Abstract
Objectives:
Assess mortality and complications in patients with chronic kidney disease (CKD) after fixation of lower extremity fractures.
Design:
Retrospective cohort.
Setting:
Single academic level 1 trauma center.
Patients:
Four hundred seventy-six patients with CKD and operative lower extremity fractures (pelvis to ankle).
Intervention:
Patients were grouped according to preoperative glomerular filtration rate (GFR) per the National Kidney Foundation Guidelines. Patients with moderate-to-severe CKD and end stage renal disease (ESRD) were compared with those with predisease state CKD.
Outcome Measurements:
Mortality and complication rates at time points up to 2 years or until death.
Results:
Patients with predisease state CKD were significantly less likely to have a complication within 2 years than patients with ESRD but not patients with moderate-to-severe CKD. One-year and 2-year mortality rate were significantly higher in the moderate-to-severe CKD (36.1%) and ESRD (49.0%) groups than in the predisease state CKD group. Compared to the patients with predisease state CKD, patients with moderate-to-severe CKD and ESRD both had increased odds of 1-year mortality, 2-year mortality, and a complication at 2 years after operative fixation of a lower extremity fracture. Patients with ESRD also had increased odds of 30-day and 90-day mortality compared with patients with predisease state CKD.
Conclusions:
Compared to patients with predisease state CKD, patients with moderate-to-severe CKD and ESRD had increased odds of 1-year mortality and 2-year mortality, as well as a complication at 2 years. Patients with ESRD had increased odds of 30-day and 90-day mortality compared with patients with predisease state CKD. Two-year mortality rate was 49% in the ESRD group.
Level of Evidence:
Therapeutic level III.
Keywords: chronic kidney disease, outcomes, lower extremity fractures
1. Introduction
The National Kidney Foundation uses glomerular filtration rate (GFR, mL/min/1.73 m) as a measure of kidney function.1 They describe 5 stages of chronic kidney disease (CKD): stage I > 90, stage II between 60 and 89, stage III between 30 and 60, stage IV between 15 and 29, and stage V < 15.2 Patients are not considered to have clinically meaningful CKD until their GFR falls below 60 (stage III). Patients with CKD stages III–IV are described as having moderate-to-severe disease, and patients with CKD stage V are known to have end-stage renal disease (ESRD).2 Patients with ESRD are generally undergoing dialysis.1,2 The clinical significance of CKD stage I‒II is unclear; however, both are thought to represent a predisease state.2
Moderate-to-severe CKD and ESRD have been shown to be independent risk factors for fragility fractures.3,4 The high risk of fragility fractures is thought to be associated with abnormalities in bone structure and function secondary to CKD.5 Clinical manifestations of this relationship have been reported in the literature on hip fractures. Multiple studies have shown a 3- to 4-fold increase in hip fracture risk in patients with ESRD on dialysis.6–10 Patients with moderate-to-severe CKD have shown a similar increased hip fracture risk, although the magnitude is reduced compared with patients with ESRD.9
The clinical implications of sustaining hip fractures in this population are severe. In patients with moderate-to-severe CKD and ESRD, 1-year mortality rates have been reported up to 37.2%–44.2% after surgical fixation of a hip fracture.8,10 In addition to increased mortality, moderate-to-severe and ESRD have also been found to be independent risk factors for postsurgical complications after hip fracture fixation, including infection, hardware failure, and nonunion.8,10–12
Aside from hip fractures, the existing literature is limited regarding other lower extremity fractures in patients with CKD, despite the presence of multiple locations in the lower extremity at risk for fragility fractures such as the distal femur and ankle.13 Endo et al11 showed that patients with operatively treated lower extremity fractures and ESRD had a 3.6 times greater risk for postoperative complications. However, they did not include patients with moderate-to-severe CKD and did not investigate the mortality. This study aimed to expand the body of literature on CKD and lower extremity fracture outcomes. The hypothesis was that patients with moderate-to-severe CKD and ESRD have increased mortality and surgical complication rates compared with patients with predisease state CKD.
2. Methods
Institutional review board approval was obtained. At a single academic level, one trauma center from 2008 to 2019, patients with CKD stages I‒IV or ESRD who sustained an operative fracture from the pelvic ring to the distal tibia/fibula were retrospectively collected. The International Classification of Diseases (ICD) 9/10 and Current Procedural Terminology (CPT) codes were used to identify eligible patients. Exclusion criteria included patients younger than 18 and greater than or equal to 90 years of age, patients with acute kidney injury at the time of injury, patients without kidney function laboratory data from the time of injury, patients treated nonoperatively, patients operated on at an outside institution, polytraumatized patients (>1 lower extremity fracture during a single injury event), pathologic fractures (other than fragility fractures secondary to underlying osteoporosis), and a diagnosis of hemiplegia or paraplegia. Only the most recent injury was included for patients who had multiple operative traumatic lower extremity injuries within the time interval for our study. Operative cases were completed by subspeciality trained attending surgeons in multiple subspecialties with the assistance of resident surgeon trainees. The minimum follow-up period was 2 years or until death.
