Hemorrhagic Cystitis after Allogeneic Bone Marrow ... - Core

7 downloads 0 Views 115KB Size Report
tricia Johnson, and Cynthia Walker for data collec- tion; Janet Davies for editorial assistance; and Sara. Williams for manuscript preparation. REFERENCES. 1.
Biology of Blood and Marrow Transplantation 9:698-705 (2003) 䊚 2003 American Society for Blood and Marrow Transplantation 1083-8791/03/0911-0005$30.00/0 doi:10.1016/S1083-8791(03)00269-6

Hemorrhagic Cystitis after Allogeneic Bone Marrow Transplantation in Children: Clinical Characteristics and Outcome G. A. Hale,1,2 R. J. Rochester,1 H. E. Heslop,1,2 R. A. Krance,1,2 J. R. Gingrich,3,4 E. Benaim,1,2 E. M. Horwitz,1,2,5 J. M. Cunningham,1,2,5 X. Tong,6 D. K. Srivastava,6 W. H. Leung,1,2 P. Woodard,1,2 L. C. Bowman,1,2 R. Handgretinger1,2 1 Division of Stem Cell Transplantation, St. Jude Children’s Research Hospital; 2Department of Pediatrics, University of Tennessee Health Science Center; 3Department of Surgery, Division of Urology, St. Jude Children’s Research Hospital; 4Department of Urology, University of Tennessee Health Science Center; 5 Division of Experimental Hematology, St. Jude Children’s Research Hospital; 6Department of Biostatistics, St. Jude Children’s Research Hospital, Memphis, Tennessee

Correspondence and reprint requests: Gregory A. Hale, MD, Department of Hematology/Oncology, St. Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105-2794 (e-mail: [email protected]). Received January 28, 2003; accepted July 30, 2003

ABSTRACT Hemorrhagic cystitis (HC) is a well-documented adverse event experienced by patients undergoing hematopoietic stem cell transplantation. When severe, HC causes significant morbidity, leads to renal complications, prolongs hospitalization, increases health-care costs, and occasionally contributes to death. We retrospectively studied the medical records of 245 children undergoing an initial allogeneic bone marrow transplantation for malignant disease at St. Jude Children’s Research Hospital between 1992 and 1999 to describe the clinical course of HC in all patients and to identify the risk factors for HC in this cohort. Conditioning regimens included cyclophosphamide, cytarabine, and total body irradiation. Grafts from unrelated or mismatched related donors were depleted of T lymphocytes, whereas matched sibling grafts were unmanipulated. All patients received cyclosporine as prophylaxis for graft-versus-host disease. Recipients of grafts from matched siblings also received pentoxifylline or short-course methotrexate. Severe HC developed in 27 patients (11.0%). The median duration of HC was 73 days (range, 5-619 days); 12 patients had ongoing HC at the time of death. In univariate analyses, patients were at increased risk of severe HC if they were male (P ⴝ .021) or had received T cell– depleted grafts (P ⴝ .017), grafts from unrelated donors (P ⴝ .021), a lower total nucleated cell dose (P ⴝ .032), or antithymocyte globulin (P ⴝ .0446). Multiple regression analysis revealed male sex (␤ ⴝ .97; P ⴝ .027) and unrelated donor graft recipients (␤ ⴝ .83; P ⴝ .039) to be significant factors. © 2003 American Society for Blood and Marrow Transplantation

KEY WORDS Hemorrhagic cystitis Pediatrics



Bone marrow transplantation

INTRODUCTION Hemorrhagic cystitis (HC) is a significant cause of morbidity and, occasionally, mortality in patients undergoing hematopoietic stem cell transplantation (HSCT) [1-3]. As many as 25% of patients experience this complication after allogeneic HSCT; lower rates are reported in autograft recipients. The severity of HC can range from mild hematuria to life-threatening 698



Adenovirus



Complications



bleeding with urinary tract obstruction or renal insufficiency [4,5]. HC is thought to be caused by bacterial or viral urinary tract infections or by certain conditioning regimens [1,2,6]. To our knowledge, this is the largest published report to describe the clinical course of HC and the risk factors for its development in children undergoing allogeneic bone marrow transplantation (BMT).

Hemorrhagic Cystitis in Children

Table 1. Patient Characteristics Feature Diagnosis

Outcome T-Cell depleted TBI ATG Busulfan Donor CMV Recipient CMV Prior autologous transplantation Donor

Donor-recipient sex match

Acute GVHD Chronic GVHD

Category

HC (n ⴝ 27)

Control (n ⴝ 218)

Total (n ⴝ 245)

ALL/NHL AML/MDS CML Dead Alive Yes No Yes No Yes No Yes No Negative Positive Negative Positive Yes No MMFM Matched sibling Unrelated Male-male Female-female Male-female Female-male Grade (II, III, IV) Grade (0,1) Yes No

