Role of total hip replacement arthroplasty between transplantation and acute kidney injury

Abstract

Osteonecrosis of the femoral head is a severely disabling complication of steroid immunosuppression in renal transplant patients. The increased number of patients undergoing transplantation has increased the number of transplant recipients undergoing total hip replacement arthroplasty (THRA). In this study, we retrospectively assessed patients who underwent THRA from May 2004 to February 2014, and evaluated their demographic and clinical characteristics, the results of peri-operative laboratory tests, the amounts of fluids transfused during surgery, and anesthesia time. Our results found that post-operative acute kidney injury (AKI) was significantly associated with transplantation, and transplantation was an independent factor predictive of post-operative AKI, so transplant recipients are at risk for AKI following THRA. Total hip replacement is a safe and effective treatment for transplant recipients and, in view of their limited life expectancy, should be considered at an early stage in their treatment.

• AKI
• AVN
• recipients
• THRA
• transplantation

Introduction

Osteonecrosis associated with steroid therapy was first described by Pietrogrande and Mastomarino in 1957, and later reported in patients after renal transplants,1,2 organ transplant recipients have a greater likelihood of undergoing elective or emergency surgery unrelated to their transplant. The femoral head is the most common site to be involved, often bilaterally and, since transplantation for end-stage renal disease became established, it is recognized as a cause of significant morbidity after otherwise successful transplant surgery, with the major cause being avascular or ischemic osteonecrosis and osteoporosis.3,4 Avascular osteonecrosis (AVN) in these patients may be due to their chronic use of immunosuppressants and glucocorticoids and changes in lipid metabolism following transplantation, resulting in osteoporosis, loss of bone mineral density, and a reduction in vascular supply to bones. Many organ transplant recipients develop AVN, with some undergoing total hip replacement arthroplasty (THRA).5–8 The development of acute kidney injury (AKI) has been associated with poor patient outcomes. Post-operative renal impairment has been shown to increase in-hospital morbidity and mortality rates and length of hospitalization, as well as to enhance the possibility of needing extended care. AKI may occur more often and be more severe, outcomes that may be due to long-term use of immunosuppressants and differences in underlying diseases. Early diagnosis and appropriate management are important in patients who develop peri-operative AKI.9,10 The few series of total hip replacements reported indicate generally good results, but a possible increase in complication rate for the relatively short follow-up times. Therefore, in this study, we reported the effect of transplanted organs on the risk of AKI in transplant and non-transplanted patients undergoing THRA.

Patients and methods

We retrospectively evaluated 3500 patients who underwent THRA from May 2004 to February 2014, assessed baseline demographic and clinical characteristics, comorbidities, medications, and pre-operative laboratory findings. We included only those who underwent general anesthesia with sevoflurane and rocuronium or atracurium. Anesthesia was induced in all patients with thiopental or propofol. Rocuronium or atracurium was used as a muscle relaxant, depending on the patient’s status. Patients were monitored during surgery by electrocardiography, pulse oximetry, and invasive surveillance of arterial blood pressure. Crystalloids were administered to preserve adequate urine output. Intraoperative blood loss was compensated by transfusing colloid solution, and packed red blood cells were given to maintain hematocrit >30%. After the completion of surgery, patients were extubated and transferred to the post-anesthesia care unit. The patients were divided into two groups: transplant recipients and non-transplanted patients. Each patient’s demographic information, anesthetic time, estimated blood loss during surgery, and amounts of fluid and red blood cells transfused were recorded.

Results

We identified 3500 patients who met the inclusion criteria, with 500 being solid organ transplant recipients and 3000 not. Of the 500 transplant recipients, 400 had received kidney and 100 liver transplants. For the analysis, we extracted a propensity score matched subset of 300 transplant recipients and 600 non-transplanted patients. We excluded 50 patients who were undergoing dialysis treatment, because of difficulties assessing the effect of transplantation on AKI based on a decrease in glomerular filtration rate. The pre-operative characteristics of the entire sample and of the groups who did and did not undergo transplantation, and of the subgroups before and after propensity matching of risk factors are presented in Tables 1 and 2, respectively. Serum Cr concentration in transplant recipients before surgery immediately increased after Trac. Table 3 summarizes the estimated effects of transplantation on AKI through risk estimates and ORs. Analysis of the matched cohort of patients showed that the incidence of AKI remained significantly higher in transplant than in matched non-transplanted patients with normal Cr. Multivariable logistic analysis showed that trans-plantation was independently associated with post-operative AKI in Table 4. Additionally, the effects of particular transplanted organs on the risk of AKI were assessed in patients undergoing THRA. Table 5 presents risk estimates and ORs of organ transplantation on the development of AKI. Post-operative AKI was significantly more frequent in the kidney transplant recipients than in the matched non-transplanted patients and in the liver transplant recipients than in the matched non-transplanted patients. Kidney and liver transplant recipients developed AKI at a similar rate. However, AKI was more severe in kidney than in liver transplant patients who developed AKl.

