End stage renal disease (ESRD) disproportionately affects older adults: approximately 60% of dialysis patients are aged e55. On average, older patients who undergo kidney transplantation (KT) have a survival benefit over dialysis. However, referral for KT in older patients is much lower than for other age groups, at almost 1/7th the rate of referral for younger patients.
The key challenge lies in identifying the appropriate candidates for KT. The inability to predict which older patients would benefit from KT, and which would be harmed, is one of the most critical dilemmas for patients, transplant physicians and referring nephrologists. To date, older KT candidates have been naively evaluated using models designed for younger patients based on registry data.
We hypothesize that risk metrics specific to older patients, but not captured in national registries or other conventional studies, will significantly improve risk prediction and thus clinical decision-making, referral, and clinical care. One such important metric is frailty, an independent syndrome of decreased physiologic reserve associated with increased hospitalizations, disability and declining cognitive function, and predictive of medical and surgical outcomes.
Additionally, systemic inflammation has been identified as a pathway linking frailty to multisystem dysregulation. We also hypothesize that evaluating outcomes specific to older patients will add clinical relevance to risk prediction models. In particular, compared with younger patients, older adults are at risk for disability, cognitive decline, and decreased quality of life, particularly i response to dialysis, major surgery, and immunosuppression.
For older adults, conventional models of patient and allograft survival may not be nearly as relevant as ones which take into account these important consequences. We will prospectively quantify the association of frailty, IL-6, other inflammatory markers, and outcomes in 600 older incident dialysis patients and 1,000 older KT recipients. We will then integrate these novel findings with a risk prediction model we previously designed based on registry data, using an innovative approach of standardizing our prospective population to the registry population. The innovative integration of our prospective data with national registry data will provide us the statistical power to identify subtle but important predictors combined with the novelty of metrics specific to older adults.
Based on this, we will design a Markov decision process model for older adults with ESRD, comparing outcomes between dialysis and KT. A successful decision process model will be immediately usable by patients, nephrologists, and transplant providers. Transplantation in older adults is a growing field, but risk prediction has been derived from population-based data utilizing age-independent measures. The incorporation of novel aging metrics, such as frailty and IL-6, and outcomes, such as disability, cognitive decline, and quality of life, will greatly improve clinical decision-making in older adults considering KT. This research will address the growing public health challenge of deciding appropriate treatment options for over 300,000 older adults on dialysis in the US.