8 (29.6%) 14 (51.9%) 5 (18.5%) 17 (63.0%) 10 (37.0%) 23 (85.2%) 4 (14.8%) 27 (100%) 0 18 (66.7%) 9 (33.3%) 0 27 (100%) 15 (55.6%) 12 (44.4%) 14 (51.9%) 13 (48.2%) 4 (14.8%) 23 (85.2%) 2 (7.4%) 4 (14.8%) 21 (77.8%) 12 (44.4%) 5 (18.5%) 2 (7.4%) 8 (29.6%) 5 (18.5%) 22 (81.5%) 6 (22.2%) 21 (77.8%)

59 (27.1%) 125 (57.3%) 34 (15.6%) 122 (56.0%) 96 (44.0%) 135 (61.9%) 83 (38.1%) 217 (99.5%) 1 (0.5%) 100 (45.9%) 118 (54.1%) 1 (0.5%) 217 (99.5%) 123 (56.4%) 95 (43.6%) 115 (52.8%) 103 (47.3%) 18 (8.3%) 200 (91.7%) 31 (14.2%) 79 (36.2%) 108 (49.5%) 53 (24.3%) 68 (31.2%) 40 (18.4%) 57 (26.2%) 53 (24.3%) 165 (75.7%) 37 (17.0%) 181 (83.0%)

67 (27.4%) 139 (56.7%) 39 (15.9%) 139 (56.7%) 106 (43.3%) 158 (64.5%) 87 (35.5%) 244 (99.6%) 1 (0.4%) 118 (48.2%) 127 (51.8%) 1 (0.4%) 244 (99.6%) 138 (56.3%) 107 (43.7%) 129 (52.7%) 116 (47.4%) 22 (9.0%) 223 (91.0%) 33 (13.5%) 83 (33.9%) 129 (52.7%) 65 (26.5%) 73 (29.8%) 42 (17.1%) 65 (26.5%) 58 (23.7%) 187 (76.3%) 43 (17.6%) 202 (82.5%)

ALL indicates acute lymphoblastic leukemia; NHL, non-Hodgkin lymphoma; AML, acute myelogenous leukemia; MDS, myelodysplastic syndrome; CML, chronic myelogenous leukemia; MMFM, mismatched family member.

PATIENTS AND METHODS Patients

We reviewed the records of 245 consecutive children who underwent allogeneic BMT for malignant disease after a myeloablative preparative regimen between 1992 and 1999 at our institution. Patients were treated on institutional protocols approved by the institutional review board. The characteristics of these

patients are shown in Tables 1 and 2. Class I HLA antigens were typed serologically, and class II antigens were typed by high-resolution DNA typing. Conditioning Regimen

Patients received cytarabine (3 g/m2 per dose; 6 doses) and cyclophosphamide (45 mg/kg per dose; 2 doses) with mesna uroprotection [7]. They also re-

Table 2. Descriptive Statistics Analysis Results Variable

HC*

Control*

P Value

Median age at BMT (y) Median donor age (y) Median T-cell dose (per kilogram) Median total cell dose (per kilogram) Median CD 34ⴙ cell dose (per kilogram) Median cyclophosphamide dose (mg/kg) Median follow-up (years after BMT) Time to ANC > 500 (d) Time to platelet > 20 (d) Time to platelet > 50 (d) Time to platelet > 100 (d)

12.6 (3.50-21.9) 30 (7.0-48) 8.89 ⴛ 105 (2.0 ⴛ 105-2.79 ⴛ 106) 1.25 ⴛ 108 (1.9 ⴛ 107-4.37 ⴛ 108) 2.3 ⴛ 106 (3.75 ⴛ 105-7.54 ⴛ 106) 89 (67-138) 0.73 (0.14-6.2) 20.5 (15-68) 33 (20-89) 37 (22-96) 43 (25-94)

11.2 (0.08-24.1) 26 (0.6-53) 1.0 ⴛ 106 (1.7 ⴛ 104-1.64 ⴛ 107) 1.59 ⴛ 108 (1.96 ⴛ 107-4.69 ⴛ 109) 2.26 ⴛ 106 (1.52 ⴛ 104-2.19 ⴛ 107) 89 (26-200) 0.98 (0.04-8.5) 19 (12-46) 33 (15-102) 33 (15-105) 38 (19-98)

.24 .19 .11 .012 .43 .47 .35 .26 .23 .17

ANC indicates absolute neutrophil count. *Numbers in parentheses indicate range.

