New-onset diabetes after transplantation: 2003 International consensus guidelines. Proceedings of an international expert panel meeting. Barcelona, Spain, 19 February 2003.

Abstract

INTRODUCTION Diabetes Mellitus Diabetes mellitus is one of the most common chronic diseases in the world, with an estimated worldwide prevalence of over 176 million in 2000. Furthermore, the prevalence of the disease is expected to increase in the coming years as industrialized societies become older, more obese, and more sedentary, and it is predicted to rise to 370 million by 2030 (1). Acute metabolic complications of diabetes are associated with high mortality rates; in particular, hyperosmolar hyperglycemia (approximately 15%) and ketoacidosis (up to 5%). The prognosis in these conditions is substantially worse in older patients and those with coma or hypotension (2). Following a review of epidemiologic and pathologic evidence, the American Heart Association has also classified diabetes a major independent risk factor for cardiovascular disease (CVD) (3–8). Manifestations of CVD include atherosclerotic coronary heart disease (CHD), heart failure (diabetic cardiomyopathy), myocardial infarction, stroke, and peripheral vascular disease (7). The risk for stroke has been reported to be two to four times higher in patients with diabetes (7,9). Furthermore, patients with diabetes who develop CVD have a worse prognosis for survival than patients with CVD but without diabetes (8–11). In fact, CVD is listed as the cause of death in approximately 65% of patients with diabetes (12). In addition to the risk for macrovascular complications, diabetes is associated with long-term microvascular complications including neuropathy, retinopathy, nephropathy, and erectile dysfunction. Diabetic nephropathy is the most common single cause of end-stage renal disease in the United States and Europe (13). With respect to retinopathy, after 20 years with the condition, almost all individuals with type 1 diabetes and over 60% with type 2 diabetes have a degree of retinopathy that may progress to loss of vision (14). In fact, diabetes is now considered the leading cause of blindness in developed countries, causing 12,000 to 24,000 new cases each year (15,16). New-Onset Diabetes after Transplantation Diabetes and impaired glucose tolerance occurring as a complication of organ transplantation have been recognized for many years. However, incidence figures for new-onset diabetes after transplantation across different studies have ranged between 2% and 53% (17). Precise figures have been difficult to determine because there has been no consensus regarding the definition of the condition and hence different clinical studies have used a variety of diagnostic criteria. Despite the apparently high incidence of new-onset diabetes after transplantation, transplant patients are not always routinely screened for hyperglycemia posttransplant and the condition is often underestimated (18). Nevertheless, it is clear that development of diabetes after transplantation has serious consequences for the patient and threatens the outcome of transplantation; studies suggest that diabetes developing after transplantation is associated with reduced graft function and patient survival, and increased graft loss (19–21). In addition to other risk factors, studies suggest that immunosuppressive regimens may account for a large degree of the increased risk for development of diabetes after transplantation (22). However, currently used immunosuppressive therapies vary in the extent to which they induce diabetes and thus the choice of immunosuppressive therapy can have a major influence on patients’ risk for developing the condition. In summary, diabetes is a serious complication that can adversely affect the survival of the transplant recipient, long-term survival of the graft, and the patient’s quality of life. It is suggested that appropriate screening and management of patients pre- and posttransplantation can minimize the risk for developing the condition. Furthermore, early detection and appropriate treatment of patients who have developed diabetes can ameliorate the long-term consequences of the condition. This publication represents the proceedings of the International Expert Panel Meeting: the first part provides the evidence on which the following management guidelines were based, and the second part outlines the consolidated New-Onset Diabetes after Transplantation International Consensus Guidelines. It is hoped that use of these guidelines may assist in avoiding or reducing the incidence and impact of new-onset diabetes mellitus after transplantation. PART I: LITERATURE REVIEW 1. The Impact of New-Onset Diabetes after Transplantation Diabetes is one of the most serious long-term complications of transplantation, particularly if the condition is poorly controlled. Data from clinical trials suggest that transplant recipients who develop diabetes are at greater risk of graft-related complications, including graft rejection, graft loss, and infection (23). Furthermore, the chronic hyperglycemia associated with the condition carries a long-term risk of both microvascular and macrovascular complications. The development of new-onset diabetes also incurs substantial additional health care costs. A recent analysis of Medicare payments in the United States between 1994 and 1998 has revealed that the costs of developing new-onset diabetes after kidney transplantation are between $12,000 and $13,000 higher than for those with no diabetes by the end of the first year posttransplant, and between $19,000 and $22,000 higher by the end of the second year (24). 