Abstract
Study Design
Retrospective cross-sectional study.
Objectives
To evaluate the association between renal osteodystrophy (ROD) and in-hospital outcomes following cervical fusion.
Methods
We performed a retrospective analysis of the National Inpatient Sample from 2016 to 2022. Adult patients undergoing cervical fusion were identified and categorized according to the presence or absence of ROD. Propensity score matching was used to balance baseline demographic, hospital, and comorbidity characteristics. Multivariable logistic regression was applied to assess in-hospital complications, and resource utilization was evaluated using length of stay and total hospital charges. A CKD-only sensitivity analysis was additionally performed by comparing patients with ROD and concomitant CKD codes against patients with CKD codes but without ROD.
Results
After matching, 576 ROD and 2,403 non-ROD patients were included. ROD was associated with higher risks of mechanical ventilation (OR 1.429, 95% CI 1.033–1.952), delirium (OR 2.043, 95% CI 1.197–3.386), thrombocytopenia (OR 1.585, 95% CI 1.148–2.163), and transfusion (OR 1.700, 95% CI 1.282–2.236) (all p < 0.05). ROD patients also had a longer length of stay (9 vs 5 days) and higher total hospital charges ($168,428 vs $133,017) (both p < 0.001). In the CKD-only sensitivity analysis, elevated risks persisted for delirium, transfusion, and thrombocytopenia, whereas the mechanical ventilation signal was attenuated.
Conclusions
ROD was associated with a higher burden of perioperative in-hospital complications and greater hospital resource utilization following cervical fusion. These findings suggest that patients with ROD may benefit from closer perioperative risk stratification and multidisciplinary monitoring, particularly with respect to bleeding risk and delirium prevention.
Keywords: renal osteodystrophy, cervical fusion, in-hospital complications, resource utilization, national inpatient sample
1. Introduction
Cervical fusion is widely used for degenerative cervical spine disorders that are refractory to conservative treatment. By restoring spinal stability and alignment after decompression, it can help relieve neural compression and improve symptoms.1,2 The utilization of cervical fusion procedures has increased over time and is projected to continue rising, particularly for more complex multi-level procedures. 3
Adequate bone quality is essential for successful fusion and long-term implant stability, whereas compromised bone stock adversely affects bone–implant integration and postoperative outcomes. 4 Patients with advanced chronic kidney disease (CKD) experience marked deterioration in bone quality, and epidemiologic data show that the global prevalence of CKD continues to rise, imposing an increasing public health burden.5,6 Renal osteodystrophy (ROD), the skeletal manifestation of CKD, is particularly prevalent in advanced disease and typically develops in CKD stages 3–5 and dialysis-dependent patients.7,8 Its pathophysiology is driven by disturbances in mineral and hormonal homeostasis—including calcium, phosphate, parathyroid hormone, and vitamin D metabolism—leading to impaired bone turnover, mineralization, bone volume, and structural integrity. 9
However, prior studies have primarily examined CKD as a broader clinical entity, whereas the present study specifically focuses on ROD to better characterize the role of CKD-related bone metabolic dysfunction in cervical fusion outcomes. In addition, the current evidence largely centers on lumbar or general spine surgery, with relatively limited attention to cervical fusion, despite its unique biomechanical characteristics and the important role of bone quality in fusion success.10,11 Accordingly, this study aims to determine whether pre-existing ROD influences perioperative complications, in-hospital outcomes, and resource utilization among patients undergoing cervical fusion. Leveraging a large national database with propensity score matching and multivariable adjustment, we seek to inform perioperative risk stratification and support targeted management strategies for this high-risk population.
2. Materials and Methods
2.1. Data Source
This retrospective study analyzed data from the National Inpatient Sample (NIS) of the Healthcare Cost and Utilization Project (HCUP) covering 2016–2022. As the largest publicly accessible all-payer inpatient database in the U.S., the NIS captures a stratified 20% sample of annual hospital discharges, including patient demographics, hospital characteristics, diagnoses, procedures, and outcomes. All information was de-identified and HIPAA-compliant; therefore, institutional review board approval was not required.
2.2. Study Population
Adult patients who underwent cervical spine fusion between 2016 and 2022 were identified using ICD-10-PCS procedure codes (prefix match: 0RG0/0RG1/0RG2/0RG4). Patients younger than 18 years, those with incomplete demographic data, and those diagnosed with spinal fracture or spinal tumors were excluded. Missing variables handled by complete-case exclusion included age, sex, race, insurance type, hospital location, length of stay, total hospital charges, and in-hospital mortality status. After applying these criteria, a total of 175,027 patients remained for analysis.
Renal osteodystrophy (ROD) was defined using relevant ICD-10-CM diagnosis codes (ICD-10-CM: N25). 12 Because ROD commonly coexists with advanced CKD and CKD-mineral and bone disorder, this definition should be interpreted as an administrative coding-based marker of CKD-related mineral and skeletal metabolism disorder rather than a laboratory-confirmed ROD phenotype. Among the 655 patients with ROD codes, 614 (93.7%) had concomitant CKD codes, whereas 41 (6.3%) did not have a recorded CKD code. Patients were categorized into an ROD population and a non-ROD population. Of this population, 655 patients (0.37%) were categorized into the ROD group, while the remaining 174,372 patients (99.63%) comprised the control group (Figure 1). To minimize confounding, propensity score matching (variable-ratio up to 1:10, caliper 0.02) was performed based on age, sex, race, hospital region, teaching status, and baseline Elixhauser comorbidities (Figure 2). After matching, the ROD group (n=576) was successfully matched with 2,403 control patients. To evaluate whether the observed associations reflected ROD-specific risk beyond CKD alone, we performed a strict CKD-only sensitivity analysis. In this analysis, the ROD group was restricted to patients with both ROD and CKD codes, and the control group was restricted to patients with CKD codes but without ROD codes. This yielded 614 patients with ROD and concomitant CKD and 10,780 CKD patients without ROD before matching. After propensity score matching, 585 ROD-with-CKD patients were matched to 2,400 CKD-without-ROD controls. Covariate balance for both the primary and CKD-only analyses is reported in Supplemental Table S1, and baseline characteristics of the CKD-only cohort are reported in Supplemental Table S2.
Figure 1.