BB&MT

699

G. A. Hale et al.

ceived total body irradiation (TBI) in 8 fractions, for a total of 12 Gy (matched sibling marrow recipients) or 14 Gy (unrelated or mismatched family member marrow recipients). One patient who could not receive TBI was treated with busulfan (1 mg/kg per dose; 16 doses) and cyclophosphamide (50 mg/kg per dose; 4 doses) with mesna uroprotection. Antithymocyte globulin (ATG) was administered as part of the pretransplantation conditioning regimen to 121 patients who received marrow from unrelated or mismatched family donors to enhance immunosuppression. In all patients, cyclophosphamide was administered as a 1-hour infusion. Hydration with intravenous fluids at 3 L/m2/d started at least 24 hours before the initial cyclophosphamide dose and continued for 24 h after the conditioning regimen was complete. Mesna (45 mg/kg divided into 5 doses) was administered before and 3, 6, 9, and 12 hours after each dose of cyclophosphamide [7]. Graft-versus-Host Disease Prophylaxis

Marrow from mismatched family members and from unrelated donors was depleted of approximately 1.5 log T lymphocytes by use of monoclonal antibodies to CD6 and CD8 plus rabbit complement [7]. Two days before transplantation, all recipients began to receive cyclosporine at dosages adjusted to yield whole-blood concentrations of 200 to 350 ng/mL. Patients given matched sibling marrow received additional graft-versus-host disease (GVHD) prophylaxis with pentoxifylline or short-course methotrexate. Toxicity

Acute and chronic GVHD were graded according to standard criteria [8]. Patients with mild or moderate GVHD (grade I or II) were treated with methylprednisolone (1-2 mg/kg/d). If GVHD remained at grade II for ⬎7 days or progressed, the steroid dosage was increased to 5 to 10 mg/kg/d or second-line therapy with ATG or to the antibody ABX-CBL was initiated [9]. Regimen-related toxicity was scored with the National Cancer Institute common toxicity criteria. Severe HC was defined as National Cancer Institute common toxicity criteria grade 2 to 4. Grade 2 HC was defined as gross hematuria without clots, treated with hydration and diuresis. Persistent gross bleeding or clots requiring transfusion or cauterization characterized grade 3 HC. Patients with grade 4 HC had resistant hematuria that required surgery. The worst clinical presentation was defined as the grade of HC. The date of onset was the first day of initial signs or symptoms of HC. Early-onset HC was defined as the onset of HC signs or symptoms before posttransplantation day 60; late-onset HC was defined as the onset of HC signs or symptoms after 60 days after transplantation. 700

HC Screening and Management

HC screening and management was standardized at our institution. Urinalysis was performed daily during cyclophosphamide administration. Surveillance urine cultures for bacteria and viruses were performed on a weekly basis and when hematuria occurred. The initial treatment for HC consisted of intravenous fluid hydration at 3 L/m2/d with diuretics. Platelet transfusions were given to patients with thrombocytopenia. Renal and bladder ultrasound was performed in patients with gross hematuria. When clinically indicated, continuous bladder irrigation with saline or glycine was performed and was titrated to keep a clear urine outflow. For patients with persistent bleeding despite these measures, cystoscopy was performed with clot evacuation, with cauterization of bleeding sites if focal. If HC persisted, progressive instillation of alum, silver nitrate, or formalin was performed. Three patients received systemic conjugate estrogens at the discretion of the transplant physician. Supportive Care

All patients with an absolute neutrophil count ⱖ1.0 ⫻ 109/L received trimethoprim/sulfamethoxazole (co-trimoxazole 5 mg/kg/d divided into 2 doses) 3 times a week for Pneumocystis carinii pneumonia prophylaxis. Patients who were seropositive for cytomegalovirus (CMV) or had a CMV-seropositive donor received ganciclovir (5 mg/kg twice daily for 14 days) when the absolute neutrophil count exceeded 0.5 ⫻ 109/L for 2 consecutive days. A reduced dose (5 mg/kg/d for 5 days per week) was then given until day 120 after transplantation. Intravenous immunoglobulin (500 mg/kg weekly) was given until day 120 and was given monthly thereafter until 12 months after transplantation. Patients had weekly viral culture and shell vial assay performed on peripheral blood as surveillance for CMV. Patients who received T cell– depleted grafts were eligible to receive donor-derived cytotoxic T cells specific for Epstein-Barr virus as prophylaxis against Epstein-Barr virus lymphoproliferative disease at approximately 60 days after transplantation [10]. No prophylaxis for fungal infections or veno-occlusive disease was routinely provided. Statistical Methods

The Student t test was used to assess the difference of distributions on the continuous variables between the HC and control groups. The cumulative incidence of HC was estimated by using the method of Prentice et al. [11]. The effects of several categorical variables were evaluated 1 at a time by using Gray’s method [12]. However, for continuous variables and multiple regression analyses, we used the proportional hazard model for competing risk proposed by Fine and Gray [13]. These variables included diagnosis, age at BMT,

