Keywords: allogeneic bone marrow transplantation; acute renal failure; hydronephrosis; adenoviral infection; necrotizing tubulointerstitial nephritis; nephrostomy.
Bone Marrow Transplantation (2003) 31, 1173–1176 & 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00
www.nature.com/bmt
Case report Acute renal failure due to adenovirus-associated obstructive uropathy and necrotizing tubulointerstitial nephritis in a bone marrow transplant recipient K Mori1, T Yoshihara1, Y Nishimura1, M Uchida2, K Katsura3, Y Kawase4, I Hatano5, H Ishida1, T Chiyonobu1, Y Kasubuchi1, A Morimoto6, T Teramura7 and S Imashuku7 1 Department of Pediatrics, Matsushita Memorial Hospital, Osaka, Japan; 2Department of Urology, Matsushita Memorial Hospital, Osaka, Japan; 3Department of Pathology, Matsushita Memorial Hospital, Osaka, Japan; 4Department of Dialysis, Matsushita Memorial Hospital, Osaka, Japan; 5Laboratory of Toxicological Pathology, Department of Safety Research on Biologics, National Institute of Infectious Diseases, Tokyo, Japan; 6Department of Pediatrics, Kyoto Prefectural University of Medicine, Kyoto, Japan; and 7Kyoto City Institute of Health and Environment, Kyoto, Japan
Summary: Management of post-transplant complications caused by severe adenoviral infection remains a major therapeutic challenge. A 17-year-old male who had undergone bone marrow transplantation for the treatment of acute lymphoblastic leukemia developed complete anuria following hemorrhagic cystitis 34 days after the transplant procedure. The computed tomogram scan revealed bilateral hydronephrosis, indicating acute renal failure because of obstructive uropathy. The emergency procedure of percutaneous nephrostomy caused massive bleeding in the left kidney, which eventually required a nephrectomy. Adenovirus-positive severe necrotizing tubulointerstitial nephritis was the histopathological diagnosis. Post-transplant acute renal failure because of hydronephrosis, which could be complicated by adenovirusinduced renal parenchymal disease, is of great concern and may cause significant problems with interventional treatment. Bone Marrow Transplantation (2003) 31, 1173–1176. doi:10.1038/sj.bmt.1704077 Keywords: allogeneic bone marrow transplantation; acute renal failure; hydronephrosis; adenoviral infection; necrotizing tubulointerstitial nephritis; nephrostomy
Severe adenoviral infection is a serious complication in immunocompromised patients, particularly in recipients of hematopoietic stem cell transplantations.1,2 To date, hemorrhagic cystitis, nephritis, pneumonitis, hepatitis, enteritis as well as disseminated infection have been reported, and the development of these conditions is closely associated with graft-versus-host disease (GVHD).1,2 Among adenovirus-induced urinary tract infections, acute necrotizing tubulointerstitial nephritis is the most critical
Correspondence: K Mori, Department of Pediatrics, Matsushita Memorial Hospital, 5-55, Sotojima-cho, Moriguchi-shi, Osaka 5708540, Japan Received 22 July 2002; accepted 13 January 2003
and fatal for transplant recipients.3 Hydronephrosis has also been described as a transplant complication of adenoviral infection.4,5 Management of adenovirus-induced urinary tract complications remains controversial because evaluation of disease progression is often difficult and the availability of effective therapeutic agents is still limited. We report here a case of acute renal failure, which may have resulted from post-transplant adenoviral infection-associated obstructive uropathy and necrotizing tubulointerstitial nephritis, in a bone marrow transplant recipient.
Case report A 17-year-old male with acute lymphoblastic leukemia in the second remission received an allogeneic bone marrow transplant (BMT) from an HLA-matched unrelated donor. The BMT was performed after a conditioning regimen of total body irradiation (12 Gy) and cyclophosphamide (60 mg/kg 2 days)/mesna. For GVHD prophylaxis, a combination of cyclosporine (CyA) and short-term methotrexate was administered, but CyA was replaced with methylprednisolone (mPSL) on day 13 because of adverse effects. The neutrophil count reached 0.5 109/l and engraftment was confirmed by microchimerism on day 20, when he began to complain of dysuria. On day 22, the patient developed gross hematuria, which was found to be positive for adenovirus by rapid enzyme-linked immunosorbent assay (ELISA). Supportive therapy of hydration with forced diuresis, intravenous vidarabine6 (600 mg/day, from day 24) and ribavirin7 (5–10 mg/kg/day, from day 26) effectively alleviated his symptoms and resulted in a negative ELISA test. However, on day 32, the acute GVHD worsened (grade IV: stage 4 for skin and stage 2 for gastrointestinal tract, stage 0 for liver) and he required high-dose mPSL therapy (1 g/day for 3 days). During this period, his urinary N-acetyl-b-d-glucosaminidase (NAG) levels were high, ranging from 13.7 to 106.7 U/g-creatinine (normal o10). On day 34, he abruptly became completely anuric (0 ml of urine/day) with elevated serum levels of urea nitrogen and creatinine. The computed tomogram (CT) and ultrasonogram showed bilateral hydronephrosis with a
Adenovirus-associated nephritis following BMT K Mori et al
1174
Figure 1
Computed tomogram on day 34 shows bilateral hydronephrosis with a fluid level of blood-contrast urine in the dilated renal pelvis.