1.1. Incidence and prevalence of new-onset diabetes after transplantation The prevalence of new-onset diabetes after transplantation has been greatly underestimated in the literature because of the lack of a standard definition for the condition. Most definitions for diabetes after transplantation are derived from random glucose testing or fasting glucose levels greater than 140 mg/dL, and clinical trials do not routinely include oral glucose tolerance tests (OGTT) to determine the exact incidence of glycemic abnormalities in transplant recipients. This heterogeneity has led to wide variations in the reported incidence of the disease (17,25). The observation periods of many studies are also too short (some are <1 year) and underestimate the true incidence of the condition; risk of diabetes increases progressively posttransplant and patients may develop the condition many years posttransplant (26). The incidence of new-onset diabetes after transplantation in adults was systematically reviewed by Montori et al. (17); 12-month cumulative incidence estimates of diabetes after transplantation for heart, liver, and kidney transplantation studies were reported to be within the range of 2% to 53% (17). In this analysis, the type of immunosuppressive regimen used was found to explain 74% of the variability in incidence, with high-dose steroids being associated with the highest incidences (17). More recently, in a retrospective analysis of Medicare beneficiaries in the United States, the cumulative incidence of new-onset diabetes after transplantation among 11,659 patients was 9.1%, 16%, and 24% at 3 months, 12 months, and 36 months, respectively (Fig. 1) (27). Figure 1: Percentage of patients surviving free of new-onset diabetes with a functioning graft after transplant declined with time posttransplant (solid line) (dashed line, 95% confidence interval). The number of patients surviving with a functioning graft and free from new-onset diabetes is shown above the x-axis. (From Kasiske BL, Snyder JJ, Gilbertson D, et al. Diabetes mellitus after transplantation in the United States. Am J Transplant 2003; 3: 178; with permission from Blackwell Publishing Ltd.)1.2. Natural history of new-onset diabetes after transplantation The natural history of diabetes after transplantation shares many similarities with type 2 diabetes in that the onset can be insidious; individuals may experience glucose intolerance and may be asymptomatic for years before symptoms clinically manifest (25,28). Furthermore, posttransplantation hyperglycemia and diabetes are not always permanent, and may normalize, sometimes without treatment, within weeks or months (29). However, abnormal OGTT may still be observed in some patients up to 26 months after remission from diabetes after transplantation (29). The potentially asymptomatic and/or transient nature of diabetes after transplantation can thus make the condition difficult to diagnose, underlining the importance of establishing a precise definition. The posttransplant development of diabetes involves two distinct phases: (1) patients are initially at greatest risk during the first 6 months posttransplant; and (2) the number of patients developing diabetes increases progressively over time thereafter (26,30). This has been illustrated by a study of 2,078 kidney allograft recipients in which 5.9% of patients developed diabetes in the first 6 months after transplantation, but then the percentage of cases increased linearly over time, leading to cumulative percentages at 1, 3, 5, 10, and 15 years of 7.1%, 10.4%, 13.2%, 20.5%, and 29.8%, respectively (Fig. 2) (26). Studies of only a few months’ duration are therefore likely to grossly underestimate the true incidence of the condition. Figure 2: Kaplan–Meier plot showing the proportion of patients developing diabetes after transplantation by time posttransplant. (From Cosio FG, Pesavento TE, Osei K, et al. Posttransplant diabetes mellitus: Increasing incidence in renal allograft recipients transplanted in recent years. Kidney Int 2001; 59: 732; with permission from Blackwell Publishing Ltd.)1.3. Graft survival In patients without diabetes, the 10-year survival of patients’ functioning grafts increased from 55% to 60% in the 1970s to 86% between 1988 and 1997. In addition to these improved survival rates, the posttransplant mortality