Hemorrhagic Cystitis in Children

Table 3. Univariate Analysis Results

Variable Patient sex Male (n ⴝ 130) Female (n ⴝ 115) Diagnosis ALL/NHL (n ⴝ 67) AML/MDS (n ⴝ 139) CML (n ⴝ 39) Donor MMFM (n ⴝ 33) Matched sibling (n ⴝ 83) Unrelated (n ⴝ 129) ATG Yes (n ⴝ 118) No (n ⴝ 127) HUSTTP Yes (n ⴝ 28) No (n ⴝ 217) TCD Yes (n ⴝ 158) No (n ⴝ 87) Donor CMV Positive (n ⴝ 107) Negative (n ⴝ 138) Recipient CMV Positive (n ⴝ 116) Negative (n ⴝ 129) Donor recipient sex match Male-male (n ⴝ 65) Female-female (n ⴝ 73) Male-female (n ⴝ 42) Female-male (n ⴝ 65) Prior autologous transplantation Yes (n ⴝ 22) No (n ⴝ 223) Disease status at BMT CP1/CR1 (n ⴝ 89) CP2/CR2 or more (n ⴝ 51) Others (n ⴝ 105) Acute GVHD Grade (0, I) (n ⴝ 187) Grade (II, III, IV) (n ⴝ 58) Chronic GVHD Yes (n ⴝ 43) No (n ⴝ 202) Cell dose Cyclophosphamide dose

2-y Cumulative Incidence of HC (% ⴞ SE) 15.4 ⴞ 3.2 6.1 ⴞ 2.2

P Value

.021

11.9 ⴞ 4.0 10.1 ⴞ 2.6 12.8 ⴞ 5.4

.85

6.1 ⴞ 4.2 4.8 ⴞ 2.4 16.3 ⴞ 3.3

.021

RESULTS

15.2 ⴞ 3.3 7.1 ⴞ 2.3

.0446

21.4 ⴞ 7.9 9.7 ⴞ 2.0

.062

14.6 ⴞ 2.8 4.6 ⴞ 2.3

.017

11.2 ⴞ 3.1 10.9 ⴞ 2.7

.95

11.2 ⴞ 2.9 10.8 ⴞ 2.7

.90

18.5 ⴞ 4.9 6.8 ⴞ 3.0 4.8 ⴞ 3.3 12.3 ⴞ 4.1 18.2 ⴞ 8.5 10.3 ⴞ 2.0 10.1 ⴞ 3.2 15.7 ⴞ 5.2 9.5 ⴞ 2.9

.082

.25

.50

11.8 ⴞ 2.4 8.6 ⴞ 3.7

.47

13.9 ⴞ 5.3 10.4 ⴞ 2.1 Continuous variable Continuous variable

.48 .032 .38

ALL indicates acute lymphoblastic leukemia; NHL, non-Hodgkin lymphoma; AML, acute myelogenous leukemia; MDS, myelodysplastic syndrome; CML, chronic myelogenous leukemia; MMFM, mismatched family member; HUSTTP, post-BMT microangiopathy; CP, chronic phase; CR, continuous remission; TCD, T-cell depleted.

disease status at the time of BMT, donor type, TBI conditioning regimen, ATG-containing conditioning regimen, donor CMV serology, recipient CMV serology, donor-recipient sex match, prior autologous transplantation, cyclophosphamide dose, platelet engraftment, and the presence of acute or chronic

BB&MT

GVHD (Table 3). Univariate analysis was performed first to assess the effect of each factor. All factors found to be significant at the level of ␣ ⫽ .10 were planned to be analyzed with a multiple regression model. The length of time at risk for HC was computed from the date of BMT to the date of the diagnosis of HC, the date of death, or the date of last contact, whichever came first. Death from any cause was considered a competing event. The criterion for significance for all analyses was P ⱕ .05. All statistical analyses used SAS release 8.1 (SAS Institute, Cary, NC) and SPLUS 2000 (Mathsoft Inc, Seattle, WA).

Transplant Patient Characteristics

Two hundred forty-five patients with hematologic malignancies underwent allogeneic BMT at our institution during the study period. The median age of patients was 11.5 years (range, 0.08-24.1 years) at the time of transplantation; 53% were male. Twenty-two (9.0%) had received a prior autologous BMT. Diseases for which BMT was performed included the following: 67 patients (27%) had acute lymphoblastic leukemia or non-Hodgkin lymphoma, 139 patients (57%) had acute myelogenous leukemia or myelodysplastic syndrome, and 39 patients (16%) had chronic myelogenous leukemia. Eighty-three (33.9%) patients received grafts from HLA-matched sibling donors, 33 (13.5%) from mismatched family member donors, and 129 (52.7%) from unrelated donors. All patients received cyclophosphamide, 1 (0.4%) patient received busulfan, and 244 (99%) patients received TBI as part of the conditioning regimen. Fifty-eight (23.7%) patients developed grade II to IV acute GVHD; 43 (17.6%) developed chronic GVHD. HC Patient Characteristics