Table 1. Patient characteristics.

Variable	Transplant patients (N = 500)	Non-transplanted patients (N = 3000)	All patients (N = 3500)	p Value
Age (y)	45.21 ± 10.57	56.41 ± 14.12	55.13 ± 14.32	< 0.0001*
Weight (kg)	65.05 ± 12.45	62.59 ± 10.91	63.20 ± 11.21	0.005*
HTN	200 (40.00)	900 (30.00)	1100 (31.43)	<0.0001*
DM	60 (12.00)	450 (15.00)	510 (14.57)	0.87
IHD	40 (8.00)	310 (10.33)	350 (10.00)	0.95
CKD	150 (30.00)	107 (3.57)	257 (7.34)	<0.0001*
CVD	50 (10.00)	116 (3.87)	166 (4.74)	0.23
Anesthetic time (min)	165.15 ± 32.41	165.30 ± 51.74	162.30 ± 52.34	0.94
EBL (mL)	435.21 ± 154.52	650.61 ± 424.24	612.71 ± 402.52	<0.0001*
Total fluid during anesthesia (mL)	1854.56 ± 586.34	2386.90 ± 1873.41	2237.24 ± 1855.43	<0.0001*

Notes: Data are expressed as mean ± SD (%). Abbreviations: HTN, hypertension; DM, diabetes mellitus; IHD, ischemic heart disease; CKD, chronic kidney disease; CVD, cerebrovascular disease; EBL, estimated blood loss.

*p < 0.05, transplant versus non-transplanted patients.

Table 2. Distribution of risk factors in a propensity-matched cohort.

Variable	Transplant patients (n = 300)	Matched non-transplanted patients (n = 600)	All patients (n = 900)	p Value
Age (y)	45.04 ± 10.24	47.25 ± 16.40	46.20 ± 15.21	0.82
Weight (kg)	65.31 ± 13.20	65.53 ± 14.12	64.75 ± 13.80	0.75
HTN	120 (40.00)	240 (40.00)	360 (40.00)	0.91
DM	35 (11.67)	90 (15.00)	115 (12.78)	0.89
IHD	15 (5.00)	25 (4.17)	40 (4.45)	0.78
CKD	30 (10.00)	72 (12.00)	102 (11.33)	0.93
CVD	9 (3.00)	24 (4.00)	33 (3.67)	0.85
Anesthetic time (min)	164.71 ± 34.24	162.13 ± 83.25	160.25 ± 71.24	0.75
EBL (mL)	442.24 ± 160.21	468.10 ± 280.31	462.52 ± 226.80	0.41
Total fluid during anesthesia (mL)	1,941.49 ± 601.62	1,928.76 ± 978.08	1,932.34 ± 870.24	0.90

Notes: Data are expressed as mean ± SD or n (%).

Abbreviations as in Table 1.

Table 3. Effects of transplantation on risk of AKI in patients undergoing THRA.

	Risk estimate	Transplant patients (n = 300)	Matched non-transplanted patients (n = 600)	Odds Ratio (95% CI)	p Value
All patients	No Cr change	220 (73.33)	592 (98.67)	–	<0.0001*
	Stage I	35 (11.67)	6 (1.00)	–	–
	Stage II	5 (1.67)	4 (0.60)	–	–
	Stage III	40 (13.33)	3 (0.50)	–	–
	AKI	80 (26.67)	13 (2.1)	28.85 (10.14–82.08)	<0.001*
Within normal range of Cr	No Cr change	106 (73.61)	595 (99.17)	–	<0.0001*
	Stage I	32 (10.67)	4 (0.67)	–	–
	Stage II	2 (0.67)	0 (0.00)	–	–
	Stage III	42 (14.00)	0 (0.00)	–	–
	AKI	76 (25.33)	4 (0.67)	51.06 (12.11–215.30)	<0.0001*
CKD	No Cr change	30 (10.00)	70 (11.67)	–	0.017*
	Stage I	5 (0.17)	0 (0.00)	–	–
	Stage II	6 (2.00)	2 (0.33)	–	–
	Stage III	0 (0.00)	2 (0.33)	–	–
	AKI	10 (33.33)	4 (0.66)	6.61 (1.15–38.07)	0.032*

Notes: Data are expresses as n (%)

Abbreviations: AKI, renal function deterioration more than stage 1; THRA, total hip replacement arthroplasty; Cr, Creatinine; CKD, chronic kidney disease; CI, confidence interval.