rate has continued to decline (31). A large proportion of these deaths can be attributed to death with graft function, which also accounts for 43% of all kidney graft losses (31). Reports have consistently shown that the development of diabetes is associated with impaired long-term graft function and survival in kidney transplant recipients (32). Roth et al. (33) have reported that the development of diabetes after transplantation was associated with a significant decrease in graft survival at 3 and 4 years, compared with control recipients (71% vs. 86% and 54% vs. 82%, respectively;P <0.05) (33). Similarly, patients with new-onset diabetes after kidney transplantation have been shown to have significantly impaired kidney function, assessed by serum creatinine level, compared with controls at 5 years (2.9±2.6 vs. 2.0±0.07 mg/dL, respectively;P =0.05). Twelve-year graft survival was also significantly worse in patients with diabetes compared with controls (48% vs. 70%, respectively;P =0.04), with new-onset diabetes being an independent predictor of graft loss (relative risk, 3.72;P =0.04) (19). The development of new-onset diabetes after transplantation has also been shown to result in a greater incidence of acute rejection in liver transplant recipients (50% vs. 30% in the control group) (34). The cause of impaired graft survival and function in transplant recipients with new-onset diabetes after transplantation is unclear. The development of diabetes-related nephropathy is one possibility because it has been clearly established that diabetic nephropathy can adversely affect allogenic kidney transplants and is associated with a high rate of allograft failure (35). However, diabetic nephropathy can take several years to develop and may not account for the early graft failure that can be associated with the development of diabetes. In kidney transplant patients, an alternative rationale may be that the presence of poorly controlled hypertension, which is common in transplant recipients with diabetes, may impact on graft function and survival by accelerating glomerular injury (25). The use of lower dosages of immunosuppressive regimens may also account for increased graft failure (19). However, it is also possible that other unmeasured factors in these retrospective studies caused both the higher incidence of new-onset diabetes and graft failure. 1.4. Patient survival In addition to producing deleterious effects on graft function and survival, some studies (28,36,37) have reported that development of diabetes after transplantation reduces long-term survival of transplant recipients. Others have reported contradictory conclusions (19,33,38). In one study, 1-year survival posttransplant was reported to be 83% for patients who developed diabetes after kidney transplantation compared with 98% for patients without diabetes (37). The survival of kidney transplant recipients developing diabetes is also reported to be reduced at 2 years compared with those without diabetes (67% vs. 83%, respectively) (36). Longer term survival of kidney transplant recipients with diabetes may also be reduced compared with nondiabetic recipients. In a study of 978 kidney transplant recipients, 6.7% of patients developed diabetes after transplantation, and the development of diabetes in these patients was found to be associated with significantly shorter patient survival compared with control patients (Fig. 3) (28). A further study has demonstrated that survival of transplant recipients is adversely affected by both preexisting diabetes and new-onset diabetes after kidney transplantation and that the development of diabetes posttransplant in recipients below the age of 55 years is associated with a particularly high risk of death (relative risk, 2.54;P <0.001) (39). Finally, a strong argument in favor of a deleterious effect of diabetes on posttransplant outcome is the fact that long-term survival of patients with diabetes who undergo simultaneous pancreas-kidney transplantation is superior to those with cadaveric kidney transplantation (8-year survival rates of 72% for pancreas-kidney recipients vs. 55% for cadaver kidney recipients) (40). The development of new-onset diabetes after transplantation has also been shown to be an independent risk factor for mortality in liver transplant recipients (41). The reason why some studies have failed to demonstrate an effect of diabetes after transplantation on patient survival is unclear, but may be related to the small numbers of patients involved (38) or improvements in patient management (19). Figure 3: Kaplan–Meier curve showing mean patient survival in kidney transplant recipients with new-onset diabetes after transplantation compared with controls. (From Jindal RM, Hjelmesæth J. Impact and management of posttransplant diabetes mellitus. Transplantation 2000; 70: S58; with permission.)The increased incidence of infections, and associated increased risk for sepsis, in transplant recipients with diabetes may contribute toward