0
Critical Care |

Computer-Assisted Electrocardiographic Modelling to Predict Severity of Hyperkalemia

Venu Madhav Velagapudi, MD; Dennis Tighe, MD; Anu Vellanki, MD; Markos Kashiouris, MD; John O Horo, MD; Stephen Baker, MS; Jeffrey Stoff, MD
Author and Funding Information

University of Massachusetts, Worcester, MA


Chest. 2013;144(4_MeetingAbstracts):357A. doi:10.1378/chest.1701530
Text Size: A A A
Published online

Abstract

SESSION TITLE: Critical Care Posters

SESSION TYPE: Original Investigation Poster

PRESENTED ON: Wednesday, October 30, 2013 at 01:30 PM - 02:30 PM

PURPOSE: Diagnosis of hyperkalemia by ECG is subjective and challenging outside of severe hyperkalemia. Little data is available to quantify ECG changes in hyperkalemia. We sought to develop an algorithm to systematically diagnose hyperkalemia using 12 lead ECG.

METHODS: Three hundred and thirty five 12-lead ECGs were retrospectively analyzed from 107 consecutive patients with hyperkalemia ([K+] >5.3 mEq/L). Each subject's baseline ECG (i.e. when normokalemic) served as matched control. The prediction rule was generated with multivariate general linear mixed models. Image analysis utilized the NIH "Image J" software. Image interpreters were blinded to serum potassium results.

RESULTS: The training set included 236 EKGs from 84 patients. Goodness of fit was obtained by multiple linear regression. Predictor variables included T width (p<0.001), new QRS widening>100 msec (p <0.0001), descending T slope (p <0.0001) and P duration (p<0.003). Receiver Operator Curve (ROC) had an area under the curve (AUC) of 0.798 with 95% confidence interval (CI) 0.738 to 0.858. Cut point analysis (maximum sensitivity and specificity product) demonstrated 63% sensitivity and 83% specificity at [K+] 5.91 mEq/L. The validation set included 99 ECGs from 23 patients. The validation AUC was 0.783 (95% CI 0.691 to 0.876).Cut point analysis was maximum at [K+] 5.78 (sensitivity 74%, specificity 76%). The maximum specificity (84%) of the validation model was achieved at [K+] of 5.91 with a respective sensitivity of 63%.

CONCLUSIONS: We developed and validated a hyperkalemia prediction and severity model from ECG measurements. When applied into the model, P-wave duration, QRS width, the width and descending slope of the T-wave can predict and quantify hyperkalemia with reasonable sensitivity and specificity.

CLINICAL IMPLICATIONS: Computer-assisted ECG image interpretation has the potential to serve as a continuous screening tool for life-threatening hyperkalemia in the intensive care unit (ICU) setting.

DISCLOSURE: The following authors have nothing to disclose: Venu Madhav Velagapudi, Dennis Tighe, Anu Vellanki, Markos Kashiouris, John O Horo, Stephen Baker, Jeffrey Stoff

No Product/Research Disclosure Information


Sign In to Access Full Content

MEMBER & INDIVIDUAL SUBSCRIBER

Want Access?

NEW TO CHEST?

Become a CHEST member and receive a FREE subscription as a benefit of membership.

Individuals can purchase this article on ScienceDirect.

Individuals can purchase a subscription to the journal.

Individuals can purchase a subscription to the journal or buy individual articles.

Learn more about membership or Purchase a Full Subscription.

INSTITUTIONAL ACCESS

Institutional access is now available through ScienceDirect and can be purchased at myelsevier.com.

Sign In to Access Full Content

MEMBER & INDIVIDUAL SUBSCRIBER

Want Access?

NEW TO CHEST?

Become a CHEST member and receive a FREE subscription as a benefit of membership.

Individuals can purchase this article on ScienceDirect.

Individuals can purchase a subscription to the journal.

Individuals can purchase a subscription to the journal or buy individual articles.

Learn more about membership or Purchase a Full Subscription.

INSTITUTIONAL ACCESS

Institutional access is now available through ScienceDirect and can be purchased at myelsevier.com.

Figures

Tables

References

NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Sign In to Access Full Content

MEMBER & INDIVIDUAL SUBSCRIBER

Want Access?

NEW TO CHEST?

Become a CHEST member and receive a FREE subscription as a benefit of membership.

Individuals can purchase this article on ScienceDirect.

Individuals can purchase a subscription to the journal.

Individuals can purchase a subscription to the journal or buy individual articles.

Learn more about membership or Purchase a Full Subscription.

INSTITUTIONAL ACCESS

Institutional access is now available through ScienceDirect and can be purchased at myelsevier.com.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543