|
|
| |||||||||||||||||||||||||||||||
Background The lysosomal storage disease cystinosis results in renal failure at approximately 10 years of age. Although oral cysteamine therapy is recognized to preserve kidney function, the extent of renal benefit has not been determined.
Methods Between 1960 and 1992, we determined 24-hour creatinine clearances in 76 children with cystinosis during 1081 admissions to the National Institutes of Health. Seventeen children were considered to have received adequate treatment with cysteamine, since they had depletion of cystine from leukocytes and began therapy before the age of 2 years; treatment lasted a mean of 7.1 years. Thirty-two children were considered to have received partial treatment, since they had poor compliance with therapy or began treatment after the age of 2; treatment lasted a mean of 4.5 years. Twenty-seven children were followed in the era before cysteamine therapy and thus never received cysteamine.
Results Of the 27 children who never received cysteamine, 16 were followed at the National Institutes of Health until renal failure occurred; their mean (±SD) creatinine clearance was 8.0 ±4.8 ml per minute per 1.73 m2 of body-surface area at a mean age of 8.3 ±1.9 years. Of the 17 children who received adequate treatment, none had renal failure; their mean creatinine clearance was 57 ±20 ml per minute per 1.73 m2 at 8.3 ±3.8 years of age. The mean creatinine clearance of the children who received partial or adequate treatment with cysteamine increased with age up to the age of five years and then declined linearly with age at a normal rate. For the children who received adequate treatment, the mean creatinine clearance was predicted to reach 0 ml per minute per 1.73 m2 at the age of 74 years, as compared with 20 years of age for the children who received partial treatment. With no therapy, the mean creatinine clearance reaches 0 ml per minute per 1.73 m2 at 10 years of age.
Conclusions Children with cystinosis who are treated early and adequately with cysteamine have renal function that increases during the first five years of life and then declines at a normal rate. Patients with poorer compliance and those who are treated at an older age do less well.
Therapy for cystinosis involves supportive care including replacement of renal tubular losses. In 1976, cysteamine (
-mercaptoethylamine) was demonstrated to deplete cystine from patients' cells in vitro and in vivo16. By 1987, oral cysteamine had proved efficacious in preserving renal function and improving growth17. Since then, bioequivalence has been demonstrated between cysteamine and phosphocysteamine,18 a more palatable phosphothioester of cysteamine, and cysteamine eyedrops have successfully removed cystine crystals from the corneas of affected patients19,20.
Despite these advances, it is not known how well cysteamine can preserve renal function. In normal children, glomerular function increases during the first three to four years of life,21 from 40 to 120 ml per minute per 1.73 m2 of body-surface area. Can any of this renal growth be salvaged with early cysteamine therapy? Normally, the glomerular filtration rate declines by 0.6 to 0.8 ml per minute per 1.73 m2 per year in older children and adults22,23. Can oral cysteamine therapy limit renal decline to this minimal rate? These questions were not addressed in previous studies,17 since renal function was evaluated with the use of creatinine clearance values based on height and the reciprocal of the serum creatinine concentration24.
We describe 32 years' experience at a single institution involving 76 children with cystinosis in whom renal function was assessed repeatedly. The results underscore the importance of early cysteamine therapy and can be used to help predict whether this treatment obviates or merely delays the need for renal transplantation in cystinosis.
Methods
Patients
From 1960 through 1992, 24-hour creatinine clearances were determined at least twice in 76 children with nephropathic cystinosis who were studied at the National Institutes of Health. Children who were not studied twice were excluded. The diagnosis was based on the finding of cystine crystals in biopsy samples, corneal crystals on slit-lamp examination, or elevated leukocyte cystine concentrations4. Informed consent was obtained from each patient's parent or guardian on first admission to the National Institutes of Health Clinical Center; return visits were usually made every four to six months.
Cysteamine
Since 1978, cysteamine hydrochloride (50 mg of free base per milliliter; Sigma Chemical, St. Louis) has been administered under an Investigational New Drug exemption held by Dr. Jess G. Thoene (University of Michigan). Until 1987, cysteamine was given every six hours at a dose intended to reduce the leukocyte cystine levels to less than 1 nmol of half-cystine per milligram of protein when measured five hours after a dose. The actual doses given averaged 50 to 60 mg of cysteamine free base per kilogram of body weight per day. Since 1987, patients have received their choice of either cysteamine or phosphocysteamine (Medea Laboratories, Port Jefferson Station, N.Y.) at a randomly assigned dose of either 1.3 or 1.95 g per square meter of body-surface area per day. In a national study,25 the outcome with respect to renal function was similar in children receiving either dosage or formulation.
The children were divided into three groups solely on the basis of their history of cysteamine treatment. The untreated group contained 67 patients with 349 admissions. Twenty-seven of the 67 patients were followed in the era before cysteamine therapy (1960 through 1978); the other 40 patients were included in this group until they began therapy, at which time they were switched to one of the two cysteamine treatment groups.
Seventeen children were considered to have been adequately treated with cysteamine because they were compliant with therapy (median leukocyte cystine levels, 
2 nmol of half-cystine per milligram of protein) and began treatment before the age of two years. These children had a total of 323 hospital admissions and received therapy for a mean of 7.1 years. Thirty-two children were considered to have been partially treated because they were poorly compliant (median leukocyte cystine value, >2 nmol of half-cystine per milligram of protein) or began cysteamine treatment after the age of two. Among these 32 children, 3 were considered to have been poorly compliant, 9 began treatment late, and 20 met both criteria. There was a total of 545 hospital admissions in this group, and therapy lasted a mean of 4.5 years. Twenty-five of the children who received cysteamine at the National Institutes of Health were included in a previous, national study17.
Although the total number of patient admissions was 1081, 1217 admissions were included in the group analyses, since the early admissions (in which patients were not treated) of some treated patients were counted.
Measurements of Creatinine Clearance
Serum creatinine and 24-hour urinary creatinine clearance were measured several times during each admission. A single creatinine clearance value was determined for each admission by dividing the mean urinary creatinine concentration by the mean serum creatinine concentration multiplied by the urinary volume per minute and then normalizing the value to a body-surface area of 1.73 m2.
Quality Control
Creatinine determinations were performed at the National Institutes of Health Clinical Center with an automated method. In the past eight years, creatinine measurements26 have been repeated on 85 urine and serum samples. The average linear deviation in creatinine clearance between the two measurements has been found to be +0.8 percent, with a correlation coefficient of 0.99.
All urine samples were collected on the same clinical-research unit under close nursing supervision. In very young children, ostomy bags, custom-fitted to prevent leakage, were used. In older children, parental cooperation was enlisted and the children were awakened regularly to urinate. With the use of such measures, one to three complete 24-hour samples were obtained for over 95 percent of admissions. A urine sample was excluded only if the collection was not completed. Completeness of urine collections was verified by comparing the urine collected with the liquid intake (consistently, approximately 125 percent of urinary output) and weight fluctuations. A second check involved nine intensively supervised collections measured on separate eight-hour shifts by one of the investigators. The creatinine content of these samples varied less (from 9 to 6 percent), but never deviated from each child's historical trend line by more than 8 percent, suggesting that routine collections were complete. Finally, nursing notes on urinary volumes, determined by weighing the urine containers, confirmed the urinary volumes determined in the laboratory.
Statistical Analysis
Two-tailed Student's t-tests were used to analyze the children's clinical characteristics. To analyze creatinine clearances in the cysteamine-treated children, we assumed that each child's values represented points on a curve having two components that reflected the normal pattern of renal development21,22,23. One component increased with age in a log-log fashion up to the age of five years, and a second component gradually decreased with age in a linear fashion after the age of five. The intercepts and slopes of these two components were assumed to vary randomly among children. The method of Jones and Boadi-Boateng27 was used to estimate the average slope and intercept in each group. This method allowed the inclusion of data on individual children irrespective of the number of occasions on which a child was studied. The estimated mean slope and intercept, standard errors, and 95 percent confidence intervals were determined and plotted. The creatinine clearance declined relentlessly before and after the age of five years in the group that received no cysteamine (n = 67), so a single linear weighted average slope and intercept were calculated.
For the yearly averaged results, all values for each group were pooled by one-year intervals centered at the integer year value, and a simple standard error was calculated. Linear regression plots for children who received no treatment and for those who were adequately treated were determined along with the standard error of the slope. The variance of each slope was averaged, and the total variance was converted to a standard error of the slope. The degrees of freedom equaled the number of admissions used in the total average minus 2N, where N individual slopes were calculated and averaged. These averaged slopes were compared with a slope of zero and each other with a two-tailed Student t-test.
Results
Patients' Characteristics
The 17 children who had received adequate treatment were given a diagnosis of cystinosis earlier than either the 32 children who had received partial treatment or the 27 children who were never treated (P<0.01); they also started cysteamine therapy earlier and had excellent depletion of cystine from leukocytes (Table 1). The children who had impending renal failure when they left the National Institutes of Health included 59 percent of the children who had never received cysteamine (16 of 27) and 66 percent of the children who had received partial treatment with cysteamine (21 of 32). None of the 17 children who had received adequate treatment had impending renal failure, and their mean serum creatinine and creatinine clearance values were nearly normal at a mean age of 8.3 years (Table 1).
|
The 67 children with untreated cystinosis (Figure 1) had creatinine clearances that fell to 0 ml per minute per 1.73 m2 by the age of 10 years (Table 2); none had an increase in renal function when very young. The creatinine clearances in the 49 cysteamine-treated children increased with age in a log-log fashion during the first five years of life and then decreased linearly with age (Figure 1). In both these respects, the treated children resembled normal children21. The higher initial creatinine clearances in the untreated as compared with the treated groups, suggested by the statistical model (Figure 1), reflect a paucity of early points; inspection of the individual (Figure 1) and mean (Figure 2) values revealed no difference in early renal function. After 5 years of age, the results in the 32 partially treated patients (Figure 1) predicted a creatinine clearance of 0 ml per minute per 1.73 m2 at 20 years of age, whereas in the 17 adequately treated children renal function declined more slowly (Figure 1); in this group the creatinine clearance was predicted to reach 0 ml per minute per 1.73 m2 at 74 years of age (Table 2). The children who received adequate treatment also grew at a normal rate, with mean height paralleling the fifth percentile for normal children during the first 12 years of life (data not shown).
|
|
|
The difference between the cysteamine-treated children and the untreated children is underscored by examining the results in individual children. In all 11 untreated children whose creatinine clearance was determined at least three times during a period of at least two years, creatinine clearance decreased with age (Figure 3). The slopes were negative even during the first few years of life, ranging from -3.2 to -14.9 ml per minute per 1.73 m2 per year. In contrast, 14 of the 17 children in whom cysteamine therapy was initiated before the age of two years had an increase in creatinine clearance during the first six years of their lives, and 1 had no change (Figure 3).
|
Previous clinical trials of cysteamine therapy in children with cystinosis,17,25 designed to compare renal function in treated children and untreated children, have not revealed how closely cysteamine-treated children resemble normal children in terms of their renal development and deterioration. This study demonstrates that cysteamine therapy can preserve renal potential during both the rapid-growth stage of renal development and the subsequent phase of slow deterioration. In the 17 children with excellent rates of compliance, selected without a priori knowledge of their renal outcomes, the mean corrected creatinine clearance doubled during the first three years of life (Figure 2). During the phase of deterioration of renal function, the response was just as striking. The mean corrected creatinine clearance in these children paralleled that in normal children, declining at a rate of 0.8 ml per minute per 1.73 m2 per year. The normal progression occurred in the presence of normal growth, which imposed increased creatinine loads on the kidney, in contrast to the untreated children, in whom growth was poor.
These results allow one to predict how long renal function can be maintained with cysteamine therapy. Suppose a six-year-old child with cystinosis has a creatinine clearance of 60 ml per minute per 1.73 m2, which is average for a well-treated patient of this age. If this child loses glomerular function at the normal rate of 0.8 ml per minute per 1.73 m2 per year (Figure 1), the creatinine clearance will still exceed 50 ml per minute per 1.73 m2 at the age of 18 years, and a renal allograft should never be needed. The results also demonstrate how early initiation of cysteamine therapy may pay dividends later in life. Starting cysteamine therapy at the age of one year instead of at the age of two years allows the glomerular filtration rate to increase, on average, by an additional 3.2 ml per minute per 1.73 m2, on the basis of the mean slope for treated children. More important, it prevents the decrease in glomerular filtration rate that would have occurred during the second year of life without cysteamine therapy (Figure 1). This degree of renal function translates into an extra 12.5 years of renal preservation. However, these projections are based on little or no data after the age of 12.
Regardless of when cysteamine therapy is initiated, some reduction in creatinine clearance is present at the start of treatment (Figure 2). This renal impairment was substantial in the children who received partial treatment and who began cysteamine therapy after the age of two years. These children had less renal growth during the first three years of life than normal children or the children who received adequate treatment, and they lost renal function during childhood faster than the adequately treated children. This lack of renal reserve creates a dilemma for children whose short stature makes them candidates for therapy with growth hormone. Growth hormone therapy may not alter the uncorrected creatinine clearance, but it does increase surface area, muscle mass, and the creatinine load, so that the corrected creatinine clearance falls. Serum creatinine levels increase, and uremia and the need for renal transplantation become more likely. This sequence of events has been documented in three children with cystinosis who had marginal renal function and were treated with growth hormone28. The pubertal growth spurt may provide a similar impetus toward uremia in children who lack renal reserve.
We used urine-based creatinine clearance determinations to estimate the glomerular filtration rate. When the equation of Schwartz et al.24 -- 0.55 x height (in centimeters)/serum creatinine (in milligrams per deciliter) -- has been used in such studies, apparent improvements in creatinine clearance could be attributed to increased growth. In contrast, growth would serve to reduce the corrected creatinine clearance values based on 24-hour urine collections, since the surface-area term (i.e., height) is in the denominator of this equation (urinary volume [in milliliters per minute] x urinary creatinine concentration [in milligrams per deciliter] x 1.73 [in square meters]/serum creatinine concentration [in milligrams per deciliter] x body-surface area [in square meters]). The accuracy and completeness of the 24-hour urine collections were supported by two findings. First, the untreated children had a monotonic decline in their creatinine clearances that predicted that renal failure would occur at 10 years of age (Figure 1), a timing consistent with published values of 9.2 and 10.1 years derived from 205 European7 and 141 American29 children with cystinosis, respectively. Second, creatinine clearance measurements of treated children 5 to 18 years of age followed the course expected for normal children21.
For many investigational and clinical purposes exogenous measures of the glomerular filtration rate, such as iothalamate or inulin clearances, are preferable to creatinine clearance measurements. These other methods are more invasive, more expensive, and less available. Furthermore, the systematic errors in using creatinine clearance to estimate the glomerular filtration rate are largely a result of tubular secretion of creatinine, which is greatly impaired in cystinosis30. Hence, we continued to use the creatinine clearance measurements begun 30 years ago in order to compare all patients with a single technique.
The possibility that the untreated children had more severe cystinosis than the cysteamine-treated children is unlikely. Our analysis included every patient who had had at least two complete urine collections at the National Institutes of Health, and only the date of admission (before or after 1978) distinguished the groups a priori. If anything, the older age of the untreated patients at diagnosis (Table 1) would suggest the presence of less severe disease in this group.
We conclude that optimal cysteamine therapy of nephropathic cystinosis beginning before the age of two years permits growth of glomerular function and that continued therapy can limit renal deterioration. Consequently, some children with cystinosis may never require a renal allograft. Since irreversible renal damage before diagnosis limits the ultimate benefit of cysteamine therapy, early diagnosis and treatment are critical in this disease.
We are indebted to the nurses, fellows, and attending physicians (in particular, Dr. Jay Seegmiller, Dr. Jerry A. Schneider, and Dr. Joseph D. Schulman) for their diligence and care of patients with cystinosis on the pediatric metabolism ward (9W) of the National Institutes of Health Clinical Center over the past 30 years, and to head nurses Ms. Kaye Van derVen and Ms. Ann McNemar for their work.
Source Information
From the Section on Human Biochemical Genetics, Human Genetics Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Md.
Address reprint requests to Dr. Gahl at Bldg. 10, Rm. 9S-242, NICHD, NIH, Bethesda, MD 20892.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
