Key words : Kidney transplant, New-onset diabetes after transplant, Triple immunosuppressant regimen, Trough levels of tacrolimus

Introduction

For the past 20 years, tacrolimus has been an integral part of immunosuppressive protocols for kidney transplant.¹ More than 90% of all transplant recipients receive tacrolimus as part of their immunosuppressant regimen.² A tacrolimus-based regimen has improved graft survival and patient survival over the recent years. However, transplant recipients continue to experience several adverse effects, such as nephrotoxicity, neurotoxicity, and posttransplant diabetes mellitus (PTDM), previously known as new-onset diabetes after transplant (NODAT).^3,4 Posttransplant diabetes mellitus is associated with increased risk of cardiovascular mortality and graft failure^5,6 and occurs in approximately one-third of renal transplant recipients, with global prevalence varying between 2% and 53%.⁷ The wide variation in the prevalence is due to differences in the definition of diabetes mellitus, study design, concomitant immunosuppressive agents, and the time from transplant.

The adverse effect profile of tacrolimus in the Indian population has been reported in only limited studies.^8-12 In addition, although evidence has shown a relationship between blood level of tacrolimus and PTDM, it has not yet been clearly defined.^13,14 Hence, to bridge this knowledge gap, we conducted a study with the objectives of estimating the frequency of PTDM and other adverse effects of tacrolimus in renal transplant patients. This study also aimed to enumerate the risk factors associated with the development of PTDM and to estimate the blood levels of tacrolimus associated with the development of PTDM.

Materials and Methods

This prospective observational study was conducted from January 2019 to June 2020 in the Department of Nephrology in a tertiary care teaching hospital in India. We enrolled renal transplant recipients receiving tacrolimus from the transplant clinic. Patients who had developed diabetes before transplant were excluded. The study was approved by the Institute Ethics Committee for Human Studies (reference number: JIP/IEC/2017/0336, dated November 20, 2017), and full written informed consent was obtained from the participants.

All study participants received a triple immuno-suppressant regimen consisting of tacrolimus, prednisolone, and mycophenolate mofetil (MMF). The enrolled participants were followed up monthly for 6 months following transplant. Participants were provided a diary and asked to note any adverse effects experienced along with the date and duration. At each visit, all participants were asked about the adverse effects experienced by them. Participants were asked and/or assessed for symptoms or signs of known adverse effects of tacrolimus-based regimens like diarrhea, neurotoxicity (headache, tremors), alopecia, and repeated infections. They were also contacted through telephone to ask about any adverse effects experienced between follow-up visits. Adherence to medication was ensured during every visit by the pill counting method.

Fasting blood sugar (FBS) and postprandial blood sugar (PPBS) levels were checked monthly for 6 months posttransplant in patients enrolled soon after transplant. In patients who had been previously transplanted (≥1 year previously), FBS and PPBS values were checked only once, at the time of recruitment into the study. Patients were said to have developed PTDM if their FBS was ≥126 mg/dL or PPBS was ≥200 mg/dL. Patients were diagnosed with calcineurin inhibitor-induced nephrotoxicity only if kidney biopsy features were suggestive. This was to avoid the confounding effects of other factors producing posttransplant renal damage. Blood levels of tacrolimus estimated by liquid chromatography-mass spectrometry were recorded every month for the first 6 months posttransplant. However, drug levels were not measured for patients who had been transplanted 1 year or more previously.

The sample size was calculated for an expected frequency of PTDM as 28% based on a previous study11 with a precision of 10% and 95% confidence intervals (95% CIs). The estimated sample size was n = 77 using the software nMaster version 2.0. Demographic and clinical characteristics were compared between patients who developed PTDM and those who did not (non-PTDM patients) by independent sample t tests for numerical data and chi-square tests for categorical data. Associations between age, sex, body mass index (BMI), family history of diabetes, and tacrolimus trough level (T₀) at month 1 posttransplant and the development of PTDM were established by multiple logistic regression. Receiver operating characteristic (ROC) curves were also constructed to find the optimal T₀ in predicting the development of PTDM. All tests were 2-tailed, and P < .05 was considered statistically significant. We performed data analysis using IBM-SPSS version 21.0.

Results

A total of 77 patients were enrolled in the study. All baseline characteristics except for donor status were comparable between PTDM and non-PTDM patients (Table 1). A greater proportion of individuals in the PTDM group received transplants from living donors compared with non-PTDM patients (81.2% vs 58.6%). The living donors were related to the recipients (first-degree relatives). The most common adverse effect of tacrolimus was PTDM (62.3%) followed by tremors (50.6%) and diarrhea (32.5%). Causality assessment using Naranjo algorithm revealed a causal relationship between tacrolimus intake and PTDM, diarrhea, and opportunistic infections (cytomegalovirus, BK virus) in all patients reporting these adverse effects as “possible” (Table 2). Tacrolimus was not discontinued for any of the participants during the course of the study.