the increased mortality. This hypothesis is supported by the findings of two studies conducted in kidney transplant recipients. Sumrani et al. (30) have reported that infections were a major complication in recipients with diabetes, with 54% experiencing infectious complications compared with 17% in the control group. Although no difference in mortality was found between the two groups, all five deaths in the group with diabetes were caused by sepsis compared with only one in the control group (30). Similarly, Miles et al. have reported that the frequency of sepsis as a cause of death was greater in kidney transplant recipients with diabetes compared with those with no diabetes (19). In a review of studies in liver transplant recipients, Benhamou and Penfornis reported that the incidence of acute rejection and the mortality rate within the first 2 years are significantly higher in patients with new-onset diabetes after transplantation (35). However, the precise association between the development of diabetes posttransplant and early mortality is difficult to define owing to the fact that rejection episodes are treated with corticosteroid boluses, which contribute substantially to the development of diabetes (35). 1.5. Diabetes as a risk factor for CVD CVD is the most common cause of death after renal transplantation in the United States (42). The incidence of myocardial infarction is between three and five times more common in transplant patients than in the general population (43). It is therefore not surprising that the overall incidence of cardiovascular mortality is also considerably higher in transplant recipients than in nontransplant patients (43). Furthermore, conventional risk factors for CVD (e.g., diabetes, hypertension, dyslipidemia) are also risk factors for chronic graft rejection (44). In addition to predisposing transplant recipients to CVD, diabetes after transplantation has a significant impact on the risk of death from cardiovascular complications. Death resulting from ischemic heart disease (IHD) has been shown to be 20.8 times higher in transplant recipients with diabetes than in the general population, compared with 6.4 times higher than the general population in transplant recipients without diabetes aged 55 to 64 years (Table 1) (45). This is not surprising because CVD is listed as the cause of death in approximately 65% of individuals with diabetes (12). Table 1: Table 1. Age-related IHD mortality after transplantation in a study of 1,347 kidney transplant recipientsIn one study conducted in kidney transplant recipients, diabetes was found to be the most important risk factor for developing both cerebrovascular disease (independent relative risk, 3.21) and peripheral vascular disease (independent relative risk, 28.18;P <0.05) (46). In a further study, diabetes mellitus carried the highest relative risk for IHD among kidney transplant patients more than 1 year posttransplant (Table 2) (47). This risk of IHD associated with diabetes was substantially higher for kidney transplant recipients than for the control population (Framingham Heart Study), particularly among women. Table 2: Table 2. Relative risk for IHD among transplant recipients more than 1 year after kidney transplantationIt therefore appears that individuals with diabetes in the general population have an increased risk for CVD mortality and that the risk of CVD is also substantially higher in transplant recipients who develop diabetes after transplantation compared with those who do not develop diabetes (45,47). The reason for the increased risk of CVD mortality and morbidity in transplant recipients developing diabetes after transplantation is not entirely clear, although both hyperinsulinemia and glucose intolerance are reported to be independent risk factors for atherosclerosis in the general population (48–50). Furthermore, fasting blood glucose values in the upper normal range (86–108 mg/dL; 4.8–6.0 mM) are associated with an increased risk for CVD mortality in nondiabetic, apparently healthy, middle-aged men (51). Transplant recipients tend to be insulin resistant because of impaired nonoxidative glucose disposal, and those who develop diabetes also have an impaired insulin secretion response (28). In addition, transplant recipients with diabetes also have an increased incidence of atherogenic dyslipidemia and hypertension. These factors are all thought to contribute to the higher risk for CVD mortality in this population (28,35). 