Kidney transplantation is the preferred treatment for the majority of patients with end-stage kidney disease, offering improved survival and quality of life at reduced cost compared with dialysis. Originally a treatment only offered to young patients free of comorbidity, transplantation is now routinely performed in patients with significant comorbidities.
The transplantation of patients with comorbidities introduces a number of challenges, including (1) patient selection (which comorbidities are acceptable and which preclude transplantation, either singly or in combination?), (2) perioperative management, and (3) increased risk of short- and long-term complications. Furthermore, observational transplant studies require adequate adjustment for comorbidities to account for their confounding effect on outcomes. The need for such adjustment is brought into sharp focus when such studies are used to rate the performance of transplant programs, as for example occurs in the United States, because inadequate adjustment for comorbidities may lead to false-positive and false-negative flagging of programs.
Therefore, a systematic method for recording the overall comorbid burden of a kidney transplant candidate or recipient is useful both clinically and epidemiologically. The Charlson Comorbidity Index (CCI), first described in 19871 and validated in many studies across a variety of clinical populations, has been widely used in the medical literature including in the field of kidney transplantation.2,3 However, being a general comorbidity index, it has some features that are not suited to kidney transplantation, such as the inclusion of metastatic solid tumor (a contraindication to transplantation) and moderate to severe renal disease. It is possible that a comorbidity index designed specifically for kidney transplantation would provide improved risk stratification compared with the CCI.
In this issue, Laging and colleagues4 describe the development of a new comorbidity index designed specifically for kidney transplantation, the Rotterdam Comorbidity in Kidney Transplantation (RoCKeT) score. The authors performed a single-center retrospective cohort study, incorporating all 1 728 kidney transplants performed at their institution over 2000 to 2013. The RoCKeT score was developed using similar methodology to Charlson; its components are similar to the CCI but with different weightings, and include cardiovascular disease, cerebrovascular accident, peripheral vascular disease, diabetes mellitus, liver disease, lung disease, prior malignancy, other organ transplantation, and human immunodeficiency virus positivity.
As might be expected, the RoCKeT score did not predict death-censored graft survival, but was a strong predictor of patient survival. Compared with a score of 0, scores of 1 to 2, 3 to 4, and 5 to 9 resulted in hazard ratios for patient death of 1.7, 2.1, and 2.7, respectively. Apart from the RoCKeT score, the patient survival model was adjusted for age, ethnicity, time on dialysis, transplantation of another organ (heart, liver, or lung), transplant year, and donor type. Thus, it is important to understand that the score by itself cannot be used to predict the overall risk of death in a transplant candidate; rather, it should be seen as a summary measure of the patient's total comorbidity burden as a component of overall risk assessment.
In this cohort, the RoCKeT score was a stronger predictor of patient survival than the CCI. At first glance, this might suggest that it is superior to the CCI in the kidney transplant population. However, because the score was both generated and tested in the same group of patients, its performance is optimized for this particular cohort. External validation, in which the score is tested using a different patient population, will be required before the RoCKeT score can be recommended over the CCI.
The authors also report that the cohort demonstrated good overall survival even in those with a high comorbid burden; the median posttransplant patient survival was greater than 10 years even in those with a RoCKeT score in the highest risk group. Citing this finding, the authors recommend considering such patients for transplantation. Although logical, this recommendation must be taken with caution for a number of reasons: the lack of a control group, the single-center nature of the study, the possibility of selection bias and unmeasured confounders, and the lack of data on quality of life. An additional consideration is that, from a utilitarian perspective, there is some rationale to restricting deceased donor transplantation to candidates with more favorable prognoses—if patients with high comorbid burden receive deceased donor transplants, the number of such organs available for patients with a better prognosis will be reduced and the overall survival benefit from transplantation is consequently likely to be reduced.
Overall, these findings are very interesting and form a useful contribution to the kidney transplant literature. However, further validation in external populations will be needed before the RoCKeT score is ready to fly.
1. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis
. 1987; 40: 373–383.
2. Jassal SV, Schaubel DE, Fenton SS. Baseline comorbidity in kidney transplant recipients: a comparison of comorbidity indices. Am J Kidney Dis
. 2005; 46: 136–142.
3. Wu C, Evans I, Joseph R, et al. Comorbid conditions in kidney transplantation: association with graft and patient survival. J Am Soc Nephrol
. 2005; 16: 3437–3444.
4. Laging M, Gestel JAK-V, van de Wetering J, et al. A high comorbidity score should not be a contraindication for kidney transplantation. Transplantation
. 2015; 100: 400–406.