Flow diagram of patient selection and inclusion
Figure 2.
Propensity score matching performed on the baseline variables
2.3. Variables and Outcomes
Demographic variables included age, sex, race, and insurance type. Hospital characteristics encompassed bed size, teaching status, and geographic region. Clinical characteristics consisted of all Elixhauser comorbidities. In-hospital complications were identified from secondary diagnosis and procedure fields recorded during the same index hospitalization as the cervical fusion procedure. Mechanical ventilation was defined as a coded in-hospital ventilation event, whereas delirium was identified based on administrative diagnosis coding. Because the NIS does not provide sufficiently precise temporal information, these outcomes were considered code-based in-hospital events rather than definitively adjudicated postoperative complications. Measures of resource utilization included length of stay (LOS) and total hospital charge. Patient demographics, hospital characteristics, comorbidities, medical and surgical complications, and outcomes were defined as shown in Table 1.
Table 1.
Variables Used in the Analysis
| Category | Specific variables |
|---|---|
| Patient demographics | Age (≥18 years), sex (male and female), race (White, Black, Hispanic, Asian or Pacific Islander, Native American and Other), household income |
| Hospital characteristics | Type of admission (non-elective, elective), bed size of hospital (small, medium, large), teaching status of hospital (nonteaching, teaching), location of hospital (rural, urban), type of insurance (Medicare, Medicaid, private insurance, self-pay, no charge, other), hospital region (northeast, Midwest or north central, south, west) |
| Comorbidities | AIDS, alcohol abuse, autoimmune condition, chronic blood loss anemia, leukemia, lymphoma, metastatic cancer, solid tumor without metastasis (malignant), cerebrovascular disease (on admission), cerebrovascular disease (sequela), coagulopathy, dementia, depression, diabetes with chronic complications, diabetes without chronic complications, drug abuse, deficiency anemias, heart failure, hypertension (complicated), hypertension (uncomplicated), liver disease (mild), liver disease (moderate to severe), chronic pulmonary disease, neurological disorders affecting movement, other neurological disorders, seizures and epilepsy, obesity, paralysis, peripheral vascular disease, psychoses, pulmonary circulation disease, renal failure (moderate), renal failure (severe), hypothyroidism, other thyroid disorders, peptic ulcer disease, valvular disease |
| Medical and Surgical complications | Acute cerebrovascular disease, acute myocardial infarction, acute respiratory failure, cardiac arrest, deep vein thrombosis, gastrointestinal bleeding, hemorrhage, mechanical ventilation, pulmonary embolism, pneumonia, delirium, shock, surgical site infection, transfusion, thrombocytopenia, urinary tract infection |
| Outcome | Length of stay, total hospital charges, and in-hospital mortality |
2.4. Statistical Analysis
Analyses were conducted using R (version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria). Continuous variables are reported as median (IQR), and group differences were assessed with the Mann-Whitney U test due to non-normal distribution. LOS and total hospital charges were compared nonparametrically as matched resource-utilization outcomes and were not formally modeled using multivariable regression because of their skewed distributions and the absence of a prespecified transformation or generalized linear modeling framework. Categorical variables are shown as counts and percentages, with comparisons performed using chi-square tests or Fisher’s exact test when expected frequencies were below five. After PSM, covariate balance was assessed using standardized mean differences (SMD). Detailed covariate balance diagnostics based on standardized mean differences are provided in Supplemental Table S1. Univariate and multivariate logistic regression models were used to evaluate the association between ROD and in-hospital complications. For analyses in the matched sample, multivariate logistic regression was fitted using the propensity score matching weights, and covariate balance was primarily assessed using standardized mean differences. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported for all outcomes. Statistical significance was defined as p < 0.05, consistent with prior NIS-based studies. 13
A strict CKD-only sensitivity analysis was conducted to assess whether adverse outcome patterns among patients with ROD persisted beyond the presence of CKD alone. This analysis restricted the control group to patients with CKD codes but without ROD codes and restricted the exposed group to patients with both ROD and CKD codes. Propensity score matching and weighted logistic regression were then repeated in this CKD-only cohort using the same analytic framework as the primary analysis.
3. Results
3.1. Prevalence of Renal Osteodystrophy in Patients Undergoing Cervical Fusion
A total of 174,372 cervical fusion patients without ROD and 655 patients with ROD were identified before PSM. After matching, 2,403 non-ROD patients and 576 ROD patients remained for comparative analysis (Table 2).
Table 2.
