Clinical Study Am J Nephrol 1995;15:473-479 Enrique Grussa Carmen Bern is3 Jose Francisco Tomasb Cesar Garcia-Cantona Angela Figuerab Jose Luis M ot ellón3 Vicente Paraísoa Juan Antonio Trover3 Jose Maria Fernandez-Rañadab Acute Renal Failure in Patients following Bone Marrow Transplantation: Prevalence, Risk Factors and Outcome Departments of a Nephrology and b Hematology, Princesa Hospital. Autónoma University, Madrid, Spain Abstract To assess the prevalence, risk factors, clinical causes and outcome of acute renal failure (ARF) following bone marrow transplantation (BMT), a retro spective analysis of 275 patients was undertaken. ARF was diagnosed in 72 patients (26%) and occurred in 81.9% within the first month. The three main clinical causes were multifactorial (36%), nephrotoxic (29%), and veno-occlusive disease of the liver (VOD) 15%. The prevalence was higher in allogeneic BMT (36%) than in autologous BMT (6.5%). Risk factors related to the devel opment of ARF were preexisting VOD and age older than 25 years. Logistic regression in allogeneic BMT confirmed this association (VOD, odds ratio 3.8; age ofer than 25, odds ratio 1.9). Underlying disease, graft-versus-host disease, sepsis, conditioning therapy, and sex were not associated with ARF. Seven teen cases of ARF required hemodialysis (24%) mainly in association with VOD (70.5%). The overall mortality from ARF was 45.8%, the dialyzed group having the highest mortality (88%). Survival in the ARF group was contin uously worse up to 3 months and the actuarial survival at 10 years was 29.7 versus 53.2%. We conclude that ARF is a common complication mainly in allogeneic BMT and carries a grave prognosis. VOD and age were risk factors for ARF. Introduction Bone marrow transplantation (BMT) has emerged in recent years as a major form of treatment for a wide range of hematologic, immunologic, metabolic and neoplastic diseases. In summary the main steps of BMT are: (1) highdose chemoradiotherapy treatment (conditioning regi men) in order to obtain a potent immunosuppressive state Received: September 2. 1994 Accepted: March 14, 1995 in the recipient that permits engraftment and also to remove unwanted malignant cell populations, and (2) in fusion of a source of stem cells from a donor (allogeneic BMT) or from the same patient (autologous BMT). In spite of advances in the technical aspects of transplanta tion, disease relapse and regimen-related toxic deaths are the main causes of BMT failure. Major complications and toxicities of BMT are immunologic problems (graft-ver- Dr. Enrique Gruss Department of Nephrology Princcsa Hospital Zazuar 8 ,9°A. E3 E—28031 Madrid (Spain) © 1995 S. Karger AG, Basel 0250-8095/95/0156-0473 $8.00/0 Downloaded by: Vanderbilt University Library 22.214.171.124 - 10/27/2017 4:19:45 AM Key Words Acute renal failure Bone marrow transplantation Veno-occlusive disease, liver Number Type of transplant Allogenic Autologous Sex Male Underlying disease AL CML AA Other Conditioning therapy Cy-TBI Bu-Cy Other Sepsis VOD Age. years ARF Non-ARF Statistics 72(26.2%) 203 66 (36%) 6(6.5%) 117 86 44 122 n.s. 35 21 8 8 100 41 20 42 n.s. n.s. n.s. 52 12 8 19 31 28.2 ±9.6 108 34 61 50 26 24.7 ± 10.8 p< 0.01 n.s. p < 0.001 n.s. p < 0.001 pcO .O l AL = Acute leukemia; CML = chronic myelogenous leukemia; AA = aplastic anemia; Cy-TBI = cyclophosphamide-total body irra diation; Bu-Cy = busulfan-cyclophosphamide. sus-host disease, GVHD), opportunistic infections, inter stitial pneumonitis and organ complications such as venoocclusive disease of the liver (VOD), cardiovascular and pulmonary problems . Although there is literature on these complications, little information is available about renal complications after BMT. In contrast to other organs, such as the liver that is a primary target organ for GVHD and chemoradiotherapy toxicity, kidneys are not directly involved. Renal impairment after BMT is pre dominantly secondary to circulatory disturbances associ ated with VOD, and to sepsis in conjunction with the wide use of nephrotoxic agents in these patients. In an attempt to define the clinical causes and charac teristics of renal complications after BMT and to evaluate