more modest growth of the former gold standard, anterior cervical discectomy and fusion (ACDF).[9] This shift in practice is supported by meta-analyses demonstrating CDA’s noninferiority in both short- and long-term outcomes.[10,11,12,13] With the rising popularity of CDA, its usage among obese patients has also increased, emphasizing the importance of evaluating outcomes in this population.

The risks associated with both CDA and ACDF have been extensively studied in obese populations. A previous study by Zeidan et al. highlighted significantly longer hospital stays and higher readmission rates among obese patients undergoing CDA.[14] Similarly, studies on the more established ACDF procedure have identified obesity as a comparable risk factor, with adverse events such as wound complications and prolonged recovery times.[8,15,16] Obesity generates unique challenges in cervical spine surgery due to increased subcutaneous fat, which can complicate surgical exposure and radiographic visualization, as well as alter implant biomechanics.[17] However, the anterior approach in cervical spine surgery generally carries little morbidity and near-zero infection rates, making it advantageous for obese patients compared to posterior approaches.[18] Given that obesity poses significant risks in both procedures, a direct comparison is needed to determine which approach is safer for this patient population. Moreover, considering the similar adverse events associated with obesity in both procedures, financial implications warrant attention. To date, CDA has been shown to be more cost-effective than ACDF in both the short and long term.[19,20] However, these cost-effectiveness analyses did not account for patient BMI, which has been demonstrated to increase costs in spine procedures.[21,22,23]

While prior studies have reported promising outcomes for CDA, its relatively recent adoption warrants further investigation to fully evaluate its effectiveness, particularly in high-risk populations such as obese patients. Given the well-documented association between obesity and increased postoperative complications, understanding the specific impact of CDA in this group is critical. This study aims to retrospectively analyze obese patients undergoing single-level CDA or ACDF to compare recovery course, healthcare costs, and rates of postoperative complications. We hypothesize that the CDA cohort will demonstrate a lower incidence of complications and improved recovery outcomes, including decreased rates of nonroutine discharge and shorter lengths of stay, all achieved at a reduced cost.

SUBJECTS AND METHODS

Data collection

The National Inpatient Sample (NIS) database was accessed for analysis and merged with the corresponding cost-to-charge ratio database, covering data from January 2016 through December 2020. Sponsored by the Healthcare Cost and Utilization Project (HCUP) under the Agency for Healthcare Research and Quality, the NIS database is the largest publicly available all-payer inpatient healthcare database, designed to provide U. S. regional and national estimates of inpatient utilization, access, cost, quality, and outcomes.[24] The database samples 20% of all inpatient encounters from U. S. acute-care hospitals, representing over 95% of the U. S. population.[24] In all analyses, weighted estimates were derived using HCUP’s discharge weights. Institutional Review Board approval was not required, as this dataset is publicly available.

Patient population

Patients who underwent single-level ACDF were identified using the International Classification of Diseases, 10^th Revision (International Classification of Diseases [ICD-10]) procedural code 0RG10A0. Patients who underwent CDA were selected using the ICD-10 procedural codes 0RR307Z, 0RR30JZ, and 0RR30KZ. Patients who underwent multi-level CDA were identified and subsequently excluded based on the presence of multiple ICD-10 procedural codes for CDA during the admission period. ICD-10 diagnosis codes were used to classify patients based on the primary admission diagnosis, including radiculopathy (codes M47.2X, M48.0X, M50.1X, M50.2X, M50.3X, M54.1X) and myelopathy (codes M47.1X, M50.0X). Those who did not have a primary admission diagnosis of radiculopathy or myelopathy were excluded. Patients younger than 18 and those without an obesity diagnosis, identified through the Elixhauser comorbidity software for ICD-10-CM from HCUP, were also excluded.[25] Cohorts were defined based on whether patients received ACDF or CDA. Additionally, nonelective patients and those with missing data on variables such as sex, race, income quartile, insurance payer, hospital bed size, teaching hospital status, total costs, and length of stay (LOS) were excluded. Discharge disposition was categorized as routine (home discharge), nonroutine (transfer to another hospital, skilled nursing facility, intermediate care, or home with health services), and other (patients who left against medical advice, died in hospital, or had an unknown destination), with those in the “other” category being excluded.