*p < 0.05, transplant versus non-transplanted patients.

Table 4. Independent predictors of AKI in patients undergoing THRA.

Predictor	Odds Ratio	95% CI	p Value
Transplantation	31.42	11.50–89.05	<0.0001*
Gender (M/F)	0.81	0.44–1.36	0.31
Age (y)	1.07	0.89–1.14	0.43
Weight (kg)	1.15	1.05–1.12	0.04*
HTN	1.83	0.91–3.80	0.11
DM	1.12	0.35–3.22	0.96
IHD	0.83	0.14–4.56	0.87
CKD	1.35	0.51–4.31	0.58
Anesthetic time (min)	1.01	0.91–1.02	0.66
EBL (mL)	1.00	1.00–1.21	0.47
Total fluid during anesthesia (mL)	1.02	1.00–1.15	0.98

Notes: Abbreviations as in Tables 1 and 3. *p < 0.05.

Table 5. Effect of transplanted organs on the risk of AKI in patients undergoing THRA.

	Risk estimate	Transplant patients	Matched non-transplanted patients	Odds Ratio (95% CI)	p Value
Kidney (n = 400)					<0.0001*
	No Cr Change	280 (70.00)	494 (99.40)	–	–
	Stage II	10 (2.5)	1 (0.20)	–	–
	Stage III	60 (15.00)	0 (0.00)	–	–
	AKI	120 (30.00)	3 (0.60)	31.44 (9.41–105.05)	<0.0001*
Liver (n = 100)					<0.0001*
	No Cr Change	80 (80.00)	98 (98.99)	–	–
	Stage II	–	–	–	–
	Stage III	2 (2.00)	1 (1.01)	–	–
	AKI	20 (20.00)	1 (1.01)	22.25 (2.72–182.05)	<0.0001*

Notes: Data are expressed as n (%).

Abbreviations as in Table 3.

*p < 0.05 transplant versus non-transplanted patients.

Discussion

Disabling osteonecrosis in transplant patients on long-term steroid immunosuppression is a major complication, which may severely impair their rehabilitation. It presents a dilemma to the surgeon since the patients are generally young, osteoporotic and on immunosuppression. This study was conducted in patients undergoing THRA for AVN. AVN is associated with abnormalities in the vascular supply of bone, particularly at the femoral head; as well as with bone diseases (bone mineral density and fracturing), changes in lipid metabolism, and post-transplantation glucocorticoid use.11,12 Abnormally elevated lipid concentrations are related to microemboli, endothelial cell changes, venous stasis, increased intraosseous pressure, and bone necrosis.13 In kidney transplant recipients, the ischemia associated with AVN occurs very early, usually within 12 weeks of transplantation.14 Symptoms of AVN seem to present later, with most cases diagnosed a mean 3.5 years after transplantation.15 Kidney transplant recipients who experience AVN, especially of the femoral head, following THRA are median 38 years old and experience AVN symptoms median 20.5 months after transplantation. In comparison, liver transplant recipients undergo THRA median 35 after transplantation,16–18 similarly to our findings. The cumulative incidence of AVN after liver transplantation has been found to range from 3 to 41%, with 80% of AVNs occurring at the femoral head. About 85% of these patients had bilateral hip disease.19,20 The incidence of AVN after transplantation was shown to be >20%, with some of these patients undergoing THRA. An earlier study found that 16% of patients who underwent THRA developed AKI. AKI may be prevented by maintaining renal perfusion and avoiding precipitating factors. Maintaining proper renal blood flow is beneficial for renal oxygenation and can prevent nephrotoxic drug-related AKI.21 Patients who have congestive heart failure, ischemic heart disease, peripheral vascular disease, and higher Revised Cardiac Risk Index score are at a greater risk of peri-operative AKI. In addition, old age, smoking history, diabetes mellitus, hypertension, and high baseline serum Cr concentration or pre-existing chronic kidney disease are risk factors for AKI, as are higher American Society of Anesthesiologists physical status and blood transfusion. Medications known to affect renal function, including angiotensin-converting enzyme inhibitors (ACE-Is), angiotensin II receptor antagonists, non-steroidal anti-inflammatory drugs (NSAIDs), anti-arrhythmic drugs, diuretics, inotropes, vasoconstrictors, and opiates, have been shown to increase the risks of AKI.22,23 An earlier study showed that risk factors for AKI in patients undergoing THRA were chronic kidney disease, diabetes, male sex, hypertension, vascular disease, and premorbid use of nephrotoxic drugs such as ACE-Is, diuretics, NSAIDs, and opiates. In addition, we found that solid organ transplantation was an independent predictor of post-operative AKI. Among the factors associated with post-operative AKI in transplant recipients is treatment with immunosuppressive regimens. All organ transplant recipients in our study had taken immunosuppressant for long periods of time and before and after surgery. The long-term use of immunosuppressive therapy has significant potential side effects, including central nervous system toxicity, neurotoxicity, diabetes, hypertension, nephrotoxicity with hyperkalemic renal tubular acidosis, thrombocytopenia, and leucopenia.24 Cyclosporine and tacrolimus, both of which are metabolized in the liver through the cytochrome P-450 system, may affect exposure to many drugs administered during anesthesia or peri-operatively, and their side effects may have a direct impact on anesthetic and peri-operative management.25 During the peri-operative period, some of our patients treated with cyclosporine or tacrolimus developed hyper-tension, diabetes, neurotoxicity, or renal insufficiency. Therefore, organ transplant recipients treated with immunosuppressants for a long period of time may experience peri-operative renal dysfunction during non-transplanted surgery, with exacerbation of renal dysfunction leading to post-operative AKI.

