CoChat
Sign Up
Cited by 24Machine Learning in HealthcareHemodynamic Monitoring and TherapyCardiac, Anesthesia and Surgical Outcomes

Prediction of Postoperative Deterioration in Cardiac Surgery Patients Using Electronic Health Record and Physiologic Waveform Data

Open access
Mathis, Michael R., Engoren, Milo C., Williams, Aaron M., Biesterveld, Ben E., Croteau, Alfred J., Cai, Lingrui, Kim, Renaid B., Liu, Gang +4 more·2022·Anesthesiology
DOI 10.1097/aln.0000000000004345PMC PMC10227693
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Postoperative hemodynamic deterioration among cardiac surgical patients can indicate or lead to adverse outcomes. Whereas prediction models for such events using electronic health records or physiologic waveform data are previously described, their combined value remains incompletely defined. The authors hypothesized that models incorporating electronic health record and processed waveform signal data (electrocardiogram lead II, pulse plethysmography, arterial catheter tracing) would yield improved performance versus either modality alone. Methods: Intensive care unit data were reviewed after elective adult cardiac surgical procedures at an academic center between 2013 and 2020. Model features included electronic health record features and physiologic waveforms. Tensor decomposition was used for waveform feature reduction. Machine learning–based prediction models included a 2013 to 2017 training set and a 2017 to 2020 temporal holdout test set. The primary outcome was a postoperative deterioration event, defined as a composite of low cardiac index of less than 2.0 ml min˗1 m˗2, mean arterial pressure of less than 55 mmHg sustained for 120 min or longer, new or escalated inotrope/vasopressor infusion, epinephrine bolus of 1 mg or more, or intensive care unit mortality. Prediction models analyzed data 8 h before events. Results: Among 1,555 cases, 185 (12%) experienced 276 deterioration events, most commonly including low cardiac index (7.0% of patients), new inotrope (1.9%), and sustained hypotension (1.4%). The best performing model on the 2013 to 2017 training set yielded a C-statistic of 0.803 (95% CI, 0.799 to 0.807), although performance was substantially lower in the 2017 to 2020 test set (0.709, 0.705 to 0.712). Test set performance of the combined model was greater than corresponding models limited to solely electronic health record features (0.641; 95% CI, 0.637 to 0.646) or waveform features (0.697; 95% CI, 0.693 to 0.701). Conclusions: Clinical deterioration prediction models combining electronic health record data and waveform data were superior to either modality alone, and performance of combined models was primarily driven by waveform data. Decreased performance of prediction models during temporal validation may be explained by data set shift, a core challenge of healthcare prediction modeling. Editor’s Perspective: What We Already Know about This Topic: What This Article Tells Us That Is New:

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1097/aln.0000000000004345

Cites

47 of 47 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • AI using EHR data has become much more general and accurate in the last few years, allowing prediction of perioperative events and learning effective treatment strategies.

    “AI using EHR data has become much more general and accurate in the last few years,35,36 allowing prediction of perioperative events37–41 and learning effective treatment strategies.42”

    From Potential uses of AI for perioperative nursing handoffs: a qualitative study (2023) · cites this paper as [39]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 3

  1. [34]Kolda TG, Bader BW: MATLAB Tensor Toolbox. Sandia National Laboratories, 2006. Available at: https://www.osti.gov/biblio/1230898. Accessed September 2, 2022.
  2. [46]Otles E, Oh J, Li B, Bochinski M, Joo H, Ortwine J, Shenoy E, Washer L, Young VB, Rao K, Wiens J: Mind the performance gap: Examining dataset shift during prospective validation, Proceedings of the 6th Machine Learning for Healthcare Conference. Edited by Jung K, Yeung S, Sendak M, Sjoding M, PMLR Ranganath R., 2021, pp 506– 34 PMC13213727 42211431
  3. [48]Sandu C, Popescu D, Popescu C: Postcardiac surgery recovery process with reinforcement learning, 2015 19th International Conference on System Theory, Control and Computing. 2015, pp 658– 61

Cited by

Sorted by citations

Related

Authors
Mathis, M. R., Engoren, M. C., Williams, A. M., Biesterveld, B. E., Croteau, A. J., Cai, L., Kim, R. B., Liu, G., Ward, K. R., Najarian, K., Gryak, J., & BCIL Collaborators Group
Affiliations
  • University of Michigan (US)
  • Michigan Medicine (US)
  • Hartford Hospital (US)
ORCID
Venue
Anesthesiology
Publisher
Ovid Technologies (Wolters Kluwer Health)
Published
2022-08-11
Volume
137
Issue
5
Pages
586-601
ISSN
  • 0003-3022 (ISSN-L)
  • 1528-1175
DOI
10.1097/aln.0000000000004345
PMC
PMC10227693
OpenAlex
W4290805428
Cited by
24
References
47
Citations by year
2026: 5 · 2025: 5 · 2024: 5 · 2023: 9
FWCI
1.56
Type
journal article
Language
en
Open access
Locations
  • Anesthesiology (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
  • Deep Blue (University of Michigan) (submittedVersion)
  • University of Michigan Library
Funders
  • National Heart, Lung, and Blood Institute (K01HL141701)
  • National Institute of General Medical Sciences (T32GM007863)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Health Sciences › Medicine › Surgery › Hemodynamic Monitoring and Therapy
  • Health Sciences › Medicine › Cardiology and Cardiovascular Medicine › Cardiac, Anesthesia and Surgical Outcomes
Keywords
cardiac surgery, early warning systems, electronic health records, electrocardiogram, tensor decomposition, cardiac index, hypotension, ICU mortality, prediction model performance, model generalization, multimodal data
MeSH terms
Machine Learning, Adult, Epinephrine, Cardiac Surgical Procedures, Humans, Hypotension, Electronic Health Records

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review