Among the 48 patients diagnosed with PTDM, 46 were recently transplanted; the remaining 2 patients were transplanted ≥1 year previously and were nondiabetic at the time of transplant but found to have diabetes after recruitment into the study. Because blood sugar levels were measured only once in patients who received transplants ≥1 year previously, it was not possible to know exactly when these 2 patients developed PTDM. Among the 46 PTDM patients who were followed up from the time of transplant, 25 patients developed PTDM during month 1 posttransplant, 14 patients during month 2 posttransplant, 5 patients during month 3 post-transplant, and 1 patient each during months 4 and 5 posttransplant. The 46 recently transplanted patients who were diagnosed with PTDM continued to have diabetes throughout study follow-up (6 months posttransplant).

The mean (SD) trough level of tacrolimus in patients who developed PTDM was significantly higher compared with non-PTDM patients in the first month following transplant (10.1 ± 2.8 vs 8.2 ± 1.8 ng/mL; P = .02). However, there was no significant difference between the groups over the next 5 months (Table 3). Trough levels of tacrolimus of patients in the first month posttransplant were grouped into specific ranges: ≤5 ng/mL, 5.1 to 10 ng/mL, 10.1 to 15 ng/mL, and 15.1 to 20 ng/mL. Figure 1 shows the number of patients who developed PTDM and did not develop PTDM for each of these specific ranges. Most patients (22 of 46 PTDM patients and 11 of 13 non-PTDM patients) had trough levels in the 5.1 to 10 ng/mL range.

Multiple logistic regression analysis was perfor-med to determine any associations between age, male sex, BMI, positive family history of diabetes, and T₀ at month 1 posttransplant and development of PTDM. Among these 5 covariates, only T₀ at month 1 posttransplant was found to be inde-pendently associated with the development of PTDM (adjusted odds ratio [OR] = 1.379; 95% CI, 1.02-1.86) (Table 4). Figure 2 shows ROC curves used to determine the optimal cutoff of T₀ level in the first month posttransplant for predicting the occurrence of PTDM. A cutoff level of 8.1 ng/mL showed sensitivity of 80% and specificity of 60% with area under the curve of 0.712. A greater number of patients with first month T0 >8.1 ng/mL developed PTDM (37/46) compared with those with first month T₀ ≤8.1 ng/mL (9/46), and this difference was found to be statistically significant (P < .01). Multiple logistic regression analysis was performed to statistically adjust for the effects of confounders (age, male sex, BMI, and positive family history of diabetes); results showed a 5 times increased risk of developing PTDM when T₀ at month 1 posttransplant was >8.1 ng/mL (adjusted OR = 5.4, 95% CI, 1.4-19.9).

The mean FBS and PPBS values for every month until 6 months posttransplant were found to be statistically significant until month 2 and 4, respectively, between PTDM and non-PTDM patients. After this time period, there was no statistically significant difference, since the PTDM group patients received treatment with metformin/insulin and their blood glucose levels began to be normalized.

Discussion

Posttransplant diabetes mellitus developed in 62.3% of our renal transplant recipients on tacrolimus. Trough level of tacrolimus >8.1 ng/mL in the first month posttransplant increased the risk of PTDM. The frequency of PTDM estimated in this study was higher than that already reported in literature (33.89% and 28%).^8,10 This might be due to the longer duration of follow-up and the difference in the definition of PTDM between the studies.

The proportion of patients who received a kidney from a living donor was higher in the PTDM group than in the non-PTDM group (81.2% vs 58.6%; P = .03). This is in contrast with a Chinese study,15 in which recipients of deceased donor kidneys were at higher risk for PTDM (OR? = ?1.58). This contrasting finding is because most patients (72.7%) in our study received a living donor transplant. However, a recent study found no significant association between donor type and PTDM.¹²

The prevalence of biopsy-proven calcineurin inhibitor-induced nephrotoxicity was 6.5% in our study, but the prevalence of nephrotoxicity has been shown to range from 17% to 44%.¹⁶ This difference might be because we limited ourselves to biopsy-proven episodes since it ensures objective evidence for establishing causality. Not all cases of nephrotoxicity in transplant recipients require a renal biopsy. Hence, the actual incidence of nephrotoxicity episodes not confirmed by renal biopsies might be higher.

