Survival Analysis (Kaplan-Meier, Cox Regression, Implant Survivorship)

Ordinary statistics struggle with follow-up data because not everyone is followed for the same time and not everyone has the event. Survival analysis solves this with censoring: a patient who is lost to follow-up, dies, or reaches the end of the study without the event is censored at their last known follow-up and still contributes the information that they were event-free up to that point. The KAPLAN-MEIER estimator uses this to build a step survival curve - cumulative survival starts at 100% and steps down at each event, with censored patients marked by ticks - from which you read survivorship at a time point (for example a 10-year implant survivorship) with a confidence interval. The number at risk is tabulated beneath the curve and falls over time, so the late part of the curve is based on few patients and is uncertain (wide confidence intervals).

• Log-rank test: a non-parametric hypothesis test comparing two or more KM curves (e.g. two implants or two patient groups) for a significant difference in survival over the whole follow-up.
• Cox proportional-hazards regression: the multivariable model of survival analysis. It estimates a HAZARD RATIO (HR) for each predictor (e.g. implant design, age, BMI, sex) while adjusting for the others; an HR above 1 means increased hazard, below 1 means protective, and 1 means no effect (e.g. an HR of 2.7 means 2.7 times the instantaneous risk of the event). It assumes the proportional- hazards assumption (the hazard ratio is roughly constant over time).
• Registries use these to compare implants and identify risk factors for failure - but interpret with care (endpoint definition, follow-up, competing risks).

• Define the endpoint: 'revision for ANY cause' gives a lower survivorship than revision for a SPECIFIC cause (e.g. aseptic loosening) - always check which is reported.
• Number at risk / follow-up: late survivorship rests on few patients; demand the number-at-risk table and beware wide late confidence intervals.
• Competing risk of death: in older patients, KM (which censors deaths) overestimates the cumulative revision risk; a competing-risks/cumulative-incidence analysis is more truthful about absolute risk.
• Implant 'camouflage': a poorly performing component combination can be hidden within an overall well-performing brand portfolio in registry averages - so brand-level KM can mask a bad combination.
• Selection/ascertainment: revision as an endpoint underestimates failure (some failing implants are not revised, e.g. unfit patients), and depends on complete registry capture.