Surgical Outcomes and Fusion Rate Following Spine Fusion Surgeries in Patients with Chronic Kidney Disease: According to Kidney Function

PURPOSE. To evaluate the surgical outcomes and fusion rate following lumbar fusion surgeries in patients with chronic kidney disease (CKD) according to kidney function. METHODS. From March 2017 to February 2021, 54 consecutive adult patients with CKD who underwent spine fusion surgery were enrolled. According to the glomerular ltration rate (GFR) categories, 35 patients were classied into the non-end-stage renal disease (ESRD) group (GFR categories 3a–4, eGFR 15–59 mL/min/1.73 m 2 ) and 19 patients into the ESRD group (GFR category 5, eGFR <15 mL/min/1.73 m 2 ). RESULTS. Baseline characteristics did not differ between the groups. The mean number of fused vertebrae (4.9 ± 2.3 vs. 4.1 ± 2.0, p = 0.122), operative time (228.4 ± 129.6 min vs. 160.5 ± 87.5 min, p = 0.113), and surgical bleeding (743.1 ± 630.5 mL vs. 539.5 ± 384.4 mL, p = 0.354) did not differ between the groups. The rates of medical complications (25.7% vs. 52.6%, p = 0.048) and 3-month readmission (8.6% vs. 35.3%, p = 0.045) were signicantly different between the groups. The 3-month mortality tended to be higher in the ESRD group (10.5%) than which in the non-ESRD group (2.9%), but the difference was not statistically signicant (p = 0.280). The rate of pseudarthrosis was signicantly higher in the ESRD group (35.3%) than in the non-ESRD group (9.1%, p = 0.047). CONCLUSIONS. Surgeons should be aware of the high morbidity and the pseudarthrosis when considering spine surgeries in patients with ESRD.


Introduction
Chronic kidney disease (CKD) is a progressive condition characterized by structural and functional changes to the kidneys present for at least 3 months 1 . The global burden of CKD is growing substantially, and approximately 10% of adults suffer from some form of CKD 2,3 . The number of patients with CKD who underwent spine surgeries has also increased due to new medications, advanced hemodialysis techniques, and kidney transplantation 4 .
Previous studies have reported increased morbidity and mortality in patients with CKD following spine surgeries [4][5][6][7][8][9][10] . However, there are few studies regarding bone union following spine fusion surgeries 4,11 , despite the fact that the kidneys and skeleton have an intimate biological relationship that can affect bone strength and metabolism 12,13 . Ho et al. reported a somewhat higher pseudarthrosis rate of approximately 40% in patients with end-stage renal disease (ESRD) following posterolateral fusion or interbody fusion 11 ; however, they did not assess the fusion rate according to kidney function.
We hypothesized that patients with CKD were associated with a lower fusion rate as well as increased postoperative morbidity and mortality. The purpose of this study was to evaluate the surgical outcomes and fusion rate following lumbar fusion surgeries in CKD patients according to kidney function based on the glomerular ltration rate (GFR).

