Variables - continuous or all-or-none
Leung and French have stated that changing the goalposts in the definition of peri-procedural MI or bleeding does not necessarily make such events more or less useful in the overall risk assessment of a patient. They have also raised the issue of whether cardiac markers should be treated as continuous rather than as discrete all-or-none variables in data analysis (1).
The relationship between a variable and a particular outcome should determine if a variable is treated as a continuous or an all-or-none parameter. If there is a convenient point on the association curve for a variable and an outcome beyond which adverse events increase significantly, then it would be prudent to have that as the cut-off value for the variable.
The relation between periprocedural CK-MB elevations and late mortality is approximately linear, and myocardial necrosis has similar prognostic significance whether it is caused by a "spontaneous" ischemic event or by PCI (2). This relation between a "surrogate" and “hard clinical outcome” would be a straight line, where any cut-off that is determined is arbitrary. A cut-off of three times the upper limit of normal has no special significance beyond which adverse outcomes increase significantly.
Therefore, moving from the all-or-none analysis for cardiac biomarkers to a more continuous model would be more valid and provide better utility for overall risk assessment.
1. Leung DY, French JK. End points in clinical trials: are they moving the goalposts? Heart 2006;92(7):870-2.
2. Akkerhuis KM, Alexander JH, Tardiff BE, et al. Minor myocardial damage and prognosis: are spontaneous and percutaneous coronary intervention- related events different? Circulation 2002;105(5):554-6.
Register for free content
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of Heart.
View free sample issue >>
The full back archive is now available for Heart. Institutional subscribers may access the entire archive as part of their subscription. Personal subscribers will also have access to all content when logged in. Non-subscribers who register have free access to all articles published before 2006, back to volume 1 issue 1.
Register to access the free archive >>
Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.