their influence on clinical outcome, we have reviewed our experience in 275 consecutice patients who received a BMT for different diseases. Patients and Methods The medical records of 275 patients who received BMT in our institution between October 1982 and December 1991 were re viewed. Minimum follow-up was 6 months. Patients were divided into 2 groups according to the presence or absence of acute renal 474 Gruss et al. failure (ARF). ARF was defined by two conditions: at least a dou bling of baseline creatinine, and also by reaching levels higher than 177 pmol/1 (2 mg/dl) . All patients had a previously normal creati nine level (below 1.5 mg/dl). The characteristics of both groups are presented in table 1. All patients were isolated either in individual high pressure fil tered air or laminar air-flow rooms from day 7 until at least neutro phil recovery. The main conditioning regimens are shown in table 1. Busulfan-cyclosphosphamide and cyclophosphamide-total body irra diation were mainly used as previously reported  for acute and chronic leukemia. Other schemes, CBV (cyclosphosphamide, BCNU. VP 16) and BEAC (BCNU, cytarabinc. VP 16. cyclosphos phamide) were used for lymphoma patients. For allogeneic BMT non-T-cell-depleted bone marrows were used in all cases and dimethylsulfoxide-cryopreserved bone marrows or peripheral stem cells were used as a source of hematopoietic progeni tors in autologous BMT. Since 1985 all allogeneic BMT patients received GVHD prophy laxis with cyclosporin-A (CsA) and a short course of methotrexate according to the Seattle protocol . Before 1985 GVHD prophylac tic regimens were either CsA or methotrexate alone as previously described . No autologous BMT patient was treated with CsA. Serum samples for the determination of CsA levels by the fluores cence polarization immunoassay method were drawn twice a week, immediately before the morning CsA dose. CsA adjustments were made to try to keep levels between 159 and 300 pg/ml. The following clinical information was gathered for each patient: age: sex; underlying disease; conditioning regimen; type of trans plant; urea and creatinine levels before BMT and during the trans plant admission, and clinical complications. In the group with ARF we also gathered: day of ARF after BMT; maximum creatinine level; clinical cause of ARF; hemodialysis (HD) necessity, and outcome of ARF. We associated ARF with 7 clinical causes: ( 1) nephrotoxicity, when ARF developed with the use of potential nephrotoxic drugs in the absence of other putative causes of ARF. Major nephrotoxic drugs used in BMT are aminoglycosides, vancomycin, amphotericin B and CsA. (2) VOD, when VOD preceded renal insufficiency. VOD is a nonthrombotic obliteration of small intrahepatic veins by loose connective tissue . VOD was diagnosed based on clinical criteria, statistically associated with the histologic abnormalities , when two of the following events occurring within 20 days of transplanta tion were presented: hyperbilirubinemia (total serum bilirubin >34.2 mmol/I, 2 mg/dl), hepatomegaly or right upper quadrant pain of liver origin and sudden weight gain (>2% of baseline body weight) because of fluid accumulation. No other explanation for these signs and symptoms could be present at the time of diagnosis. (3) Sepsis, when ARF developed with fever, hypotension, tachycardia and posi tive blood cultures. (4) Systemic capillary leak syndrome (SCLS), when ARF developed according to the criteria previously reported . (5) Hemolytic uremic syndrome (HUS). ARF developed with hemolytic microangiopatic anemia, thrombopenia and hypertension. (6) Hemoglobinuria (HB), ARF developed on the same day of mar row infusion due to the presence of freeze-broken red cells on the cryopreserved product. (7) Multifactorial, ARF was assigned to this category when more than one of the above causes was present in the same patient. ARF was analyzed first in all the patients and a separate analysis was performed in the allogeneic BMT group because of the special circumstances that occurred in this group (greater incidence of ARF, use of cyclosporin and greater incidence of VOD). Acute Renal Failure in Bone Marrow