Patients were stratified into 3 groups based on the CKD staging established by the National Kidney Foundation. Preoperative GFR from routine laboratory work, obtained at the time of presentation, was used to stratify patients into groups. Specifically, group one included patients with GFR ≥60 (CKD I/II), group 2 included patients with GFR 15–59 (CKD III/IV), and group 3 included patients with GFR <15 (CKD V). Based on prior work by Kim et al,9 group one was designated as predisease state CKD, group 2 was designated as moderate-to-severe CKD, and group 3 was designated as ESRD.
Patient demographics, pertinent medical comorbidities, and the Charlson comorbidity index (CCI) were collected. Fracture characteristics, including date of injury, anatomic location, and closed versus open injury, were collected.
In each group, patients' fractures were classified using the OTA/AO 2018 fracture and dislocation classification compendium.14 The patients' fractures were then grouped into one of 6 subgroups: pelvic, hip, tibial or femoral shaft, periarticular, patella, and ankle. These 6 groups were chosen based on institution-specific postoperative weight-bearing protocols. Specifically, at our institution, periarticular and ankle fractures were made non–weight bearing or limited (touch down) weight bearing postoperatively, whereas tibial/femoral shaft and hip fractures were made weight bearing as tolerated. The pelvic subgroup consisted of OTA/AO type 61 and 62 fractures of the pelvic ring or acetabulum. The hip subgroup consisted of OTA/AO type 31 fractures of the proximal femur (from the femoral head to the subtrochanteric region). The shaft subgroup consisted of OTA/AO type 32 and 42 fractures of the femoral and tibial shafts, respectively. The periarticular subgroup consisted of OTA/AO type 33, 41, and 43 fractures of the distal femur, tibial plateau, and tibial plafond, respectively. The patellar subgroup consisted of OTA/AO type 34 fractures. The ankle subgroup comprised OTA/AO type 4F/44 fractures of the medial, lateral, and posterior malleoli.
Each patient's electronic medical record was reviewed for any postoperative complications or deaths within 2 years of their respective lower extremity surgery. The primary outcomes of interest were mortality and surgical complication rates at 2 years. Surgical complications were subdivided into major and minor. Major surgical complications included hardware failure, nonunion, malunion, deep infection, exposed hardware, peri-implant fracture, dislocation, or amputation. Minor surgical complications include deep venous thromboembolism, superficial infection, or delayed wound healing.
Categorical variables were compared using analysis of variance, chi-square, and Fisher exact tests. Continuous variables were compared using analysis of variance and Student t-test. The alpha level was set at 0.05. Subsequently, a multivariate logistic regression model was used to determine the odds ratio of surgical complications or mortality in relation to CKD severity. The covariates for these models were selected from the comorbidity and demographic data that were significant (P < 0.05) or nearly significant (P < 0.05 + 0.025) among the groups.
3. Results
Four hundred seventy-six patients with CKD who met the inclusion criteria were identified. Their average age was 70.0 years old (22–89 years). There were 40.5% male and 59.5% female. Two hundred six patients had predisease state CKD, 219 patients had moderate-to-severe CKD, and 51 patients had ESRD. The mean length of follow-up was 33.6 weeks (0.0–537.7). As shown in Table 1, age and race differed significantly among the groups (P < 0.001). Specifically, the predisease state CKD group was significantly younger than the moderate-to-severe CKD group by 6.5 years (P < 0.001), and the predisease state CKD group differed significantly from the ESRD group in race categorization (P < 0.001). There were no statistically significant differences in gender, BMI, tobacco use, other drug use, or alcohol use between the groups (P > 0.05).
Table 1.
Patient demographics.
