Diabetes mellitus was first described as a complication of kidney transplantation over 30 years ago (1). Since that time, both impaired glucose tolerance and overt diabetes mellitus have been recognized as common metabolic derangements associated with kidney transplantation. The reported incidence of posttransplant diabetes mellitus (PTDM) has ranged widely from 2% to 53%, reflecting wide variations in the definition of the disorder (2, 3, 4, 5, 6, 7, 8, 9, 10, 11). In recent clinical trials, the most commonly used definition of PTDM is the requirement for insulin therapy for an arbitrarily minimum period of time (e.g., 30 days). Such a definition underestimates the prevalence of PTDM because it excludes patients treated with oral hypoglycemic agents and those with impaired glucose tolerance. Examples of other definitions and the reported incidence of PTDM in various studies are shown in Table 20.1.

In order to better define the natural history of PTDM, some authors have favored adoption of guidelines for the diagnosis of diabetes mellitus and glucose intolerance recommended for the general population by the World Health Organization (WHO) and recently endorsed by the American Diabetes Association (ADA) (12) (see Table 20.2). Until such strict criteria are embraced by the community of transplant physicians, it must be recognized that lack of a standard definition obscures interpretation of the literature dealing with the incidence of PTDM and its effects on allograft and patient outcomes. Varying definitions also probably account for the fact that some (13,14), but not all (15) recent reviews indicate that the incidence of PTDM has increased during the past 3 decades. It seems likely that the current epidemic of obesity in the United States will ultimately result in a higher incidence of diabetes mellitus in patients with end-stage renal disease, including those who receive kidney transplants.

While the exact incidence of PTDM remains unclear, there is a general consensus that the prevalence of diabetes mellitus increases with time after transplantation. Using Medicare claims as the source of the diagnosis, Kasiske et al studied 11,659 Medicare beneficiaries who received first
kidney transplants between 1996 and 2000 and reported a cumulative incidence of PTDM of 9%, 16%, and 24% at 3, 12, and 36 months, respectively (14). When reviewing such data, it is important to keep in mind that a significant percentage of patients with PTDM might have developed diabetes mellitus even if they remained on dialysis. Thus, the incidence of new onset diabetes mellitus attributable to factors related to transplantation per se is the incremental difference between the baseline rate among wait-listed patients and the observed rate after transplantation. Using Medicare claims as the basis for a diagnosis, Woodward et al estimated the true incremental incidence of PTDM to be 8% to 10% during the first posttransplant year (16).

Patients are at greatest risk for developing PTDM during the first 6 months following kidney transplantation. As noted above, however, the prevalence of diabetes mellitus increases progressively thereafter. The time of developing diabetes mellitus following transplantation may depend, in part, on the patient’s immunosuppressive regimen. Posttransplant hyperglycemia and diabetes mellitus may be transient and even resolve without treatment, usually as a consequence of reduction in the doses of incriminating immunosuppressive drugs (see below). For most patients, the onset of hyperglycemia is insidious and asymptomatic. Presentation with diabetic ketoacidosis is rare. The asymptomatic nature of early PTDM makes the disorder difficult to diagnose without proper laboratory screening.

A number of reports indicate that the development of PTDM adversely effects the survival and function of renal allografts. Using PTDM as a time-dependent covariate in a Cox regression analysis, Kasiske et al. showed that new onset diabetes mellitus was associated with an increased risk of graft failure (relative risk 1.63, p <0.0001) and death-censored graft failure (relative risk 1.87, p <0.0001) (14). Similarly, Roth et al. reported that, compared to nondiabetic controls, kidney transplant recipients with PTDM exhibited a significant decrease in graft survival at 4 years (54% vs 82%, respectively; p <0.05) (17). Patients with PTDM have been shown to have significantly impaired renal function, based on serum creatinine concentrations, compared with nondiabetic controls at 5 years (18).

Although it has been shown that PTDM increases the risk of infection (19) and cardiovascular disease (20) after kidney transplantation, there is no consensus on whether PTDM adversely effects patient survival. Some (2,7,14,15, 21) but not all (17,19,22) studies have reported higher mortality rates in patients with this disorder. In a large single
center experience, Revanur et al showed that 10-year patient survival in patients with PTDM was significantly lower than that of nondiabetic patients and comparable to that of patients with preexisting type 1 diabetes mellitus (21) (see Fig. 20.1). Although its effect on patient mortality remains controversial, PTDM almost certainly increases the economic burden of kidney transplantation, not only because of the cost of therapy to control hyperglycemia, but also because of the recognized association between diabetes mellitus and cardiovascular complications. In a recent economic analysis of patients transplanted between 1994 and 1998, Woodward et al. estimated that Medicare paid an extra $21,500 per newly diabetic patient by 2 years posttransplant (16).