fluid level of blood and contrast contaminated urine in the dilated renal pelvis (Figure 1), markedly thick-walled ureters and bladder showing no urine retention, suggesting acute renal insufficiency because of obstructive uropathy. Percutaneous nephrostomy tubes were promptly inserted into both renal pelvises with interventional ultrasound when his hemostasis/coagulation parameters showed no evidence of disseminated intravascular coagulation except for mildly reduced platelet counts (59 000/ml) and a prolonged APTT (44%). However, the following day, massive bleeding occurred into and around the left kidney as well as down into the sacral cavity. Since the renal arteriogram did not reveal any bleeding spots and because arterial coil embolization8 of the left renal artery did not stop the bleeding, an emergency left nephrectomy9 was performed. Subsequently, the patient was maintained on hemodialysis but died on day 56 of multiorgan failure.
Laboratory studies Retrospective analysis of adenovirus genome levels in the blood by real-time polymerase chain reaction (PCR)10 revealed a positive result for adenovirus 25 days after the transplant procedure (data not shown), suggesting that the patient actually had systemic adenovirus disease at this stage. Adenovirus isolated from the urine on day 22 was later determined to be type 11. Grossly, the resected kidney had a subcapsular hematoma around the nephrostomy tube and blood clots were noted in the pelvis, the ureter and the surrounding fatty tissues. No large vessel damage from the tube insertion was identified. Many microscopic foci of tubular epithelial cell degeneration and necrosis associated with necrotic cell debris were present and hemorrhaging in the tubular lumens was apparent (Figure 2a). Besides edema and hemorrhage, inflammatory infiltration was barely present in the surrounding interstitial tissue. The affected tubular epithelial cells contained intranuclear inclusion bodies of three types: Cowdry A, full-type and smudge-type (Figure 2b). The smudge-type inclusion is Bone Marrow Transplantation
Figure 2 (a) Nephrectomized kidney tissue shows degeneration and necrosis of tubular epithelial cells with associated necrotic cell debris and hemorrhage in the lumen (hematoxylin–eosin stain, magnification 25). (b) The three types of inclusion bodies: Cowdry A (long arrow), full-type (middle-length arrow) and smudge-type (short arrow) are identified in the affected tubules (hematoxylin–eosin stain, magnification 100).
known to be specific for adenovirus and is characterized by nuclei with dark basophilic material and blurred margins. Adenoviral particles were clearly observed by electron microscopy of the nuclei of the affected tubular epithelial cells (Figure 3). Taken together, the diagnosis of necrotizing tubulointerstitial nephritis induced by type 11 adenovirus was confirmed.
Discussion Common causes of post-transplant acute renal failure include GVHD, adverse effects of immunosuppressive agents (CyA, tacrolimus) and adenoviral infection-induced urinary tract diseases. Previously, an association between adenovirus-induced hydronephrosis and necrotizing tubulointerstitial nephritis has not been clearly described. In the case described here, anuria abruptly occurred on day 36 post-transplantation during treatment for adenovirusinduced hemorrhagic cystitis. At the time, we believed that percutaneous nephrostomy11 was the best treatment.
Adenovirus-associated nephritis following BMT K Mori et al
procedures, as reported by Echavarria et al,13 is crucial. In particular, real-time quantification of the adenovirus genome in the blood10 would help to determine if the infection is systemic and likely to progress into the renal parenchyma. Early detection of disease progression enables prompt reduction of immunosuppression14 or prompt treatment with antiviral agents such as cidofovir, gancyclovir, ribavirin, or vidarabine.15 Early administration of cidofovir may be effective for adenoviral infection,16,17 if started just after the positive serum PCR results. Donor lymphocyte infusion would be another choice of treatment, although it may worsen GVHD.14,18 In conclusion, posttransplant adenoviral infections should be assessed more extensively in a prospective study employing quantitative PCR approaches in order to develop more appropriate management procedures. Figure 3 Electron microscopy reveals adenoviral particles in the nucleus of a tubular epithelial cell. The particles measure 75–80 nm in diameter and are arranged in crystalline arrays or are randomly scattered.
