Background and aim of the study: Different standards for the reporting of morbid events and different follow up techniques have a profound impact on reported morbidity after prosthetic valve replacement. Most studies follow the guidelines of The American Association of Thoracic Surgery (AATS) and The Society of Thoracic Surgeons (STS); the present authors’ group has now developed an adapted Karnofsky scale which allows a more precise grading of the severity of morbid events.
Methods: The AATS/STS criteria and the adapted Karnofsky criteria were applied to the database of the German Experience with Low-Intensity Anticoagulation (GELIA) study. In a study population of 2,735 patients, GELIA compared three different intensities of oral anticoagulation in a prospective and randomized design. Patients registered morbid events

prospectively by means of documentation cards.
Results: The overall rate of complications was comparable when utilizing the two classification systems. However, use of the AATS/STS criteria resulted in the counting of fewer bleeding complications, because only major bleedings were recorded. In contrast, the incidence of embolic complications was higher compared to the Karnofsky criteria because all events were counted, irrespective of their severity, while clinically insignificant (transient, reversible within 24 h) events were disregarded when using the Karnofsky grading.
Conclusion: The adapted Karnofsky criteria provide a precise and easily understandable framework for the assessment of complications, with equal weighting of both hemorrhagic and embolic events.
The Journal of Heart Valve Disease 2005;14:467-471