Baseline Characteristics Before and After Propensity Score Matching
| | Before matching | After matching | ||||
|---|---|---|---|---|---|---|
| Without renal osteodystrophy (n=174372) | With renal osteodystrophy (n=655) | P | Without renal osteodystrophy (n=2403) | With renal osteodystrophy (n=576) | P | |
| Age | 60 (52–68) | 65 (58–72) | <0.001 | 67 (59–74) | 65 (59–72) | 0.001 |
| Age group | ||||||
| <65 | 112106 (64.3%) | 317 (48.4%) | <0.001 | 966 (40.2%) | 266 (46.2%) | 0.065 |
| ≥65 | 62266 (35.7%) | 338 (51.6%) | 1437 (59.8%) | 310 (53.8%) | ||
| Gender | ||||||
| Male | 88872 (51.0%) | 397 (60.6%) | <0.001 | 1349 (56.1%) | 344 (59.7%) | 0.636 |
| Female | 85500 (49.0%) | 258 (39.4%) | 1054 (43.9%) | 232 (40.3%) | ||
| Race | ||||||
| White | 131385 (75.3%) | 264 (40.3%) | <0.001 | 1432 (59.6%) | 249 (43.2%) | <0.001 |
| Black | 21955 (12.6%) | 280 (42.7%) | 601 (25.0%) | 229 (39.8%) | ||
| Hispanic | 12399 (7.1%) | 76 (11.6%) | 209 (8.7%) | 69 (12.0%) | ||
| Asian or Pacific Islander | 3040 (1.7%) | 19 (2.9%) | 74 (3.1%) | 16 (2.8%) | ||
| Native American | 1000 (0.6%) | 8 (1.2%) | 23 (1.0%) | 8 (1.4%) | ||
| Other | 4593 (2.6%) | 8 (1.2%) | 64 (2.7%) | 5 (0.9%) | ||
| Household income | ||||||
| Q1 | 45130 (26.3%) | 216 (33.5%) | <0.001 | 669 (28.3%) | 189 (33.3%) | 0.103 |
| Q2 | 45073 (26.3%) | 164 (25.4%) | 600 (25.4%) | 147 (25.9%) | ||
| Q3 | 43123 (25.1%) | 164 (25.4%) | 620 (26.2%) | 148 (26.1%) | ||
| Q4 | 38154 (22.2%) | 101 (15.7%) | 473 (20.0%) | 83 (14.6%) | ||
| Type of insurance | ||||||
| Medicare | 72301 (41.5%) | 517 (78.9%) | <0.001 | 1702 (70.8%) | 452 (78.5%) | 0.263 |
| Medicaid | 19201 (11.0%) | 39 (6.0%) | 185 (7.7%) | 32 (5.6%) | ||
| Private insurance | 66444 (38.1%) | 80 (12.2%) | 415 (17.3%) | 75 (13.0%) | ||
| Self-pay | 2837 (1.6%) | 9 (1.4%) | 24 (1.0%) | 8 (1.4%) | ||
| No charge | 237 (0.1%) | 0 (0.0%) | 2 (0.1%) | 0 (0.0%) | ||
| Other | 13352 (7.7%) | 10 (1.5%) | 75 (3.1%) | 9 (1.6%) | ||
| ECI | ||||||
| 0 | 31396 (18.0%) | 2 (0.3%) | <0.001 | 20 (0.8%) | 2 (0.3%) | 0.813 |
| 1 | 41510 (23.8%) | 5 (0.8%) | 63 (2.6%) | 5 (0.9%) | ||
| 2 | 40359 (23.1%) | 44 (6.7%) | 274 (11.4%) | 44 (7.6%) | ||
| ≥3 | 61107 (35.0%) | 604 (92.2%) | 2046 (85.1%) | 525 (91.1%) | ||
3.2 Patient Demographics and Comorbidity Characteristics
3.2.1. Before Matching
Significant differences were observed across multiple demographic characteristics between ROD and non-ROD patients. ROD patients were older (65 vs. 60 years; p < 0.001) and exhibited distinct racial distributions, including lower proportions of White patients and higher proportions of Black and Hispanic patients (all p < 0.001). Type of insurance also differed significantly (p < 0.001), with ROD patients more frequently covered by Medicare.
Before matching, ROD patients demonstrated a markedly higher comorbidity burden. Comorbidities significantly more prevalent among ROD patients included: AIDS (1.5% vs. 0.4%; p < 0.001), chronic blood loss anemia (0.8% vs. 0.2%; p = 0.012), lymphoma (2.7% vs. 0.4%; p < 0.001), cerebrovascular disease on admission (4.3% vs. 1.2%; p < 0.001), cerebrovascular disease sequela (4.9% vs. 0.9%; p < 0.001), coagulopathy (15.6% vs. 2.7%; p < 0.001), dementia (2.0% vs. 1.0%; p = 0.013), diabetes with chronic complications (56.6% vs. 9.7%; p < 0.001), drug abuse (4.0% vs. 2.6%; p = 0.025), deficiency anemia (66.4% vs. 6.4%; p < 0.001), heart failure (35.6% vs. 3.7%; p < 0.001), complicated hypertension (20.0% vs. 4.7%; p < 0.001), liver disease—mild (9.9% vs. 3.0%; p < 0.001), liver disease—moderate-to-severe (2.0% vs. 0.3%; p < 0.001), neurological disorders affecting movement (5.2% vs. 3.3%; p = 0.007), other neurological disorders (11.1% vs. 3.2%; p < 0.001), seizures/epilepsy (4.1% vs. 2.2%; p = 0.001), obesity (30.5% vs. 21.7%; p < 0.001), paralysis (21.8% vs. 5.6%; p < 0.001), peripheral vascular disease (9.9% vs. 3.0%; p < 0.001), pulmonary circulation disorders (7.9% vs. 0.7%; p < 0.001), moderate renal failure (14.4% vs. 4.6%; p < 0.001), severe renal failure (82.4% vs. 1.1%; p < 0.001), peptic ulcer disease (1.1% vs. 0.4%; p = 0.016), and valvular disease (10.5% vs. 2.5%; p < 0.001) (Table 3). Conversely, several conditions were significantly less common in ROD patients, including: depression (13.7% vs. 17.1%; p = 0.022), diabetes without complications (3.4% vs. 13.9%; p < 0.001), and uncomplicated hypertension (7.8% vs. 49.5%; p < 0.001) (Table 3). Other comorbidities without statistically significant differences included: alcohol abuse, autoimmune conditions, leukemia, metastatic cancer, solid tumor without metastasis, chronic pulmonary disease, psychoses, hypothyroidism, and other thyroid disorders (Table 3).
Table 3.
