Do Optimal Prognostic Thresholds in Continuous Physiological Variables Really Exist? Analysis of Origin of Apparent Thresholds, with Systematic Review for Peak Oxygen Consumption, Ejection Fraction and BNP

Abstract

BackgroundClinicians are sometimes advised to make decisions using thresholds in measured variables, derived from prognostic studies.ObjectivesWe studied why there are conflicting apparently-optimal prognostic thresholds, for example in exercise peak oxygen uptake (pVO2), ejection fraction (EF), and Brain Natriuretic Peptide (BNP) in heart failure (HF).Data Sources and Eligibility CriteriaStudies testing pVO2, EF or BNP prognostic thresholds in heart failure, published between 1990 and 2010, listed on Pubmed.MethodsFirst, we examined studies testing pVO2, EF or BNP prognostic thresholds. Second, we created repeated simulations of 1500 patients to identify whether an apparently-optimal prognostic threshold indicates step change in risk.Results33 studies (8946 patients) tested a pVO2 threshold. 18 found it prognostically significant: the actual reported threshold ranged widely (10–18 ml/kg/min) but was overwhelmingly controlled by the individual study population's mean pVO2 (r = 0.86, p<0.00001). In contrast, the 15 negative publications were testing thresholds 199% further from their means (p = 0.0001). Likewise, of 35 EF studies (10220 patients), the thresholds in the 22 positive reports were strongly determined by study means (r = 0.90, p<0.0001). Similarly, in the 19 positives of 20 BNP studies (9725 patients): r = 0.86 (p<0.0001).Second, survival simulations always discovered a “most significant” threshold, even when there was definitely no step change in mortality. With linear increase in risk, the apparently-optimal threshold was always near the sample mean (r = 0.99, p<0.001).LimitationsThis study cannot report the best threshold for any of these variables; instead it explains how common clinical research procedures routinely produce false thresholds.Key FindingsFirst, shifting (and/or disappearance) of an apparently-optimal prognostic threshold is strongly determined by studies' average pVO2, EF or BNP. Second, apparently-optimal thresholds always appear, even with no step in prognosis.ConclusionsEmphatic therapeutic guidance based on thresholds from observational studies may be ill-founded. We should not assume that optimal thresholds, or any thresholds, exist.

Body composition measured by computed tomography is associated with colorectal cancer survival, also in early-stage disease

“…However, the risk of attaining spurious correlations could increase when using optimal stratification, and generalization of results can be limited. [45] We, therefore, used both sex-specific tertiles and optimal stratification to define cutoffs for sarcopenia and myosteatosis, and the results were similar but predictably stronger for optimal stratification. Our results regarding both SMI and SMR were comparable to the large C-SCANS cohort, in which the lowest tertile of muscle area and sarcopenia defined by optimal stratification and the lowest quartile of SMR were significantly associated with poor cancer-specific survival [9,20].…”

Thoughts modulate the expression of inflammatory genes and may improve the coronary blood flow in patients after a myocardial infarction

“…This suggests that clinically it may be more beneficial to rely on the subjective variation of the parameters analyzed over time rather than trying to establish a reference cut-off. 44 , 45 Indeed, when taking a blood sample, it is like capturing a still frame from a movie this is a dynamic process. Each molecule that circulates in our bodies continuously brings a message that has no particular limits or thresholds.…”

Echocardiographic assessment of maximum and minimum left atrial volumes: a population-based study of middle-aged and older subjects without apparent cardiovascular disease

“…Most biological risk markers measured on a continuous scale are likely to display a risk continuum without an exact point above which the risk suddenly increases, and an increased risk can even be detected at levels that are considered to be within the distribution of the healthy population (e.g., serum cholesterol [ 25 ] and blood pressure [ 26 ] ). In fact, the concept of “optimal” prognostic thresholds has recently been questioned [ 27 ]. Risk stratification is not straightforward, as it requires multifactorial considerations in a clinical context and cannot be simplified axiomatically into a threshold of a single parameter.…”