Transplantation Downloaded by: Vanderbilt University Library 126.96.36.199 - 10/27/2017 4:19:45 AM Table 1. Data base and statistical comparison for groups Statistical A nalysis Table 2. Characteristics of ARF Results The main characteristics of ARF are presented in table 2. ARF occurred in 72 of 275 patients who under went BMT (26.2%) and in most cases appeared during the first month after transplant (81.9%). The median dura tion of ARF was 14.4 days and in 89.7% of the patients creatinine levels returned to normal. The major causes of ARF, according to the clinical classification given above, were as follows: (1) multifacto rial, 26 patients (36.1%); (2) nephrotoxicity, 21 patients (29.1%); (3)VOD, 11 patients (15.3%); (4) sepsis, 6 pa tients (8.3%); (5) SCLS, 4 patients (5.5%); (6) HUS, 2 patients (2.7%), and (7) HB, 2 patients (2.7%). All cases in the multifactorial group had nephrotoxic drugs, other associated principal causes were: VOD (69%), and sepsis (50%). CsA was the most common drug responsible for renal injury in the nephrotoxic group. Higher levels were found in most of the patients and the mean serum level was 588 ± 47 ng/dl (79-1,182) for this group of patients. Table 3 shows clinical data for 17 of the 72 patients with ARF that required HD. The mean number of ses sions was 4.5 and most of them were carried out in the first 2 weeks. The mean creatinine concentration was 4.6 mg/dl. Seventy percent of the dialyzed patients had severe VOD preceding ARF. The type of BMT was an important determinant for the development of ARF, being more common in allo geneic (36%) than in autologous (6.5%) (p< 0.001) BMT. The differences between both types of transplants are pre sented in table 4. Table 4. Differences between allogeneic BMT and autologous BMT Allogeneic BMT Autologous BMT Total patients 275 72 (26.2%) With ARF Without ARF 203 Creatinine level, mg/dl 0.8 ±0.2 (0.4-1.5) Initial Maximum 3.19± 1.3 (2-7,2) 1.1 ± 0.3 (0.5-1.8) Final1 Time of ARF after BMT 59(81.9%) 1st month 2nd month 9(12.5%) 3rd month 4 (5.5%) Evolution ARF during transplant hospitalization Death with ARF 33(45.8%) Survival 39 35(89.7%) Recovery of total renal function 4 No recovery of total renal function 14.4 ± 8 (3-39) Mean duration ARF, days 1 At discharge in surviving patients. Table 3. Characteristics of hemodialysis Number of patients Mean age, years Sex Males Underlying disease AL CML Other Indication for initiating dialysis Volume overload Severe azotemia (cr > 6 mg/dl) Number of sessions Day after BMT Mean level of Cr, mg/dl Evolution Death Survival Associated disease VOD Sepsis 17(23.6%) 27.2 ± 8 9 8 6 3 13(72%) 4 4.5 ± 3.7 (1-16) 8.7 ± 5.4 ( 1—23) 4.6 ±1.6 (3-72) 15 2 12(70.6%) 6(35.3%) Abbreviation are as in table 1. Number ARF VOD CsA GVHD 183 92 66 (36%) 6(6.5%) 51 (27.9%) 6 (6.5%) yes no yes no 47Í Downloaded by: Vanderbilt University Library 188.8.131.52 - 10/27/2017 4:19:45 AM Values are expressed as mean ± standard deviation or percent age. Univariate analyses were performed on the data employing '/} statistics for discrete variables and the Student’s t test was used for continuous variables. Using the statistical analysis system package (SAS Institute 1984, Cary, N.C., USA) in allogeneic transplants a stratified analysis by Mantel-Haenszel method was conducted. Confidence intervals were determined by the Cornfield method. After multivariate analysis, multiple logistic regression was performed in allogeneic transplant. Actuarial survival was analyzed by the Kaplan-Meier method. Table 5. Data base and statistical comparison in allogogous BMT ARF Non-ARF Analysis univariate multivariate Age, years 27.6 ±9.2 117 73 n.s. n.s. 48 41 20 8 n.s. n.s. n.s. n.s. n.s. n.s. 80 17 8 19 24 20 n.s. n.s. n.s. n.s. n.s. n.s. p < 0.001 n.s. n.s. p < 0.0001 odds ratio 3.8 (1.95-7.38) p < 0 .0 5 odds ratio 1.9 (1.09-4.16) 22.7 ± 9.9 p < 0.001 Abbreviations are as in table 1. 