Methods
All methods were carried out in accordance with relevant guidelines and regulations. This study was approved by the institutional review board in Kyung Hee University Hospital at Gangdong (KHNMC 2021-09-022) including waiver of informed consent. We retrospectively reviewed the medical records of adult patients with CKD who underwent spine fusion surgeries at a single academic hospital from March 2017 to February 2021. Patients who underwent simple decompression, discectomy, or local procedures were excluded from this study. Patients who were lost to follow-up within six months following surgery were also excluded. CKD was de ned as abnormalities of kidney structure or function, present for >3 months, and determined by an estimated GFR (eGFR) <60 mL/min/1.73 m 2 (GFR categories G3a-G5) 14 . Based on the GFR categories, 54 patients were evaluated in this study. Among them, 35 patients were classi ed into the non-ESRD group (GFR categories 3a-4, eGFR 15-59 mL/min/1.73 m 2 ) and 19 patients to the ESRD group (GFR category 5, eGFR <15 mL/min/1.73 m 2 ). None of the patients in the non-ESRD group received hemodialysis (HD) or peritoneal dialysis (PD). In the ESRD group, 18 of 19 patients underwent dialysis (17 patients on HD, one patient on PD). One patient in the group did not initiate dialysis. The medical records and radiographic ndings of enrolled patients were reviewed, and comorbidities, medications, laboratory ndings, surgical parameters, perioperative course, clinical outcomes, and complications were evaluated. Additionally, a 3-month readmission (from any cause) and 3-month mortality after discharge were also investigated. Perioperative complications were divided into operative complications (complications directly related surgical procedure such as dural tear, wound problem, and neurological de cit), medical complications (medical illness not directly related to the surgical procedures), and mechanical complications (complications related to instrument failure, such as screw pullout or loosening, loss of correction, cage subsidence, and junctional problem). Fusion was assessed using computed tomography (CT) and radiography. Fusion on CT was de ned as evidence of trabecular bone bridging between the vertebral bodies. Fusion on the radiograph was de ned as less than 5° of angular motion on exion and extension radiographs or where radiolucency lines, which exceed 50% of the upper or lower surface of the implant, with a width of more than 2 mm, did not appear 15,16 .
The mean body mass index in the non-ESRD group was greater than that in the ESRD group (25.8 ± 3.9 kg/m 2 vs. 23.0 ± 3.7 kg/m 2 , p = 0.023). Comorbidities such as hypertension (82.9% vs. 78.9%, p = 0.728), diabetes mellitus (57.1% vs. 57.9%, p = 0.957), coronary artery disease (25.7% vs. 15.8%, p = 0.506), cerebrovascular disease (17.1% vs. 10.5%, p = 0.698), and use of antiplatelet medications (51.4% vs. 47.4%, p = 0.776) did not differ between the groups. Forty patients (74.1%) were examined for bone mineral density (BMD) measured by dual energy X-ray absorptiometry preoperatively, and the mean Tscores were lower in patients in the ESRD group (-2.9 ± 1.3) than in the non-ESRD group (-1.9 ± 1.11, p = 0.020). However, there was no signi cant difference in the proportion of osteoporosis between the groups (34.6% vs. 50.0%, p = 0.343). Regarding laboratory ndings, the mean preoperative values of blood urea nitrogen (27.0 ± 8.8 mg/dL vs. 53.4 ± 17.7 mg/dL, p < 0.001), creatinine (1.5 ± 0.4 mg/dL vs. 7.2 ± 2.8 mg/dL, p < 0.001), eGFR (44.0 ± 10.6 mL/min/1.73 m 2 vs. 8.4 ± 3.5 mL/min/1.73 m 2 , p < 0.001), and hemoglobin (11.6 ± 1.8 g/dL vs. 10.5 ± 1.7 g/dL, p = 0.047) showed signi cant differences between the groups. The mean preoperative potassium level did not differ between the groups (4.4 ± 0.5 mEq/L vs. 4.5 ± 0.8 mEq/L, p = 0.594). The most commonly operated segments were the lumbar spine (12 patients in the non-ESRD group, eight patients in the ESRD group) and thoracolumbar spine (nine patients in the non-ESRD group, four patients in the ESRD group) spine in both groups (  Complications and Outcomes Perioperative complications and outcomes are summarized in Table 3 and the Supplementary Table. There were ve operative complications in the non-ESRD group (dural tear in three patients and wound dehiscence in one patient) and two in the ESRD group (dural tear in one patient, wound dehiscence in three patients). The rate of operative complications did not differ between the groups (11.4% vs. 21.1%, p = 0.431). Approximately half (52.6%) of patients in the ESRD group experienced postoperative medical complications, which were signi cantly greater than those in the non-ESRD group (25.7%, p = 0.048). The most common medical complications were delirium (three patients in the non-ESRD group and ve patients in the ESRD group), followed by urinary problems (four patients in the non-ESRD group). Three patients (two patients in the non-ESRD group, one patient in the ESRD group) suffered from acute deterioration of kidney function, and one of them (in the ESRD group) required initiation of hemodialysis. The 3-month readmission rate was signi cantly higher in the ESRD group (35.3%), than in the non-ESRD group (8.6%, p = 0.045). There were three cases of death within 3 months postoperatively. One patient in the non-ESRD group underwent sudden cardiac arrest immediately after the operation, and died within a week despite resuscitation. In the ESRD group, one patient suffered from status epilepticus associated with uremic encephalopathy and died within 3 months. Another patient died from an unknown cause within 2 months of discharge. The 3-month mortality rate tended to be higher in the ESRD group (10.5%) than in the non-ESRD group (2.9%); however, the difference was not statistically signi cant (p = 0.280). Approximately half of the patients experienced mechanical complications in both groups postoperatively. The most common mechanical complications were screw loosening (16 patients in the non-ESRD group and seven patients in the ESRD group), followed by cage subsidence (seven patients in the non-ESRD group, three patients in the ESRD group). The rate of mechanical complications did not differ between the groups (54.5% vs. 52.9%, p = 0.914). Pseudarthrosis occurred in three patients (9.1%) and six patients (35.3%) in the non-ESRD group and the ESRD group, respectively. The rate of pseudarthrosis was signi cantly higher in the ESRD group than in the non-ESRD group (p = 0.047). The reoperation rate did not differ between the groups (8.3% vs. 0%, p = 0.542). There were three cases of reoperation due to mechanical complications, all of which were in the non-ESRD group. A patient refused revision surgery for mechanical complications in the ESRD group. The rate of revision surgery did not differ between the groups (9.1% vs. 0%, p = 0542).