1.6. Summary In summary, development of new-onset diabetes after transplantation is associated with impaired long-term graft function and survival, reduced long-term survival of transplant recipients and increased risk of mortality and morbidity associated with CVD (Fig. 4). Figure 4: Consequences of development of new-onset diabetes after transplantation.2. Diagnosis of New-Onset Diabetes after Transplantation 2.1. Diagnosis of new-onset diabetes after transplantation in clinical studies The major difficulty in estimating precisely the incidence for diabetes after transplantation has been the lack of consensus regarding the definition and diagnosis of the condition. Most definitions of the condition in the literature are derived from fasting or random glucose testing greater than 140 mg/dL (7.8 mM) or OGTT (25,28). In clinical trials, the most commonly used definition is the requirement of insulin for a minimum period (usually 30 days) posttransplantation. This definition has resulted in an underestimate of the prevalence of diabetes after transplantation because it excludes patients treated with oral antidiabetic agents, as well as those with asymptomatic hyperglycemia and impaired fasting glucose (IFG) or impaired glucose tolerance (IGT). In addition, this definition fails to address the long-term implications of diabetes after transplantation, as it does not identify patients with preexisting diabetes whose glycemic control worsens posttransplant. Finally, it does not distinguish between patients with new onset of disease from those with worsening of disease. It has been suggested that the American Diabetes Association (ADA) criteria described below should also be recommended for the diagnosis of diabetes after transplantation, and that adoption of the criteria described in Table 3 may be the first step toward standardizing the definition of the condition (28). In this population, it may be particularly important to consider both IGT and IFG because these are likely to be important predictive factors for the development of diabetes. Table 3: Table 3. WHO and ADA criteria for the diagnosis of diabetes mellitus2.2. Diagnosis of diabetes in the general population 2.2.1. Diagnostic criteria The ADA has recently endorsed revised diagnostic guidelines for diabetes mellitus in accordance with the World Health Organization (WHO) in 1999 (52), which recommend that all types of diabetes be diagnosed through the criteria outlined in Table 3 (53). These guidelines are also in line with the criteria recommended by the International Diabetes Foundation (IDF) and the American College of Endocrinology (ACE) (54,55). A number of significant changes have been made to the diagnostic criteria and classification of diabetes over the past 20 years. The major and most recent changes to the diagnostic criteria include the following: Lowering of fasting plasma glucose values for diagnosis of diabetes from greater than or equal to 140 mg/dL (7.8 mM) to greater than or equal to 126 mg/dL (7.0 mM). Lowering of the plasma glucose level for diagnosis of impaired fasting glucose from greater than or equal to 120 mg/dL (6.7 mM) to greater than or equal to 110 mg/dL (6.1 mM). The rationales for these revisions were to avoid the discrepancy between the fasting plasma glucose (FPG) and OGTT in identifying individuals at high risk for developing adverse effects of diabetes, such as microvascular complications, and to facilitate and encourage the use of a simpler and equally accurate test—FPG—for diagnosing diabetes (53). The new FPG criteria represent the upper end of the range that corresponds with the 2-hr postload concentration (that is unchanged in the revised recommendations). Thus, it is considered that the risk for microvascular complications is increased in individuals with an FPG level greater than or equal to 126 mg/dL (7.0 mM) (52,53). Furthermore, the WHO considers the risk for macrovascular disease to be increased in individuals with FPG values greater than or equal to 126 mg/dL (7.0 mM) even if 2-hr values are less than 140 mg/dL (7.8 mM) (52). An intermediate group of individuals whose glucose levels do not meet the criteria for diabetes but cannot be considered normal have also been recognized in the recent changes to the diagnostic criteria (Table 3) (53). Although the FPG levels of people with impaired glucose tolerance or impaired fasting glucose are below diabetic thresholds, these individuals have a higher risk for the development of diabetes and CVD than the general population. 3. Risk Factors for Developing Diabetes after Transplantation The ability to predict a patient’s risk for developing diabetes after transplantation would be of considerable benefit in selecting appropriate immunosuppressive regimens for individuals, and in identifying those who may need more intensive monitoring and risk-factor intervention. Although there are currently no clearly established risk factors for diabetes after transplantation, a number of characteristics have been identified that appear to predispose patients to the development of the condition (Fig. 5) (22). Consideration of such factors is an important step in the clinical assessment of patients before transplantation and may be used to individualize therapy and reduce the risk of developing diabetes after transplantation. Obviously, some of these risk factors, such as age and ethnicity, are not modifiable, although risk may have an additive effect and the evaluation of individual risk factors should play a significant