BTS 2012 Report – How long have I got? Up to date estimates of patient survival

by Maria Dalby reporting on the presentation by Dave Collett, Donation Advisory Group, NHS Blood and Transplant.  Patient survival is by far the most important outcome in any transplantation and it is important that the clinician is able to give a realistic estimate of the expected survival. However, the standard method of estimating survival through Kaplan-Meier analysis is based on transplantations carried out many years ago, and thus represents clinical practice and surgical techniques that may be more or less out of date. Professor Dave Collett from NHSBT outlined alternative options of estimating survival and suggested that some methods may be more accurate than others.

The most common way of estimating survival data when Kaplan-Meier survival analysis is not possible is by extrapolating the most recent available data by fitting a Cox regression model with cohort effect. However, validation shows that the extrapolation model often does not fit very well with known data and thus estimates based on extrapolation cannot be regarded as reliable. An alternative method, period analysis, has been used for estimating the population-based survival amongst cancer patients, and was recently proposed for use in transplantation.¹ Period analysis considers the number of patients at risk and the number of deaths within a given, recent calendar period. This way, estimates of survival are based on and reflect the most recent data available. Validation of survival estimates obtained in this way show a good fit with known NHSBT data on liver transplantation, and showed clear differences in prognosis for patients undergoing liver transplantation due to primary biliary cirrhosis (PBC) or primary sclerosing cholangitis (PSC) compared with those transplanted due to hepatitis C infection or cancer. In addition, period analysis of survival following kidney transplantation showed a significant benefit for patients receiving kidneys from a live donor, but there was no difference in estimated survival between kidneys from donors after brain death or circulatory death.

As the survival estimates obtained with period analysis reflect up-to-date clinical practice, they are generally greater, ie more optimistic, than those obtained with Kaplan-Meier analysis – an important aspect, not least for counselling patients. The NHSBT are currently working to introduce the period analysis methodology on their website, and information and tools will be available in the near future.

Reference:

1. Gondos A, Brenner H. Period analysis for more up-to-date graft and patient survival estimates in transplantation: an evaluation using united network for organ sharing data. Transplantation 2010; 89(5): 580-588