Hemopoietic stem cell transplantation (HSCT) has evolved as an accepted treatment modality for a diverse spectrum of diseases in children such as hematological malignancies, bone marrow failure syndromes, immunodeficiencies and inborn errors of metabolism. The 5-year survival rates depend on the disease for which HSCT is performed and vary from 90% for immunodeficiencies to 25% for high-risk hematological malignancies.1 As long-term survival has improved over the years, assessment of late complications becomes increasingly important. Chronic renal failure (CRF) following allogeneic HSCT is reported in children, although data on incidence and etiology in long-term survivors of HSCT are still scarce.2, 3 Understanding of the possible risk factors and the pathogenesis of acute and chronic renal failure after HSCT is required in order to reduce its incidence. We have previously shown a prevalence of 28% CRF in children 1-year post-HSCT in a retrospective study, and more recently a lower prevalence of only 11% in a prospective study.4, 5 In the current study, we investigated how renal function of these patients evolved after an extended follow-up period.
Patients and methods
General Outline of the studies
Study I: retrospective study 1991–1998
At the department of Pediatrics of Leiden University Medical Center, 207 children received an allogeneic HSCT from January 1991 to January 1998. Glomerular filtration rate (GFR) of all long-term survivors pre-HSCT, 1-year post-HSCT and at the most recent outpatient clinic visit was studied retrospectively from medical records. Long-term survivors are defined as having survived at least 2 years after HSCT without recurrence of disease.
Study II: prospective study 1998–2000
All children who underwent allogeneic HSCT from January 1998 to January 2000 were asked to participate in the study. Glomerular function was studied before, 1 and 5 years post-HSCT; tubular function was studied 5 years post-HSCT. Renography was performed before and 1 year post-HSCT. The ethics committee of the hospital approved the study protocol.
In both studies possible risk factors for CRF after HSCT were evaluated, that is, age at HSCT, primary diagnosis, conditioning with total body irradiation (TBI) or thoraco-abdominal irradiation (TAI), donor type, veno-occlusive disease (VOD) of the liver and acute renal failure within 3 months after HSCT. The kidneys were not shielded during irradiation. VOD of the liver was diagnosed if two of three symptoms (jaundice, painful hepatomegaly and fluid retention) were present before day 30 after BMT and other possible causes of these manifestations had been excluded.6 During the first 30 days, patients received cyclosporin A intravenously (2 mg/kg/day) as graft-versus-host disease prophylaxis. Subsequently, cyclosporin was given orally for 3 months in study I and 3–6 months in study II. Surveillance of possible cyclosporin A toxicity was performed once a week on a trough level not exceeding 300 
g/l blood in study I and 200 g/l blood in study II. In patients who received a non-T-cell-depleted graft, methotrexate 10 mg/m2 on days 1, 3 and 6 was added.
Renal function
Glomerular function
The estimated GFR was used to study renal function using a modified Schwartz formula (40 
length (in cm)/serum creatinine (in mol/l)).7 CRF was defined as an estimated GFR for patients of 3 months, 6 months, 1 year and more than 2 years old of <30, <44, <70 and <85 ml/min/1.73 m2, respectively.8 Weekly measurement of serum creatinine was used to study renal function of the patients in the first 3 months after HSCT. Acute renal failure was defined as at least doubling of pre-HSCT serum creatinine concentration in the first 3 months post-HSCT. Serum creatinine concentration was measured using a photometric method on an automatic analyzer (Hitachi 747–100, Hitachi, Tokyo, Japan). Antihypertensive medication was recorded.
Proximal tubular function
Proximal tubular function was studied by means of tubular reabsorption of phosphate, plasma bicarbonate, urine 
2-microglobulin excretion and presence of glucosuria. Urine anion gap (normal value <-20 to -50 mmol/l for adequate distal H+ production) was used to differentiate between proximal and distal tubular acidosis. Serum and urine phosphate concentration was measured on an automatic analyzer (Hitachi 747–100). Plasma bicarbonate was measured with a colorimetric assay on a fully automated Hitachi 911, with a variation coefficient of less than 2%. Urine 2-microglobulin was analyzed by microparticle enzyme immunoassay on an Imx (Abbot, Illinois, USA). Urine glucose was tested semiquantitatively with a dipstick, a positive result corresponding to glucose >0.3 g/l.
