Prevalence and risk factors for perioperative complications of CKD patients undergoing elective hip surgery

Chronic kidney disease (CKD) is a common and complex disease defined as eGFR (glomerular filtration rate) less than 60 ml/min per 1.73 m² [1]. It is reported that the incidence of CKD doubled in the past decade, and the prevalence of end stage renal disease (ESRD) is expected to increase by 70% from 2000 to 2015 [2]. Therefore, patients with CKD need to be managed carefully during the perioperative period to avoid severe complications.

CKD patients often have disturbances in bone and mineral metabolism due to the impairment of renal function, such as secondary hyperparathyroidism, and lack of erythropoietin (EPO) synthesis and Vitamin D [3]. Previous studies indicate that patients with CKD would suffer a higher occurrence of related complications than patients with normal renal function after orthopedic surgery [4, 5]. Ackland et al. [6] reported that CKD would increase the risk of morbidity in elective orthopedic surgery, and that preoperative glomerular filtration rate (GFR) is correlated with the occurrence of postoperative complications. Deegan et al. [7] reported that end stage renal disease would increase the risk of perioperative complications in patients undergoing total hip arthroplasty (THA) and total knee arthroplasty (TKA), and that the severity of CKD is related to the mortality rate. Kuo et al. [8] concluded that CKD would reflect an increased risk for obtaining fracture nonunion and subsequent surgical revision for an intracapsular hip fracture. However, few studies have focused on the effect of CKD on perioperative complications in patients undergoing hip surgery.

In this study, we conducted a retrospective study aiming to investigate whether CKD would affect morbidities and complications after hip surgery.

All 230 patients received hip surgery successfully, such that no patients had fatal complications after surgery. Preoperative demographic data which included age, sex, congestive heart failure, alcohol, osteoporosis, hepatitis, hyperlipidemia, and albumin showed no significant differences between CKD and non-CKD patients (Table 1). There were significant differences in the number of people with hypertension (90% vs 27.3%, P < 0.001), diabetes (33.3% vs 8.7%, P = 0.01), coronary heart disease (20% vs 2%, P = 0.001), smoking habits (56.7% vs 22.7%, P = 0.016), preoperative anemia (90% vs 19.3%, P < 0.001), and preoperative hemoglobin (94.1 ± 19.7 vs 121.3 ± 18.8, P < 0.001). There was no significant difference in operation time and surgical methods between CKD and non-CKD patients; however, the fraction of patients who received a blood transfusion was significantly higher in CKD patients (Table 2).

Postoperative complication analysis indicated that there were significant differences in the percentage of ICU transfers (73.3% vs 19.3%, P < 0.001), levels of postoperative hemoglobin (92.5 ± 16.8 vs 107.5 ± 18.3, P < 0.001), and levels of postoperative albumin (32.4 ± 4.1 vs 34.9 ± 5.5, P < 0.02) between CKD and non-CKD patients (Table 3). In subgroup analysis, we found that patients received dialysis treatment had a higher proportion of blood transfusion and higher value of postoperative creatinine compared with non-dialysis CKD patients (Table 4). Logistic regression analysis indicated that diabetes, alcohol, and anemia were all independent risk factors for acquiring a blood transfusion (P < 0.05) (Table 5), while age (greater than 60 years), CKD, and osteoporosis were all independent risk factors for an ICU transfer (P < 0.05) (Table 6).

