|
|
||||||||
Pediatric Imaging |
1 From the Departments of Hematology-Oncology (A.E.H., L.C.B.), Diagnostic Imaging (S.C.K.), and Biostatistics and Epidemiology (O.G.G., X.P.X.), St Jude Childrens Research Hospital, 332 N Lauderdale, Memphis, TN 38105-2794; and Department of Radiology, University of Tennessee, Memphis (S.C.K.). Received September 4, 2001; revision requested October 29; final revision received May 24, 2002; accepted June 24. Supported in part by grant P01 CA-20180, Cancer Center Support (CORE) grant P30 CA-21765 from the National Cancer Institute, and the American Lebanese Syrian Associated Charities. Address correspondence to S.C.K. (e-mail: sue.kaste@stjude.org).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: A retrospective review of records of 120 consecutive pediatric patients who underwent allogeneic BMT in 1997 or 1998 was performed. Eighty-nine patients (median age, 8.1 years) fulfilled study eligibility criteria. Cumulative doses of nephrotoxic antimicrobial drugs were recorded, as well as serum creatinine and blood urea nitrogen concentrations from 24 hours before to 72 hours after each administration of iopamidol during a computed tomographic examination performed within 100 days after BMT. Random coefficient models were used to estimate nephrotoxic effects.
RESULTS: Mean baseline glomerular filtration rate was 130.2 mL/min/1.73 m2, and mean baseline creatinine concentration was 0.51 mg/dL (45 µmol/L). Older age at BMT (P < .001), use of foscarnet (P = .003), and receipt of iopamidol (P = .073) each prompted a rise in serum creatinine concentration. The antiviral drug foscarnet was associated with the largest increase in the creatinine level; the use of iopamidol effected a relatively small rise in creatinine level.
CONCLUSION: Iopamidol nephrotoxicity was negligible in this cohort of pediatric patients who had undergone allogeneic BMT, even in the presence of elevated renal function levels.
© RSNA, 2003
Index terms: Abdomen, CT, 70.12112, 80.12112 Bone marrow, transplantation Contrast media, effects Contrast media, toxicity Kidney, effects of drugs on, 81.64 Thorax, CT, 50.12112, 60.12112
| INTRODUCTION |
|---|
|
|
|---|
When searching for occult infection in patients who have undergone BMT, we routinely performed contrast-enhanced abdominal CT and nonenhanced chest and sinus CT. For patients with elevated serum creatinine concentrations, the decision of whether or not to administer contrast material follows an assessment by the diagnostic imaging and transplantation teams that takes into account the potential for contrast materialinduced renal impairment, the need to initiate or continue empiric antifungal or other nephrotoxic medications, and the probability of life-threatening complications, namely invasive fungal disease. After these factors have been considered, contrast material is often administered to patients with impaired renal function so that imaging studies can be optimized.
Our anecdotal experience not only has justified continuing this practice but also has prompted us to objectively substantiate this approach. Thus, the purpose of our study was to assess the effect of iopamidol on renal function as measured with blood urea nitrogen (BUN) and serum creatinine levels in pediatric, adolescent, and young adult patients who have undergone allogeneic BMT.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Twenty-one patients were excluded for not having undergone contrast-enhanced CT examination during the 100 days after transplantation. Ten patients with a history of BMT who had also undergone contrast-enhanced CT after that BMT were also excluded. Thus, the study cohort included a total of 89 patients.
Patients were categorized into three groups according to physiologic development at the time of BMT, corresponding to normative ranges used at our institution to evaluate renal function: infants younger than 1 year (n = 5), children between 1 and 12 years of age (n = 51), and adolescents older than 12 years (n = 33).
Primary diagnoses in the patients included (a) acute lymphoblastic leukemia (n = 20); (b) acute myelogenous leukemia (n = 27); (c) myelodysplastic syndrome or therapy-related myeloid malignancies (n = 17); (d) chronic myelogenous leukemia (n = 11); (e) other malignancies, including biphenotypic leukemia (n = 1) and malignant hemophagocytic lymphohistiocytosis (n = 1); and (f) nonmalignant diseases, including severe aplastic anemia (n = 2), Wiskott-Aldrich syndrome (n = 4), severe combined immunodeficiency syndrome (n = 1), Hurler syndrome (n = 1), and osteogenesis imperfecta (n = 4).
BMT Preparative Regimens
A human leukocyte antigen (HLA)-identical sibling served as the graft donor in 25 (28%) of the 89 patients; grafts were provided by alternative donors (matched unrelated volunteers or mismatched family members) for the other 64 patients (72%) and were T-lymphocyte depleted. As a preparative regimen, all but three patients received high-dose cytarabine (3 g/m2/dose, six doses), cyclophosphamide (45 mg/kg/dose, two doses) with 2-mercaptoethane sulphonate (45 mg/kg in five divided doses), and fractionated total body irradiation (TBI) to 12 Gy (for matched sibling marrow recipients) or 14 Gy (for alternative donor marrow recipients). Three patients received a non-TBI preparative regimen of busulfan and cyclophosphamide.
