Articles
Subphenotypes in acute respiratory distress syndrome: latent class analysis of data from two randomised controlled trials

https://doi.org/10.1016/S2213-2600(14)70097-9Get rights and content

Summary

Background

Subphenotypes have been identified within heterogeneous diseases such as asthma and breast cancer, with important therapeutic implications. We assessed whether subphenotypes exist within acute respiratory distress syndrome (ARDS), another heterogeneous disorder.

Methods

We used data from two ARDS randomised controlled trials (ARMA trial and ALVEOLI trial), sponsored by the National Heart, Lung, and Blood Institute. We applied latent class modelling to identify subphenotypes using clinical and biological data. We modelled data from both studies independently. We then tested the association of subphenotypes with clinical outcomes in both cohorts and with the response to positive end-expiratory pressure (PEEP) in the ALVEOLI cohort.

Findings

We analysed data for 1022 patients: 473 in the ARMA cohort and 549 in the ALVEOLI cohort. Independent latent class models indicated that a two-class (ie, two subphenotype) model was the best fit for both cohorts. In both cohorts, we identified a hyperinflammatory subphenotype (phenotype 2) that was characterised by higher plasma concentrations of inflammatory biomarkers, a higher prevalence of vasopressor use, lower serum bicarbonate concentrations, and a higher prevalence of sepsis than phenotype 1. Participants in phenotype 2 had higher mortality and fewer ventilator-free days and organ failure-free days in both cohorts than did those in phenotype 1 (p<0·007 for all). In the ALVEOLI cohort, the effects of ventilation strategy (high PEEP vs low PEEP) on mortality, ventilator-free days and organ failure-free days differed by phenotype (p=0·049 for mortality, p=0·018 for ventilator-free days, p=0·003 for organ-failure-free days).

Interpretation

We have identified two subphenotypes within ARDS, one of which is categorised by more severe inflammation, shock, and metabolic acidosis and by worse clinical outcomes. Response to treatment in a randomised trial of PEEP strategies differed on the basis of subphenotype. Identification of ARDS subphenotypes might be useful in selecting patients for future clinical trials.

Funding

National Institutes of Health.

Introduction

Acute respiratory distress syndrome (ARDS) is a heterogeneous syndrome first identified in 1967 and defined by the clinical criteria of bilateral pulmonary opacities on chest radiograph, arterial hypoxaemia (partial pressure of arterial oxygen [PaO2] to fraction of inspired oxygen [FiO2] ratio <300), and exclusion of cardiac failure as the primary cause of the syndrome.1, 2, 3 This definition was derived empirically on the basis of clinical experience, with the hypothesis that it would identify patients with non-cardiogenic pulmonary oedema, characterised by increased protein permeability of the alveolar–capillary membrane. Since the time of the original identification of ARDS, and increasingly during the past two decades, there has been recognition of the clinical and biological heterogeneity within ARDS;4, 5 this heterogeneity might reflect our incomplete understanding of the biology of ARDS and probably contributes to the poor track record of phase 2 and 3 trials of new treatments for patients with ARDS.6 As a result, some investigators have proposed subdividing ARDS on the basis of clinical risk factors, or by direct versus indirect cause of lung injury. However, no consensus exists on the appropriate approach to reduce ARDS heterogeneity.

By contrast with ARDS, research in airways disease and cancer has made substantial progress towards identifying subphenotypes of disease, with important therapeutic implications. For example, subphenotypes based on the presence or absence of Th2-dependent inflammation have been identified in asthma, with important mechanistic and therapeutic implications.7 This insight has led to new targeted treatments, such as a monoclonal antibody to interleukin-13, which is especially effective in individuals with Th2-predominant inflammation.8 Despite widespread recognition of the heterogeneity within common critical illness syndromes such as sepsis and ARDS, and some evidence suggesting that subphenotypes might exist within severe sepsis,6, 9, 10 little data are available for whether such subphenotypes exist in ARDS.

Latent class analysis is a well-validated statistical technique that uses mixture modelling to find the best-fitting model for a set of data, based on the hypothesis that the data contain several unobserved groups or classes. The statistical approaches underlying this method were originally developed more than a century ago by investigators analysing whether a population of crabs consisted of two subspecies.11 By contrast with traditional regression analyses, in which the goal is to understand the association between pre-specified independent variables and a known outcome, latent class analysis models ask whether there are subgroups of patients defined by a combination of the baseline variables, without mandating consideration of the outcome. Latent class-based methods have been extensively used in the social sciences and in other medical disciplines,12, 13 for instance in the identification of asthma subphenotypes,14 but have not been used extensively in critical care. We sought to capitalise on the wealth of clinical and biological data available from two National Heart, Lung, and Blood Institute (NHLBI)-sponsored ARDS Network randomised controlled trials by using latent class analysis methods to attempt to identify and validate novel subphenotypes of ARDS and to test their association with clinical outcomes and response to treatment.

Section snippets

Study design

Clinical and biological data were obtained from patients enrolled in the NHLBI ARDS Network's randomised controlled trials of lower versus higher tidal volume ventilation (trial referred to here as ARMA)15, 16, 17 and higher versus lower positive end-expiratory pressure (PEEP; trial referred to here as ALVEOLI).18 Some patients in the ARMA trial were co-enrolled in a trial of lisofylline versus placebo;17 after 194 patients were enrolled in that trial, the ventilator trial was discontinued

Results

Baseline clinical characteristics for patients in both cohorts are shown in table 1. We noted several statistically significant differences between the two cohorts, including primary ARDS risk factor, severity of illness as measured by Acute Physiology and Chronic Health Evaluation (APACHE III) scores, and prevalence of vasopressor use at enrolment. Additionally, several ventilator variables at randomisation differed substantially between the two cohorts, probably indicative of changes in

Discussion

Our findings suggest the existence of two different subphenotypes in patients with ARDS. These two subphenotypes have very different natural histories, clinical and biological characteristics, clinical outcomes, and response to treatment, fulfilling the criteria necessary to define a subphenotype.7 Clinicians caring for patients with ARDS and researchers studying ARDS have long appreciated the heterogeneity within this complex syndrome, but the critical care community has not had empirical data

References (41)

  • SE Wenzel

    Asthma phenotypes: the evolution from clinical to molecular approaches

    Nat Med

    (2012)
  • J Corren et al.

    Lebrikizumab treatment in adults with asthma

    N Engl J Med

    (2011)
  • HR Wong et al.

    Validation of a gene expression-based subclassification strategy for pediatric septic shock

    Crit Care Med

    (2011)
  • KN Iskander et al.

    Sepsis: multiple abnormalities, heterogeneous responses, and evolving understanding

    Physiol Rev

    (2013)
  • G McLachlan et al.

    Finite mixture models

    (2000)
  • K Ezzedine et al.

    Latent class analysis of a series of 717 patients with vitiligo allows the identification of two clinical subtypes

    Pigment Cell Melanoma Res

    (2014)
  • R Rindskopf et al.

    The value of latent class analysis in medical diagnosis

    Stat Med

    (1986)
  • M Depner et al.

    Clinical and epidemiologic phenotypes of childhood asthma

    Am J Respir Crit Care Med

    (2014)
  • Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome

    N Engl J Med

    (2000)
  • Ketoconazole for early treatment of acute lung injury and acute respiratory distress syndrome: a randomized controlled trial

    JAMA

    (2000)
  • Cited by (923)

    View all citing articles on Scopus
    View full text