In the present study, CKD patients showed more cardiac diseases and diabetes preoperatively. Furthermore, CKD patients had a higher risk of receiving a blood transfusion and being transferred to the ICU. It was known that patients with CKD were always accompanied with several comorbidities including cardiac, pulmonary, and liver disease preoperatively because the kidney plays an important role in regulating homeostasis. Therefore, it was difficult to manage these patients during the perioperative period. In a retrospective study by Kuo et al. [9], patients with CKD had significantly higher occurrences of cardiovascular diseases, diabetes, and gout before total knee arthroplasty. This association between CKD and other comorbidities was also verified in another study, which enrolled 13,844 patients with a 2-year follow-up [10]. In this present study, there were significant differences in the prevalence of hypertension, diabetes, coronary heart disease, smoking habits, and anemia between CKD and non-CKD patients. In addition, the impairment of renal function and related comorbidities would increase the risk of the operation itself and increase morbidity and mortality after orthopedic surgery. De la Garza Ramos et al. [11] reported that CKD or end-stage renal disease would increase the incidence of postoperative major complications for patients undergoing anterior cervical fusion, and that CKD patients had a 15-fold longer hospital stay. Puvanesarajah et al. [12] reported that CKD patients older than 65 years of age had a higher risk of experiencing major medical complications within 3 months of surgery, and that they had an increased mortality rate by a factor of 2.8 within 12 months after one or two level lumbar fusion surgery. Bains et al. [13] reported that compared with normal renal function dialysis-independent, CKD patients had a 1.78-fold increase in overall mortality rate, while dialysis-dependent CKD patients had a 3.98-fold increase in overall mortality rate after spine surgery. This indicated that preoperative renal disease as an independent risk factor for mortality after surgery.

In this study, there was no significant difference in operation time and surgical methods between CKD and non-CKD patients. However, 50% of CKD patients received an intraoperative blood transfusion compared to 28.5% in non-CKD patients (P = 0.018). In addition, from subgroup analysis, we found that CKD patients who received dialysis treatment had a higher postoperative creatinine level and higher rates of blood transfusion than non-dialysis CKD patients. This result indicated dialysis may have a negative influence on perioperative complications. Augustin et al. [14] reported that patients with GFR, less than 30 ml/min/1.73 m², were at greater odds of receiving a blood transfusion in the hospital (OR 4.42, P < 0.001), meanwhile CKD, operation time, and anemia were all independent risk factors for obtaining a blood transfusion (P < 0.05). Tan et al. [15] also reported that CKD patients (> stage 2) showed an increased likelihood of receiving a blood transfusion. This can be seen in the linear relationship obtained between eGFR and blood transfusion. Unfortunately, in logistic regression analysis, not CKD but diabetes, alcohol, and anemia proved to be independent risk factors for blood transfusion. This phenomenon might attribute to the fact that some CKD patients did not show a severe lack in hemoglobin concentration, while some non-CKD patients lost large amounts of blood due to trauma or other systematic diseases. Moreover, a previous study demonstrated that CKD patients usually had iron-deficient anemia rather than erythropoietin deficiency, and hemoglobin levels could be improved by correcting the iron deficiency effectively. In contrast, anemia or preoperative hemoglobin was a direct indicator that strongly correlated with patient blood transfusions [16]. Graves et al. [17] also reported that CKD alone without anemia did not predispose an increased risk of obtaining a blood transfusion in patients with mild-to-moderate CKD.

CKD patients were likely to have several comorbidities after surgery, which contribute to a slow and difficult recovery. Some patients with complicated comorbidities needed to be transferred to an ICU for subsequent treatment. Adogwa et al. [18] reported that CKD patients had a higher incidence of being transferred to an ICU (52.9% vs 29.4%, P = 0.04) than non-CKD patients after lumbar decompression and fusion. Porter et al. [19] reported that an impairment in kidney function would increase the length of critical care stay, and therefore the severity of kidney impairment was related to the duration of critical care stay. In our study, there was a significant difference in the number of ICU transfers between CKD and non-CKD patients (P < 0.001). Furthermore, logistic regression analysis indicated that CKD was an independent risk factor for ICU transfer (OR 7.60, P < 0.001). We suspect that in CKD patients, the combination of increased mobility, higher mortality [20], and increased burden to the kidney resulted from surgery. The stress from surgery might be responsible for the higher rate of ICU transfer after hip surgery. Furthermore, CKD was not the only risk factor for ICU transfer, our results also showed that patients older than 60 and patients with osteoporosis were both independent risk factors. This indicated that anti-osteoporosis treatment was also important for these patients.

CKD patients showed higher occurrences of preoperative cardiac disease compared with non-CKD patients. CKD patients had a higher risk of receiving a blood transfusion and being transferred to the ICU after hip surgery. Preoperative anemia was an independent risk factor for obtaining a blood transfusion and, therefore, should be focused on during the perioperative period.