Discussion
In the present study, we evaluated the surgical outcomes and fusion rates following spine fusion surgeries in patients with CKD according to the severity of kidney function. As expected, the ESRD group demonstrated signi cantly worse outcomes regarding the length of hospital stay, discharge rate to home, medical complications, 3-month readmission, and pseudarthrosis, as compared to those in the non-ESRD group. Other parameters including the ICU admission rate, operative complications, and 3-month mortality did not reach statistical signi cance, although the ESRD group showed worse outcomes than those in the non-ESRD group.
There are ample studies regarding surgical outcomes in patients with CKD in various surgical elds.
Cloyd et al. evaluated the impact of CKD and ESRD on outcomes following major abdominal surgeries 17 . They reported that 30-day mortality (12.8% vs. 1.8%, p < 0.0001) and overall complications (23.5% vs. 12.3%, p < 0.0001) were signi cantly higher in the HD group than in the non-HD group based on the National Surgical Quality Improvement Program database. They also found that even moderate CKD was approaximately 40% in both groups 11 . These results were also true for the present study that ESRD patients demonstrated signi cant higher rate of pseudarthrosis (35.3%) than non-ESRD patients (9.1%).
The exact mechanism of defective bony fusion in patients with CKD following spine surgery has not been documented yet. However, it is well-known that CKD has adverse effects on the skeletal system, including disruptions in mineral metabolism, bone microarchitecture, and increased risk of fractures 23 . Such changes occur early in the course of CKD and worsen with the progression of renal impairment 23 . Moreover, CKD patients may have a signi cant co-prevalence of osteoporosis, although the utility of BMD is undetermined in the later stages of CKD [23][24][25] . Altered bone growth and remodeling processes resulting from combinations of metabolic abnormalities and impairment of bone quality could explain the defective bony fusion in CKD patients following spine surgeries in this study.
The strength of this study was that we conducted a detailed evaluation and comparison of fusion rates between early CKD patients and late CKD (ESRD) patients, in addition to evaluating the clinical outcomes following spine fusion surgeries. Therefore, patients should be informed about the high rate of pseudarthrosis and morbidity following surgery.
The limitations of this study should be noted. First, this study was retrospective in nature, with a relatively short-term follow-up period. Because of the poor general condition associated with CKD, some patients could not make regular follow-ups despite requests for a visit. Second, we classi ed the patients into dichotomous fashion, rather than the degree of CKD stage. This was because of the small number of patients in this study (there were only ve patients with GFR category 4). Therefore, future research with a larger number of patients is necessary to validate this study.

Conclusions
Among CKD patients, ESRD patients were associated with worse outcomes regarding postoperative medical complications, length of hospital stay, and 3-month readmission than those in non-ESRD patients. The pseudarthrosis rate was also signi cantly higher in ESRD patients than in non-ESRD patients. Surgeons should be aware of the high morbidity and the possibility of pseudarthrosis when considering spine surgeries in patients with ESRD.