Severe HC developed in 27 patients, yielding a cumulative incidence of 11% ⫾ 2.0%. Of note, there was no difference in time to neutrophil or platelet engraftment between transplant recipients who did and did not develop HC (Table 2). For HC patients, the median age was 12.6 years (range, 3.5-21.9 years); 74% were male. The characteristics of patients who did and did not experience HC are compared in Tables 1 and 2. HC began a median of 52 days (range, 10-347 days) after BMT, and 14 patients had earlyonset HC (Table 4). Patients with HC received a graft with a lower total nucleated cell dose than the control group (P ⫽ .012). Cultures of urine were positive for adenovirus in 3 patients who had cystitis; CMV was isolated from the urine of 1 patient who had HC, and adenovirus was isolated from the throat of the same patient. Adenovirus was identified in the stool of 1 patient, but this patient’s urine contained no patho701

G. A. Hale et al.

Table 4. Clinical Course Acute/ Patient Chronic Silver Renal No. Onset GYHD CBI Nitrate* Alum* Formalin* Estrogen Failure 1

Late

II/yes

Yes

2 3 4 5 6 7

Late Late Late Late Late Early

I/no I/no I/no II/yes I/no I/no

No Yes Yes No Yes Yes

8 9

Early Early

I/no 0/yes

Yes Yes

10 11 12 13 14 15 16

Early Late Early Early Early Early Late

0/no I/no 0/no I/no I/no 0/no I/no

Yes No No Yes Yes No Yes

17

Early

I/yes

Yes

18 19

Late Late

0/no I/no

Yes Yes

20

Early

I/yes

Yes

21 22 23

Late Early Early

I/no II/yes 0/no

Yes Yes Yes

24 25 26 27

Late Late Early Early

II/no 0/no I/no II/no

Yes Yes Yes Yes

2

1 2

2

1

1

1

No

No

176

No

Died

2

No Yes No No No No

No No No No No No

10 107 104 95 117 144

No Yes No No No No

Died Died Died Died Died Died

1

No Yes

No No

61 80

No Yes

Died Alive

No No No No No No Yes

No No No No No No No

24 48 12 20 69 45 47

No No Yes Yes Yes Yes No

Died Died Alive Alive Alive Died Died

No

Yes

619

Yes

Alive

No No

No Yes

146 89

Yes Yes

Alive Alive

No

No

73

Yes

Alive

No No No

No No No

8 129 87

Yes Yes Yes

Died Died Alive

No No No No

No No Yes No

220 5 66 53

No Yes Yes No

Died Died Alive Died

2

3

1

2 1

Culture

HC Duration Resolution (d) of HC Outcome

2

CMV urine, adenovirus throat

Adenovirus urine Adenovirus urine

Adenovirus urine

Adenovirus stool

Comments Fulguration, suprapubic catheter, cystoscopy

Perforation

Ureteral stents Cystoscopy Cystoscopy

Ureteral stents, suprapubic catheter, perforation Fulguration Nephrostomy, suprapubic catheter Ureteral stents, cystoscopy

Cystoscopy Nephrostomy Cystoscopy

CBI indicates continuous bladder irrigation. *Number of instillations.

gens. Cystoscopy did not reveal any cytologic findings suggestive of viral infection in any patient. Five (18.5%) patients had grade II to IV acute GVHD, and 6 (22%) had chronic GVHD. Clinical Course

Treatment interventions and outcomes are outlined in Table 4. All patients were treated with intravenous hydration and diuresis. Patients with thrombocytopenia were given platelet transfusions, and no patient had evidence of immune-mediated platelet consumption. These interventions alone led to reso702

lution of cystitis in 5 patients. Twenty-two (81%) patients required continuous bladder irrigation (4 with glycine), and 9 underwent cystoscopy with clot evacuation or cauterization of the sites of bleeding. Thirteen patients were treated with irrigation alone. Other therapeutic interventions included instillation of silver nitrate (n ⫽ 4), alum (n ⫽ 4), or formalin (n ⫽ 7) or administration of estrogens (n ⫽ 3). Hydronephrosis developed in 9 patients, of whom 3 patients required ureteral stent placement and 2 required nephrostomy tubes. Three patients required suprapubic catheters. Two patients who received continuous bladder irriga-

Hemorrhagic Cystitis in Children

tion had perforated bladders, 1 of which was discovered at autopsy. Renal failure occurred in 3 patients, and 1 required hemodialysis. Urologists recommended that 1 patient (patient 17) undergo cystectomy; however, the family refused, and the patient eventually recovered. No patient died as a direct result of HC. Five of the patients who had resolution of HC subsequently died of other causes not related to HC. Twelve patients had ongoing HC at the time of death. For these patients, HC lasted a median of 78 days (range, 10-220 days). For survivors, the median duration of HC was 77 days (range, 12-619 days). For all patients, the median duration of cystitis was 73 days (range, 5-619 days). HC resolved in 15 patients, including 2 of 3 patients treated with estrogens and 3 of 4 with documented viral infection. Early-onset HC occurred in 14 patients (Table 4): 8 of these patients survived, and all had resolution of HC. Of the 6 who died, only 2 had resolution of HC before death. For patients with early-onset HC, causes of death included infection (n ⫽ 2), relapse (n ⫽ 2), respiratory failure (n ⫽ 1), and multisystem organ failure (n ⫽ 1). Late-onset HC occurred in 13 patients (Table 4); only 2 of these patients survived. The survivors experienced resolution of cystitis, but only 3 of the 11 patients who died had resolution of HC before death. For patients with late-onset HC, causes of death included relapse (n ⫽ 4), respiratory failure (n ⫽ 3), multisystem organ failure (n ⫽ 1), adult respiratory distress syndrome (n ⫽ 2), and hepatic failure (n ⫽ 1). Risk Factor Analysis

