Predictive factors for development of diabetes mellitus post-heart transplant

It is known that there is a high incidence of diabetes mellitus (DM) among heart transplant (HT) patients, which may be up to 30% at 5 years. The presence of DM has been associated with increased morbidity (infections, renal dysfunction, or graft vascular disease), and its development has been related primarily to immunosuppressive therapy. The objective of this study was to determine, in our experience, the presence of predictive variables for the development of DM following HT.

We studied 315 consecutive non-DM patients (88.6% men, mean age 51.5 years) who underwent HT in our hospital from November 1987 to May 2003, analyzing all variables that could be related to the development of DM during follow-up. Student t-test and chi(2) test were used for univariate statistical analysis and logistic regression for multivariate analysis.

Of the 315 patients, 64 developed DM (20.3%) during a mean follow-up of 3.3 years. The univariate analysis showed that patients developing DM are older (54.9 +/- 8.7 versus 50.7 +/- 11.8 years, P = .008), have a higher body mass index (BMI) (27.3 +/- 3.8 versus 25.7 +/- 3.7, P = .003), a higher prevalence of HT (37.5% versus 23.5%, P = .023), a lower frequency of urgent HT (9.4% versus 26.2%, P = .004), are more often treated with steroids (85.9% versus 70.1%, P = .011) and tacrolimus (12.5% versus 4.4%, P = .015), and have a higher frequency of rejection episodes (71.2% versus 44.6%, P = .001). Multivariate analysis identified the following as predictive factors for the development of DM: age (OR = 1.04, P = .013), urgent HT (OR = 0.36, P = .031), treatment with tacrolimus (OR = 3.89, P = .012), and number of rejections (OR = 2.34, P = .002).

In our population, age, urgent HT (which had a protective effect), treatment with tacrolimus, and number of rejections were independent predictive variables for the development of DM during follow-up.