Distal tubular function
Distal tubular function was evaluated by means of urine calcium/creatinine ratio and plasma bicarbonate. Urine calcium was measured on an automatic analyzer (Hitachi 747–100). For the difference between proximal and distal tubular acidosis see proximal tubular function.
Miscellaneous
Blood pressure was compared to values for healthy children of the same age and gender.9 Antihypertensive medication was started if blood pressure repeatedly exceeded 10 mmHg above the 95th percentile.
Urinary microalbumin was measured with an immunoturbidimetric assay on a fully automated Hitachi 911; the variation coefficient ranged from 1.5 to 3.1% at different levels. Urine analysis was performed by dipstick; if the dipstick was positive for hemoglobin or leukocytes, microscopy was performed.
Renography was performed with a GCA 7100 or GCA 7200 gamma camera (Toshiba, Nasu, Japan) equipped with a low energy general-purpose collimator. The patients were hydrated half an hour before the scintigraphic study. After application of 50 MBq technetium-99 m labeled Mertiatide (99mTc-MAG3®) images were obtained with patients in supine position with the lumbar region in the field of view of the gammacamera under the examination table. Digital information was acquired in a 64 64 matrix size during 120 s and one image per second. After these 2 min the registration was continued during 1800 s, taking 90 images of 20 s each. The renogram was generated with the aid of a region-of-interest around the kidneys as a time-activity curve.
Statistics
A multiple logistic regression model was used to predict the effect of risk factors for CRF. The statistical significance of the variables in the model was tested using the Wald test at a 5% significance level. For the associated odds ratios, 95% confidence intervals were computed. All computations were carried out in SPSS version 12.0.1.
Results
Study I: retrospective study 1991–1998
Of 207 children who underwent HSCT from 1991 to 1998, 81 (39%) patients died. Causes of death were recurrence of primary disease (n=41, 51%), and transplant-related mortality (n=40, 49%). None of the patients died of renal failure. Renal function of five (4%) patients could not be obtained, four patients returned to their home country abroad and one refused further follow-up. Patient characteristics, conditioning regimen and donor type of 121 evaluable survivors are given in Table 1. VOD of the liver developed in 12 of 121 (10%) patients.
Table 1 - Patient characteristics, diagnosis, conditioning regimen and donor type of long-term survivors of hemopoietic stem cell transplantation studied retrospectively (study I, 1991–1998) and prospectively (study II, 1998–2000).
Full table Renal function of long-term survivors of HSCT is depicted in Table 2 and Figure 1. Of 121 long-term survivors, 29 (24%) children had CRF at a mean of 7.6 (2.4 s.d.) years post-HSCT. However, only five (4%) patients had a GFR <70 ml/min/1.73 m2 (36, 40, 60, 67 and 69 ml/min/1.73 m2, respectively). The two children with CRF pre-HSCT (GFR 84 and 68 ml/min/1.73 m2) also had low GFR at last follow-up (36 and 67 ml/min/1.73 m2, respectively). CRF persisted at the last outpatient clinic visit in 13 out of 24 patients with CRF 1-year post-HSCT. Median (range) GFR pre- 1-year post-, and at last outpatient clinic visit was 118 (68–194), 97 (28–156) and 97 (36–166) ml/min/1.73 m2, respectively. GFR 1-year post-HSCT was significantly lower compared to pre-HSCT (P<0.0001). None of the patients received antihypertensive medication or dialysis.
Figure 1.
Study I, 1991–1998. Glomerular filtration rate of children before (n=121), 1 year (n=107) and at last outpatient clinic visit (n=121) after hemopoietic stem cell transplantation. Median (range) is given pre- and 1-year post-HSCT; regression line (r=-0.13
0.36, NS); GFR: glomerular filtration rate; HSCT: hemopoietic stem cell transplantation.
Table 2 - Renal failure in long-term survivors of hemopoietic stem cell transplantation (HSCT). chronic renal failure (CRF), acute renal failure (ARF).
