Thromboembolism in renal transplant artery due to atrial fibrillation

Background: Central and peripheral arterial thromboembolisms are well known complications of atrial fibrillation. We report the first case of thromboembolism in a renal transplant due to atrial fibrillation after changing anticoagulation treatment. Case presentation: A 79-year-old woman who had undergone kidney transplantation in 2001 presented herself with abdominal pain and oliguria. Serum creatinine had been stable (130 – 150 µmol/l) since transplantation, and, because of atrial fibrillation, the patient received oral anticoagulation treatment for many years until it was switched to aspirin 100 mg due to a cholecystectomy in 2012. Three weeks thereafter is when the patient was admitted to our unit. In a computed tomography scan of the abdomen, multiple renal infarcts were detected. The thromboembolism resulted in a complete loss of transplant function. As the patient remained anuric, the transplant was declared to be lost, immunosuppression was reduced, and renal replacement therapy was commenced over a central catheter. However, the patient died the next day. Conclusion: Our case highlights the fact that changing anticoagulation treatment should be performed carefully and can be detrimental. The location of thromboembolism in renal transplant artery makes this case unique and has not ever been described thus far. As renal transplantation and risk factors for thromboembolism i.e., atrial fibrillation are increasing, embolization in renal transplant artery is a potential reason for abdominal pain and anuria; physicians should take this into consideration when treating transplanted patients.

Abstract. Background: Central and peri pheral arterial thromboembolisms are well known complications of atrial fibrillation. We report the first case of thromboembolism in a renal transplant due to atrial fibrillation after changing anticoagulation treatment. Case presentation: A 79yearold woman who had undergone kidney transplantation in 2001 presented herself with abdominal pain and oliguria. Serum creatinine had been stable (130 -150 µmol/l) since transplanta tion, and, because of atrial fibrillation, the patient received oral anticoagulation treat ment for many years until it was switched to aspirin 100 mg due to a cholecystectomy in 2012. Three weeks thereafter is when the pa tient was admitted to our unit. In a computed tomography scan of the abdomen, multiple renal infarcts were detected. The throm boembolism resulted in a complete loss of transplant function. As the patient remained anuric, the transplant was declared to be lost, immunosuppression was reduced, and renal replacement therapy was commenced over a central catheter. However, the patient died the next day. Conclusion: Our case high lights the fact that changing anticoagulation treatment should be performed carefully and can be detrimental. The location of throm boembolism in renal transplant artery makes this case unique and has not ever been de scribed thus far. As renal transplantation and risk factors for thromboembolism i.e., atrial fibrillation are increasing, embolization in renal transplant artery is a potential reason for abdominal pain and anuria; physicians should take this into consideration when treating transplanted patients.

Background
Stroke and less common peripheral arte rial thromboembolism are well known com plications of atrial fibrillation [1,2]. The reason for unequal incidence of thromboem bolism in different organs is unknown. Renal infarcts of native kidney due to embolism are often difficult to diagnose and are probably underestimated [3,4]. Reviewing the litera ture, peripheral arterial thromboembolism in a kidney transplant artery has never been de scribed to date.

Case presentation
We report the case of a 79yearold wom an who received a kidney transplant in 2001 and maintained good function thereafter. The primary kidney disease was not known. In October 2012, she presented to our emer gency unit with a history of abdominal pain, fever (38 °C), and shivering. Two days prior, urinary excretion had decreased to 600 ml/d. On the day of presentation, the patient was anuric. She described her abdominal pain as being similar to that of cholecystitis from which she had suffered before. A cholecys tectomy had been performed 3 weeks prior in another hospital.
Her past medical history included atrial fibrillation, coronary heart disease (with coronary artery bypass graft in 2006), stroke, epilepsy (since childhood), and hyperten sion. Her immunosuppressive medication consisted of mycophenolate and cyclospo rine. She was taking candesartan, torasemid, metoprolol, and amlodipine for hyperten sion, and as an antiepileptic prophylaxis, la motrigine. Oral anticoagulation was discon tinued and replaced by aspirin 100 mg prior to the cholecystectomy.
Upon admission, her vital signs were as follows: blood pressure of 123/70 mmHg and heart rate of 61 beats per minute. There was tenderness and pain on palpitation in the region of the kidney transplant in the left in guinal area. The other examination was un remarkable.
Laboratory results revealed a white blood count of 17.5 × 10 9 /l (being in the normal range before), erythrocytes of 3.49 × 10 9 /l and platelets of 148 × 10 9 /l. Creactive protein was 33 mg/l, potassium was in the normal range, but creatinine was elevated to 304 µmol/l (3.45 mg/dl). Previous mean creatinine levels had been ~ 140 µmol/l (1.58 mg/dl) since renal transplantation in 2001. Lactate dehydrogenase (LDH) was strongly elevated to 1,382 U/dl (nl < 250 U/ dl). International normalized ratio (INR) was in normal range as oral anticoagulation with Marcumar was stopped, pending cholecys tectomy. Before, the patient's INR had been in the therapeutic range for years ( Figure 1).
Electrocardiogram showed atrial fibrilla tion with 70 beats per minute. Xray of the chest and abdomen could neither detect infil trates nor free gas or free fluid. An antibiosis with ertapenem was started after blood cul tures had been taken. A computed tomogra phy (CT) scan of the abdomen, with contrast, revealed multiple infarct areas affecting at least 50 -60% of the transplanted paren chyma (Figure 2). We began anticoagulation therapy with heparin in therapeutic doses. Over the following days, creatinine levels increased to 522 µmol/l (5.9 mg/dl). As the patient remained anuric, the transplant was declared to be lost, immunosuppression was reduced, and renal replacement therapy was commenced over a central catheter.
The following day, the patient developed a status epilepticus that did not respond to high doses of antiepileptic drugs. A cranial CT ruled out intracranial bleeding or a mass. No fresh infarct was noted. Cranial MRI was not possible because of the patient's agita tion. The patient died within 24 hours in our intensive care unit after a decision was made in consensus with her relatives not to inten sify the treatment. An autopsy was refused by the family.

Discussion
The incidence of peripheral arterial thromboembolic events in patients with atrial fibrillation is much lower than the incidence of strokes [1].
The most common sites for peripheral arterial thromboembolism are the extremi ties (61%), followed by the mesenteric ar teries (29%), the pelvic arteries (9%), the aorta (7%), and the renal arteries (2%) [2]. Peripheral arterial thromboembolic events causing renal transplant damage seem to be exceptional as we could not find any simi lar published cases. One explanation for the rarity might be the unlikelihood for thrombi reaching renal transplant artery due to ana tomical reasons. Renal infarction of native kidney occurs in patients with high morbidity and mortality and is associated with diffuse atherosclerosis and atrial fibrillation [3]. It is often difficult to diagnose because the symptoms are simi lar to those of nephrolithiasis, pyelonephri tis, ruptured abdominal aortic aneurysm, tes ticular/ovarian torsion, incarcerated hernia, intestinal perforation, mesenteric ischemia, diverticulitis, spinal disease, or myocardial infarction. Thus, as it shown in postmortem studies, the real incidence of renal artery embolism is probably underestimated [4]. In our case, abdominal CT together with the pa tient's medical history and laboratory results led to the diagnosis.
As renal artery was perfused in our pa tient, we had no evidence for local stenosis. "Cortical rim sign" (1 to 3mm rim of sub capsular enhancement paralleling the renal margin) is a common feature found with renal artery obstruction by CT [5]. "Arterial cutoffsign" (abrupt termination of contrast material in the renal artery) [6] and "flip-flop enhancement" (delayed enhancement of the initial hypo attenuations due to extravasation of contrast material in areas of ischemia) [7] are other possible features in CT.
Frequently, patients with renal infarction of native kidney are initially hypertensive, which is thought to be reninmediated due to decreased renal blood supply [3]. Our pa tient's blood pressure was normal under mul tiple antihypertensive agents. Urine analysis after renal infarction of native kidney is as sociated with hematuria [8]. Proteinuria was seen in 45% of the patients in one retrospec tive case review [9]. Elevations of LDH are almost universal in renal infarction, and LDH can also be detected in urine [10,11]. The rise in LDH has been considered as a good screen ing marker in patients suffering from embo lism [12]. Our patient was anuric. Therefore, no urine assessment for LDH, hemoglobin, or protein was possible. Another typical finding in renal artery embolism is the elevation in white blood cell count [5], a fact that was also seen in our patient.
Typical signs, symptoms, and laboratory findings associated with atheroembolic renal disease (AERD) due to cholesterol embo lism such as livedo reticularis, gangrene of the extremities, and neurological symptoms like amaurosis fugax or strokelike episodes or eosinophilia were not observed. Further more, the patient had not received an inter vention that would have triggered the oc currence of AERD. Therefore, we have no evidence that kidney failure was caused by cholesterol embolism.
In prior cases of artery embolism of the native kidney, serum creatinine was not markedly elevated at presentation. In a study of patients with 44 cases of renal infarction with atrial fibrillation by Hazanov et al. [9], 61% had normal renal function on follow up. In our case, the patient was dependent on renal transplant since native kidneys had been removed years before. Thus, severe thromboembolism of renal transplant artery A known treatment option of artery em bolism is anticoagulation. Thrombolysis or thrombectomy can be considered, but no sig nificant clinical evidence supports these inter ventions. For renal transplant embolism result ing in renal transplant infarction, no special treatment option has yet been described.

Conclusions
Our case highlights the fact that chang ing anticoagulation treatment should be per formed carefully as it can be detrimental. In our patient, the location of thromboembolism was the renal transplant artery; this resulted in complete loss of kidney transplant.
As renal transplantation and risk factors for thromboembolism i.e., atrial fibrillation are increasing, embolization in renal trans plant artery is a potential reason for abdomi nal pain and anuria, physicians should take this into consideration when treating trans planted patients.
Until now, the only known treatment op tion is anticoagulation, and thus far no spe cial treatment option has been described. Therefore, research in this field is highly requested.

Consent
Written informed consent for publication of the clinical details and clinical images was obtained from the relatives of the patient.