The risks of post-operative AKI, however, may depend on the organ transplanted. For example, the characteristics of kidney transplant recipients depend on the history of individual donors and recipients. Most kidney transplant recipients have suffered from chronic kidney disease for a long time and have undergone peritoneal dialysis or hemodialysis. These patients have functional and structural vascular abnormalities, as shown by their high rates of increased pulse pressure and elevated systolic blood pressure.26,27 Their cardiovascular status may also be affected, and many of these patients have diabetes mellitus. Indeed, diabetes mellitus is the most common endocrine abnormality in these patients before and after kidney transplantation. Up to 32% of kidney transplant recipients have diabetes mellitus before transplantation, and up to 34% develop new-onset diabetes mellitus during the 1 year after transplantation,28 with almost 50% developing diabetes within 3 years after transplantation. Diabetes may affect vascularity and may be associated with poor transplant outcomes, such as AKI. In addition, effective renal plasma flow and the glomerular filtration rate may be significantly lower in recipients with well functioning kidney grafts and normal plasma creatinine concentrations than in healthy patients,29,30 Our findings suggested that kidney transplant recipients are more likely to develop post-operative AKI than non-transplanted patients and even than liver transplant recipients. Accordingly, the possibility of AKI should be monitored carefully in kidney transplant recipients undergoing THRA. Post-operative AKI was also more frequent in liver transplant recipients than in non-transplanted patients. Patients with liver cirrhosis are more susceptible to AKI owing to reduced effective blood volume, a progressive vasodilatory state, and stimulation by vasoconstrictor hormones. Prerenal azotemia, acute tubular necrosis, and hepatorenal syndrome can more easily cause AKI in patients with than without liver cirrhosis. Patients with end-stage liver cirrhosis have decreased mean arterial pressure, enhanced renal adrenergic tone, decreased cardiac performance owing to chronic heart overload, reduced renal synthesis of vasoprotective peptides, and increased intra-abdominal pressure due to accumulating ascites, all of which can enhance the risk of post-operative AKI.31

Decreased urine output or post-operative AKI is an important complication of surgery, especially in transplant recipients. Patients experiencing post-operative AKI may not only develop transient kidney disease, but also show progression of chronic kidney disease. Development of post-operative AKI may be associated with poor patient outcomes, including longer hospital stays, increased costs, and greater risk of discharge to extended-care facilities. Early diagnosis and appropriate management are therefore required for patients who develop post-operative AKI.

In conclusion, our findings show that transplant recipients undergoing THRA are at greater risk for AKI than non-transplanted patients. It is important to promptly recognize the risks of post-operative AKI, thus preventing poor patient outcomes. Future research on organ-specific stepwise therapies and interventions may reveal methods to prevent and treat post-operative AKI.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.