role in the management of transplant recipients (27). Risk factors may also differ depending on the immunosuppressant agent used; immunosuppression therapies have changed greatly over the last two decades, which may account for discrepancies in risk factor assessment in the literature. Figure 5: Factors associated with increased risk for developing new-onset diabetes after transplantation.3.1. Patient age The incidence of diabetes after transplantation appears to be greater in older age patients. When taking the findings of a number of studies into account, risk of developing diabetes after transplantation appears to increase in patients over the age of 40 years (22,30,37,56). Increased age is also reported to be associated with decreased patient and graft survival and increased morbidity from infections (30). In contrast, age does not appear to be a significant risk factor for the development of diabetes after liver transplantation (57). 3.2. Ethnic background There is strong evidence that African American and Hispanic populations have a greater risk of developing new-onset diabetes after transplantation than white populations (Table 4) (21,30). The differing incidence of new-onset diabetes after transplantation in patients of different ethnicity may reflect differential pharmacokinetics and diabetogenic effects of immunosuppressive agents (17). Compared with whites, African Americans require 37% higher doses of tacrolimus to achieve comparable blood concentrations, yet this agent is reported to be up to five times more diabetogenic than cyclosporine and has particularly potent diabetogenic effects in African Americans compared with whites (27,58). Table 4: Table 4. Incidence of new-onset diabetes after kidney transplantation according to ethnicity3.3. Family history Type 2 diabetes is a complex condition involving a combination of genetic and environmental factors and studies suggest that such factors may also be involved in the development of new-onset diabetes after transplantation (59). For example, there have been some reports that a family history of diabetes may be a predictor for the development of diabetes after kidney transplantation, with one study showing a sevenfold increase in the condition in patients with a positive family history (29,30). Similarly, significantly more heart transplant recipients who developed diabetes were found to have a family history of diabetes in first-degree relatives compared with those who remained free of the condition (46% vs. 15%, respectively;P <0.05) (60). These findings suggest that individuals with a history of diabetes among first-degree relatives should be identified early in the course of their treatment to monitor the development of diabetes and adapt their immunosuppressive therapy accordingly. The incidence of diabetes after transplantation has been reported to be higher in individuals with certain histocompatibility leukocyte antigen (HLA) phenotypes; however, the results from these studies are contradictory and involve only small numbers of patients (30,56,61). Consequently, HLA phenotype cannot be considered as a reliable risk factor for new-onset diabetes after transplantation at this stage. 3.4. Patient body weight Obesity frequently develops after transplantation and is associated with both reduced graft and patient survival (20,62). Truncal obesity is also a risk factor for developing insulin resistance (63). Body weight has been shown to be associated with the development of diabetes after transplantation in most studies (19,21,27,29,37). However, some studies have found the association between development of diabetes posttransplantation and either body weight or body mass index (BMI) to be weak (17,30). Nevertheless, obesity is a known risk factor for type 2 diabetes and it is possible that other indices such as intra-abdominal fat or waist-to-hip ratio may be more important risk factors for diabetes after transplantation than total body weight or BMI. 3.5. Hepatitis C virus status The development of diabetes after liver transplantation appears to be associated with a pretransplant diagnosis of hepatitis C virus (HCV) infection (Fig. 6) (41,64). Furthermore, new-onset diabetes after liver transplantation in patients with HCV infection is associated with an increase in both overall mortality and infection-related mortality (41). Figure 6: Incidence of pretransplant diabetes and diabetes after liver transplantation in patients with and without hepatitis C virus infection. (From Baid S, Cosimi AB, Farrell ML, et al. Posttransplant diabetes mellitus in liver transplant recipients: Risk factors, temporal relationship with hepatitis C virus allograft hepatitis, and impact on mortality. Transplantation 2001; 72: 1066; with permission.)HCV infection is also a significant comorbidity in kidney transplant recipients, occurring in 10% to 40% of patients, and again is associated with an increased risk of both graft failure and mortality (65). Furthermore, a strong association has bee