| Predisease state CKD (N = 206) | Moderate-to-severe CKD (N = 219) | ESRD (N = 51) | Overall (N = 476) | P | |
| Gender | 0.631 | ||||
| Male | 90 (43.73%) | 82 (37.4%) | 21 (41.2%) | 193 (40.5%) | 0.423 |
| Female | 116 (56.3%) | 137 (62.6%) | 30 (58.8%) | 283 (59.5%) | 0.949 |
| Age | <0.001 | ||||
| Mean | 67.1 (15.9) | 73.6 (12.3) | 66.4 (14.4) | 70.0 (14.6) | <0.001 |
| Median | 69.0 [22.0, 89.0] | 76.0 [31.0, 89] | 71.0 [31.0, 89.0] | 73.0 [22.0, 89] | 0.952 |
| Race | |||||
| White | 158 (76.7%) | 180 (82.2%) | 21 (41.2%) | 359 (75.47%) | |
| African American | 38 (18.4%) | 30 (13.7%) | 23 (45.1%) | 91 (19.1%) | <0.001 |
| Other | 9 (4.4%) | 7 (3.2%) | 7 (13.7%) | 23 (4.8%) | 0.684 |
| Missing | 1 (0.5%) | 2 (0.9%) | 0 (0%) | 3 (0.6%) | <0.001 |
| BMI | |||||
| Mean | 28.3 (8.59) | 28.7 (7.17) | 28.0 (7.53) | 28.5 (7.84) | 0.922 |
| Median | 26.4 [13.0, 63.0] | 28.2 [15.8, 57.3] | 25.5 [16.1, 51.4] | 27.0 [13.0, 63.0] | 0.863 |
| Missing | 1 (0.5%) | 0 (0%) | 1 (2.0%) | 2 (0.4%) | 0.975 |
| Tobacco | |||||
| Current use | 39 (18.9%) | 25 (9.6%) | 8 (15.7%) | 72 (15.1%) | 0.553 |
| Former use | 69 (33.5%) | 85 (38.8%) | 18 (35.3%) | 172 (36.1%) | 0.296 |
| Never | 98 (47.6%) | 109 (49.8%) | 25 (49.0%) | 232 (48.7%) | 0.999 |
| Other drugs | |||||
| Current use | 8 (3.9%) | 6 (2.7%) | 1 (2.0%) | 15 (2.6%) | 0.951 |
| Former use | 4 (1.9%) | 4 (1.8%) | 2 (3.9%) | 10 (1.9%) | 0.978 |
| Never | 194 (94.2%) | 209 (95.4%) | 48 (94.1%) | 451 (95.5%) | 0.891 |
| Alcohol | |||||
| Current use | 54 (26.2%) | 36 (16.4%) | 10 (19.6%) | 100 (21.0%) | 0.054 |
| Former use | 4 (1.9%) | 12 (5.5%) | 5 (9.8%) | 21 (4.4%) | 0.064 |
| Never | 148 (71.8%) | 171 (78.1%) | 36 (70.6%) | 355 (74.6%) | 0.093 |
| Injury type | 0.988 | ||||
| Isolated fracture | 196 (95.1%) | 207 (94.5%) | 48 (94.1%) | 451 (94.7%) | 0.959 |
| Multiple fractures | 10 (4.9%) | 12 (5.5%) | 3 (5.9%) | 25 (5.3%) | 0.956 |
Bold entries represent any p-value <0.05.
(SD), [Min, Max]. Order of P-values: 1. Predisease State CKD versus Moderate-to-Severe CKD versus ESRD; 2. Predisease State CKD versus Moderate-to-Severe CKD; 3. Predisease State CKD versus ESRD.
The medical comorbidities and CCI are summarized in Table 2. As represented by the CCI scores, the groups differed significantly in comorbidity burden (P < 0.001). Specifically, the predisease state CKD group had significantly lower CCI scores than the moderate-to-severe CKD group by 0.52 (P = 0.007) and the ESRD group by 1.28 (P < 0.001). When adjusted for age, the predisease state CKD groups CCI score remained significantly lower than the moderate-to-severe CKD group by 1.13 (P < 0.001) and the ESRD group by 1.2 (P = 0.005).
Table 2.
Patient comorbidities.
| Predisease state CKD (N = 206) | Moderate-to-severe CKD (N = 219) | ESRD (N = 51) | Overall (N = 476) | P | |
| Cardiovascular disease | 45 (21.8%) | 61 (27.9%) | 25 (49%) | 131 (27.5%) |
0.002 0.359 <0.001 |
| Chronic heart failure | 32 (15.5%) | 68 (31.1%) | 18 (35.3%) | 118 (24.8%) |
<0.001
<0.001 0.006 |
| Chronic pulmonary disease | 53 (25.7%) | 57 (26.0%) | 12 (23.5%) | 122 (25.6%) | 0.987 0.998 0.949 |
| Peripheral vascular disease | 12 (5.8%) | 15 (6.8%) | 8 (15.7%) | 35 (7.4%) | 0.112 0.911 0.063 |
| Cerebrovascular accident | 23 (11.2%) | 34 (15.5%) | 14 (27.5%) | 71 (14.9%) |
0.034 0.419 0.012 |
| Dementia | 26 (12.6%) | 33 (15.1%) | 6 (11.8%) | 65 (13.71%) | 0.870 0.767 0.986 |
| Liver disease | 11 (5.3%) | 15 (6.8%) | 3 (5.9%) | 29 (6.1%) | 0.935 0.810 0.988 |
| Diabetes | 76 (36.9%) | 110 (50.2%) | 27 (52.9%) | 213 (44.7%) |
0.027 0.022 0.097 |
| Diabetes complication | 23 (11.2%) | 37 (16.9%) | 23 (45.1%) | 83 (17.4%) |
<0.001
0.002 |
| Missing | 69 (33.5%) | 118 (53.9%) | 4 (7.8%) | 191 (40.1%) | <0.001 |
| CCI | <0.001 | ||||
| Mean | 1.99 (1.72) | 2.51 (1.68) | 3.27 (1.92) | 2.37 (1.76) | 0.007 |
| Median | 2.00 [0, 7.00] | 2.00 [0, 8.00] | 3.00 [0, 7.00] | 2.00 [0, 8.00] | <0.001 |
| Age-adjusted CCI | <0.001 | ||||
| Mean | 4.29 (2.33) | 5.42 (2.10) | 5.49 (2.27) | 4.94 (2.29) | <0.001 |
| Median | 4.00 [0, 11.0] | 6.00 [0, 11.0] | 5.00 [0, 11.0] | 5.00 [0, 11.0] | 0.005 |
Bold entries represent any p-value <0.05.
