Aims To assess different aspects of the definition of valvular/non-valvular atrial fibrillation (AF) used in clinical practice by physicians who usually treat this condition. Methods and results We prospectively conducted a web-based survey including cardiologists and internists who attended continuing medical education courses on cardiovascular medicine. A questionnaire was drawn up, containing 17 questions clustered into five main topics: (A) known rheumatic aetiology; (B) site/type of valve involvement; (C) prosthetic heart valve; (D) haemodynamic relevance; (E) miscellaneous. The overall response rate was 22.4% (21.1% for cardiologists and 24% for internists). Coexistence of both medical history of rheumatic disease and clinical signs of valvular involvement were considered as essential prerequisites for the diagnosis of rheumatic AF by half of the respondents, and one-third assumed that lone aortic valve disease was sufficient for AF to be defined as valvular. A similar proportion of respondents considered that in the presence of mitral regurgitation, AF had to be defined as valvular. The majority of responding physicians considered the degree of valvular defect of lesser importance for the definition of valvular or non-valvular origin of AF. Conclusion We found important heterogeneity and uncertainties in the answers given by physicians who usually treat patients with AF, as evidence of the lack of precise and unique definitions of the origin of AF (valvular/non-valvular). It is urgent to issue clear widely accepted definitions of the origin of AF, which should improve clinical practice and research.

Cystatin C (CC) has been proposed to play a role in atherosclerosis. We aimed to review the prognostic value of CC serum/plasma levels in patients with acute coronary syndromes (ACS). Fifteen observational longitudinal studies were selected by Medline. Increased CC over threshold values ranging from 0.93 to 1.3 mg/L were prognostic for death (hazard ratio; HR: 2.04-3.6) and for the occurrence of any fatal and non-fatal cardiovascular events (HR: 1.7-9.6) for patients with either ACS only or coronary heart disease and prevalent ACS. Only one study showed an increased risk for future myocardial infarction (MI) in patients with marker levels higher than 1.0 mg/L. Three studies reported the risk associated with a change of one unit of CC for long-term death (HR ranging from 1.9 to 6.3) and for the composite end point of 1 year MI and death (HR 2.15). Some studies showed the additional prognostic value contributed from CC measurements to other markers and to conventional risk scores. Despite low to moderate evidence, there is a general agreement on the significant prognostic value of CC in ACS that might encourage further research focused on risk assessment for patients with MI.

The assessment of prognostic markers is key to the improvement of therapeutic strategies for cancer patients. Some promising markers may fail to be applied in clinical practice, or some useless markers may be applied, because of misleading results ensuing from inadequate planning of the study and/or from an oversimplified statistical analysis. This commentary illustrates and discusses the main issues involved in planning an effective clinical study and the subsequent statistical analysis for the prognostic evaluation of a cancer marker. Another aim is to extend the most applied statistical models (ie, those using Kaplan-Meier and Cox) to enable the choice of the best-suited methods for study endpoints. Specifically, for tumor-centered endpoints like tumor recurrence, the issue of competing risks is highlighted. For markers measured on a continuous numerical scale, a loss of relevant prognostic information may occur by setting arbitrary cutoffs; thus, the methods to analyze the original scale are explained. Furthermore, because the P-value is not a sufficient criterion to assess the usefulness of a marker in clinical practice, measures for evaluating the ability of the marker to discriminate between “good” and “bad” prognoses are illustrated. Several tumor markers are considered both in human and veterinary medicine. Given the similarity between markers for human breast cancer and canine mammary cancer, an application of the statistical methods discussed within the article to a public dataset from human breast cancer patients is shown.