Variable selection

Collected independent variables included age, sex, race, hospital bed size, teaching hospital/location status, region, charges, and LOS. Charges were converted to costs using cost-to-charge ratios from the respective database. To account for data spanning multiple years, all cost figures were inflation-adjusted to 2020 U. S. dollar values using specific weighting factors.[26] Patient comorbidities were identified with the Elixhauser comorbidity software for ICD-10-CM, provided by HCUP, which detects 38 distinct preexisting conditions based on secondary diagnoses.[25]

Primary outcome variables

The primary outcomes analyzed encompassed postoperative complications, including acute posthemorrhagic anemia, wound disruption, surgical site infection, dysphagia, mechanical complications, hematoma, nervous system complications, acute deep vein thrombosis, myocardial infarction, cerebrovascular accident, venous thromboembolism, pneumonia, acute kidney injury, sepsis, and anesthesia-related complications. Additional outcomes considered were LOS, total hospital admission cost, and discharge disposition.

Statistical analysis

In this analysis, 1:3 propensity score matching was carried out between cohorts, matching for age, sex, race, Elixhauser Comorbidity Index (ECI), and diagnosis. Statistical analyses were performed using R software (version 4.4.0; R Project for Statistical Computing, Vienna, Austria). Chi-square tests and Student’s t-tests were utilized to assess differences in categorical and continuous outcomes, respectively, across age groups, with statistical significance defined at the P < 0.05 level.

RESULTS

Demographics

A total of 26,996 unweighted cases of single-level ACDF and 3,857 cases of CDA were identified in the initial query, corresponding to 134,980 weighted cases of single-level ACDF and 19,285 weighted cases of CDA. After filtering for cases with a diagnosis of obesity, restricting to single-level CDA procedures, and applying all other exclusion criteria, 3,515 weighted cases of ACDF and 485 weighted cases of CDA remained. Prior to propensity score matching, baseline demographic differences were noted between the ACDF and CDA cohorts. The mean age of the ACDF cohort was significantly higher at 54.4 ± 0.4 years compared to 46.6 ± 1.0 years for the CDA cohort (P < 0.001). Racial distribution varied slightly between groups, with a higher proportion of White patients in the CDA group (81.4%) compared to the ACDF group (75.7%; P = 0.087). Female patients made up a slightly larger proportion of the CDA group compared to the ACDF group (60.8% vs. 51.6%, respectively; P = 0.090). Medicare coverage was more common among ACDF patients (33.1% vs. 13.4% in the CDA cohort), whereas private insurance was more prevalent among CDA patients (63.9% vs. 43.5%; P < 0.001; [Table 1]).

Table 1.

Baseline demographic characteristics of patients prior to propensity score-matching

	ACDF (n=3515), n (%)	CDA (n=485), n (%)	P
Age (mean±SD)	54.4±0.4	46.6±1.0	<0.001
Race
White	2660 (75.7)	395 (81.4)	0.087
Black	515 (14.7)	50 (1.4)
Hispanic	220 (6.3)	n=<10*
Asian or Pacific Islander	n=<10*	n=<10*
Native American	n=<10*	n=<10*
Other	70 (2.0)	n=<10*
Sex
Male	1700 (48.4)	190 (39.2)	0.090
Female	1815 (51.6)	295 (60.8)
Insurance
Medicare	1165 (33.1)	65 (13.4)	<0.001
Medicaid	420 (11.9)	n=<10*
Private insurance	1530 (43.5)	310 (63.9)
Self-pay	n=<10*	n=<10*
Other	360 (10.2)	55 (11.3)
Hospital bedsize†
Small	825 (23.5)	115 (23.7)	0.938
Medium	925 (26.3)	135 (27.8)
Large	1765 (50.2)	235 (48.5)
Diagnosis
Myelopathy	1005 (28.6)	75 (15.5)	0.007
Radiculopathy	2510 (71.4)	410 (11.7)
ECI	3.0±0.1	2.5±0.2	<0.001