1 GVHD only in engrafted patients. The data base and statistical comparison, by univar iate analysis, between the patients that developed ARF and those without ARF are presented in table 1. The sex distribution, underlying disease, and sepsis did not differ between both groups. ARF was more common among patients cyclophosphamide-total body irradiation as the conditioning regimen (p < 0.01). The main factors associ ated with ARF were older ager (over 25 years) and VOD preceding renal insufficiency. The prevalence of ARF among patients older than 25 was 33.8% (43 of 127); in those younger than 25, it was 19.6% (29 of 148; p < 0.01). VOD was present in 43% (31 of 72) of patients with renal failure, whereas it was present in only 12.8% (26 of 203) patients without ARF (p < 0.001). As ARF was especially prevalent among allogeneic BMT patients a separate univariate and multivariate analysis was performed in this group (table 5). Both stud ies confirmed the strong association between older age, VOD and ARF. In this sense the global incidence of ARF in patients older than 25 was 45.7% (37 of 81) vs. 28.4% (29 of 102) (p < 0.0 T, odds ratio 1.9). On the other hand 476 Grusset al. Fig. 1. Different actuarial survival rates (months) for the ARF and non-ARF groups. VOD was present in 47% (31 of 66) of the patients with ARF vs. only 17% of those without ARF (20 of 177; p < 0.0001; odds ratio 3.8). Conditioning therapy was not sig nificant. Survival was significantly lower for those patients who developed ARF because the transplant-related mortality was 45.8% in the ARF group (33 of 72) vs. 17.7% (36 of 203) in the rest. A higher mortality rate was observed for those patients who required HD: 88% (15 of 17). Among those patients with ARF the highest mortality rate was found in the VOD group (10 of 11) and the lowest in the nephrotoxicity group (2 of 21). The overall survival was worse in the ARF group at 1 month (69.4 vs. 88.1%) and 3 months (50 vs. 77.8%). With a minimum follow-up of 6 months the actuarial overall projected survival at 10 years was 29.7% for the ARF group vs. 53.2% in those without ARF (p < 0.001; fig. 1). Discussion Organ toxicities other than bone marrow are common side effects in BMT recipients. A lot of information is available concerning liver, cardiopulmonary, neurologic, and endocrine toxicities. This study shows that ARF is also a very common complication of BMT because 26% of the 275 patients analyzed developed renal damage (36% in allogeneic transplant and 6% in autologous). This experience is not unique. A retrospective study from the Hutchinson Center conducted in 1989 found a 53% inci Acute Renal Failure in Bone Marrow Transplantation Downloaded by: Vanderbilt University Library 184.108.40.206 - 10/27/2017 4:19:45 AM Number 66(36%) Sex Male 40 Underlying disease 33 AL CML 21 AA 8 Other 4 Conditioning therapy Cy-TBI 52 Bu-Cy 10 4 Other 12 Sepsis GV HD 1 15 VOD 31 (ATN) the incidence was relatively low in our group and in other series . It has been postulated that this low incidence is due to the emergency of counter-balancing cytoprotective pathways that may be the cytoprotectant heat-shock proteins in relation with common hyperther mia following BMT . ATN is also associated with HB developing on the same day of marrow infusion in autolo gous BMT due to the presence of freeze-broken red cells in the cryopreserved product , However, this complica tion can usually be avoided with good previous hydration, and this could explain why only 2 of 92 autologous BMT patients developed HB. ATN is also associated with tumor lysis syndrome but we have not found a single case because most transplanted patients were in remission or routinely received prophylaxis with volume expansion. Tubular dysfunction associated with high-dose cyclophos phamide and radiotherapy has been proposed . The influence of a conditioning regimen on renal function is controversial. We found a higher incidence of ARF by univariate analysis among patients that were treated with cyclophosphamide-total body irradiation in the total number of patients. This was not confirmed by multivar iate analysis in the more homogeneous group of allogeneic BMT. It has been described as a belated impairment of renal function associated with irradiation in autologous BMT , Also some authors found that total body irra diation is the single most important risk factor for renal damage at 6 months , but they conducted the study in a small