Diabetes complication refers to the presence of end-organ damage, and CCI was calculated without renal disease. (SD), [Min, Max] Order of P-values: 1. Predisease State CKD versus Moderate-to-Severe CKD versus ESRD; 2. Predisease State CKD versus Moderate-to-Severe CKD; 3. Predisease State CKD versus ESRD.
As summarized in Table 3, pelvic and patellar fractures represented the smallest cohort of fractures for all groups, with overall occurrences of 2.7% and 0.8%, respectively. Hip fractures represented the largest cohort across all groups, with an overall occurrence of 55.5%. There was a significantly larger proportion of hip fractures by 11.7% (P = 0.043) and a smaller proportion of shaft fractures by 6.3% (P = 0.035) in the moderate-to-severe CKD group than in the predisease state CKD group. The rate of open fracture was significantly higher in the predisease state CKD group by 4.1% than in the moderate-to-severe CKD group (P = 0.018) but not the ESRD group (P = 0.233).
Table 3.
Patient fracture characteristics.
| Predisease state CKD (N = 206) | Moderate-to-severe CKD (N = 219) | ESRD (N = 51) | Overall (N = 476) | P | |
| Fracture group | |||||
| Pelvic | 5 (2.4%) | 8 (3.7%) | 0 (0%) | 13 (2.7%) | 0.469 0.578 0.586 |
| Hip | 102 (49.5%) | 134 (61.2%) | 26 (51.0%) | 262 (55.0%) |
0.043 0.019 0.876 |
| Shaft | 27 (13.1%) | 15 (6.8%) | 7 (13.7%) | 49 (10.3%) | 0.059 0.035 1.00 |
| Periarticular | 31 (15.0%) | 27 (12.3%) | 8 (15.7%) | 66 (13.9%) | 0.635 0.480 1.00 |
| Ankle | 38 (18.4%) | 34 (15.5%) | 10 (19.6%) | 82 (17.2%) | 0.630 0.440 1.00 |
| Patella | 3 (1.5%) | 1 (0.5%) | 0 (0%) | 4 (0.8%) | 0.593 0.359 0.234 |
| Open fracture |
0.022 0.018 0.233 |
||||
| Closed | 191 (92.7%) | 212 (96.8%) | 49 (96.1%) | 452 (95.0%) | |
| Type-1 | 7 (3.4%) | 0 (0%) | 0 (0%) | 7 (1.5%) | |
| Type-2 | 5 (2.4%) | 6 (2.7%) | 0 (0%) | 11 (2.3%) | |
| Type-3a | 3 (1.5%) | 1 (0.5%) | 2 (3.9%) | 6 (1.3%) |
Bold entries represent any p-value <0.05.
Order of P-values: 1. Predisease State CKD versus Moderate-to-Severe CKD versus ESRD; 2. Predisease State CKD versus Moderate-to-Severe CKD; 3. Predisease State CKD versus ESRD.
The complication rates are summarized in Table 4. There was a significant difference in the overall complication rate at 2 years between the groups. Patients with predisease state CKD were significantly less likely to have a complication within 2 years than the patients with ESRD (13.6% vs. 32.4%, P = 0.011). However, there was no significant difference in 2-year complication rate between patients with predisease state and patients with moderate-to-severe CKD (13.6% vs. 21.5%, P = 0.104). Furthermore, the 2-year major complication rate was not significantly different in the predisease state CKD group compared with the ESRD group (11.2% vs. 23.5%, P = 0.07) or the moderate-to-severe CKD group (11.2% vs. 17.8%, P = 0.153). Comparing the individual complications, patients with ESRD were significantly more likely to develop a DVT by 10.3% (P = 0.002), have a superficial wound dehiscence/infection by 5.4% (P = 0.021), or develop a nonunion/malunion by 12.3% (P = 0.013) than patients with predisease state CKD. However, patients with moderate-to-severe CKD were not significantly more likely to develop a DVT (P = 0.822), have a superficial wound dehiscence/infection (P = 0.440), or develop a nonunion/malunion (0.501) than patients with predisease state CKD. In addition, there was no difference among the predisease state CKD group and either the moderate-to-severe CKD group or the ESRD group in the rate of deep infection (P = 0.870), exposure of hardware (P = 0.656), hardware failure (P = 0.368), new fracture (P = 0.091), peri-implant fracture (P = 0.805), or dislocation (P = 0.754). The overall 30-day readmission rate was 9.2%, and the 90-day readmission rate was 12.8%; neither was significantly different among the groups (P = 0.414 and P = 0.730, respectively).