*Signifies that the count number is <10 and cannot be reported, ^†Hospital bed size designated by the HCUP based on geographic location, teaching hospital status, and number of beds. Bold values are statistically significant. ACDF - Anterior cervical discectomy and fusion; CDA - Cervical disc arthroplasty; n - Number; SD - Standard deviation; ECI - Elixhauser comorbidity index; HCUP - Healthcare cost and utilization project

After 1:3 propensity score matching, 485 cases of single-level CDA were compared against 1,455 cases of single-level ACDF. The mean age was similar, with 47.0 ± 0.6 years in the ACDF group and 46.6 ± 1.0 years in the CDA group (P = 0.720). Racial composition was also comparable, with 75.3% of the ACDF group and 81.4% of the CDA group identifying as White (P = 0.112). The sex distribution remained similar between the groups, with females comprising 61.2% of the ACDF group and 60.8% of the CDA group (P = 0.952; [Table 2]).

Table 2.

Matched demographic characteristics of patients after propensity score-matching

	ACDF (n=1455), n (%)	CDA (n=485), n (%)	P
Age (mean±SD)	47.0±0.6	46.6±1.0	0.720
Race
White	1095 (75.3)	395 (81.4)	0.112
Black	240 (16.5)	50 (10.3)
Hispanic	90 (6.2)	n=<10*
Asian or Pacific Islander	n=<10*	n=<10*
Native American	n=<10*	n=<10*
Other	n=<10*	n=<10*
Sex
Male	565 (38.8)	190 (39.2)	0.952
Female	890 (61.2)	295 (60.8)
Insurance
Medicare	270 (18.6)	65 (13.4)	0.244
Medicaid	235 (16.2)	n=<10*
Private insurance	755 (51.9)	310 (63.9)
Self-pay	n=<10*	n=<10*
Other	170 (11.7)	55 (11.3)
Hospital bedsize†
Small	385 (26.5)	115 (23.7)	0.865
Medium	395 (27.1)	135 (27.8)
Large	675 (46.4)	235 (48.5)
Diagnosis
Myelopathy	250 (17.2)	75 (15.5)	0.695
Radiculopathy	1205 (82.8)	410 (84.5)
ECI	2.5±0.1	2.5±0.2	0.627

*Signifies that the count number is <10 and cannot be reported, ^†Hospital bed size designated by the HCUP based on geographic location, teaching hospital status, and number of beds. P values < 0.05 are statistically significant and bolded. ACDF - Anterior cervical discectomy and fusion; CDA - Cervical disc arthroplasty; n - Number; SD - Standard deviation; ECI - Elixhauser comorbidity index; HCUP - Healthcare cost and utilization project

Perioperative complications and hospital course

The rate of adverse events was comparable between the matched cohorts, with 7.6% in the ACDF group and 8.2% in the CDA group (P = 0.827). The LOS was also similar across both groups, averaging 1.4 ± 0.1 days for each cohort (P = 0.931). The total cost of hospitalization was significantly higher in the CDA group, averaging $19,400 ± 700, compared to $16,400 ± 700 in the ACDF group (P = 0.003). In addition, the rate of nonroutine discharge was substantially lower in the CDA cohort at 4.1%, compared to 11.3% in the ACDF cohort (P = 0.037; [Table 3]).

Table 3.