The following factors were identified as being associated with an increased risk of HC by univariate analysis: (1) male gender (P ⫽ .021); (2) a graft with a lower total nucleated cell dose (P ⫽ .032); (3) an unrelated donor graft (P ⫽ .021); (4) a T cell– depleted graft (P ⫽ .017); and (5) use of ATG (P ⫽ .0446; Table 3). However, donor type, total cell dose, use of ATG, and the use of T-cell depletion were statistically confounded with one another. For example, recipients of unrelated and mismatched family member grafts received T cell– depleted grafts and grafts with lower nucleated cell doses than those who received matched sibling grafts. In addition, ATG was not administered to recipients of matched sibling grafts. Donor type determined the use of ATG and whether a T cell– depleted graft was used, and, thus, a lower cell dose was given. Therefore, donor type was entered into the multiple regression model. A multiple regression analysis then identified male sex (␤ ⫽ .97; P ⫽ .027) and unrelated donor grafts (␤ ⫽ .83; P ⫽ .039) as significant factors.

BB&MT

DISCUSSION To our knowledge, this article describes the largest series that describes the risk factors for and the clinical course of HC in children after allogeneic BMT. We found that children who received a graft with a lower total nucleated cell count, received a T cell– depleted graft, received an unrelated donor graft, or underwent an ATG-containing conditioning regimen were at increased risk for HC after allogeneic BMT. The incidence of HC in our series is comparable to that reported in other series of patients undergoing allogeneic HSCT [1-3]. In our study, all patients who received cyclophosphamide also received mesna uroprotection and intravenous hydration—interventions that have been shown to reduce the incidence of HC caused by cyclophosphamide treatment [14]. In our series, the cyclophosphamide dose did not affect the risk of HC. Other studies have likewise not demonstrated any relationship between HC and the dose of cyclophosphamide [14,15]. The use of busulfan in the conditioning regimen has been reported to be a risk factor for HC in several series of predominantly adult patients [1-3], but in our series, only 1 patient received busulfan and did not experience HC. In a previous study at our institution [16], TBI was identified as a risk factor for adenovirus infection; however, we were not able to investigate TBI in this study. This may not have been corroborated in this study because adenoviral infection does have other manifestations that may occur without HC. Although other investigators have identified GVHD as a risk factor for HC, our study did not corroborate this finding [1-3]. This discrepancy may be the result of the low rates of acute and chronic GVHD observed with the method of T-cell depletion at our institution [7]. In the univariate analysis, patients who received grafts with a lower total number of nucleated cells or T cell– depleted grafts were more likely to develop HC. This observation may be attributable to the more rapid hematologic and immunologic recovery seen in patients who receive grafts that contain more nucleated cells [17]. Our study identified several factors (unrelated donor grafts and male sex) that were associated with an increased risk of HC. Our series found that recipients of unrelated donor grafts were more likely to develop HC. In our study, recipients of unrelated donor grafts received T cell– depleted grafts and ATG-containing conditioning regimens, both of which are known to result in significant immunosuppression. Considered in toto, these may be surrogate markers of the immunocompromised state after BMT. Unfortunately, immune reconstitution studies were not serially performed in these patients. Other series have described HC resulting from viral infections in immunocompro703