Association Between Renal Osteodystrophy and Comorbidities
| Comorbidity | Before matching | After matching | ||||
|---|---|---|---|---|---|---|
| Without renal osteodystrophy (n=174372) | With renal osteodystrophy (n=655) | P | Without renal osteodystrophy (n=2403) | With renal osteodystrophy (n=576) | P | |
| AIDS | 766 (0.4%) | 10 (1.5%) | <0.001 | 24 (1.0%) | 8 (1.4%) | 0.943 |
| Alcohol abuse | 3162 (1.8%) | 6 (0.9%) | 0.086 | 61 (2.5%) | 5 (0.9%) | 0.395 |
| Autoimmune condition | 8467 (4.9%) | 38 (5.8%) | 0.261 | 143 (6.0%) | 32 (5.6%) | 0.601 |
| Chronic blood loss anemia | 358 (0.2%) | 5 (0.8%) | 0.012 | 12 (0.5%) | 3 (0.5%) | 0.622 |
| Leukemia | 314 (0.2%) | 2 (0.3%) | 0.331 | 7 (0.3%) | 2 (0.3%) | 0.352 |
| Lymphoma | 746 (0.4%) | 18 (2.7%) | <0.001 | 30 (1.2%) | 12 (2.1%) | 0.057 |
| Metastatic cancer | 2218 (1.3%) | 5 (0.8%) | 0.246 | 23 (1.0%) | 5 (0.9%) | 0.585 |
| Solid tumor without metastasis, malignant | 2209 (1.3%) | 10 (1.5%) | 0.553 | 42 (1.7%) | 7 (1.2%) | 0.449 |
| Cerebrovascular disease, on admission | 2076 (1.2%) | 28 (4.3%) | <0.001 | 84 (3.5%) | 26 (4.5%) | 0.525 |
| Cerebrovascular disease, sequela | 1631 (0.9%) | 32 (4.9%) | <0.001 | 94 (3.9%) | 27 (4.7%) | 0.824 |
| Coagulopathy | 4759 (2.7%) | 102 (15.6%) | <0.001 | 257 (10.7%) | 86 (14.9%) | 0.224 |
| Dementia | 1764 (1.0%) | 13 (2.0%) | 0.013 | 53 (2.2%) | 12 (2.1%) | 0.651 |
| Depression | 29871 (17.1%) | 90 (13.7%) | 0.022 | 394 (16.4%) | 79 (13.7%) | 0.419 |
| Diabetes with chronic complications | 16918 (9.7%) | 371 (56.6%) | <0.001 | 1256 (52.3%) | 316 (54.9%) | 0.049 |
| Diabetes without chronic complications | 24213 (13.9%) | 22 (3.4%) | <0.001 | 102 (4.2%) | 21 (3.6%) | 0.543 |
| Drug abuse | 4494 (2.6%) | 26 (4.0%) | 0.025 | 102 (4.2%) | 18 (3.1%) | 0.171 |
| Deficiency anemias | 11214 (6.4%) | 435 (66.4%) | <0.001 | 1113 (46.3%) | 359 (62.3%) | 0.226 |
| Heart failure | 6528 (3.7%) | 233 (35.6%) | <0.001 | 610 (25.4%) | 190 (33.0%) | 0.995 |
| Hypertension, complicated | 8280 (4.7%) | 131 (20.0%) | <0.001 | 848 (35.3%) | 124 (21.5%) | 0.019 |
| Hypertension, uncomplicated | 86229 (49.5%) | 51 (7.8%) | <0.001 | 281 (11.7%) | 48 (8.3%) | 0.826 |
| Liver disease, mild | 5304 (3.0%) | 65 (9.9%) | <0.001 | 178 (7.4%) | 50 (8.7%) | 0.495 |
| Liver disease, moderate to severe | 450 (0.3%) | 13 (2.0%) | <0.001 | 41 (1.7%) | 12 (2.1%) | 0.682 |
| Chronic pulmonary disease | 34309 (19.7%) | 113 (17.3%) | 0.119 | 428 (17.8%) | 104 (18.1%) | 0.956 |
| Neurological disorders affecting movement | 5741 (3.3%) | 34 (5.2%) | 0.007 | 119 (5.0%) | 29 (5.0%) | 0.782 |
| Other neurological disorders | 5527 (3.2%) | 73 (11.1%) | <0.001 | 252 (10.5%) | 67 (11.6%) | 0.962 |
| Seizures and epilepsy | 3888 (2.2%) | 27 (4.1%) | 0.001 | 82 (3.4%) | 25 (4.3%) | 0.790 |
| Obesity | 37830 (21.7%) | 200 (30.5%) | <0.001 | 712 (29.6%) | 163 (28.3%) | 0.560 |
| Paralysis | 9739 (5.6%) | 143 (21.8%) | <0.001 | 435 (18.1%) | 108 (18.8%) | 0.344 |
| Peripheral vascular disease | 5177 (3.0%) | 65 (9.9%) | <0.001 | 188 (7.8%) | 49 (8.5%) | 0.783 |
| Psychoses | 6524 (3.7%) | 32 (4.9%) | 0.124 | 129 (5.4%) | 28 (4.9%) | 0.839 |
| Pulmonary circulation disease | 1211 (0.7%) | 52 (7.9%) | <0.001 | 99 (4.1%) | 38 (6.6%) | 0.520 |
| Renal failure, moderate | 8049 (4.6%) | 94 (14.4%) | <0.001 | 777 (32.3%) | 87 (15.1%) | 0.336 |
| Renal failure, severe | 1911 (1.1%) | 540 (82.4%) | <0.001 | 1274 (53.0%) | 461 (80.0%) | 0.859 |
| Hypothyroidism | 21753 (12.5%) | 94 (14.4%) | 0.147 | 426 (17.7%) | 86 (14.9%) | 0.470 |
| Other thyroid disorders | 2439 (1.4%) | 12 (1.8%) | 0.346 | 43 (1.8%) | 12 (2.1%) | 0.861 |
| Peptic ulcer disease | 684 (0.4%) | 7 (1.1%) | 0.016 | 31 (1.3%) | 4 (0.7%) | 0.548 |
| Valvular disease | 4423 (2.5%) | 69 (10.5%) | <0.001 | 172 (7.2%) | 49 (8.5%) | 0.911 |
3.2.2. After Matching
After propensity score matching, covariate balance was assessed primarily using standardized mean differences (SMDs), and the matched groups were considered adequately balanced overall (Supplemental Table S1).
3.3. Hospital Characteristics and Outcomes Before and After Matching
Prior to matching, ROD patients were more likely to be treated in teaching hospitals (88.1% vs. 77.5%; p < 0.001) and urban hospitals (98.8% vs. 97.0%; p = 0.007). LOS and total hospital charges were significantly higher in ROD patients (LOS 9 vs. 2 days; charges $171,312 vs. $87,363; both p < 0.001). In-hospital mortality was also higher among ROD patients (3.5% vs. 0.4%; p < 0.001) (Table 4). ROD patients were more likely to be treated in large hospitals (66.1% vs. 53.2%; p < 0.001) and less likely to be admitted to small hospitals (11.8% vs. 21.1%; p < 0.001). Regional distribution of hospitals likewise showed significant variation (p < 0.001) (Table 4). After matching, no significant differences remained in teaching status, hospital location, bed size of hospital, or mortality. Although in-hospital mortality was higher among patients with ROD before matching, the difference was no longer statistically significant after propensity score matching. However, LOS (9 vs. 5 days; p < 0.001) and total hospital charges ($168,428 vs. $133,017; p < 0.001) continued to be significantly higher in ROD patients (Table 4).
