This may seem rather obvious, but these researchers found
little formal evidence to support this notion. As such, they performed this decent
but small study.
From their laboratory database, they were able to pull the
records of all adult patients over a few years that had a potassium of >6.5 mEq/L. They included patients that had ECG’s
done within one hour of the blood test and had no treatment for hyperkalemia.
Two blinded emergency specialists reviewed the ECG’s to record
the rate, rhythm, peaked T’s, PR intervals and QRS duration.
They defined short
term adverse events as symptomatic
bradycardia, VT, VF, CPR and/or death within 6 hours of the ECG. Relative risk was calculated to determine the
association between the ECG changes and adverse events.
Results?
They found 28 short
term adverse events in 188 patients with severe hyperkalemia. Most of these
adverse events (22) were bradycardia. There were 4 deaths, and 2 episodes each
of VT & CPR.
An increased likelihood for adverse events were found for:
- Bradycardia RR 12.29
- QRS prolongation RR 4.47
- Junctional rhythm RR 7.46
There was no
statistically significant correlation between isolated peaked T’s and short term adverse events. But all adverse events
were preceded by ECG abnormalities.
So, it looks like bradycardia is the most powerful
predictor?
Not so fast. (Get it…
not so fast…)
In this study, bradycardia was both a predictor and outcome
variable. Therefore, it comes as no surprise that bradycardia is associated
with bradycardia. I’m not sure what to do with this…
The biggest limitation of this study was the small numbers
of meaningful adverse events. As such there are wide confidence intervals. No
study is really “definitive” and this research would officially be far from this standard.
Nevertheless, what
are we to conclude?
An ugly ECG in
the setting of severe hyperkalemia is a
bad thing. But don't go bananas about isolated peaked T’s (in the short term.)
Ok… this is not earth shattering, but does help fill in the research gap.

Covering:
No comments:
Post a Comment