Comparison of complications and hospital course between cohorts after propensity score-matching

	ACDF (n=1455), n (%)	CDA (n=485), n (%)	P
Adverse events	7.6	8.2	0.827
LOS	1.4±0.1	1.4±0.1	0.931
Cost	$16,424.12±715.48	$19,413.45±691.39	0.003
Nonroutine discharge	11.3	4.1	0.037

P values < 0.05 are statistically significant and bolded. $ = United States Dollars ACDF - Anterior cervical discectomy and fusion; CDA - Cervical disc arthroplasty; n - Number; LOS - Length of stay

DISCUSSION

Obesity presents significant challenges in the treatment of cervical spine disease, as it contributes to higher rates of perioperative complications, prolonged recovery, and increased healthcare costs. These risks make it imperative to evaluate surgical approaches tailored to this high-risk population. CDA has emerged as a preferred alternative to the traditional ACDF for conditions like disc herniation and degeneration, owing to its potential benefits in preserving motion and reducing adjacent segment disease.[9,10,11,12,13] However, the impact of CDA compared to ACDF in obese patients remains underexplored, despite obesity’s growing prevalence and its influence on surgical outcomes. This study revealed that single-level CDA was associated with significantly lower rates of nonroutine discharge, comparable lengths of stay, and higher costs compared to single-level ACDF, while complication rates did not differ significantly between the two groups.

Clinical outcomes and total hospital costs varied between the CDA and ACDF cohorts in our study. After controlling for key confounders such as age, race, diagnosis, and ECI through propensity score matching, we found that the incidence of nonroutine discharge was nearly three times higher in the ACDF cohort compared to the CDA cohort among patients with obesity. This finding underscores the importance of considering a surgical approach when treating obese patients, as CDA appears to offer distinct advantages in this population. Our results align with prior unmatched analyses by Saifi et al. and Lu et al., which demonstrated significantly lower rates of nonroutine discharge in CDA patients. Lu et al. reported routine discharge rates of 95.2% for CDA versus 90.4% for ACDF, while Saifi et al. found rates of 96% for CDA and 89% for ACDF.[27,28] Similarly, a matched analysis by Nunna et al. found the incidence of nonroutine discharge to be nearly twice as high in the ACDF cohort compared to the CDA cohort.[29] By contrast, Ostrov et al., using the nearest neighbors matching algorithm, reported no significant differences in nonhome discharge rates between CDA and ACDF cohorts, highlighting the potential variability in findings based on study design and patient populations.[30] The disparity observed in our analysis may be due to the enhanced postoperative mobility provided by CDA. By preserving cervical motion and avoiding vertebral fusion, CDA reduces stiffness and eases rehabilitation. In addition, CDA’s shorter operative time may contribute to its benefits. In contrast, ACDF involves fusion, which can increase stress on adjacent levels, particularly in obese patients, whose higher body weight amplifies these demands. These findings underscore the importance of tailored surgical planning for obese patients and suggest that CDA may better optimize discharge outcomes and recovery.

Rising healthcare costs remain a critical concern in the American healthcare system, and reducing hospital expenses is a primary goal. In our study, CDA was significantly more expensive than ACDF by approximately $3,000, despite comparable lengths of stay. This finding suggests that factors beyond LOS, such as the cost of the prosthetic disc and the surgical technique itself, may drive the higher expenses associated with CDA. Hospital costs for cervical spine procedures vary widely in the literature, with some studies reporting CDA to have higher short-term costs, while others emphasize its long-term cost-effectiveness.[19,20,29,31] Medicare coverage, which was more common among ACDF patients, may also influence cost differences, as Medicare reimbursements tend to be lower than those of private insurers. In addition, the long-term cost implications of CDA versus ACDF merit consideration. Ghori et al., using a Markov model, found that CDA offers significant long-term cost savings compared to ACDF in younger patients (aged 45-65), likely due to better postoperative recovery and reduced incidence of adjacent segment disease.[19] The impact of obesity on hospital and long-term costs must also be acknowledged, as higher BMI is associated with increased surgical complexity and resource utilization.[21,22,23] Future studies should explore how payer type, procedural costs, and patient-specific factors such as obesity intersect to shape both short- and long-term healthcare expenditures for cervical spine surgeries.

