Journal of Clinical & Experimental Nephrology

ECG-based Potassium Measurement is Unaffected by Errors in Blood Potassium Measurement during Hemodialysis

Background: Potassium abnormalities can cause lifethreatening arrhythmias. Measuring potassium requires access to blood. We have developed methods measuring potassium noninvasively using the processed, signalaveraged ECG. Four patients, in a larger study, were found to have unexpected discrepancies between measured blood potassium and ECG-derived estimated potassium values.
Methods: Of 240 patients enrolled at 17 sites in the PORTEND (REVEAL-HD) study, 200 wore a continuouslyrecording, single-lead, wireless ECG patch. Blood for chemistries was obtained once before, twice during and once after dialysis. Complete blood test and ECG data were available for 142 subjects. The general potassium pattern during dialysis was an exponential decay throughout the treatment. Four subjects, whose blood potassium values, but not ECG-based potassium values, deviated from this pattern, are the subjects of this analysis.
Findings: Among 4 patients, at least one blood potassium value declined to 2.6 mmol/l or less during dialysis, and then rebounded unexpectedly, while the ECG-based potassium values were consistent with the expected exponential delay. Three of these four patients were at a single site, suggesting site-specific likelihood of pattern deviation (p=0.04). In each case, BUN and phosphorous blood levels were markedly low, with albumin and calcium unaffected.
Conclusions: These results are compatible with blood drawing errors in which dialyzed blood was obtained from the venous return, rather than from the arterial tubing. A physiologic, ECG-based test that estimates potassium on the basis of the concentration of potassium in the blood surrounding the heart is free from local aberrations and might be a useful potassium monitoring tool in dialysis patients.

Author(s):

Zachi Attia, Paul A Friedman, Dorothy J Ladewig, Guarav C Satam, Jennifer L Dugan, Yehu Sepir, Dan Sadot, Amir Geva, Virend K. Somers, Samuel J. Asirvatham, Michael J Ackerman, Kevin E Bennet, Bhupinder Singh and John J Dillon

Abstract | Full-Text | PDF

Share this