G. A. Hale et al.

mised patients [18-22]. A recent case report observed an inverse association between the onset of GVHD and systemic adenovirus infection [23]. In addition, the use of ATG in preparative regimens or as treatment for GVHD has been associated with infectious complications, particularly viruses [24-27]. The low nucleated cell dose in HC patients did not result in delayed neutrophil or platelet engraftment compared with controls in our series; therefore, prolonged neutropenia is unlikely to have resulted in an increased period of infection risk. It is possible that the recovery in lymphocyte number and function may have been delayed in recipients of T cell– depleted grafts, thus putting the patients at risk for infection. Pathogens known to cause HC were identified from urine samples in only 4 patients: 3 patients were infected with adenovirus, and 1 was infected with CMV [28]. No fungal infection was detected. Because polymerase chain reaction as surveillance testing of peripheral blood or urine samples was not routinely performed during the study period, all pathogens were likely not detected. Many viral infections, including polyomaviruses such as BK and JC viruses, can be extremely difficult to detect by typical viral culture [29,30]. The use of ganciclovir prophylaxis likely resulted in a low rate of CMV reactivation and may have reduced other viral infections as well. The routine use of sensitive screening and diagnostic assays to detect potential pathogens may increase the ability to detect pathogens. Various therapeutic interventions for patients with HC have been reported, with varying degrees of success [24-28,31-37]. For most of the patients in our study, conservative management achieved resolution of HC: the HC resolved in 5 patients without continuous bladder irrigation or surgical intervention and in an additional 13 patients with continuous bladder irrigation alone. However, for some patients with HC that did not respond to these measures, interventions such as cystoscopy with cauterization or instillation were successful. Although we recommended cystectomy for 1 patient, no patient in our series underwent cystectomy, an intervention that others have reported as necessary for the treatment of extraordinarily severe HC [38]. However, HC was refractory to these interventions in 12 patients. Two patients experienced perforation of their bladders. In summary, the incidence of HC in our series of children is similar to that reported in other large series of predominantly adult recipients. Although our study did not confirm risk factors for HC identified in other studies of cohorts comprised primarily of adult patients undergoing autologous or allogeneic BMT, it did show that older patients were more likely to develop HC. We found that male patients were at higher risk of HC. We also identified recipients of unrelated donor grafts as being high risk of severe HC, likely 704

because of the use of T-cell depletion, the low cell content of the graft, and the use of ATG. These factors are reflective of the immunocompromised state after BMT. Improved diagnostic and screening methods, such as polymerase chain reaction assays, that are now available to identify infectious agents should be incorporated into evaluations of patients who have undergone HSCT and are at risk for HC, allowing treatment with appropriate medications [39].

ACKNOWLEDGMENTS This work was supported in part by National Institutes of Health Cancer Center Support grant no. P 30CA 21765 and by the American Lebanese Syrian Associated Charities. The authors would like to express their appreciation to Frances Curran and Nancy Wright for data management; Margaret Aymond, Patricia Johnson, and Cynthia Walker for data collection; Janet Davies for editorial assistance; and Sara Williams for manuscript preparation.

REFERENCES 1. Seber A, Shu XO, Defor T, et al. Risk factors for severe hemorrhagic cystitis following BMT. Bone Marrow Transplant. 1999;23:35-40. 2. Trotman J, Nivison-Smith I, Dodds A. Haemorrhagic cystitis: incidence and risk factors in a transplant population using hyperhydration. Bone Marrow Transplant. 1999;23:797-801. 3. Turkeri LN, Lum LG, Uberti JP, et al. Prevention of hemorrhagic cystitis following allogeneic bone marrow transplant preparative regimens with cyclophosphamide and busulfan: role of continuous bladder irrigation. J Urol. 1995;153:637-640. 4. Azzi A, Cesaro S, Laszlo D, et al. Human polyomavirus BK (BKV) load and haemorrhagic cystitis in bone marrow transplantation patients. J Clin Virol. 1999;14:79-86. 5. Leung AYH, Suen CKM, Lie AKW, et al. Quantification of polyoma BK viruria in hemorrhagic cystitis complicating bone marrow transplantation. Blood. 2001;98:1971-1978. 6. Talar-Williams C, Hijazi YM, Walther MM, et al. Cyclophosphamide-induced cystitis and bladder cancer in patients with Wegener granulomatosis. Ann Intern Med. 1996;124:477-484. 7. Hongeng S, Krance RA, Bowman LC, et al. Outcomes of transplantation with matched sibling and unrelated-donor bone marrow in children with leukemia. Lancet. 1997;350:767-770. 8. Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors. Transplantation. 1974;18: 295-304. 9. Heslop HE, Benaim E, Brenner MK, et al. Response of steroidresistant graft-versus-host disease to lymphoblast antibody CBL1. Lancet. 1995;346:805-806. 10. Rooney CM, Smith CA, Ng CY, et al. Infusion of cytotoxic T cells for the prevention and treatment of Epstein-Barr virusinduced lymphoma in allogeneic transplant recipients. Blood. 1998;92:1549-1555. 11. Prentice RL, Kalbfleisch JD, Peterson AV Jr, et al. The analysis