In addition to the drugs listed above, alternative donor marrow recipients also received antithymocyte globulin. For graft-versus-host disease (GVHD) prophylaxis, all patients received cyclosporine A beginning 2 days before transplantation at doses designed to maintain serum levels of 150250 ng/mL. Additionally, patients treated with grafts from matched sibling donors received short-course methotrexate (15 mg/m2, then three doses of 10 mg/m2).
Radiologic Findings
All CT examinations were performed with a Somatom Plus 4 (Siemens, Iselin, NJ) helical scanner with 1:1 pitch and a section thickness of 5 or 8 mm, depending on the patients age and size. CT examinations of the chest in this study were performed with intravenous contrast material enhancement and lung and mediastinal windows. Abdominal studies were performed with the administration of 23 mL (maximum dose, 150 mL) of the intravenous contrast agent iopamidol 61% (Isovue 300; Bracco Diagnostics, Princeton, NJ) per kilogram of body weight. When contrast-enhanced studies of the chest and abdomen were performed together, 3 mL/kg (maximum dose, 150 mL) of iopamidol was used to provide adequate contrast enhancement.
Before contrast material administration, each patients renal function was assessed with measurements of BUN and serum creatinine levels; these were compared with age-adjusted normal values. The contrast agent was withheld when the patients serum creatinine level exceeded 1.8 mg/dL (159 µmol/L).
For each patient, we recorded (S.C.K.) the timing of contrast-enhanced CT studies relative to the first (index) BMT procedure, the number of contrast-enhanced CT studies performed during the year before the index BMT, and the number of contrast-enhanced CT studies performed during the 100 days after transplantation, which is the conventional time frame used for acute complications following BMT.
Radiology reports were reviewed (S.C.K.) for the presence or absence of abnormal findings on CT scans. We recorded (S.C.K.) the following clinically relevant CT findings: pulmonary nodules, pulmonary parenchymal opacities, pleural and pericardial effusions, solid organ lesions, and ascites. Any CT examinations performed during the study period but without intravenous administration of a contrast agent were excluded from this study.
Laboratory Values
The glomerular filtration rate (GFR) was measured during the pretransplantation evaluation with a technetium 99m plasma clearance study or a 24-hour creatinine clearance study for a baseline indicator of renal function. Follow-up GFR measurements were not obtained during the study period of 100 days after transplantation, because these are routinely obtained at yearly follow-up evaluations. For each CT study, we (S.C.K.) recorded BUN and serum creatinine levels 24 hours before each study, at the time of the study but preceding intravenous administration of contrast material, and 24, 48, and 72 hours after each study (± 4 hours).
In the laboratory at our institution, normal age-adjusted serum creatinine levels are as follows: 09 years, less than 0.7 mg/dL (62 µmol/L); older than 9 to 13 years, less than 1.0 mg/dL (88 µmol/L); older than 13 to 19 years, less than 1.2 mg/dL (106 µmol/L); and older than 19 years, less than 1.5 mg/dL (133 µmol/L) for men and less than 1.2 mg/dL (106 µmol/L) for women.
Nephrotoxic Drug Regimens
Standard antineoplastic and immunosuppressant drugs used for BMT conditioning and GVHD prophylaxis were administered as described above. The study patients were all receiving additional nephrotoxic drugs for prophylaxis or treatment of infections. Thus, we (A.E.H.) calculated the number of days (beginning 10 days prior to the date of BMT) the patients received acyclovir, aminoglycosides, foscarnet, and/or vancomycin; drug doses were recorded for amphotericin B and amphotericin B lipid complex (ABLC) (Abelcet; The Liposome Co, Princeton, NJ), the lipid formulation of amphotericin B used at our institution.
Cumulative exposure to each of these drugs during the study period was paired with results of each of the posttransplantation CT examinations. Each patients cumulative exposure to antimicrobial agents at the time of the CT study was classified as minor or major: Exposure to aminoglycosides for more than 7 days was considered major, exposure to vancomycin and/or acyclovir for more than 14 days was also considered major, and exposure to foscarnet in any amount was considered major. Exposure to antifungal agents was recorded as a cumulative dose at the time of the CT study. In the case of a patients exposure to foscarnet at any cumulative dose, a score of 2 was assigned; for drug exposure classified as major, a score of 1 was assigned; and for drug exposure classified as minor, a score of 0 was assigned.
