Clinically significant hypokalaemia and the utilisation of cardiac telemetry in hospitalised adults

Abstract

Hypokalaemia is one of the most common electrolyte abnormalities in hospitalised patients, with moderate (2.5–3.0 mmol/L) to severe (< 2.5 mmol/L) deficiency associated with an increased risk of cardiac arrhythmias, prolonged hospitalisation, and inpatient mortality [1]. While clinical guidelines recommend the use of continuous cardiac monitoring (telemetry) for patients with moderate to severe hypokalaemia, there is limited evidence of adherence to these recommendations, and a lack of understanding of the resultant patient outcomes. This study aimed to characterise the clinical features, pattern of telemetry utilisation, and outcomes of adults admitted with clinically significant hypokalaemia at a major Australian tertiary hospital. A retrospective cohort study was conducted at Townsville University Hospital (TUH) between January 2017 and August 2023. Adult inpatients (≥ 18 years) with an initial serum potassium concentration ≤ 3.0 mmol/L were included. Data was extracted from electronic medical records, including demographics, comorbidities, inpatient telemetry use, and hospitalisation outcomes. The primary outcome was telemetry utilisation, with secondary outcomes including time to normalisation of serum potassium concentration, length of stay (LOS), ICU transfer, and inpatient mortality. Statistical analysis included Chi-square tests, logistic regression, log-rank testing, and survival analysis. A total of 5,035 unique patients with a total of 11,143 admissions (1.6% of total admissions) were included, with a mean age of 64.0 years and a median admission potassium of 2.9 mmol/L. Cardiac telemetry was utilised in 3.7% of patients. Cox regression demonstrated that telemetry use was more likely amongst patients with higher Charlson Comorbidity Index (CCI) scores ( p < 0.001) and prior arrhythmia diagnosis ( p = 0.028), but its use was not significantly associated with hypokalaemia severity. Telemetry use was independently associated with longer hospital stay ( p = 0.016) but not with mortality. No demographic factor predicted shorter LOS. Survival analysis showed no significant difference between inpatient mortality across hypokalaemia severity groups (moderate or severe) ( p = 0.190). Faster correction of serum potassium was weakly associated with shorter LOS. In this large cohort of hospitalised adults with moderate to severe hypokalaemia, cardiac telemetry was underutilised and not consistently applied according to the severity of electrolyte derangement. Instead, its use was more closely tied to patient comorbidities. Hypokalaemia severity alone was not an independent predictor of inpatient mortality or LOS. These findings support the need for standardised risk-stratified approaches to application of cardiac telemetry monitoring in hypokalaemia management, particularly in tertiary centres serving broad and remote populations.