group without separately analyzing autologous and allogeneic BMT. Some cases of HUS following BMT have been pub lished  and we observed 2 cases. This syndrome is associated with the vascular and thrombotic effects of CsA in early BMT renal dysfunction  and with irra diation in late renal failure . Another uncommon cause of ARF is SCLS. It is char acterized by attacks of a marked increase in capillary per meability in relation to liberation of leukotrienes  that can cause intravascular volume depletion and hence ARF. Four cases with this diagnosis were observed before 1985. Nowadays we consider this syndrome as volume depletion in severe presentation of GVHD. The study of HB has identified hepatic dysfunction and weight gain as risk factors of ARF following BMT , but these factors are included in the definition of VOD. We have identified several factors that either contributed to or correlated with the subsequent development of ARF. In our series the main factor, as assessed by risk factor analysis, was VOD. Forty-three percent of patients with ARF had associated VOD versus 12.8% in the group 477 Downloaded by: Vanderbilt University Library 220.127.116.11 - 10/27/2017 4:19:45 AM dence  and recently this center reported a 40% inci dence . The Johns Hopkins University School of Medicine reported a 64% incidence of ARF after BMT . The lower incidence in our group can be explained because there were fewer allogeneic BMTs, less mis matched grafts, no cases of tumor lysis syndrome, and because of our more restricted definition of ARF (not only by doubling the baseline creatinine but also by reaching levels higher than 2 mg/dl). Most of the ARFs developed in the first month after BMT (81%) and nearly all the patients that recovered renal function did so completely. This experience is simi lar to other series [9,11]. The main complications develop during the first month after transplant: life-threatining infections and potential use of nephrotoxic drugs; hepatic toxicity (mainly VOD); acute GVHD, and the necessity of prophylaxis with CsA (only in allogeneic transplant). Therefore, it is understandable that ARF mainly develops in this period, and that the first clinical cause was attribut ed to multifactoriality (VOD, sepsis, nephrotoxicity). It can be extremely difficult to identify the underlying cause of ARF in some patients. The important role of VOD and its relationships with ARF will be discussed below. The second clinical cause of ARF is nephrotoxicity. Of the diverse nephrotoxic agents commonly employed in BMT, CsA is one of the most relevant. Nephrotoxicity with CsA was clearly dose-dependent with improvement after the dose was lowered. Two forms of CsA nephrotoxicity are recognized: an acute reversible form, and a chronic irre versible one (beyond the scope of this study). There is now a consensus that acute CsA nephrotoxicity is the result of alterations in intrarenal hemodynamics  due mainly to an increase in renal vascular resistance associated with a reduction in the glomerular filtration rate [ 13]. There is also evidence that acute CsA nephrotoxicity is rapidly reversible when CsA is stopped . When vascular renal lesions occur during CsA treatment, they are probably triggered, enhanced or act with other nephrotoxic drugs [14-16] or with other causes of endothelial damage present in BMT as ischemia, infection and VOD . Although other authors have not found this association between CsA and BMT nephrotoxicity, the final end point analysis (renal function at 6 months) and the ARF criteria definition were different [9, 18]. In most cases ARF caused by nephrotoxicity was reversible, HD was not necessary, and mortality was low. Less frequent clinical causes of ARF were sepsis, SCLS, HUS and HB. Sepsis is very common in immunosuppressed BMT patients. Although BMT patients have many potential risk factors for acute tubular necrosis without ARF. By multivariate analysis in allogeneic trans plants VOD carried an odds ratio of 3.8. VOD occurs principally as a complication of chemoradiation particu larly following BMT. Although jaundice is a risk factor for the development of ATN, it probably was not the underly ing cause