Hemorrhagic Cystitis in Children

12. 13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

of failure times in the presence of competing risks. Biometrics. 1978;34:541-554. Gray RJ. A class of k-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141-1154. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496509. Cox PJ. Cyclophosphamide cystitis—identification of acrolein as the causative agent. Biochem Pharmacol. 1979;28:2045-2049. Al-Rawithi S, El-Yazigi A, Ernst P, et al. Urinary excretion and pharmacokinetics of acrolein and its parent drug cyclophosphamide in bone marrow transplant patients. Bone Marrow Transplant. 1998;22:485-490. Hale GA, Heslop H, Krance R, et al. Adenovirus infection after pediatric bone marrow transplantation. Bone Marrow Transplant. 1999;23:277-281. Rocha V, Labopin M, Gluckman E, et al. Relevance of bone marrow cell dose on allogeneic transplantation outcomes for patients with acute myeloid leukemia in first complete remission: results of a European survey. J Clin Oncol. 2002;20:43244330. Rabenau HF, Preiser W, Franck S, et al. Polyomavirus viruria in bone marrow transplant recipients: lack of correlation with clinical symptoms. Infection. 2002;30:91-93. Baldwin A, Kingman H, Darville M, et al. Outcome and clinical course of 100 patients with adenovirus infection following bone marrow transplantation. Bone Marrow Transplant. 2000;26: 1333-1338. Chakrabarti S, Collingham KE, Stevens RH, et al. Isolation of viruses from stools in stem cell transplant recipients: a prospective surveillance study. Bone Marrow Transplant. 2000;25:277282. Kondo M, Kojima S, Kato K, et al. Late-onset hemorrhagic cystitis after hematopoietic stem cell transplantation in children. Bone Marrow Transplant. 1998;22:995-998. Longergan TA, Walzak MP. Hemorrhagic cystitis due to adenovirus infection following bone marrow transplantation. J Urol. 1994;151:1013-1014. Echavarria M, Herrera F, Solimano J, et al. Cyclic recovery of adenovirus in a stem cell transplant recipient: an inverse association with graft-verus-host disease. Bone Marrow Transplant. 2003;31:301-303. Cragg L, Blazar BR, Defor T, et al. A randomized trial comparing prednisone with antithymocyte globulin/prednisone as an initial systemic therapy for moderately severe acute graftversus-host disease. Biol Blood Marrow Transplant. 2000;6:441447. Bashey A. Immunosuppression with limited toxicity: the characteristics of nucleoside analogs and anti-lymphocyte antibodies used in non-myeloablative hematopoietic cell transplantation. Cancer Treat Res. 2002;110:39-49. Nakai K, Kanda Y, Mineishi S, et al. Suspected delayed immune

BB&MT

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

recovery against cytomegalovirus after reduced-intensity stem cell transplantation using anti-thymocyte globulin. Bone Marrow Transplant. 2002;29:237-241. Bacigalupo A, Lamparelli T, Bruzzi P, et al. Antithymocyte globulin for graft-versus-host disease prophylaxis in transplants from unrelated donors: 2 randomized studies from Gruppo Italiano Trapianti Midollo Osseo (GITMO). Blood. 2001;98: 2942-2947. Spach DH, Bawens JE, Myerson D, et al. Cytomegalovirusinduced hemorrhagic cystitis following bone marrow transplantation. Clin Infect Dis. 1993;16:142-144. Bedi A, Miller CB, Hanson JL, et al. Association of BK virus with failure of prophylaxis against hemorrhagic cystitis following bone marrow transplantation. J Clin Oncol. 1995;13:11031109. Rice SJ, Bishop JA, Apperley J, et al. BK virus as cause of haemorrhagic cystitis after bone marrow transplantation. Lancet. 1985;2:844-845. Vianelli N, Renga M, Azzi A, et al. Sequential vidarabine infusion in the treatment of polyoma virus-associated acute haemorrhagic cystitis late after allogeneic bone marrow transplantation. Bone Marrow Transplant. 2000;25:319-320. Lui YK, Harty JI, Steinbock GS, et al. Treatment of radiation or cyclophosphamide induced hemorrhagic cystitis using conjugated estrogen. J Urol. 1990;144:41-43. Vogeli TA, Pienemann F, Burdach S, Ackermann R. Urological treatment and clinical course of BK polyomavirus-associated hemorrhagic cystitis in children after bone marrow transplantation. Eur Urol. 1998;36:252-257. Ordemann R, Naumann R, Geissler G, et al. Encouraging results in the treatment of haemorrhagic cystitis with estrogen—report of 10 cases and review of the literature. Bone Marrow Transplant. 2000;25:981-985. Gonzalez-Fraile MI, Canizo C, Caballero D, et al. Cidofovir treatment of human polyomavirus-associated acute haemorrhagic cystitis. Transpl Infect Dis. 2001;3:44-46. Vela-Ojeda J, Tripp-Villanueva F, Sanchez-Cortes E, et al. Intravesical rhGM-CSF for the treatment of late onset hemorrhagic cystitis after bone marrow transplant. Bone Marrow Transplant. 1999;24:1307-1310. Demesmay K, Tissot E, Bulabois CE, et al. Factor XIII replacement in stem-cell transplant recipients with severe hemorrhagic cystitis: a report of four cases. Transplantation. 2002;74:11901192. Garderet L, Bittencourt H, Sebe P, et al. Cystectomy for severe hemorrhagic cystitis in allogeneic stem cell transplant recipients. Transplantation. 2000;70:1807-1811. Handgretinger R, Klingebiel T, Lang P, et al. Megadose transplantation of purified peripheral blood CD34(⫹) progenitor cells from HLA-mismatched parental donors in children. Bone Marrow Transplant. 2001;27:777-783.

705