Table 4.
Patient complications at 2 years and readmissions at 30 days and 90 days.
| Predisease state CKD (N = 206) | Moderate-to-severe CKD (N = 219) | ESRD (N = 51) | Overall (N = 476) | P | |
| Complications | |||||
| All complication (1–12) | 28 (13.6%) | 47 (21.5%) | 16 (32.4%) | 91 (19.1%) |
0.020 0.104 0.011 |
| Major complications (4–11) | 23 (11.2%) | 39 (17.8%) | 12 (23.5%) | 74 (15.5%) | 0.096 0.153 0.07 |
| 1. DVT | 3 (1.5%) | 5 (2.3%) | 6 (11.8%) | 14 (2.9%) |
0.001 0.822 0.002 |
| 2. Drainage | 1 (0.5%) | 6 (2.7%) | 0 (0%) | 7 (1.5%) | 0.206 0.189 0.883 |
| 3. Superficial infection | 1 (0.5%) | 4 (1.8%) | 3 (5.9%) | 8 (1.7%) | 0.064 0.440 0.021 |
| 4. Deep infection | 7 (3.4%) | 8 (3.7%) | 3 (5.9%) | 18 (3.8%) | 0.870 0.990 0.714 |
| 5. Exposed hardware | 1 (0.5%) | 1 (0.5%) | 1 (2.0%) | 3 (0.6%) | 0.656 0.999 0.562 |
| 6/7. Nonunion or malunion | 11 (5.3%) | 18 (8.2%) | 9 (17.6%) | 38 (8.0%) |
0.037 0.501 0.013 |
| 8. Hardware failure | 7 (3.4%) | 16 (7.3%) | 3 (5.9%) | 26 (2.6%) | 0.368 0.205 0.714 |
| 9. New fracture | 4 (1.9%) | 0 (0%) | 2 (3.9%) | 6 (1.3%) | 0.091 0.117 0.704 |
| 10. Peri-implant fracture | 4 (1.9%) | 4 (1.8%) | 0 (0%) | 8 (1.7%) | 0.805 0.996 0.605 |
| 11. Dislocation | 2 (1.0%) | 1 (0.5%) | 1 (2.0%) | 5 (0.8%) | 0.754 0.819 0.841 |
| 12. Other | 2 (1.0%) | 1 (1.8%) | 1 (2.0%) | 7 (1.5%) | 0.889 0.756 0.841 |
| Readmissions | |||||
| 30-day readmission | 18 (8.7%) | 18 (8.2%) | 8 (15.7%) | 44 (9.2%) | 0.414 0.982 0.338 |
| 90-day readmission | 24 (11.7%) | 32 (14.6%) | 5 (9.8%) | 61 (12.8%) | 0.730 0.666 0.933 |
Bold entries represent any p-value <0.05.
Order of P-values: 1. Predisease State CKD versus Moderate-to-Severe CKD versus ESRD; 2. Predisease State CKD versus Moderate-to-Severe CKD; 3. Predisease State CKD versus ESRD.
The mortality rates at 30 days, 90 days, 1 year, and 2 years are shown in Table 5. At all time points, the mortality rate increased as CKD severity increased. The 30-day mortality rate was significantly higher in the moderate-to-severe CKD group by 4.9% (P = 0.035) and in the ESRD group by 6.3% (P = 0.043) compared with the predisease state CKD group. The 90-day mortality rate was significantly higher in the moderate-to-severe CKD group by 6.6% (P = 0.040) and in the ESRD group by 11.3% (P = 0.014) compared with the predisease state CKD group. The 1-year mortality rate was significantly higher in the moderate-to-severe CKD group by 12.5% (P = 0.003) and in the ESRD group by 18.2% (P = 0.005) compared with the predisease state CKD group. The 2-year mortality rate was significantly higher in the moderate-to-severe CKD group by 18.6% (P < 0.001) and in the ESRD group by 31.5% (P < 0.001) compared with the predisease state CKD group.
Table 5.
Patient mortality rates at 30 days, 90 days, 1 year, and 2 years.
| Predisease state CKD (N = 206) | Moderate-to-severe CKD (N = 219) | ESRD (N = 51) | Overall (N = 476) | P | |
| Mortality | |||||
| 30 days | 3 (1.5%) | 14 (6.4%) | 4 (7.8%) | 21 (4.4%) | 0.052 0.035 0.043 |
| 90 days | 9 (4.4%) | 24 (11.0%) | 8 (15.7%) | 41 (8.6%) |
0.024
0.040 0.014 |
| 1 year | 23 (11.2%) | 52 (23.7%) | 15 (29.4%) | 90 (18.9%) |
0.002
0.003 0.005 |
| 2 years | 36 (17.5%) | 79 (36.1%) | 25 (49.0%) | 140 (29.4%) |
<0.001
<0.001 <0.001 |
Bold entries represent any p-value <0.05.