of renal failure. The observed hepatic dysfunc tion plays an important pathophysiologic role in the induction of renal failure by decreasing renal perfusion and producing prerenal azotemia . Brief renal insuffi ciency is mainly hemodynamic and in many ways mimicks the hepatorenal syndrome. Investigators have found that pretransplant levels of tumor-growth factor-P may predict subsequent VOD and the probability of ARF . Preventive and newer treatment measures in lessening VOD and ARF, such as prostaglandins of the El group , recombinant tissue plasminogen factor  or block ing tumor-necrosis factor-L (pentoxifylline) , are espe cially interesting but require further evaluation. An age of more than 25 years was the second strongest association with ARF. In spite of important advances in BMT over the last 10 years, age persists as one of the most important factors related to survival . The higher inci dence of ARF in older patients could probably contribute to this finding. This higher incidence of ARF in the oldest patients could be explained by a lower renal function reserve or because of a higher number of medical compli cations. ARF was significantly more frequent in allogeneic transplants than in autologous BMT. The major incidence of VOD and the use of CsA definitely contributed to this finding (table 4). Another observation of our study was that develop ment of renal insufficiency had ominous implications for patient survival. The overall mortality in the ARF group was 45.8 versus 17% in the non-ARF group. However, in patients who needed dialysis mortality increased to 88%. The higher mortality for ARF dialyzed patients is not related to the dialysis technique. Other authors found mortality rates in the dialyzed group as high as ours . A potential explanation exists: 70% of the patients had severe hepatic VOD preceding ARF, and the substantial mortality and morbidity from VOD are significantly asso ciated with the subsequent onset of multiorgan failure. This high mortality rate is similar to intensive care unit multiorgan ARF . Also this high incidence of VOD could explain why volume overload, not severe azotemia, is the usual indication for initiating dialysis. Approxi mately two thirds of the patients are monoliguric; how ever, after dialysis is begun oliguria frequently super venes. Although nearly all cases with ARF that survived recovered renal function completely at 1 month, renal insufficiency is a bad prognostic marker for up to 3 months because the mortality rate in this group was signif icantly worse. In summary ARF is a common and early complication following BMT especially in allogeneic BMT and carries a grave prognosis. Neither sex, nor underlying disease, nor sepsis, nor GVHD, nor conditioning therapy correlated with ARF. Both age and VOD are the main risk factors that can increase the probability of ARF following BMT. ARF associated with VOD is the main indication of HD. Acknowledgements The authors wish to thank Dr. F. Rodriguez Salvanés from the Investigation Unit for his statistical assistance. This work was supported in part by grant number 93/0522 awarded by the Fondo de Investigaciones Sanitarias de la Seguridad Social o f the National Institute of Health of Spain. References 478 4 Storb R, Deeg HJ, Whitehead J, et al: Metho trexate and cyclosporine compared with cyclo sporine alone for the prophylaxis of acute graft versus host disease after marrow transplanta tion. N Engl J Med 1986;314:729-735. 5 Thomas ED. Storb R. Clift RA. et al: Bone marrow transplantation. N Eng J Med 1975; 292:832-843. 6 Rollins BJ: Hepatic veno-occlusive disease. Am J Med 1986;81:297-304. 7 Shulman HM. Hinterberger W: Hepatic venoocclusive disease-liver toxicity syndrome after bone marrow transplantation. Bone Marrow Transplant 1992;10:197-214. Grusset al. 8 Atkinson JP. Waldman TA. Stein SF. et al: Sys temic capillary leak syndrome and monoclonal IgG gammopathy. Medicine (Baltimore) 1977; 56:225-239. 9 Zager RA, O'Quigley, Zager BK. et al: Acute renal failure following bone marrow transplan tation. A retrospective study of 272 patients. Am J Kidney Dis 1989;13:210-216. 