The causal relationship between tacrolimus intake and development of PTDM, diarrhea, and oppor-tunistic infections (cytomegalovirus, BK virus) as assessed by Naranjo scale was categorized as “possible” in all patients experiencing these adverse effects. They are more so attributed to a tacrolimus-based immunosuppressant regimen rather than to tacrolimus alone. This is because concomitant steroids contribute to increased risk of PTDM and oppor-tunistic infections. Similarly, concomitant MMF contributes to gastrointestinal disturbances in the form of diarrhea. Because the concomitant drugs and underlying disease does not cause adverse effects like neurotoxicity, biopsy-proven nephrotoxicity, and alopecia, these attained a higher level in the Naranjo scale (“probable”).

Multiple logistic regression analysis showed that the higher T₀ at month 1 posttransplant indepen-dently increased the risk of PTDM (adjusted OR = 1.379; 95% CI, 1.02-1.86). The cutoff point of T₀ in the first month posttransplant, identified by ROC curve analysis for predicting PTDM occurrence, was 8.1 ng/mL. There was a 5 times increased risk of developing PTDM when T₀ at month 1 posttrans-plant was >8.1 ng/mL (adjusted OR = 5.4; 95% CI, 1.4-19.9). This is in accordance with a recent observational study in which it was found that the best cutoff of T₀ to predict PTDM was 8.5 ng/mL.⁹ The same study also reported that a higher T₀ increases PTDM risk by 1.3 times (OR = 1.316; 95% CI, 1.119-4.796). A recent study conducted by Turgut and colleagues investigated the intrapatient variability in blood level of tacrolimus and reported that high intrapatient variability increases risk of BK virus nephropathy and chronic calcineurin inhibitor toxicity.¹⁷

Our study could not establish an increased risk of PTDM with older age. Patel and colleagues⁸ reported that patients aged >45 years have a greater chance to develop NODAT than patients aged <45 years (P < .0001). However, our study and the study from Patel and colleagues are not comparable. The study by Patel and colleagues was a retrospective record-based study with a sample size of 537 that included 99 recipients aged ≥45 years. In our study, we had only 17 patients who were ≥45 years of age, of which 8 were found to develop PTDM. We could not establish an increased risk of PTDM with older age because the patients in our study were predominantly younger with only a small proportion (22%) of patients above 45 years of age.

A meta-analysis (that included 55 eligible studies involving renal transplant patients) showed that BMI is an independent risk factor of NODAT (mean difference of 1.88; 95% CI, 1.48-2.27).¹⁸ In our study, only 6 patients were found to be overweight at the time of transplant, of which 4 patients developed PTDM. However, the limited number of obese patients in our study may have contributed to the association not being significant. There was an overall male preponderance in this study, and only 28.6% of patients were found to have family history of diabetes. It was impossible to study the association between sex and PTDM in our study as the proportion of female patients was low (20%).

The strengths of our study were its prospective nature, which added validity to the results and helped in avoiding recall bias. Reporting of adverse effects by patients was increased by using tools like diaries and telephone inquiries. This ensured complete retrieval of information regarding adverse effects experienced by the patients and avoidance of memory or recall bias. The follow-up of patients for a considerably long period of 6 months after transplant provided a complete picture regarding the adverse effect profile of tacrolimus. Our study had several limitations. Because patients were adminis-tered other drugs like prednisolone and MMF, the impact of tacrolimus individually over PTDM development could not be established. In addition, it was not possible to stop the drug and see whether the adverse effect was reversible, since discontinuing the drug can lead to loss of immunosuppression, thereby increasing the risk of graft rejection. We had to limit the number of risk factors (covariates) in multiple logistic regression to predict occurrence of PTDM to 5 only because of sample size constraints.

Conclusions

Posttransplant diabetes mellitus occurred in 62.3% of renal transplant patients receiving tacrolimus-based immunosuppression regimen. A first month T₀ >8.1 ng/mL was an independent predictor of PTDM in our Indian population. There was a 5 times increased risk of developing PTDM when T₀ at month 1 posttransplant was >8.1 ng/mL (adjusted OR = 5.4; 95% CI, 1.4-19.9). It is recommended that the trough levels of tacrolimus in the first month after renal transplant should be maintained below 8.1 ng/mL to reduce the risk of PTDM development. Physicians should also actively investigate the laboratory markers, signs, and symptoms suggestive of PTDM and other adverse effects of tacrolimus to prevent them.