Order of P-values: 1. Predisease State CKD versus Moderate-to-Severe CKD versus ESRD; 2. Predisease State CKD versus Moderate-to-Severe CKD; 3. Predisease State CKD versus ESRD.
The results of the multivariate logistic regression model for mortality rate are presented in Table 6. The predictor variables included in the analysis were age, race, cardiovascular disease, chronic heart failure, peripheral vascular disease, cerebrovascular accidents, and diabetes mellitus and diabetes-related complications. When comparing the moderate-to-severe CKD group with the predisease state CKD group, moderate-to-severe CKD was associated with increased odds of mortality at 1 year (OR 2.08 [95% CI 1.17, 3.69]) and 2 years (OR 2.22 [95% CI 1.36, 3.64]). In addition, ESRD was also associated with increased odds of 1-year mortality (OR 2.84 [95% CI 1.22, 6.65]) and 2-year mortality (OR 4.94 [95% CI 2.28, 10.69]), as well as 30-day mortality (OR 5.44 [95% CI 1.01, 29.32]) and 90-day mortality (OR 3.38 [95% CI 1.09, 10.50]) compared with the predisease state CKD group. Specifically, the odds of mortality at 30 days were 5.44 times and at 2 years were 4.94 times higher in the ESRD group than in the predisease state CKD group.
Table 6.
Multivariate logistic regression model for mortality rate.
| 30 days | 90 days | 1 year | 2 years | |
| Moderate-to-severe CKD versus predisease state CKD | 3.11 (0.84, 11.48) | 1.99 (0.87, 4.55) | 2.08 (1.17, 3.69) | 2.22 (1.36, 3.64) |
| ESRD versus predisease state CKD | 5.44 (1.01, 29.32) | 3.38 (1.09, 10.50) | 2.84 (1.22, 6.65) | 4.94 (2.28, 10.69) |
Table 7 shows the results of the multivariate logistic regression model for complication rates. Both moderate-to-severe CKD and ESRD were associated with significantly higher odds of any complication at 2 years (OR 2.15 [95% CI 1.23, 3.76], OR 3.46 [95% CI 1.54, 7.77]). Specifically, the odds of complication were 3.46 times higher in the ESRD group and 2.15 times higher in the moderate-to-severe CKD group than in the predisease state CKD group. Neither the moderate-to-severe CKD nor ESRD group was associated with significantly higher odds of return to the OR, 30-day readmission, or 90-day readmission, compared with the predisease state CKD group.
Table 7.
Multivariate logistic regression model for complications.
| Complication at 2 years | Return to the OR | 30-day readmission | 90-day readmission | |
| Moderate-to-severe CKD versus predisease state CKD | 2.15 (1.23, 3.76) | 1.89 (0.85, 4.20) | 1.07 (0.52, 2.22) | 1.16 (0.63, 2.13) |
| ESRD versus predisease state CD | 3.46 (1.54, 7.77) | 1.30 (0.38, 4.44) | 2.02 (0.74, 5.50) | 0.63 (0.21, 1.88) |
4. Discussion
CKD is a life-altering diagnosis with significant medical sequelae as a patient progresses from a predisease state to moderate-to-severe disease, and finally ESRD.8,10,11,15,16 Complications, such as fractures, are a known occurrence. Goto et al described the relationship between CKD and fractures. They reported that renal osteodystrophy combined with a propensity for falls increased the risk of fracture in patients with CKD. Specifically, they reported that a decreasing GFR was associated with an increased risk of fracture and a graded risk of falls as kidney function worsened.16 However, their review mainly focused on hip and vertebral body fractures. This study aimed to investigate the lower extremity, including the pelvis, in its entirety.
Patients with predisease state CKD were used as comparison group9 to better reflect the prodrome of CKD to provide a more robust comparison to patients in more severe disease states. In this study, the groups differed significantly in age, race, and comorbidity burden (P < 0.001). Patients with moderate-to-severe CKD were, on average, older than the patients with predisease state CKD by 6 and a half years and the predisease state CKD group was disproportionately White compared with the ESRD group. To determine whether a separate subgroup analysis by age was warranted, we tested for interaction between CKD group and age group (18–35, 35–65, and 65–90) in the multivariate logistic regression model. This was done using Wald tests, which assess whether the interaction terms significantly improve model fit. Age group served as the moderator variable. The test results indicated no significant interaction, suggesting that the effect of CKD group on the outcome was consistent across age groups. Furthermore, although there was no significant difference between the moderate-to-severe CKD group and the predisease state CKD group in race, as CKD severity increased, the proportion of each group represented by African American patients increased. African American patients represented 45.1% of patients with ESRD, despite representing 19.1% of all patients. This high representation is in line with the high rates of kidney failure among African Americans in the United States. Furthermore, CKD affects other organ systems in the body, and this was likely captured by the difference in comorbidity burden, represented by CCI, even after adjusting for age. Ultimately, it is not unexpected that a patient with a more severe form of the disease with known multiorgan system manifestations would have a higher comorbidity burden.