10 Zager RA: Nephrology forum: acute renal fail ure in the setting of bone marrow transplanta tion. Kidney Ini 1994;46:1443-1458. Acute Renal Failure in Bone Marrow Transplantation Downloaded by: Vanderbilt University Library 18.104.22.168 - 10/27/2017 4:19:45 AM 1 Champlin RE. Gale RP: Bone marrow trans plantation for acute leukemia: Recent ad vances and comparison with alternative thera pies. Semin Hematol 1987;24:55-67. 2 Liano F, Gallego A, Pascual J, et al: Prognosis of acute tubular necrosis: An extended prospec tively contrasted study. Nephron 1993:63:2131. 3 Lopez J, Vazquez L, Serrano F, et al: Fractioned TBI and methotrexate-cyclosporin do not seem to increase relapses in BMT for first chronic phase CML patients: Results of a single centre study. Bone Marrow Transplant 1992; 10:235-239. 18 Deray G. Lebland V, Rafficks S. Sutton L, Ray mond F, Jacobs C: Evaluation of renal function after bone marrow transplantation (BMT): Outcome of 56 patients (abstract). 30th Congr EDTA, Glasgow 1993. 19 Lindquist S: The heat shock response. Annu Rev Biochem 1986:55:1151-1191. 20 Smith DM. Wiscnberger DD. Bicrman P. ct al: Acute renal failure associated with autologous bone marrow transplantation. Bone Marrow Transplant 1987;2:195-201. 21 Tarbell NJ, Guinan EC. Chin L, et al: Renal insufficiency after total body irradiation for pediatric bone marrow transplantation. Radiother Oncol 1990:18(suppl 1): 139—142. 22 Lonnerkolm G. Carlson K. Bratteby L, et al: Renal function after autologous bone marrow transplantation. Bone Marrow Transplant 1991:8:129-134. 23 Craig JF. Sheehan T, Bell K: The haemolytic uremic syndrome and bone marrow transplan tation. Br Med J 1987:295:887. 24 Rennuzi G, Bertani T: Renal vascular and thrombotic effects of cyclosporine. Am J Kid ney Dis 1989;13:261-272. 25 Rondeau E. Sraer J, Bens M, et al: Production of 5-lipoxygenase by peripheral leucocytes in capillarv leak svndrome (Clarkson disease). E urJ Clin Invest'1987;17:53-57. 26 Castill A. Prieto J, Fausto N: Transforming growth factors beta and alpha in chronic liver disease. Effects of interferon alpha therapy. N Engl J Med 1991;324:933-940. 27 Ibrahim A, Pico JL, Marahinchi D: Hepatic veno-occlusive disease following bone marrow transplantation treated by prostaglandin El. Bone Marrow Transplant 1991;7(suppl 2):53. 28 Bagin TP. Horber P, Marcus RE: Veno-occlu sive disease of the liver complicating ABMT successfully treated with recombinant tissue plasminogen activator (rt-PA). Bone Marrow Transplant 1990;5:439-441. 29 Bianco J A, Appelbaum FR. N em unaitisJ.et al: Phase I-II Trial of Pentoxifylline for the pre vention of transplant related loxicities follow ing bone marrow transplantation. Blood 1991; 78:1205-1211. 30 Bortiw MM, Horowitz MM. Rimm AA: Progress report from the International Bone Marrow Transplant Registry. Bone Marrow Transplant 1992;10:113-122. 31 Tran DD, Groeneveld ABJ, Van dcr Meulen J, Nauta JJP: Age. chronic disease, sepsis, organ system failure, and mortality in a medical in tensive care unit. Crit Care Med 1990; 18:474— 479. 479 Downloaded by: Vanderbilt University Library 22.214.171.124 - 10/27/2017 4:19:45 AM 11 Kone BC. Whelton A, Santos G, et al: Hyper tension and renal dysfunction in bone marrow transplant recipients. QJ Med 1988:260:985— 995. 12 Thicld G: Experimental cyclosporin A nephro toxicity: A summary of the international work shop, Basle, April 24-26, 19S5. Clin Nephrol 1985;25(suppl 1):205-210. 13 Chapman JR. Griffiths D. Hardling NGL, Morris PJ : Reversibility of cyclosporin nephro toxicity after three months treatment. Lancet 1985;i: 128—130. 14 Kennedy MS, Deeg HJ. Siegel M, et al: Acute renal nephrotoxicity with combined use of amphotericine B and cyclosporine after marrow transplantation. Transplantation 1983;35: 211-215. 15 Whiting PH. Simpson JG: The enhancement of cyclosporine A nephrotoxicity by gentamicin. Biochem Pharmacol 1983;32:2025. 16 Churchill DN. Seely J: Nephrotoxicity associ ated with combined gentamicin-amphotericinB therapy. Nephron 1977;19:176-181. 17 Thiru S: Pathological effects of cyclosporine A in clinical practice: in Thomson AW (ed): Cy closporin: Mode of Action and Clinical Appli cations. London. Kluwer Academic, 1989. pp 324-364.