Statistical Analysis
The main objective of this analysis was to assess the effect of the nonionic contrast agent iopamidol on renal function as measured with BUN and serum creatinine levels and urinary output rate in the 24 hours preceding and 72 hours following the administration of iopamidol. We used the linear mixed-effects model for analyses in which the effect of the contrast agent on renal function was measured by the rates of change (ie, the slopes) of renal function indexes (eg, per-hour increase or decrease in serum creatinine levels) during the time ranging from 24 hours before to 72 hours after the infusion of contrast agent (810). Hence, time was the primary covariate for estimating the effect of contrast material infusion on serum creatinine and BUN levels and urinary output rate in their respective models.
The following plausible predictors of renal function characteristics were also included in the full model: the actual amount (in milligrams) and the duration (in days) of the administration of nephrotoxic drugs, patient age (in years) at the time of BMT, diagnosis before BMT, sex, nature of BMT preparative regimen (ie, with or without TBI), and donor type (ie, matched sibling or alternative).
To test for the possible interactions between drugs and the contrast agent, we used an alternative process of fitting the model in which each patient was assigned a drug score according to all drugs taken by the individual and the dosage or duration of administration for each of these drugs. Interaction measured the influence of the antimicrobial drugs on the effect of the contrast agent. From the resulting models, we estimated changes in renal function indexes attributable to the contrast agent and those attributable to the individual antimicrobial drugs. For the covariance structures of the mixed models for analysis, we used autoregressive structure for within-patient covariance and unstructured covariance for the random effects of intercept and time. All analyses performed were two-tailed tests; P values
.05 were considered to represent a statistically significant difference.
We also evaluated changes in serum creatinine values between the time of BMT and the first contrast-enhanced CT examination to determine the acute effect of BMT on renal function before contrast material administration.
| RESULTS |
|---|
|
|
|---|
|
|
|
Radiologic Findings
A total of 351 contrast-enhanced CT studies obtained in the patients were reviewed. Table 1 lists the number of CT examinations per patient and the frequency distributions for these numbers. The mean time from BMT to first CT examination was 20.8 days (median, 16 days; range, 483 days).
Nephrotoxicity
Results in all 89 patients who underwent contrast-enhanced CT were considered for linear model fitting. The drug scores of the patients ranged from 0 to 7. Patient sex, diagnosis before BMT, use of TBI, donor type, and patient age category at the time of BMT were not associated with statistically significant differences in the rate of change in creatinine or BUN concentration (P = .49). However, when included in the model as a continuous variable, age at the time of transplantation had a significant effect on creatinine concentration (P < .001) and BUN concentration (P = .002).
The increase in serum creatinine concentration can be predicted by the resulting linear regression model: creatinine = [0.18 + (0.002 x mg/kg of ABLC) - (0.002 x days of acyclovir therapy) + (0.003 x days of aminoglycoside therapy) + (0.008 x mg/kg of amphotericin B) + (0.66 x days of foscarnet therapy) + (0.006 x days of vancomycin therapy) + (0.58 x patient age in years)] + 0.0007 (time + 24), where time is the number of hours after the administration of iopamidol. This single equation determines the effect of contrast agent, the age of a patient at BMT, and all included drugs on average serum creatinine level. Estimated effects of the covariates are listed in Table 4. The values of the estimated parameters allow comparison of their effects on the average serum creatinine level.
|
Further, of the 89 patients evaluated, 76 (85%) received 2 mL/kg of intravenous contrast material and seven (8%) received 3 mL/kg of contrast material. Information regarding contrast agent dose was missing for six patients. The dose of contrast agent used did not have a significant effect (P = .22) on creatinine levels. However, because the number of children who received 3 mL/kg of contrast agent was small in our study, the statistical power was too low to enable detection of any meaningful difference between the two doses.
| DISCUSSION |
|---|
|
|
|---|
A profound and prolonged immunocompromised state is produced by the BMT preparative regimen, by the immunosuppressive medications used to facilitate engraftment and prevent or treat GVHD, and by other factors affecting the rate of immune reconstitution such as infections, graft source, and graft manipulation (1618). As a result, patients who undergo BMT are at substantial risk for life-threatening opportunistic infections with bacteria, viruses, and fungi. Invasive fungal infections are the leading cause of death due to infection after BMT (2).
These infections appear primarily in the lungs and solid organs of the abdomen and less frequently in the brain and paranasal sinuses (2,1922). The more common primary Candida species have been replaced in frequency in the past decade by Aspergillus species and fluconazole-resistant Candida species such as Candida krusei; reports of previously rarely encountered organisms such as Fusarium and Pseudallescheria species are also becoming more common (2,17,2325). Due to difficulties in early detection of these organisms and limitations of currently available treatments, mortality rates associated with these challenging organisms exceed 80% and increase when the organisms disseminate (24,26). Thus, although a culture-proved infectious cause of fever is often not found, an expeditious and exhaustive search for the etiology of fever in patients who undergo allogeneic BMT is imperative so that aggressive therapy can be initiated promptly.