As previously reported, when stratifying operatively treated lower extremity fractures that occurred in this study, hip fractures represented the largest cohort. The proportion of hip fractures was significantly higher (P = 0.006) and the proportion of shaft fractures was significantly lower (P = 0.008) in the moderate-to-severe CKD group than in the predisease state CKD group. To evaluate whether a subgroup analysis by hip fracture status was needed, we tested for an interaction between CKD group and hip fracture status in the logistic regression model. The interaction was not statistically significant (P > 0.05), suggesting that the effect of CKD on the outcome was consistent between patients with and without hip fractures. Furthermore, the higher occurrence of hip fractures in the moderate-to-severe group than in the ESRD group may be related to more stringent metabolic monitoring and management, which is absent in patients with earlier stages of CKD. However, the lower occurrence of shaft fractures in the moderate-to-severe CKD group did not support this finding. It is possible that as renal osteodystrophy progresses, the location of bone demineralization in the lower extremity changes, contributing to fewer hip fractures. Further investigations are required to better understand this observation.
The goal of this study was to assess the mortality and complication rates of surgically treated lower extremity fractures in CKD patients. At all time points, the mortality rate increased as CKD severity increased. Moreover, multivariate logistic regression modeling demonstrated that compared with the predisease state CKD group, both the moderate-to-severe CKD group and ESRD group had significantly increased odds of a 1- and 2-year mortality event after operative treatment of a lower extremity fracture (OR 2.08 [95% CI 1.17, 3.69], OR 2.22 [95% CI 1.36, 3.64] and OR 2.84 [95% CI 1.22, 6.65], OR 4.94 [95% CI 2.28, 10.69]). This corroborates the conclusions of other studies that have identified moderate-to-severe CKD and ESRD as risk factors for increased mortality after hip fractures while allowing for increased generalizability to other lower extremity fractures.7,8 Of note, patients with ESRD had increased odds of having 30-day and 90-day mortality events compared with patients with predisease state CKD. The odds ratios for mortality in ESRD patients were highest at 30 days and 2 years. Specifically, a 5.44 [95% CI 1.01, 29.32] and 4.94 [95% CI 2.28, 10.69] increased odds of a mortality event in patients with ESRD, respectively; however, regarding the 30-day mortality OR, this was also accompanied by the widest confidence interval because of the relatively low overall mortality events at this time point.
Similar to mortality, the overall complication rate at 2 years increased with increasing CKD severity. The ESRD group, but not the moderate-to-severe CKD group, had a significantly higher 2-year overall complication by 18.8% (P = 0.011). The 2-year major complication rate in the ESRD group was 12.3% higher but not statistically significant (P = 0.070). Assessed individually, the ESRD group had significantly higher rates of DVT (P = 0.001), superficial infection (P = 0.021), nonunion (P = 0.013), and malunion (P = 0.013) than those in the predisease state CKD group. Interestingly, the moderate-to-severe CKD group complication rates were not significantly higher than those of the predisease state CKD groups for any of the individually investigated complications at up to 2 years. However, multivariate logistic regression modeling demonstrated that compared with the predisease state CKD group, both the moderate-to-severe CKD group and ESRD groups had significantly increased odds for any complication at up to 2 years (OR 2.15 [95% CI 1.23, 3.76], 3.46 [95% CI 1.54, 7.77]). Compared to the moderate-to-severe group, the ESRD group had a 38% increase in the odds of a complication at 2 years. Patients with ESRD encompass a more significant disease burden, likely contributing to this observed difference. The significantly increased odds of a complication did not coincide with the significantly increased odds of returning to the operating room, 30-day readmission, or 90-day readmission. Therefore, it is possible that the increased odds of a complication represent more minor and not major complications necessitating readmission or return to the operating room. This is supported by the observed major complication rate being increased, but not significantly increased, in the ESRD group by 12.3% compared with the predisease state CKD group.
This study was limited in its ability to make inferences regarding how the various lower extremity fracture types in patients with CKD relate to complications and mortality, given the relatively small number of patients in each group. This could be improved with a larger number of patients using registry data or data from multiple centers. Large datasets would allow for the matching of specific OTA/AO fracture types between groups, which was lacking in this study. Moreover, this study was limited by its retrospective design and was subject to errors associated with incomplete data that are inherent to retrospective reviews. Relying solely on the electronic medical records within a single hospital system for mortality tracking presented a limitation, as patients may have died outside the system. The addition of a nonoperative group comparison group would also help to expand the understanding of the risks associated with both nonoperative and operative treatment. The outcomes of nonoperatively treated pelvic ring and ankle fractures are of particular interest. Literature regarding the nonoperative treatment outcomes of ankle fractures in patients with diabetes mellitus exists, but there is a void in the literature regarding patients with CKD.