The rate of adverse events in this study did not differ significantly between ACDF and CDA patients. Previous studies have reported lower complication rates for CDA compared with ACDF.[11,29] However, these studies did not specifically account for the effect of obesity on complication rates, a critical consideration given obesity’s well-established role as a risk factor for adverse outcomes.[14] Singh et al. found that CDA may be a safer alternative to ACDF in obese patients, demonstrating lower rates of 90-day emergency department visits and fewer hardware removal procedures at two years postoperatively compared to ACDF. Additionally, Singh et al.’s findings align with the current study, suggesting that CDA provides a comparable complication profile across various levels of obesity.[32] Berry et al. conducted a similar analysis and found that obesity increased the risk of neurologic complications, pulmonary embolism, and wound infection in patients undergoing CDA, while ACDF was associated with higher risks of dysphagia, durotomy, hardware failure, and complications involving cardiac, respiratory, hematologic, and genitourinary systems.[17] Altogether, our study and Berry et al. provide evidence that CDA is noninferior compared to ACDF in terms of the overall complication profile for obese individuals. Our findings, combined with Berry et al.’s results, suggest that CDA is noninferior to ACDF in terms of complication rates in obese individuals, despite differing complication profiles. Importantly, these findings do not suggest that CDA is universally preferable for obese patients. Instead, they highlight that obese patients undergoing ACDF may face higher risks and could benefit from additional preoperative optimization and nutritional interventions to mitigate these risks. Future studies should explore whether preoperative optimization and targeted management strategies can further reduce complications. Given the nuanced differences in complication profiles, surgical decisions should be tailored to the patient’s overall health and risk factors, emphasizing individualized strategies to improve outcomes in obese patients undergoing cervical spine surgery.

This study has several limitations. The retrospective design introduces potential unmeasured confounding and limits causal inference. While propensity score matching controlled for factors such as age, insurance type, and comorbidity burden, residual confounding remains possible, particularly regarding the severity of preoperative symptoms. Younger patients were more likely to receive CDA than ACDF, which may have influenced complication rates, LOS, and discharge outcomes, as younger patients generally experience better recovery. The reliance on the NIS database, though robust, is subject to coding errors and lacks detailed clinical data on surgical techniques, indications, and severity of complications. Self-reported data on costs and complications may introduce reporting bias, while variations in surgeon expertise and institutional practices further limit generalizability. In addition, Medicare coverage was more common in ACDF patients, potentially affecting cost comparisons due to lower Medicare reimbursements. Finally, this study focused on short-term outcomes and did not evaluate long-term complications or quality of life, which are critical for a comprehensive comparison of CDA and ACDF. Prospective, multicenter studies with standardized protocols are needed to better understand the impact of obesity on surgical outcomes and costs.

CONCLUSIONS

Obese patients undergoing ACDF experienced significantly higher rates of nonroutine discharge compared to those undergoing CDA, despite similar lengths of stay. While CDA incurred higher hospital costs, its comparable complication profile and improved discharge outcomes suggest it may be just as safe and should be considered for use in obese patients. These findings highlight the importance of individualized surgical planning, particularly in populations with higher baseline risk factors like obesity. Future studies should explore long-term functional outcomes and cost-effectiveness to further inform decision-making and optimize care for this high-risk population.

Conflicts of interest

Mitchell K. Ng is a paid consultant at Johnson and Johnson Ethicon Inc., CurvaFix Inc., Pacira BioSciences Inc., Sage Products Inc., Alafair Biosciences Inc., Next Science LLC, Bonutti Technologies Inc., Hippocrates Opportunities Fund LLC, and Ferghana Partners Inc. All other authors have no other relevant financial disclosures. The author(s) declare that there is no conflict of interest and received no financial support for the research, authorship, and/or publication of this article. No IRB approval was required for performance of this review.

Funding Statement

Nil.