AK+ Guard™ Pilot Study in Chronic Kidney Disease: Outpatient Diagnostic Accuracy and Remote Monitoring

Official title

Diagnostic Accuracy, Usability, Patient Compliance, and System Reliability of the AK+ Guard™ ECG Application for Hyperkalemia Prediction in Ambulatory Chronic Kidney Disease Patients (Outpatient and Remote Monitoring Arms)

Overview

The goal of this observational pilot study is to evaluate the investigational AK+ Guard™ software as a medical device (SaMD) for detection of moderate to severe hyperkalemia (serum potassium (K+) ≥ 6.5 mmol/L) in adults with chronic kidney disease (CKD). Study objectives are: * Arm 2A (Outpatient Diagnostic Accuracy): To generate a preliminary, real-world signal of the diagnostic performance of AK+ Guard™ when used in an ambulatory CKD cohort for identifying clinically significant hyperkalemia episodes (serum K+ ≥ 6.5 mmol/L) at the time of an outpatient laboratory draw. * Arm 2B (Remote Patient Monitoring): To assess participant compliance, usability, and end-to-end system reliability of AK+ Guard™ when deployed for daily remote monitoring of CKD patients outside the clinical environment for up to four weeks.

Detailed description

The AK+ Guard™ application is an investigational Software as a Medical Device (SaMD) developed by AccurKardia, Inc. (New York, USA). It is designed to aid in the diagnosis of moderate to severe hyperkalemia (serum potassium ≥ 6.5 mmol/L) in adults using artificial intelligence analysis of Lead I electrocardiogram (ECG) data. The software is intended as a clinical decision support tool for adults at risk of hyperkalemia, including but not limited to those with end-stage renal disease (ESRD), chronic kidney disease (CKD) on renin-angiotensin-aldosterone system inhibitors, heart failure, adrenal disorders, and patients taking aldosterone synthase inhibitors. The AK+ Guard™ application analyzes only Lead I ECG signals, which may be obtained from standard 12-lead ECGs, clinical-grade ECG devices with Lead I capability, or consumer wearable devices (e.g., Apple Watch).

This pilot study is designed to generate preliminary real-world evidence on both the diagnostic performance of the application and its usability in daily life. The study includes two arms:

Arm 2A - Outpatient Diagnostic Accuracy:

* Estimate sensitivity, specificity, PPV, and NPV of AK+ Guard™ in ambulatory CKD stages III-IV * Evaluate interoperability across selected Lead I ECG capture devices (Apple Watch, HeartBeam, and 12 lead reference) * Characterize system reliability metrics (upload success, algorithm runtime) in the outpatient setting

Arm 2B - Remote Patient Monitoring

* Quantify participant compliance, data completeness, and attrition during four week remote monitoring * Evaluate user experience using System Usability Scale (SUS) score, Net Promoter Score (NPS), and semi structured feedback * Capture technical performance metrics in a real world, unsupervised context

Interventions

• Device AK+ Guard™ ECG Application
  An investigational Software as a Medical Device (SaMD) that analyzes Lead I electrocardiogram (ECG) data using artificial intelligence to detect moderate to severe hyperkalemia (serum potassium ≥ 6.5 mmol/L). ECG inputs may be obtained from standard 12-lead ECGs, clinical-grade ECG devices with Lead I capability, or consumer wearable devices (e.g., Apple Watch, HeartBeam).

Primary outcome measures

• Arm 2A: Sensitivity and specificity of AK+ Guard™ for detecting serum potassium greater than or equal to 6.5 mmol/L [Time frame: Baseline (Day 0)]
• Arm 2B: (a) Daily compliance rate (proportion of completed ECGs out of scheduled ECGs) [Time frame: Daily monitoring up to 4 weeks (Day 28)]
• Arm 2B: (b) Mean System Usability Scale (SUS) score [Time frame: Study completion at Week 4 (Day 28)]
• Arm 2B: (c) Net Promoter Score (NPS) [Time frame: Study completion at Week 4 (Day 28)]

Secondary outcome measures (9)

• Arm 2A: Positive Predictive Value (PPV) [Time frame: Baseline (Day 0)]
• Arm 2A: Negative Predictive Value (NPV) [Time frame: Baseline (Day 0)]
• Arm 2A: Device specific concordance [Time frame: Baseline (Day 0)]
• Arm 2A: Algorithm runtime [Time frame: Baseline (Day 0)]
• Arm 2A: Upload success rate [Time frame: Baseline (Day 0)]
• Arm 2B: Data completeness [Time frame: Daily monitoring up to 4 weeks (Day 28)]
• Arm 2B: System reliability (application ("app") crash rate, upload failure less than 2 percent (%)) [Time frame: Daily monitoring up to 4 weeks (Day 28)]
• Arm 2B: Number of service calls per participant [Time frame: Daily monitoring up to 4 weeks (Day 28)]
• Arm 2B: False positive alarm rate (less than 20%) [Time frame: Daily monitoring up to 4 weeks (Day 28)]

Eligibility criteria

Inclusion criteria

(Arm 2A - Outpatient Diagnostic Accuracy)

• Age 22 years or older
• CKD stages III-IV managed at Central Jersey Kidney Care outpatient clinic
• Scheduled outpatient serum potassium laboratory test
• On RAASi therapy or documented hyperkalemia (K+ ≥ 5.5 mmol/L) within the past 12 months
• Able to provide written informed consent

(Arm 2B - Remote Patient Monitoring)

• Completion of Arm 2A visit
• Owns an iPhone compatible with the study application

Exclusion Criteria (Both arms):

• Age 21 years or younger
• Pacemaker or implantable cardioverter defibrillator
• Pre existing Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), Intraventricular Conduction Delay (IVCD), or clinically significant hypocalcemia
• Potassium lowering treatment administered before Lead I ECG acquisition
• Trauma, acute events, or active interventions altering potassium homeostasis
• Physical limitation precluding ECG acquisition

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: No

Study design

Observational model

Cohort

Study locations

United States · 1 center

• Central Jersey Kidney Care - Hypertension & Nephrology Associates — Eatontown

Identifiers

NCT: NCT07210021 · AKG CKD 002A, AKG CKD 002B

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