Table 4.
Hospital Characteristics and Outcomes Before and After Propensity Score Matching
| | Before matching | After matching | ||||
|---|---|---|---|---|---|---|
| Without renal osteodystrophy (n=174372) | With renal osteodystrophy (n=655) | P | Without renal osteodystrophy (n=2403) | With renal osteodystrophy (n=576) | P | |
| Bed size of hospital | ||||||
| Small | 36860 (21.1%) | 77 (11.8%) | <0.001 | 370 (15.4%) | 68 (11.8%) | 0.308 |
| Medium | 44805 (25.7%) | 145 (22.1%) | 594 (24.7%) | 131 (22.7%) | ||
| Large | 92707 (53.2%) | 433 (66.1%) | 1439 (59.9%) | 377 (65.5%) | ||
| Region of hospital | ||||||
| Northeast | 31588 (18.1%) | 68 (10.4%) | <0.001 | 346 (14.4%) | 62 (10.8%) | 0.067 |
| Midwest or north central | 33912 (19.4%) | 160 (24.4%) | 533 (22.2%) | 143 (24.8%) | ||
| South | 73960 (42.4%) | 275 (42.0%) | 1039 (43.2%) | 237 (41.1%) | ||
| West | 34912 (20.0%) | 152 (23.2%) | 485 (20.2%) | 134 (23.3%) | ||
| Teaching hospital | 135223 (77.5%) | 577 (88.1%) | <0.001 | 2102 (87.5%) | 503 (87.3%) | 0.925 |
| Urban hospital | 169121 (97.0%) | 647 (98.8%) | 0.007 | 2367 (98.5%) | 568 (98.6%) | 0.800 |
| LOS (median, d) | 2 (1–4) | 9 (4–16) | <0.001 | 5 (2–11) | 9 (4–16) | <0.001 |
| TOTCHG (median, $) | 87,363.00 (55,270.25–143,071.25) | 171,312.00 (105,262.00–290,393.00) | <0.001 | 133,017.00 (77,656.00–233,797.00) | 168,428.00 (105,269.00–286,210.75) | <0.001 |
| In-hospital mortality | 684 (0.4%) | 23 (3.5%) | <0.001 | 52 (2.2%) | 23 (4.0%) | 0.091 |
LOS: Length of stay, TOTCHG: Total hospital charges.
3.4. In-Hospital Complications
3.4.1. Before Matching
Before matching, ROD patients demonstrated significantly higher incidences of multiple in-hospital complications compared with non-ROD patients, including acute cerebrovascular disease (2.0% vs. 0.5%, p < 0.001), acute myocardial infarction (2.6% vs. 0.3%, p < 0.001), acute respiratory failure (13.3% vs. 2.8%, p < 0.001), cardiac arrest (0.6% vs. 0.2%, p = 0.043), deep vein thrombosis (1.1% vs. 0.4%, p = 0.024), gastrointestinal bleeding (1.4% vs. 0.1%, p < 0.001), mechanical ventilation (9.6% vs. 1.9%, p < 0.001), pneumonia (5.0% vs. 1.1%, p < 0.001), delirium (3.7% vs. 0.6%, p < 0.001), shock (2.7% vs. 0.5%, p < 0.001), transfusion (14.5% vs. 1.9%, p < 0.001), thrombocytopenia (10.7% vs. 1.6%, p < 0.001), and urinary tract infection (8.1% vs. 2.6%, p < 0.001) (Figure 3). Complications did not show statistically significant differences, including hemorrhage, pulmonary embolism, and surgical site infection (Figure 3).
Figure 3.
Association between renal osteodystrophy and complications
3.4.2. After Matching
Logistic regression identified several complications associated with ROD after PSM, including mechanical ventilation (OR = 1.429; 95% CI 1.033–1.952; p = 0.027), delirium (OR = 2.043; 95% CI 1.197–3.386; p = 0.007), transfusion (OR = 1.700; 95% CI 1.282–2.236; p < 0.001), and thrombocytopenia (OR = 1.585; 95% CI 1.148–2.163; p = 0.004). ROD was associated with increased risks of mechanical ventilation, delirium, transfusion, and thrombocytopenia (Figure 3). Other in-hospital complications without statistically significant associations included acute cerebrovascular disease, acute myocardial infarction, acute respiratory failure, cardiac arrest, deep vein thrombosis, gastrointestinal bleeding, hemorrhage, pulmonary embolism, pneumonia, shock, surgical site infection, and urinary tract infection (Figure 3).
3.5. CKD-Only Sensitivity Analysis
To determine whether the observed associations were attributable solely to underlying CKD, we performed a strict CKD-only sensitivity analysis. Among the 655 patients with ROD codes, 614 (93.7%) had concomitant CKD codes, whereas 41 (6.3%) did not have a recorded CKD code. Before matching, the CKD-only cohort included 10,780 CKD patients without ROD and 614 patients with both ROD and CKD codes. After propensity score matching, 2,400 CKD-without-ROD patients and 585 ROD-with-CKD patients were included. Covariate balance improved after matching, with all post-matching SMDs below 0.10 in the CKD-only sensitivity analysis (Supplemental Table S1). Baseline, hospital, and comorbidity characteristics of the CKD-only cohort are presented in Supplemental Tables S2 and S3.
In the matched CKD-only cohort, elevated odds persisted for delirium (OR = 1.780; 95% CI, 1.051–2.919; p = 0.026), transfusion (OR = 1.656; 95% CI, 1.263–2.156; p < 0.001), and thrombocytopenia (OR = 1.474; 95% CI, 1.077–1.994; p = 0.013). In contrast, the association between ROD and mechanical ventilation was attenuated and was no longer statistically significant (OR = 1.241; 95% CI, 0.892–1.700; p = 0.189). In-hospital mortality did not differ significantly between groups after matching in the CKD-only cohort (3.4% vs. 3.0%; p = 0.942). LOS and total charges remained higher among patients with ROD and CKD than among CKD patients without ROD (9 vs. 6 days and $170,732 vs. $139,279, respectively; both p < 0.001) (Supplemental Table S2).
