Every guideline recommendation carries two separate grades. The Class says how strongly it is recommended. The Level says how good the evidence behind it is. They are independent: a recommendation can be strong on weak evidence, or weak on strong evidence. ACC/AHA and the ESC use different scales, so they are shown side by side and never converted.
ACC/AHA
- Class 1Strong. Benefit far outweighs risk. Worded as "is recommended".
- Class 2aModerate. Benefit outweighs risk. Worded as "is reasonable" or "can be useful".
- Class 2bWeak. Benefit is about equal to or greater than risk. Worded as "may be considered".
- Class 3: No BenefitBenefit equals risk. Worded as "is not recommended".
- Class 3: HarmStrong. Risk outweighs benefit. Worded as "potentially harmful".
- Level AHigh-quality evidence from more than one randomized trial, meta-analyses of high-quality trials, or trials corroborated by high-quality registries.
- Level B-RModerate-quality evidence from one or more randomized trials, or meta-analyses of them.
- Level B-NRModerate-quality evidence from well-designed nonrandomized, observational or registry studies, or meta-analyses of them.
- Level C-LDLimited data: observational or registry studies with design limitations, or physiological and mechanistic studies in people.
- Level C-EOConsensus of expert opinion based on clinical experience.
ESC
- Class IEvidence and general agreement that it is beneficial, useful and effective. Worded as "is recommended".
- Class IIaWeight of evidence or opinion favors usefulness. Worded as "should be considered".
- Class IIbUsefulness is less well established. Worded as "may be considered".
- Class IIIEvidence or agreement that it is not useful or effective, and may be harmful. Worded as "is not recommended".
- Level AData from multiple randomized trials or meta-analyses.
- Level BData from a single randomized trial or large nonrandomized studies.
- Level CExpert consensus, small or retrospective studies, or registries.