Article Text

Download PDFPDF
Latent class analysis to predict outcomes after surgery for primary mitral regurgitation: a scientific validation of common sense
  1. David Messika-Zeitoun1,
  2. Vincent Chan2,
  3. Ian G Burwash1
  1. 1 Division of Cardiology, University of Ottawa Heart Institute, Ottawa, Ontario, Canada
  2. 2 Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Ontario, Canada
  1. Correspondence to Dr David Messika-Zeitoun, University of Ottawa Heart Institute, Ottawa, ON K1Y 4W7, Canada; DMessika-zeitoun{at}ottawaheart.ca

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Primary mitral regurgitation (MR) is one of the most common valvular heart lesions and a leading cause for cardiac surgery. Over the last several decades, the management of patients with primary MR has evolved. Moreover, the epidemiology of mitral valve disease has shifted with a marked decrease in the prevalence of rheumatic valve disease and an increase in myxomatous (degenerative) mitral valve disease, which is nowadays the most common aetiology of primary MR. Observational studies have described the negative sequelae of MR such as atrial fibrillation (AF), heart failure, left ventricular and atrial remodelling, left ventricle dysfunction and pulmonary hypertension, which has encouraged surgery to be performed earlier in the disease course.1 2 Thus, surgery is recommended in asymptomatic patients with preserved ejection fraction (EF) in sinus rhythm at low surgical risk if the probability of repair is high and surgery is performed in experienced hands. However, such patients represent only a small proportion of those requiring an intervention, and risk stratification for most patients with primary MR remains an important question.3

Kwak et al have used an interesting methodology, namely latent class analysis (LCA), to identify phenotype groups with different outcomes after mitral valve surgery.4 LCA is a statistical method used to identify subsets of patients (latent groups or classes) within populations that share certain characteristics and experienced similar outcomes. To identify these latent groups, LCA uses study participants’ responses to various indicators (in the present study, mortality and surgical risks factors, respectively). In other words, LCA profiles the population to define distinct classes with similar indicators and responses. Although relying on different statistical principles, LCA shares significant similarities with cluster analysis, another statistical method designed to achieve a similar goal.

The authors retrospectively collected data on 2321 patients with severe, isolated, pure primary MR who …

View Full Text

Footnotes

  • Twitter @none

  • Contributors All authors have contributed to the drafting of this manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests DM-Z received research grant and is a consultant for Edwards Lifesciences. VC and IGB have no disclosures.

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles