The number of pediatric renal transplants has risen significantly across the past decade with overall success rates that equal or exceed those of adults (20
). Our understanding of pediatric transplantation, however, is gleaned largely from research on adult patients whose clinical situation and experience may be quite different. This issue is particularly salient for the development of NODAT which has been shown to significantly increase the risk of graft failure and death among adult transplant patients (1
The current study is the largest to date to examine the development and consequences of NODAT among pediatric renal transplant recipients. We found that the overall rate of NODAT for these 2168 patients was 7.1% which is notably higher than the 2.6% rate previously observed in the North American Pediatric Renal Transplant Cooperative Study, the only other large-scale study conducted to date (11
). Smaller studies have yielded inconsistent results with rates of NODAT ranging from 2–20% (12
). Direct comparison of these studies is challenging due to notable differences in methodology, variables examined, sample size, patient age, and perhaps most importantly the criteria used to indicate NODAT.
Consistent with earlier research (11
), we found that NODAT is much less common in children (7.1%) than in their adult counterparts (as high as 25% by year 3) (1
), although these differences narrowed as the age of the child at transplant increased. For children younger than 13 years at transplant, only 4.4% developed NODAT compared to 10.1% of children ages 13–18 years. This increased risk may be a result of differences in body mass, immunosuppression/steroid dosing and primary cause of end-stage renal disease (ESRD).
Not surprisingly, we found that the rate of NODAT (7.1%) is considerably higher than the rate of diabetes among non-transplant children in the general population (0.18%) (24
). Given the reports of escalating rates of Type 2 diabetes in the general pediatric population (25
), we tested to see whether there have been similar increases in NODAT. Between 1995 and 2002 we found numerical increases in these rates annually, but they failed to reach statistical significance. This was at the same time, however, that there were concerted efforts to reduce the maintenance dose of immunosuppression and steroids in part to lower the rate of NODAT (8
Previous research of adult transplantation has paid considerable attention to identifying the risk factors of NODAT which have included older recipient age, black race, deceased donor type, hepatitis C antibody status, use of TAC (vs. CsA), acute rejection episodes, and recipient body weight (1
). Similar to these findings, we found that children are at increased risk of NODAT if they were of older age, obese (BMI ≥ 30) and had Cytomegalovirus D+/R−. As was mentioned earlier, comparison to prior adult research is challenging because of differences in criteria and methodology. To date, no two studies have examined the same set of potential risk factors. For example, the single center study by Greenspan et al. found that NODAT was associated with family history of diabetes and peri-transplant hyperglycemia, neither of which were available in the USRDS data we analyzed (17
). The retrospective analysis of the North American Pediatric Renal Transplant Cooperative Study identified race, steroid dose and use of TAC as risk factors for NODAT (11
). Other previous pediatric research (12
) had smaller sample sizes which limited statistical power to identify risk factors. Nonetheless, the consistency of risk factors suggests that the same physiological mechanisms may be at play in children and adults.
Although the impact of NODAT on graft and death is well recognized in the adult population (1
), we failed to find any such association among pediatric transplant patients. These null results, consistent with earlier research, have important implications for clinical decision-making after transplant. Although minimizing the risk of NODAT is certainly a goal of post-transplant care, the primary objective has always been to prevent graft failure and death. The results of this study suggest that even if NODAT does develop, it does not have the impact on these ultimate patient outcomes that exists for adult patients. The large sample (i.e., 430 cases of graft failure and 78 deaths among our 2168 pediatric transplant patients) in this study provides confidence these null results are not due to low statistical power as had been suggested for prior research. It is possible, however, that the lack of association between NODAT and graft failure and death may be due to the fact that pediatric patients have overall lower mortality rates, fewer co-morbid conditions and better overall health state than their adult counterparts. It is also possible that the 3-year follow-up period is inadequate for identifying the development of NODAT in children.
A number of studies have previously examined the accuracy of administrative data such as Medicare for analyses of this type, and have found strong concordance between the codes in these data and the information present in patient medical records (29
). Nonetheless, it is important to recognize that the data were collected for payment purposes, not for research. Consequently, there are certain limitations that should be taken into account when considering these results.
First, the absence of blood glucose levels and patient symptoms in the USRDS makes it impossible to use the gold standard definitions of diabetes established by the American Diabetes Association and the World Health Organization. The definition that was used in this study had been developed and shown accurate in several validation studies (32
), but this definition still does not provide the definitive diagnosis possible with laboratory results. Second, there are a number of important factors (e.g., use of insulin, regimen adherence, glycemic control, family history) that are unavailable in these data. Third, it is quite possible that some patients had undiagnosed diabetes prior to transplantation, although this risk is lower in a pediatric population than in an adult population due to the relatively low incidence of non-insulin requiring diabetes. Fourth, children with primary Medicare coverage are a unique group and their results may not reflect that of the full population of pediatric transplant recipients. Fifth, as an observational study, causality cannot be established. Despite these limitations, registry analyses, with large national samples, provide a unique mechanism for examining complicated relationships among treatments and outcomes which would be impractical or unethical to examine in randomized clinical trials.
The results of this study suggest an increasing incidence of NODAT for children and adolescents following kidney transplantation. Although the rate of NODAT is lower than for adults, it is not insignificant. The existence of NODAT complicates the care delivery for these patients, places additional physiological stress on the body and provides already-worried parents and children with yet another health issue to worry about. At the same time, however, our study failed to find any significant association between NODAT, graft failure and death. This information is critical in guiding clinical decisions that must balance risk of all these outcomes. Choice of immunosuppressant and BMI were the only modifiable risk factors reaching significance. It is important to note that the choice of immunosuppression regimen, however, is often a balancing act between multiple factors that may be associated with risk of graft failure (e.g., risk of acute rejection, patient’s prior success, side-effects and the protocol for a particular transplant center) (35
). The risk of NODAT may be overshadowed by other risks in making this choice. Additional research, particularly as additional years of data are added to the USRDS, is needed to more fully understand these relationships.