More recently, clinicians have increasingly prescribed empiric and, in some cases, prophylactic antifungal therapy (18,27,28). An important disadvantage of these approaches is the associated toxicity caused by amphotericin B, which can lead to renal impairment and infusion reactions. Liposomal formulations may be better tolerated but are considerably more expensive and are also associated with toxic effects, including renal and hepatic damage (29).
Until the mid 1980s, intravenously administered contrast agents contributed substantially to renal impairment and were believed to be contraindicated in patients with renal insufficiency or failure (30). The nephrotoxic effects of contrast agents are believed to be due to their charge and osmolarity. Damage to the proximal tubule and changes in systemic and renal hemodynamics also play a role in the pathophysiology of the nephrotoxic effects of contrast agents (31).
Although newer contrast agents such as iopamidol are costly, they contain sufficient amounts of iodine to allow their use in imaging studies. Because they are nonionic and have lower osmolarity, they reportedly may be associated with a lower rate of toxicity (7,32,33); a decrease in nephrotoxicity has been observed in patients with underlying renal failure who have received nonionic contrast agents (34). A rise in serum creatinine concentration can be predicted by using patient-specific clinical indicators, which may be useful in the development of objective criteria for iopamidol use in this patient population.
Because CT examinations are used to both detect and rule out infections, CT results guide clinical decisions to initiate, continue, or stop potentially nephrotoxic antimicrobial therapy (35). For example, CT findings suggestive of infection indicate the need for antimicrobial therapy even in the presence of renal compromise; when CT findings indicate improvement or a return to normal, the administration of nephrotoxic antimicrobial agents may be limited or discontinued. Alternatively, if CT studies of the chest, abdomen, and sinuses demonstrate no evidence of sites of infectious disease, antimicrobial therapy in many instances may be limited or discontinued when the patients clinical condition improves. In both cases, the insult to renal function should be minimized. CT can also help identify noninfectious pulmonary or abdominal complications such as bronchiolitis obliterans organizing pneumonia (1) or diffuse alveolar hemorrhage.
Neutropenia typically persists for 2128 days after BMT and is almost always accompanied by fever. This period of extreme vulnerability to opportunistic infections before neutrophil engraftment explains why the median time to the first CT examination in this study was 21 days after BMT.
CT is not the only imaging method that is useful after BMT, but it is often superior to other methods used in the detection of invasive fungal infections. Although magnetic resonance (MR) imaging may provide excellent resolution, completing the MR imaging examination is logistically more challenging, particularly during the early posttransplantation period when patients may be critically ill. The added time to examination completion, the limited ability to assess the lungs (resulting in the fact that two MR imaging studiesone for evaluation of the lungs and one for evaluation of solid abdominal organsare often required), and the fact that patient monitoring is more difficult at MR imaging limit its usefulness in this patient cohort. The ready availability of ultrasonography (US), its lack of ionizing radiation, and its portability to the bedside make this a useful imaging modality for the detection of solid organ fungal disease. However, the sensitivity of US for depicting such lesions may be limited compared with the sensitivity of CT or MR imaging (36).
In our study population, increased age adversely affected serum creatinine and BUN levels. The reason for this finding is unclear. Several studies of renal insufficiency in children after BMT did not reveal older age to be a predisposing factor (3739). The median age of patients experiencing intravascular hemolysis and renal insufficiency in the study by Guinan et al (37) was 6 years. However, young patients with neuroblastoma and a history of chemotherapy with cisplatin were heavily represented in that study cohort in contrast to our cohort, in which none of the patients had a history of platinum exposure.
Although the results of our study validate our clinical practice of using iopamidol in patients who have undergone BMT and have altered renal function, our study was not without the limitations of a retrospective study. Our investigation was limited to a very focused pediatric patient cohort. CT technique was chosen by several pediatric radiologists and was adjusted to address patient-specific situations, thus introducing variability to scanning parameters.
The results of this study support the continuing use of intravenously administered iopamidol for contrast-enhanced CT studies in pediatric, adolescent, and young adult patients who have undergone BMT. We found the nephrotoxic effects of iopamidol to be negligible, even in patients with elevated renal function values; this agent can be used in contrast-enhanced CT examinations in pediatric, adolescent, and young adult patients who have undergone BMT.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, S.C.K.; study concepts, S.C.K.; study design, all authors; literature research, S.C.K., A.E.H., O.G.G.; clinical studies, S.C.K., A.E.H., L.C.B.; data acquisition, S.C.K., A.E.H.; data analysis/interpretation, all authors; statistical analysis, O.G.G., X.P.X.; manuscript preparation, all authors; manuscript definition of intellectual content, S.C.K., A.E.H., O.G.G., L.C.B.; manuscript editing, S.C.K., A.E.H., O.G.G.; manuscript revision/review and final version approval, all authors.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |