Behavioural patterns of electrolyte repletion in intensive care units: lessons from a large electronic dataset

Thomas T Joseph,Krzysztof Laudanski,Matthew Dimeglio,Annmarie Huffenberger

doi:10.1038/s41598-018-30444-3

Abstract

Repletion of electrolytes often depends on provider-specific behavior and hospital policy. We examined the pattern of electrolyte repletion across several intensive care units (ICU) in a large healthcare system from 2010–2015. This included 109 723 potassium repletions, 51 833 magnesium repletions, 2 306 calcium repletions, 8 770 phosphate repletions, and 3 128 249 visit-days over 332 018 visits. Potassium, magnesium, and calcium were usually repleted within the institutional reference range. In contrast, the bulk of phosphate repletion was done with pre-repletion serum level below the reference range. The impact of repletion on post-repletion levels was significant but uniformly small. The pre-repletion serum level had a significant inverse correlation with the post-repletion level of each electrolyte. Potassium, magnesium and phosphate follow-up labs were scheduled in 9–10 hours after their repletion. In contrast, calcium was rechecked in less than 20 minutes. Routine repletion of potassium, magnesium and calcium had no effect on the incidence of tachyarrhythmias. We estimated the expense from electrolyte repletion within the reference range was approximately $1.25 million. Absent a specific clinical indication, repleting electrolytes when the serum concentration are within normative values may represent an avenue for cost savings, staff burden unload and potential reduction in frequency of complications in the ICUs.

Highlights

Repletion of electrolytes often depends on provider-specific behavior and hospital policy
The dataset used in this study was drawn from electronic medical record (EMR) data consisting of vital signs, orders, lab results, and admission dates and times for all intensive care units (ICUs) across three major hospitals in the University of Pennsylvania Health System
Carrying a chronic kidney disease (CKD) diagnosis code was related to an increased pre-repletion serum level of potassium

Summary

Introduction

Repletion of electrolytes often depends on provider-specific behavior and hospital policy. We examined the pattern of electrolyte repletion across several intensive care units (ICU) in a large healthcare system from 2010–2015 This included 109 723 potassium repletions, 51 833 magnesium repletions, 2 306 calcium repletions, 8 770 phosphate repletions, and 3 128 249 visit-days over 332 018 visits. The clinical benefits of these protocols are often much higher than observed, while the frequency of side effects is often underestimated All these factors may affect how routine electrolyte abnormalities are treated, but most studies examining the patterns of electrolyte repletions were performed using small data set[7,8,9,10]. Magnesium and calcium are commonly measured as total serum concentration, even though the ionized forms are the biologically active species[18] These are the reasons why repletion of electrolytes often fails to provide sustainable and tangible clinical effects[12]. We hypothesized that electrolyte repletion practices are linked to significant economic expenses

Methods

Results

Conclusion