Full table Pre-HSCT serum creatinine was the only independent predictor of CRF (odds ratio 1.064 (95% C.I. 1.017–1.113), P=0.007), indicating an odds ratio of 1.86 for every 10 mol/l increase in pre-HSCT serum creatinine. None of the other risk factors studied, including TBI/TAI (P=0.10) and acute renal failure within 3 months post-HSCT (P=0.08), was associated with long-term CRF.
Study II: prospective study 1998–2000
Of the 66 patients, 23 (35%) who received an HSCT between 1998 and 2000 died. Causes of death were recurrence of primary disease (n=13, 57%) and transplant-related mortality (n=10, 43%). None of the patients died of acute or CRF.
Data on two (5%) patients' renal function could not be obtained; one moved abroad and one child was treated in another hospital. Patient characteristics, conditioning regimen and donor type of 41 evaluable long-term survivors of HSCT are given in Table 1. VOD of the liver developed in five (12%) patients. Two (5%) patients had chronic graft-versus-host disease.
Renal function of long-term survivors after HSCT is depicted in Table 2 and Figure 2. None of the patients had CRF pre-HSCT. Of 41 long-term survivors four (10%) children had CRF 5.0 (0.7 s.d.) years post-HSCT. Of these patients only two (5%) had a GFR <70 ml/min/1.73 m2 (62 and 50 ml/min/1.73 m2, respectively). Of the five patients with CRF 1-year post-HSCT, two continued to have CRF after 5 years. Median (range) GFR pre-, 1-year post- and 5.0 (0.7 s.d.) years post-HSCT was 116 (64–161), 98 (32–165) and 104 (50–212) ml/min/1.73 m2, respectively. GFR 1-year post-HSCT was significantly lower as compared to GFR pre-HSCT (P<0.01). None of the patients had hemolytic uremic syndrome or glomerulonephritis. Although five out of 30 (17%) patients had microalbuminuria (20–200 mg/l), none had albuminuria (>200 mg/l). One child had microscopic hematuria, one leucocyturia and one had both microscopic hematuria and leucocyturia. At 5 years after HSCT none of the children received cyclosporine A or antihypertensive medication. All children had systolic and diastolic blood pressure within normal values for age and gender at long-term (5 years) follow-up.
Figure 2.
Study II, 1998–2000. Glomerular filtration rate (GFR) of children before (n=41), 1 year (n=39) and 5 years (n=41) after hemopoietic stem cell transplantation (HSCT). Bar signifies median glomerular filtration rate.
Full figure and legend (12K)Two of 25 children had asymmetrical renal function on renography pre-HSCT (38 : 62 and 44 : 56%, respectively). A repeat renography of the former was not performed; the latter had normal renography 1-year post-HSCT. Only one of 16 performed renographies 1-year post-HSCT was abnormal (60 : 40%). This patient had normal renography pre-HSCT but developed severe hemorrhagic cystitis post-HSCT with unilateral hydroureteronephrosis due to obstruction of the ureter. A renography of the other patients (n=25) could not be obtained post-HSCT.
None of the possible risk factors (age at HSCT, pre-HSCT GFR, primary diagnosis, conditioning with TBI/TAI, VOD and acute renal failure within 3 months after HSCT) were significant predictors of CRF post-HSCT; however, the number of patients studied was relatively small.
Table 3 depicts proximal and distal tubular function 5 years after HSCT. None of the patients had clinical consequences or required medication for tubular dysfunction. Low plasma bicarbonate can signify either proximal or distal tubular dysfunction; in two of three patients with low plasma bicarbonate, urine anion gap was inappropriate, suggestive of distal tubular acidosis. Plasma pH was not measured; therefore, this conclusion is not decisive. Urine of the third patient could not be obtained. Only three children received ifosfamide as treatment of their hematological malignancy. None of these children had low tubular phosphate reabsorption. Furthermore, the two children with low tubular reabsorption of phosphate had high normal values for plasma phosphate (Table 3). Of four patients with CRF 5-years post-HSCT two had mild proximal tubular dysfunction (No 1: 2-microglobulin 2173 g/l; No 2: tubular phosphate reabsorption 81% and 2-microglobulin 352 g/l), one child had mild distal tubular dysfunction (plasma bicarbonate 20 mmol/l) and tubular function of one patient was not evaluated.