4. Discussion
4.1. Overview and Principal Findings
This study examined the impact of ROD on outcomes after cervical fusion. After propensity-score matching, patients with ROD had longer hospital stay, higher charges (both p < 0.001), increased delirium (OR = 2.043; 95% CI 1.197–3.386; p = 0.007), greater need for mechanical ventilation (OR = 1.429; 95% CI 1.033–1.952; p = 0.027), higher transfusion rates (OR = 1.700; 95% CI 1.282–2.236; p < 0.001), and higher odds of thrombocytopenia (OR = 1.585; 95% CI 1.148–2.163; p = 0.004). Importantly, the strict CKD-only sensitivity analysis showed that the associations with delirium, transfusion, and thrombocytopenia persisted among patients with CKD, suggesting that the excess risks of delirium, transfusion, and thrombocytopenia were not fully explained by CKD coding alone. In contrast, the mechanical ventilation association was attenuated in the CKD-only cohort, indicating that this respiratory signal may reflect broader CKD-related illness severity or perioperative vulnerability rather than an ROD-specific effect. After matching, ROD was not significantly associated with in-hospital mortality.
4.2. Possible Mechanisms and Interpretation of Principal Findings
Although variables capturing renal failure severity were included as a covariate in the propensity score matching to minimize its direct confounding effect on the outcomes, ROD itself represents a late-stage skeletal manifestation of CKD-related metabolic disturbances. Accordingly, referencing certain classical pathophysiological processes of CKD in the mechanistic discussion remains scientifically justified, as these processes constitute the biological foundation of ROD rather than merely reflecting global renal dysfunction. The CKD-only sensitivity analysis further suggests that the excess risks of delirium, transfusion, and thrombocytopenia are not fully attributable to CKD coding alone.
4.2.1. ROD and Perioperative Physiological Vulnerability
ROD reflects disordered bone and mineral metabolism in the setting of CKD-MBD, including abnormalities of PTH, calcium, phosphate and FGF23 that alter bone quality and systemic physiology.14,15 Our observations align with prior reports that CKD and related bone disease are associated with higher perioperative morbidity and resource utilization after spine and orthopedic surgery.16-18 Altered bone microarchitecture and decreased mechanical competence have been linked with higher fracture risk and impaired skeletal healing; these pathophysiologic changes plausibly contribute to a more complicated perioperative course after spinal instrumentation and fusion.15,19 Case reports and neurosurgical series have also described instrument failure and nonunion in ROD patients, consistent with biologic bone fragility in this population. 20 Taken together, compromised bone quality and systemic metabolic disturbance likely underlie part of the increased resource use (longer LOS, higher charges) observed in our ROD population.16,21
4.2.2. Transfusion and Thrombocytopenia
The transfusion and thrombocytopenia findings were among the most consistent results across the primary and CKD-only analyses. In addition to impaired bone quality and systemic metabolic derangements, the propensity of CKD patients toward perioperative thrombocytopenia likely reflects a constellation of overlapping pathogenic mechanisms. Chronic accumulation of uremic toxins in CKD has been shown to impair platelet receptor expression, granule release, and aggregation responses, producing a functional platelet defect and reduced platelet lifespan.22,23 CKD patients frequently have hematologic abnormalities (anemia of CKD, uremic platelet dysfunction) and are more likely to bleed or require transfusion during major orthopedic procedures24,25; prior series demonstrate higher transfusion rates and associated resource use in CKD populations undergoing arthroplasty and spine surgery. 21
4.2.3. Mechanical Ventilation and Respiratory Vulnerability
Higher odds of mechanical ventilation were observed in the primary matched analysis. Because this association was attenuated in the CKD-only sensitivity analysis, the ventilation finding should be interpreted cautiously and may reflect broader CKD-related perioperative illness severity in addition to any ROD-related skeletal or metabolic vulnerability. CKD patients are known to have higher rates of perioperative complications and ICU-level care after spine procedures.17,18 A case report has described severe thoracic skeletal deformity related to CKD-MBD resulting in restrictive pulmonary dysfunction, suggesting a biologically plausible pathway for increased respiratory vulnerability in selected patients. 26 On the other hand, pulmonary dysfunction in CKD patients may result directly from circulating uremic toxins or indirectly from factors such as volume overload, anemia, immunosuppression, extra-skeletal calcification, malnutrition, electrolyte disturbances, and/or acid-base imbalances. 27
4.2.4. Delirium
Delirium was more frequent in ROD patients both before and after matching, and remained associated on multivariable analysis (OR = 2.043; p = 0.007). Delirium in patients with ROD may, in part, reflect CKD-related neurotoxicity, including mechanisms implicated in uremic encephalopathy, as the accumulation of various guanidino compounds (GCs) in the body is believed to play an important role in the pathogenesis of uremic encephalopathy. 28 Delirium after spine surgery strongly correlates with advanced age, comorbidity burden and baseline cognitive status.29,30 Prior spine populations likewise report markedly elevated delirium rates among CKD patients and identify similar risk factors.17,29
4.3. Clinical Implications — Preoperative, Intraoperative, and Postoperative Strategies
Our results have several practical implications. First, preoperative optimization of ROD (targeting calcium-phosphate balance, PTH control, correction of anemia and management of platelet dysfunction) should be considered to reduce transfusion and delirium risk and to optimize bone biology for fusion.14,15,19 Second, intraoperative planning — such as meticulous hemostasis, anticipation of increased bleeding risk, and strategies to minimize blood loss — may reduce transfusion dependence in ROD patients. 31 Third, postoperative management should prioritize early delirium screening and prevention, aggressive respiratory care, and close hematologic monitoring; multidisciplinary perioperative pathways may improve outcomes and shorten LOS.17,29,32 Finally, given the demonstrable increase in hospital charges and LOS, institutions should recognize patients with ROD coding as a group with greater observed resource utilization and plan perioperative resources accordingly.21,32
4.4. Limitations
Several limitations should be acknowledged. First, ROD and CKD-related characteristics were identified using administrative codes. Although indicators of renal failure severity and dialysis status were available, their limited clinical granularity precluded reliable severity-based subgroup analyses, and residual confounding related to renal disease severity and CKD-MBD heterogeneity may remain. Second, the NIS cannot establish the precise temporal sequence of diagnoses and procedures; therefore, the assessed complications should be interpreted as code-based in-hospital events. Mechanical ventilation codes may not distinguish routine anesthesia-related ventilation from postoperative or prolonged support, and delirium was identified using administrative diagnosis codes rather than a validated algorithm or prospective assessment. Third, complete-case exclusion may have introduced selection bias, while the nonparametric comparisons of LOS and hospital charges may remain affected by residual confounding. Finally, the absence of radiographic and longitudinal outcomes, including fusion status, hardware failure, pseudarthrosis, and revision surgery, limits the assessment of long-term surgical success.
