Send to

Choose Destination
J Clin Monit Comput. 2018 Feb;32(1):117-126. doi: 10.1007/s10877-017-0001-7. Epub 2017 Feb 22.

Cardiorespiratory instability in monitored step-down unit patients: using cluster analysis to identify patterns of change.

Author information

The University of Texas at Austin, 1710 Red River St., Austin, TX, 78701, USA.
Department of Critical Care Medicine, University of Pittsburgh Schools of Medicine, Pittsburgh, USA.
Auton Lab, Carnegie Mellon University Robotics Institute, Pittsburgh, USA.
Department of Acute and Tertiary Care, University of Pittsburgh Schools of Nursing, Pittsburgh, USA.


Cardiorespiratory instability (CRI) in monitored step-down unit (SDU) patients has a variety of etiologies, and likely manifests in patterns of vital signs (VS) changes. We explored use of clustering techniques to identify patterns in the initial CRI epoch (CRI1; first exceedances of VS beyond stability thresholds after SDU admission) of unstable patients, and inter-cluster differences in admission characteristics and outcomes. Continuous noninvasive monitoring of heart rate (HR), respiratory rate (RR), and pulse oximetry (SpO2) were sampled at 1/20 Hz. We identified CRI1 in 165 patients, employed hierarchical and k-means clustering, tested several clustering solutions, used 10-fold cross validation to establish the best solution and assessed inter-cluster differences in admission characteristics and outcomes. Three clusters (C) were derived: C1) normal/high HR and RR, normal SpO2 (n = 30); C2) normal HR and RR, low SpO2 (n = 103); and C3) low/normal HR, low RR and normal SpO2 (n = 32). Clusters were significantly different based on age (p < 0.001; older patients in C2), number of comorbidities (p = 0.008; more C2 patients had ≥ 2) and hospital length of stay (p = 0.006; C1 patients stayed longer). There were no between-cluster differences in SDU length of stay, or mortality. Three different clusters of VS presentations for CRI1 were identified. Clusters varied on age, number of comorbidities and hospital length of stay. Future study is needed to determine if there are common physiologic underpinnings of VS clusters which might inform clinical decision-making when CRI first manifests.


Cardiorespiratory instability; Cluster analysis; Non-invasive monitoring

[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Springer Icon for PubMed Central
Loading ...
Support Center