Unfortunately, however, it caused massive hemorrhaging. Although the CT findings clearly suggested obstructive uropathy, the patient actually had severe necrotizing tubulointerstitial nephritis caused by systemic adenoviral infection. The underlying nephritis and severe GVHD also must have contributed to the development of renal failure. Although bleeding complications from nephrostomy are rare when the kidney tissues are intact,8,9 we learned that interventional procedures, such as those decribed here, are risky in patients with severe adenoviral infections because the necrotic renal tissue through which the catheter penetrates may be extremely fragile and prone to bleeding. Alternatively, Kapelushnick et al 4 suggested that treatment with pentosan-polysulfate or intravesical installation of alum, silver nitrate or prostaglandin E2 should be attempted before using aggressive procedures to treat children with post-transplant hydronephrosis. To date, almost all cases of acute necrotizing tubulointerstitial nephritis have resulted in renal failure and death.12 Either systemic or local adenoviral infection ascending from the bladder accounts for disease development.3,12 According to a study by Ito et al,3 acute necrotizing tubulointerstitial nephritis is associated with prior hemorrhagic cystitis and the presence of the type 11 adenoviral strain. Clinically, urinary NAG analysis is useful to estimate the damage to renal tubular epithelial cells. The observations of Hiraoka et al are particularily pertinent with respect to the prediction of an association between obstructive uropathy and underlying necrotizing tubulointerstitial nephritis. They reported the presence of an unexplained ureteric obstruction and a hypertrophic kidney cortex despite the persistent presence of hydronephrosis, suggesting the involvement of an inflammatory process in the renal parenchyma. In our case also, hypertrophic cortexes were observed in both kidneys (Figure 1). Regardless, management of adenovirus-induced necrotizing tubulointerstitial nephritis is very difficult. Therefore, early diagnosis of adenoviral disease is mandatory. For that purpose, detection of the viral genome in the sera by PCR
1175
References 1 Shields AF, Hackman RC, Fife KH et al. Adenovirus infections in patients undergoing bone-marrow transplantation. N Engl J Med 1985; 312: 529–533. 2 Runde V, Ross S, Trenschel R et al. Adenoviral infection after allogeneic stem cell transplantation (SCT): report on 130 patients from a single SCT unit in a prospective multi center surveillance study. Bone Marrow Transplant 2001; 28: 51–57. 3 Ito M, Hirabayashi N, Uno Y et al. Necrotizing tubulointerstitial nephritis associated with adenovirus infection. Hum Pathol 1991; 22: 1225–1231. 4 Kapelushnik J, Verstandig A, Or R et al. Hydronephrosis in children after bone marrow transplantation: case reports. Bone Marrow Transplant 1996; 17: 873–875. 5 Hiraoka A, Teshima H, Mitsui H et al. Ureteric obstruction after allogeneic bone marrow transplantation: an unusual complication. Bone Marrow Transplant 1989; 4: 449–451. 6 Kawakami M, Ueda S, Maeda T et al. Vidarabine therapy for virus-associated cystitis after allogeneic bone marrow transplantation. Bone Marrow Transplant 1997; 20: 485–490. 7 Murphy CF, Wood DP, McRoberts JW et al. Adenovirusassociated hemorrhagic cystitis treated with intravenous ribabirin. J Urol 1993; 149: 565–566. 8 Peene P, Wilms G, Beart AL. Embolization of iatrogenic renal hemorrhage following percutaneous nephrostomy. Urologic Radiol 1990; 12: 84–87. 9 von der Recke P, Nielsen MB, Pedersen JF. Complications of ultrasound-guided nephrostomy. Acta Radiol 1994; 35: 452–454. 10 Teramura T, Naya M, Yoshihara T et al. Quantitative detection of serum adenovirus in a transplant recipient. Lancet 2002; 359: 1945. 11 Debbagh A, Dassouli B, Hafiani M et al. Acute renal insufficiency due to hydronephrosis. Ann Urol (Paris) 2001; 35: 26–29. 12 Erdogan O, Bulbul M, Demircin G et al. Acute necrotizing tubulointestinal nephritis due to systemic adenoviral infection. Pediatr Nephrol 2001; 16: 265–268. 13 Echavarria M, Forman M, van Tol MJD et al. Prediction of severe disseminated adenovirus infection by serum PCR. Lancet 2001; 358: 384–385. 14 Einsele H, Roosnek E, Rufer N et al. Infusion of cytomegalovirus(CMV)-specific T cells for the treatment of CMV Bone Marrow Transplantation
Adenovirus-associated nephritis following BMT K Mori et al
1176 infection not responding to antiviral chemotherapy. Blood 2002; 99: 3916–1922. 15 Chen FE, Liang RHS Lo JY, Yuen KY et al. Treatment of adenovirus-associated haemorrhagic cystitis with ganciclovir. Bone Marrow Transplant 1997; 20: 997–999. 16 Ribaud P, Scieux C, Freymuth F, Morinet F et al. Successful treatment of adenovirus disease with intravenous cidofovir in an unrelated stem-cell transplant recipient. Clin Infect Dis 1999; 28: 690–691.
Bone Marrow Transplantation
17 Legrand F, Berrebi D, Houhou N et al. Early diagnosis of adenovirus infection and treatment with cidofovir after bone marrow transplantation in children. Bone Marrow Transplant 2001; 27: 621–626. 18 Haque T, Taylor C, Wilkie GM et al. Complete regression of posttransplant lymphoproliferative disease using partially HLA-matched Epstein-Barr virus-specific cytotoxic T cells. Transplantation 2001; 72: 1399–1402.
