A 69-year-old man with chronic kidney disease comes to the office to discuss options for dialysis. The patient wishes to choose the dialysis modality that will allow him to live the longest. The following Kaplan-Meier survival curve is obtained from a prospective study on the survival of 1800 patients who began dialysis with either hemodialysis (HD) or peritoneal dialysis (PD) and were followed for 5 years. The event of interest was death while on dialysis. Based on the findings of this study, which dialysis option offers improved survival?

Question

A 69-year-old man with chronic kidney disease comes to the office to discuss options for dialysis.  The patient wishes to choose the dialysis modality that will allow him to live the longest.  The following Kaplan-Meier survival curve is obtained from a prospective study on the survival of 1800 patients who began dialysis with either hemodialysis (HD) or peritoneal dialysis (PD) and were followed for 5 years.  The event of interest was death while on dialysis.  Based on the findings of this study, which dialysis option offers improved survival?

Quizplus · Accepted Answer

Explanation: When subjects in a clinical trial or observational study are followed over time, they can experience only one of several possible events: Death or other primary event of interest (eg, hospitalization, development of cancer) Completion of the study Loss to follow-up Different subjects experience these events at different points in time; therefore, comparing event rates at the end of the study is not appropriate, especially when comparing survival rates in groups using different drugs or with different disease states. More helpful information is obtained if the rate at which these events occur during the entire study period can be compared. This comparison is commonly done by survival analysis, often represented graphically by a Kaplan-Meier survival curve. The Kaplan-Meier survival curve depicts the probability of survival at various time points during the study. This probability is calculated based on the proportion of subjects who are alive (or "at risk" for an event of interest) at a given time. The x-axis denotes time, and the y-axis denotes cumulative event-free survival starting from 1.0 (ie, 100% of patients are alive without having experienced any event) at the beginning of the study. As the study progresses, events occur and the proportion of subjects who are still alive or "at risk" for an event of interest decreases, demonstrated as vertical downward dips in the curve. Depending on the number of study groups, >2 curves can be displayed simultaneously and compared. Survival of different groups can be compared by the log-rank test, which reports a p-value. In most studies, a p-value <0.05 is considered significant. A simplified explanation is that there is a <5% probability that the event-free survivals are different due to chance. The p-value reported in this study (p = 0.45) was not statistically significant as it was >0.05; therefore, survival on HD or PD was found to be not significantly different. (Choices A, B, D, and E) Median survival (follow-up time when 50% of subjects are alive) is determined by drawing a horizontal line at 0.5 on the y-axis and looking at the corresponding time point on the x-axis. In this study, median survival on HD was ~3.0 years and median survival on PD was ~3.6 years. Similarly, survival at arbitrary time points (eg, end of study) can be determined by the curves. Although survival on PD appeared to be more than on HD at all these time points, the p-value was not significant; therefore, patients on PD were not significantly more likely to live longer than those on HD. Educational objective: The Kaplan-Meier survival curve is a commonly used graphical representation of the probabilities that subjects survive in a study without an event. The event-free survival rates of >2 study groups can be compared; they are statistically different only if the p-value is <0.05. References: A practical guide to understanding Kaplan-Meier curves. (http://www.ncbi.nlm.nih.gov/pubmed/20723767)

A 69-Year-Old Man with Chronic Kidney Disease Comes to the Office