Table 3 - Tubular function 5.0 (0.7 s.d.) years after hemopoietic stem cell transplantation (HSCT).
Full table Discussion
Glomerular function
Studies on long-term renal function post-HSCT in children are scarce and concern small numbers of patients. In the eighties Berg and Bolme2 studied 28 children before and 5 years post-HSCT using inulin clearance. In this study 8 of 28 (29%) patients had a GFR <85 ml/min/1.73 m2.2 In contrast, Kumar et al3 found normal glomerular function in 17 long-term survivors 796.6 months after allogeneic HSCT performed between 1983 and 1992, and in only two (11%) children hypertension, glucosuria and/or hematuria was found. Interestingly, the prevalence of CRF in long-term survivors of HSCT is lower (10%) in our more recent cohort (study II, 1998–2000) compared to the older cohort (24%) (study I, 1991–1998). We can only speculate that the decrease in time of CRF post-HSCT may be caused by improved patient care combined with less frequent use of nephrotoxic medication (ie i.v. amphotericin, aminoglycosides and target levels of cyclosporine A). Furthermore, numbers in the more recent study are smaller (n=41 vs n=121) and the mean follow-up time is shorter 5.0 (s.d. 0.7) vs 7.6 (s.d. 2.4) years compared to the older cohort.
It appears that renal function stabilizes after 1-year post-HSCT (Table 2, Figures 1 and 2). This is in agreement with the results of Berg and Bolme2 who also found no further deterioration of GFR after the first year post-HSCT. Recently, Patzer et al10 found a GFR <90 ml/min/1.73 m2 in one patient before and in only two of 36 (6%) patients 2 years after HSCT performed between 1992 and 1998. Then again, half of the patients in this study, comprising children as well as adults, were treated with autologous HSCT, which may be less toxic compared to allogeneic HSCT. Likewise, Frisk et al11 found seven out of 40 (17%) children with GFR <70 ml/min/1.73 m2 at a median of 10 years postautologous HSCT.
Occasionally children reach end stage renal disease after HSCT. Recently, successful living related kidney transplantation was reported in three children post-HSCT.12
Summarizing, the prevalence of 24% and in a more recent cohort of 10% CRF post-HSCT in our study compares well to similar studies performed in children (0–29%).2, 3, 10 However, only 4–5% had GFR <70 ml/min/1.73 m2. Consistently, 5–20% of adults surviving long term after HSCT will develop CRF.13, 14
CRF post-HSCT in children as well as in adults has predominantly been associated with TBI.11, 13, 15, 16, 17, 18, 19, 20, 21, 22, 23 Nevertheless, in accordance with our previous studies 1-year post-HSCT, in the present study we did not find a relation between TBI/TAI and development of CRF.4, 5 The irradiation dose in our study was slightly different (4–8 Gy in one fraction or 12 Gy in two fractions, Table 1) compared to other pediatric studies: 10 Gy in one fraction,2, 19 12–14 Gy in 6–8 fractions15 and 13 Gy in eight fractions.22 This might influence the incidence of CRF after HSCT. Similar to our study, Patzer et al10 did not find a correlation between CRF and high-dose fractionated TBI (12 Gy in 6 fractions) in a 2-year follow-up study of 36 patients after HSCT.10
Other possible factors influencing the development of CRF are the occurrence of hemolytic uremic syndrome and polymorphism of the gene for angiotensin-converting enzyme.14, 24 None of our patients developed hemolytic uremic syndrome, and we are not informed of polymorphisms for angiotensin-converting enzyme gene in our patients.
We found that high serum creatinine pre-HSCT was the only predictor of CRF in long-term survivors, indicating that pre-existent kidney damage possibly due to previous nephrotoxic therapy for hematological malignancy is a risk factor for long-term CRF. We could not confirm our earlier finding that acute renal failure post-HSCT predicted CRF 1-year post-HSCT; however, a trend (P=0.08) in long-term survivors can be discerned.