In summary, compared to patients with predisease state CKD, patients with moderate-to-severe CKD and patients with ESRD had significantly increased odds of 1-year and 2-year mortality, as well as higher odds of complication at 2 years. Patients with ESRD had increased odds of 30-day and 90-day mortality compared with patients with predisease state CKD. These data can be used to counsel patients with CKD and operative lower extremity fractures.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
Footnotes
Previous Presentations: Poster presentation at OTA 2022.
The authors report no conflict of interest.
Source of funding: Nil.
Contributor Information
Alexander D. Jeffs, Email: alexander.jeffs2@unchealth.unc.edu.
Kristin A. Toy, Email: kristintoy@gmail.com.
Beau R. White, Email: beau.white@unchealth.unc.edu.
Feng-Chang Lin, Email: flin33@email.unc.edu.
Di Hu, Email: dihu0103@live.unc.edu.
Margaret D. Fisher, Email: maggiefisher225@gmail.com.
Jason McCartney, Email: mccartneyj@uthscsa.edu.
Judith A. Siegel, Email: jodi_siegel@med.unc.edu.
Andrew T. Chen, Email: andrew_chen@med.unc.edu.
References
- 1.Webster AC, Nagler EV, Morton RL, et al. Chronic kidney disease. Lancet. 2017;389:1238–1252. [DOI] [PubMed] [Google Scholar]
- 2.Levey AS, De Jong PE, Coresh J, et al. The definition, classification, and prognosis of chronic kidney disease: a KDIGO controversies conference report. Kidney Int. 2011;80:17–28. [DOI] [PubMed] [Google Scholar]
- 3.Nickolas TL, McMahon DJ, Shane E. Relationship between moderate to severe kidney disease and hip fracture in the United States. J Am Soc Nephrol. 2006;17:3223–3232. [DOI] [PubMed] [Google Scholar]
- 4.Coco M, Rush H. Increased incidence of hip fractures in dialysis patients with low serum parathyroid hormone. Am J Kidney Dis. 2000;36:1115–1121. [DOI] [PubMed] [Google Scholar]
- 5.Kazama JJ. Chronic kidney disease and fragility fracture. Clin Exp Nephrol. 2017;21(suppl 1):46–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Maravic M, Ostertag A, Torres PU, et al. Incidence and risk factors for hip fractures in dialysis patients. Osteoporos Int. 2014;25:159–165. [DOI] [PubMed] [Google Scholar]
- 7.Alem AM, Sherrard DJ, Gillen DL, et al. Increased risk of hip fracture among patients with end-stage renal disease. Kidney Int. 2000;58:396–399. [DOI] [PubMed] [Google Scholar]
- 8.Lin JCF, Liang WM. Mortality and complications after hip fracture among elderly patients undergoing hemodialysis. BMC Nephrol. 2015;16:100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kim SH, Yi SW, Yi JJ, et al. Chronic kidney disease increases the risk of hip fracture: a prospective cohort study in Korean adults. J Bone Miner Res. 2020;35:1313–1321. [DOI] [PubMed] [Google Scholar]
- 10.Suh YS, Won SH, Choi HS, et al. Survivorship and complications after hip fracture surgery in patients with chronic kidney disease. J Korean Med Sci. 2017;32:2035–2041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Endo A, McTague MF, Allen E, et al. End-stage renal disease increases risk of postoperative complications after lower extremity fracture. Arch Orthop Trauma Surg. 2021;141:925–928. [DOI] [PubMed] [Google Scholar]
- 12.Karaeminogullari O, Demirors H, Sahín O, et al. Analysis of outcomes for surgically treated hip fractures in patients undergoing chronic hemodialysis. J Bone Joint Surg Am. 2007;89:324–331. [DOI] [PubMed] [Google Scholar]
- 13.Court-Brown CM, Caesar B. Epidemiology of adult fractures: a review. Injury. 2006;37:691–697. [DOI] [PubMed] [Google Scholar]
- 14.Meinberg EG, Agel J, Roberts CS, et al. Fracture and dislocation classification Compendium-2018. J Orthop Trauma. 2018;32(suppl 1):S1–S170. [Google Scholar]
- 15.Tentori F, McCullough K, Kilpatrick RD, et al. High rates of death and hospitalization follow bone fracture among hemodialysis patients. Kidney Int. 2014;85:166–173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Goto NA, Weststrate ACG, Oosterlaan FM, et al. The association between chronic kidney disease, falls, and fractures: a systematic review and meta-analysis. Osteoporos Int. 2020;31:13–29. [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.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.