References:

1. Banas B, Kramer BK, Kruger B, Kamar N, Undre N. Long-term kidney transplant outcomes: role of prolonged-release tacrolimus. Transplant Proc. 2020;52(1):102-110. doi:10.1016/j.transproceed.2019.11.003
   CrossRef - PubMed
   
2. Gonzales HM, McGillicuddy JW, Rohan V, et al. A comprehensive review of the impact of tacrolimus intrapatient variability on clinical outcomes in kidney transplantation. Am J Transplant. 2020;20(8):1969-1983. doi:10.1111/ajt.16002
   CrossRef - PubMed
   
3. Shrestha BM. Two decades of tacrolimus in renal transplant: basic science and clinical evidences. Exp Clin Transplant. 2017;15(1):1-9. doi:10.6002/ect.2016.0157
   CrossRef - PubMed
   
4. Sharif A, Hecking M, de Vries AP, et al. Proceedings from an international consensus meeting on posttransplantation diabetes mellitus: recommendations and future directions. Am J Transplant. 2014;14(9):1992-2000. doi:10.1111/ajt.12850
   CrossRef - PubMed
   
5. Solhjoo M, Kumar SC. New onset diabetes after transplant. In: StatPearls. StatPearls Publishing; 2022.
   CrossRef - PubMed
   
6. Kumar S, Sanyal D, Das P, Bhattacharjee K, Rungta R. An observational prospective study to evaluate the outcomes of new onset diabetes after renal transplantation (NODAT) in a tertiary care centre in eastern India. Diabetes Res Clin Pract. 2020;159:107948. doi:10.1016/j.diabres.2019.107948
   CrossRef - PubMed
   
7. Balla A, Chobanian M. New-onset diabetes after transplantation: a review of recent literature. Curr Opin Organ Transplant. 2009;14(4):375-379. doi:10.1097/MOT.0b013e32832dbb98
   CrossRef - PubMed
   
8. Patel DD MK, Patel AK, Chaudhary VI. New onset of diabetes mellitus in Indian renal transplant recipient–A retrospective study. Int J Pharm Pharm Sci. 2015;7(11):228-232.
   CrossRef - PubMed
   
9. Choudhury PS, Mukhopadhyay P, Roychowdhary A, Chowdhury S, Ghosh S. Prevalence and predictors of “new-onset diabetes after transplantation” (NODAT) in renal transplant recipients: an observational study. Indian J Endocrinol Metab. 2019;23(3):273-277. doi:10.4103/ijem.IJEM_178_19
   CrossRef - PubMed
   
10. Guruprasad P, Kishore K, Mahajan S, Aggarwal S. Active surveillance for adverse events among patients who underwent renal transplantation: A prospective observational study. Perspect Clin Res. 2017;8(3):118-123. doi:10.4103/2229-3485.210447
    CrossRef - PubMed
    
11. Memon SS, Tandon N, Mahajan S, Bansal VK, Krishna A, Subbiah A. The prevalence of new onset diabetes mellitus after renal transplantation in patients with immediate posttransplant hyperglycemia in a tertiary care centre. Indian J Endocrinol Metab. 2017;21(6):871-875. doi:10.4103/ijem.IJEM_309_17
    CrossRef - PubMed
    
12. Jayam J BM, Srinivasaprasad ND, Sujit S, Fernando ME. Prevalence and risk factors for posttransplant diabetes mellitus: data from government tertiary care center. Indian J Transplant. 2018;12(2):119-124.
    CrossRef - PubMed
    
13. Maes BD, Kuypers D, Messiaen T, et al. Posttransplantation diabetes mellitus in FK-506-treated renal transplant recipients: analysis of incidence and risk factors. Transplantation. 2001;72(10):1655-1661. doi:10.1097/00007890-200111270-00014
    CrossRef - PubMed
    
14. Yagmurdur MC, Sevmis S, Emiroglu R, Moray G, Bilgin N, Haberal M. Tacrolimus conversion in kidney transplant recipients: analysis of 107 patients. Transplant Proc. 2004;36(1):144-147. doi:10.1016/j.transproceed.2003.11.044
    CrossRef - PubMed
    
15. Lv C, Chen M, Xu M, et al. Influencing factors of new-onset diabetes after a renal transplant and their effects on complications and survival rate. PLoS One. 2014;9(6):e99406. doi:10.1371/journal.pone.0099406
    CrossRef - PubMed
    
16. Bentata Y. Tacrolimus: 20 years of use in adult kidney transplantation. What we should know about its nephrotoxicity. Artif Organs. 2020;44(2):140-152. doi:10.1111/aor.13551
    CrossRef - PubMed
    
17. Turgut D, Sayin B, Soy EA, Topcu DI, Ozdemir BH, Haberal M. Tacrolimus intrapatient variability in BK virus nephropathy and chronic calcineurin toxicity in kidney transplantation. Saudi J Kidney Dis Transpl. 2021;32(2):348-354. doi:10.4103/1319-2442.335446
    CrossRef - PubMed
    
18. Chang S, Jiang J. Association of body mass index and the risk of new-onset diabetes after kidney transplantation: a meta-analysis. Transplant Proc. 2018;50(5):1316-1325. doi:10.1016/j.transproceed.2018.02.075
    CrossRef - PubMed