5. Conclusion
In conclusion, this study provides a comprehensive evaluation of the impact of renal osteodystrophy on perioperative outcomes following cervical fusion. Using a large national population with rigorous propensity-score matching, we found that ROD is associated with several adverse in-hospital outcomes in the primary matched analysis, including increased risks of mechanical ventilation, delirium, thrombocytopenia, and transfusion. ROD was also linked to substantially greater hospital resource utilization, with significantly prolonged length of stay and higher total charges. In the strict CKD-only sensitivity analysis, the associations with delirium, thrombocytopenia, and transfusion persisted, whereas the mechanical ventilation signal was attenuated. These findings suggest that ROD coding may serve as a marker of increased perioperative vulnerability among patients undergoing cervical fusion. Given the growing burden of CKD and its skeletal complications, recognition of ROD as a marker of increased perioperative vulnerability may help inform surgical planning and resource allocation.
Supplemental Material
Supplemental Material for Impact of Renal Osteodystrophy on In-Hospital Outcomes Following Cervical Fusion: A National Inpatient Sample Study by Rongxin Zhang, Yanda Du, Ying Xu, Jian Wang, Keyun Zhang and ZhiGang Ai in Global Spine Journal.
Author Contributions: Rongxin Zhang, Yanda Du, and Ying Xu contributed equally to this work. Rongxin Zhang contributed to study conception, data curation, formal analysis, interpretation of the data, and manuscript drafting. Yanda Du contributed to study design, data analysis, figure and table preparation, and drafting of the manuscript. Ying Xu served as the primary investigator and contributed to study design, data interpretation, and critical revision of the manuscript. Keyun Zhang and Jian Wang contributed to supervision, interpretation of the findings, and critical revision of the manuscript for important intellectual content. ZhiGang Ai contributed to overall supervision, project administration, and critical revision of the manuscript. All authors reviewed the manuscript and approved the final version.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Supplemental Material: Supplemental material for this article is available online.
ORCID iD
ZhiGang Ai https://orcid.org/0009-0000-0836-179X
Ethical Considerations
This study used the National Inpatient Sample, a de-identified publicly available database. In accordance with our institutional policy, ethical approval was not required for this study.
Data Availability Statement
The data that support the findings of this study are available from the Healthcare Cost and Utilization Project National Inpatient Sample (HCUP-NIS) and are obtainable upon purchase and completion of the required data use agreement.*
References
- 1.Rhee JM, Ju KL. Anterior Cervical Discectomy and Fusion. JBJS Essent Surg Tech. 2016;6(4):e37. doi: 10.2106/JBJS.ST.15.00056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Song KJ, Choi BY. Current Concepts of Anterior Cervical Discectomy and Fusion: A Review of Literature. Asian Spine J. 2014;8(4):531-539. doi: 10.4184/asj.2014.8.4.531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mastrokostas PG, Said M, Yusupov D, et al. Trends in Anterior Cervical Discectomy and Fusion: Medicare Projections Through 2060. Global Spine J. 2026;16:2337-2344. doi: 10.1177/21925682251405773. Published online November 29 2025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lechtholz-Zey EA, Ayad M, Gettleman BS, et al. Systematic Review and Meta-Analysis of the Effect of Osteoporosis on Fusion Rates and Complications Following Surgery for Degenerative Cervical Spine Pathology. Int J Spine Surg. 2024;18(3):277-286. doi: 10.14444/8568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.GBD Chronic Kidney Disease Collaboration . Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2020;395(10225):709-733. doi: 10.1016/S0140-6736(20)30045-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ginsberg C, Ix JH. Diagnosis and Management of Osteoporosis in Advanced Kidney Disease: A Review. American Journal of Kidney Diseases. 2022;79(3):427-436. doi: 10.1053/j.ajkd.2021.06.031. [DOI] [PubMed] [Google Scholar]
- 7.Damasiewicz MJ, Nickolas TL. Rethinking Bone Disease in Kidney Disease. JBMR Plus. 2018;2(6):309-322. doi: 10.1002/jbm4.10117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Moe S, Drüeke T, Cunningham J, et al. Definition, evaluation, and classification of renal osteodystrophy: A position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney International. 2006;69(11):1945-1953. doi: 10.1038/sj.ki.5000414. [DOI] [PubMed] [Google Scholar]
- 9.Shah A, Hashmi MF, Aeddula NR. Chronic Kidney Disease-Mineral Bone Disorder (CKD-MBD). In: StatPearls [Internet]. StatPearls Publishing; 2024. https://www.ncbi.nlm.nih.gov/books/NBK560742/. Accessed 7 January 2026. [Google Scholar]
- 10.Pinter ZW, Monsef JB, Salmons HI, et al. Does Preoperative Bone Mineral Density Impact Fusion Success in Anterior Cervical Spine Surgery? A Prospective Cohort Study. World Neurosurg. 2022;164:e830-e834. doi: 10.1016/j.wneu.2022.05.058. [DOI] [PubMed] [Google Scholar]
- 11.Huang H, Liu J, Wang L, Fan Y. A critical review on the biomechanical study of cervical interbody fusion cage. Medicine in Novel Technology and Devices. 2021;11:100070. doi: 10.1016/j.medntd.2021.100070. [DOI] [Google Scholar]
