Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 27, 2026Open Access

A Smartphone-Based Predictor of Hypokalemia in Pediatric Acute Gastroenteritis: A Pilot Diagnostic Accuracy Study

View Full Paper
Ask AI
Bookmark
Share

Key result

SE-PO predictor detects hypokalemia in pediatric acute gastroenteritis with ~92% sensitivity.

  • AUC 0.988
  • n=530

Why the study?

In low- and middle-income settings, serum potassium results may not be available when initial treatment decisions are made for pediatric acute gastroenteritis admissions.

Does the SE-PO Serum Potassium Predictor accurately predict hypokalemia in children admitted for acute gastroenteritis?

Comparison

SE-PO Serum Potassium Predictor vs venous serum potassium reference standard

Design

Single-center cross-sectional diagnostic accuracy study

Authors

JCJohn Paul T. CuevasBMBernadette C. MacrohonNJNorvie T. Jalani

Discussion

Loading...

Member takes

Overview

Should not yet change practice; leaves open prospective validation before clinical use.

Key Points

  • To evaluate the diagnostic accuracy of the SE-PO Serum Potassium Predictor for hypokalemia in children with acute gastroenteritis.
  • Cross-sectional diagnostic accuracy design
  • Involved 530 children aged 5 months to 17 years
  • Bedside assessments conducted before intravenous fluids and lab results were obtained.
  • Application sensitivity was 92.1% and specificity was 97.3%, indicating high accuracy.
  • Detection rate for moderate and severe hypokalemia was 100%, while mild hypokalemia detection was 76.1%.
  • Mean underestimation by the application was 0.18 mEq/L according to Bland-Altman analysis.

Study Design

Type

Cross-Sectional (n=530)

Multicenter

No

Structured PICO

Does the SE-PO Serum Potassium Predictor accurately predict hypokalemia in children admitted for acute gastroenteritis?

P
Population
530 children aged 5 months to 17 years admitted for acute gastroenteritis at a single tertiary government hospital in the southern Philippines.
E
Exposure
SE-PO Serum Potassium Predictor (an Android application that converts bedside clinical and electrocardiographic findings into a predicted serum potassium value)
C
Comparator
Venous serum potassium measurement (reference standard)
O
Outcome
Diagnostic accuracy for detecting hypokalemia (defined as serum potassium < 3.5 mEq/L)surrogate

Main Result

Effect estimate: AUC 0.988

A smartphone application utilizing bedside clinical and ECG findings demonstrated high preliminary diagnostic accuracy for predicting hypokalemia in pediatric acute gastroenteritis, though external validation is required.

Limitations

  • Single-center pilot estimates evaluated in the same institution where the model was developed
  • Prospective external validation is required before clinical adoption
  • Single-center pilot estimates
  • Model was evaluated in the same institution in which it was developed
  • Detection was lower (76.1%) for mild hypokalemia
  • Specificity decreased to 81.5% among children with hypovolemic shock

Cite This Study

Cuevas et al. (2026) conducted a cross-sectional in pediatric acute gastroenteritis (n=530). SE-PO Serum Potassium Predictor vs. Venous serum potassium was evaluated on Diagnostic accuracy for hypokalemia (<3.5 mEq/L) (AUC 0.988). The SE-PO Serum Potassium Predictor accurately detected hypokalemia in pediatric acute gastroenteritis, demonstrating 92.1% sensitivity, 97.3% specificity, and an AUC of 0.988.

synapsesocial.com/papers/6a67008840bca442e0d4a420https://doi.org/10.5281/zenodo.21475950
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Development and validation of a dynamic inpatient risk prediction model for clinically significant hypokalemia using electronic health record data2019 · 6 citations
  2. 2Prediction of hypokalemia in patients with ST-segment elevation myocardial infarction: development and validation of a model with prehospital applicability2026 · 1 citations
  3. 3Avoidable Serum Potassium Testing in the Cardiac ICU: Development and Testing of a Machine-Learning Model2020 · 2 citations
  4. 4Artificial intelligence-based model for automatic real-time and noninvasive estimation of blood potassium levels in pediatric patients2024 · 1 citations
  5. 5Predicting potassium trajectories for risk monitoring in outpatients with heart failure, diabetes mellitus or chronic kidney disease2025