No patient required long-term antihypertensive medication in either of our studies (study I 1991–1998 and study II 1998–2000). Kumar et al3 noted hypertension in two of 17 (12%) children after a mean follow-up period of 796.6 months. Frisk et al11 found hypertension in two out of 40 (5%) children with CRF at a median of 10 years after autologous HSCT.11 The other long-term studies on renal function post-HSCT in children did not include evaluation of hypertension.2, 10
Tubular function
In all, 3–12% of children had proximal tubular dysfunction (low tubular reabsorption of phosphate, and/or low plasma bicarbonate and/or 2-microalbuminuria) 5 years after HSCT. Likewise, 9–13% of patients had mild distal tubular dysfunction (high calcium : creatinine ratio and/or low plasma bicarbonate, Table 3). However, none of the proximal and distal tubular abnormalities had clinical consequences. There is only one study in children evaluating tubular function before and 1 and 2 years post-HSCT.10 Tubular dysfunction was found in 14–45% of children 1–2 years after allogeneic or autologous HSCT, depending on the parameter investigated. Patzer et al10 found more patients with impaired proximal tubular function as compared to our study. Phosphate reabsorption was below the normal range in 15 of 33 (45%) patients 2 years post-HSCT. In contrast, our study 5 years after HSCT revealed only two of 29 (7%) children with low phosphate reabsorption; in addition these children had plasma phosphate values in the high normal range. More children received ifosfamide in Patzers' study, 23 of 44 (52%) vs three of 41 (7%) in our group, and the shorter follow-up time of 2 years vs 5 years in our study, with the possibility of repair of proximal tubules may both have contributed to this discrepancy. Besides, Patzer et al10 found elevated 
2-microglobulin excretion in 13 of 33 (39%) patients, whereas in our patient group only four of 35 (11%) had excessive 2 microglobuline excretion, both signifying proximal tubular dysfunction. Distal tubular function studied by means of urinary calcium excretion seemed to occur less in Patzers' compared to our study. He found hypercalciuria in only three of 40 (8%) patients before and one of 33 (3%) patients after HSCT. This is in contrast to our study where four of 32 (13%) children had hypercalciuria post-HSCT. Distal tubular function assessed by concentrating capacity by Frisk et al11 after autologous HSCT remained unaffected through a 5-year follow-up period.11 We did not measure the concentrating capacity of the kidneys. The authors do not know of similar studies concerning tubular function post-HSCT in adults.
In conclusion, there appears to be a decline in time of the prevalence of CRF in long-term survivors post-HSCT. This may be caused by improved patient care in general combined with less frequent use of nephrotoxic medication. The sole predictor of CRF in our study was high serum creatinine pre-HSCT, while acute renal failure in the first 3 months after HSCT only showed a trend towards predicting CRF. We cannot confirm TBI/TAI as a risk factor for CRF post-HSCT, as was shown in several pediatric and adult studies. Proximal and distal tubular dysfunction occurred only in a minority of long-time survivors of HSCT and had no clinical consequences. Further studies are necessary to elucidate pathophysiological mechanisms and prevent CRF in long-term survivors of HSCT.
References
| 1. | Vossen JMJJ. Bone marrow transplantation in children. Baillière's Clin Paediatr 1994; 1: 1029−1059. |
| 2. | Berg U & Bolme P. Renal function in children following bone marrow transplantation. Transplant Proc 1989; 21 (Part 3): 3092−3094. |
| 3. | Kumar M, Kedar A & Neiberger RE. Kidney function in long-term pediatric survivors of acute lymphoblastic leukemia following allogeneic bone marrow transplantation. Pediatr Hematol Oncol 1996; 13: 375−379. |
| 4. | Kist-van Holthe JE, van Zwet JM & Brand R et al.. Bone marrow transplantation in children: consequences for renal function shortly after and 1 year post-BMT. Bone Marrow Transplant 1998; 22: 559−564. |
| 5. | Kist-Van Holthe JE, Goedvolk CA & Brand R et al.. Prospective study of renal insufficiency after bone marrow transplantation. Pediatr Nephrol 2002; 17: 1032−1037. |
| 6. | McDonald GB, Hinds MS & Fisher LD et al.. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients. Ann Intern Med 1993; 118: 255−267. |
| 7. | Counahan R, Chantler C & Ghazali S et al.. Estimation of glomerular filtration rate from plasma creatinine concentration in children. Arch Dis Child 1976; 51: 875−878. |
| 8. | Peters AM & Gordon I. Quantitative assessment of the urinary tract with radionucleotides. In: Barrett TM, Avner ED, Harmon WE (eds.).Pediatric Nephrology 4th edn. Williams & Wilkins: Baltimore, MA 1999; pp 365−375. |
| 9. | Update on the 1987 Task Force Report on High Blood Pressure in Children and Adolescents: a working group report from the National High Blood Pressure Education Program. National High Blood Pressure Education Program Working Group on Hypertension Control in Children and Adolescents. Pediatrics 1996; 98 (Part 1): 649−658. |
| 10. | Patzer L, Kentouche K, Ringelmann F & Misselwitz J. Renal function following hematological stem cell transplantation in childhood. Pediatr Nephrol 2003; 18: 623−635. |
| 11. | Frisk P, Bratteby LE, Carlson K & Lonnerholm G. Renal function after autologous bone marrow transplantation in children: a long-term prospective study. Bone Marrow Transplant 2002; 29: 129−136. |
| 12. | Thomas SE, Hutchinson RJ, DebRoy M & Magee JC. Successful renal transplantation following prior bone marrow transplantation in pediatric patients. Pediatr Transplant 2004; 8: 507−512. |
| 13. | Cohen EP. Radiation nephropathy after bone marrow transplantation. Kidney Int 2000; 58: 903−918. |
| 14. | Cohen EP. Renal failure after bone-marrow transplantation. Lancet 2001; 357: 6−7. |
| 15. | 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. |
| 16. | Miralbell R, Bieri S & Mermillod B et al.. Renal toxicity after allogeneic bone marrow transplantation: the combined effects of total-body irradiation and graft-versus-host disease. J Clin Oncol 1996; 14: 579−585. |
| 17. | Guinan EC, Tarbell NJ & Niemeyer CM et al.. Intravascular hemolysis and renal insufficiency after bone marrow transplantation. Blood 1988; 72: 451−455. |
| 18. | Bergstein J, Andreoli SP, Provisor AJ & Yum M. Radiation nephritis following total-body irradiation and cyclophosphamide in preparation for bone marrow transplantation. Transplantation 1986; 41: 63−66. |
| 19. | Antignac C, Gubler MC & Leverger G et al.. Delayed renal failure with extensive mesangiolysis following bone marrow transplantation. Kidney Int 1989; 35: 1336−1344. |
| 20. | Leblond V, Sutton L & Jacquiaud C et al.. Evaluation of renal function in 60 long-term survivors of bone marrow transplantation. J Am Soc Nephrol 1995; 6: 1661−1665. |
| 21. | Lawton CA, Cohen EP & Barber-Derus SW et al.. Late renal dysfunction in adult survivors of bone marrow transplantation. Cancer 1991; 67: 2795−2800. |
| 22. | Van Why SK, Friedman AL, Wei LJ & Hong R. Renal insufficiency after bone marrow transplantation in children. Bone Marrow Transplant 1991; 7: 383−388. |
| 23. | Lawton CA, Cohen EP & Murray KJ et al.. Long-term results of selective renal shielding in patients undergoing total body irradiation in preparation for bone marrow transplantation. Bone Marrow Transplant 1997; 20: 1069−1074. |
| 24. | Moulder JE, Fish BL & Cohen EP. ACE inhibitors and AII receptor antagonists in the treatment and prevention of bone marrow transplant nephropathy. Curr Pharm Des 2003; 9: 737−749. |
MORE ARTICLES LIKE THIS
These links to content published by NPG are automatically generated
REVIEWS
Bone Marrow Transplantation Review
Bone Marrow Transplantation Review
RESEARCH
A prospective study of respiratory ciliary structure and function after stem cell transplantation
Bone Marrow Transplantation Original Article
Bone Marrow Transplantation Original Article
Long-term renal function following bone marrow transplantation
Bone Marrow Transplantation Original Article