- 12.Tain YL, Li LC, Kuo HC, Chen CJ, Hsu CN. Gestational Exposure to Nonsteroidal Anti-Inflammatory Drugs and Risk of Chronic Kidney Disease in Childhood. JAMA Pediatr. 2025;179(2):171-178. doi: 10.1001/jamapediatrics.2024.4409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Huang X, Dong B, Liu Z, et al. Post-operative pulmonary complications in total knee arthroplasty: Incidence, risk factors, and impact on patient outcomes. Orthopedics and Tissue Engineering. 2025;1(1):6662. doi: 10.36922/ote.6662. [DOI] [Google Scholar]
- 14.Ketteler M, Evenepoel P, Holden RM, et al. Chronic kidney disease–mineral and bone disorder: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney International. 2025;107(3):405-423. doi: 10.1016/j.kint.2024.11.013. [DOI] [PubMed] [Google Scholar]
- 15.West SL, Patel P, Jamal SA. How to predict and treat increased fracture risk in chronic kidney disease. J Intern Med. 2015;278(1):19-28. doi: 10.1111/joim.12361. [DOI] [PubMed] [Google Scholar]
- 16.Karsalia R, Xu E, Hejazi-Garcia C, et al. The effect of chronic kidney disease on short-term single-level lumbar fusion outcomes. Clinical Neurology and Neurosurgery. 2024;244:108459. doi: 10.1016/j.clineuro.2024.108459. [DOI] [PubMed] [Google Scholar]
- 17.Adogwa O, Elsamadicy AA, Sergesketter A, et al. The Impact of Chronic Kidney Disease on Postoperative Outcomes in Patients Undergoing Lumbar Decompression and Fusion. World Neurosurg. 2018;110:e266-e270. doi: 10.1016/j.wneu.2017.10.147. [DOI] [PubMed] [Google Scholar]
- 18.Jo DJ, Choi HY. Surgical Outcomes and Fusion Rate Following Spinal Fusion in Patients with Chronic Kidney Disease: Impact of Kidney Function. World Neurosurg. 2022;168:e587-e594. doi: 10.1016/j.wneu.2022.09.068. [DOI] [PubMed] [Google Scholar]
- 19.Hampson G, Elder GJ, Cohen-Solal M, Abrahamsen B. A review and perspective on the assessment, management and prevention of fragility fractures in patients with osteoporosis and chronic kidney disease. Endocrine. 2021;73(3):509-529. doi: 10.1007/s12020-021-02735-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Veeravagu A, Ponnusamy K, Jiang B, et al. Renal osteodystrophy: neurosurgical considerations and challenges. World Neurosurg. 2012;78(1-2):191.E23-191.E33. doi: 10.1016/j.wneu.2011.09.027. [DOI] [PubMed] [Google Scholar]
- 21.Burns KA, Robbins LM, LeMarr AR, Morton DJ, Wilson ML. Chronic kidney disease increases cost of care and readmission risk after shoulder arthroplasty. Journal of Shoulder and Elbow Surgery. 2024;33(4):e208-e214. doi: 10.1016/j.jse.2023.08.018. [DOI] [PubMed] [Google Scholar]
- 22.Baaten CCFMJ, Sternkopf M, Henning T, Marx N, Jankowski J, Noels H. Platelet Function in CKD: A Systematic Review and Meta-Analysis. J Am Soc Nephrol. 2021;32(7):1583-1598. doi: 10.1681/ASN.2020101440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sreedhara R, Itagaki I, Hakim RM. Uremic patients have decreased shear-induced platelet aggregation mediated by decreased availability of glycoprotein IIb-IIIa receptors. American Journal of Kidney Diseases. 1996;27(3):355-364. doi: 10.1016/S0272-6386(96)90358-3. [DOI] [PubMed] [Google Scholar]
- 24.Friedman AL. Management of the surgical patient with end-stage renal disease. Hemodial Int. 2003;7(3):250-255. doi: 10.1046/j.1492-7535.2003.00045.x. [DOI] [PubMed] [Google Scholar]
- 25.Martin CT, Pugely AJ, Gao Y, Mendoza-Lattes SA, Weinstein SL. The Impact of Renal Impairment on Short-term Morbidity Risk Following Lumbar Spine Surgeries. Spine (Phila Pa 1976). 2015;40(12):909-916. doi: 10.1097/BRS.0000000000000890. [DOI] [PubMed] [Google Scholar]
- 26.Yaxley J, Scott T. Respiratory Failure: A Rare Complication of Chronic Kidney Disease Mineral and Bone Disorder. Ochsner J. 2019;19(3):282-285. doi: 10.31486/toj.18.0177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Gembillo G, Calimeri S, Tranchida V, et al. Lung Dysfunction and Chronic Kidney Disease: A Complex Network of Multiple Interactions. J Pers Med. 2023;13(2):286. doi: 10.3390/jpm13020286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.De Deyn PP, Vanholder R, Eloot S, Glorieux G. Guanidino compounds as uremic (neuro)toxins. Semin Dial. 2009;22(4):340-345. doi: 10.1111/j.1525-139X.2009.00577.x. [DOI] [PubMed] [Google Scholar]
- 29.Kimura A, Shiraishi Y, Sawamura H, Sugawara R, Inoue H, Takeshita K. Predictors of Postoperative Delirium in Older Patients Undergoing Elective Spine Surgery. Spine Surg Relat Res. 2022;7(1):13-18. doi: 10.22603/ssrr.2022-0118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bilotta F, Russo G, Verrengia M, et al. Systematic review of clinical evidence on postoperative delirium: literature search of original studies based on validated diagnostic scales. J Anesth Analg Crit Care. 2021;1(1):18. doi: 10.1186/s44158-021-00021-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Nagrebetsky A, Al-Samkari H, Davis NM, Kuter DJ, Wiener-Kronish JP. Perioperative thrombocytopenia: evidence, evaluation, and emerging therapies. British Journal of Anaesthesia. 2019;122(1):19-31. doi: 10.1016/j.bja.2018.09.010. [DOI] [PubMed] [Google Scholar]
- 32.Jin H, He M, Yang G, et al. A Retrospective Study of the Perioperative Period Management of Joint Arthroplasty in Patients with Chronic Kidney Disease. Orthop Surg. 2022;15(2):591-605. doi: 10.1111/os.13589. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Material for Impact of Renal Osteodystrophy on In-Hospital Outcomes Following Cervical Fusion: A National Inpatient Sample Study by Rongxin Zhang, Yanda Du, Ying Xu, Jian Wang, Keyun Zhang and ZhiGang Ai in Global Spine Journal.
Data Availability Statement
The data that support the findings of this study are available from the Healthcare Cost and Utilization Project National Inpatient Sample (HCUP-NIS) and are obtainable upon purchase and completion of the required data use agreement.*