Podcasts

Join Now

How accurate are point-of-care glucometers in hemodiluted and hemoconcentrated canine blood samples? | VETgirl Veterinary Continuing Education Podcasts

Need Help?

In this VETgirl online veterinary continuing education podcast, we review the importance of anemia or hemoconcentration on blood glucose measurements when using point-of-care (POC) gluometers in our veterinary patients.

Most veterinary emergency hospitals utilize a triage protocol for the initial workup of unstable emergencies. An essential part of this triage workup includes limited blood work [what VETgirl calls the BIG 4: PCV, TS, blood glucose (BG) and BUN/AZO strip)] to evaluate some of the key blood work indicators of disease. The ability to quickly measure blood glucose (BG) on triage in our canine and feline patients can help trigger the administration of life-saving measures such as dextrose administration to a hypoglycemic patient. The portable hand-held glucometers have become widely used in veterinary hospitals for their ability to measure BG levels quickly and with a small volume (one drop) of blood.

The first point-of-care (POC) glucometers utilized by veterinary professionals were from the human medical field and lacked the mathematical algorithms to account for species differences. These devices produced error in measuring canine and feline BG levels. With the newer POC glucometers, we can now account for differences between canine and feline blood samples. However, a continued limitation of the POC glucometer is that blood samples with PCVs higher or lower than the reference range produce a clinically significant error in glucose measurement. In hemoconcentrated samples, the lower ratio of plasma to red blood cells means that less plasma interfaces with the test reagents; this results in a falsely lowered BG measurement compared to lab values. Vice versa, in hemodilute samples, there is a higher ratio of plasma to red blood cells so more plasma interfaces with the reagents, producing a falsely elevated BG level compared to lab values (1).

You’re probably thinking – how in the world am I going to remember that? Well, I clinically think of two cases I see a lot in the ER: the HGE patient and the hemoabdomen patient. Before these POC glucometer studies came out, I always saw a low blood glucose in my HGE patients (making me concerned they were really hemoconcentrated and septic); likewise, in my anemic hemoabdomen patients, I always saw a high blood glucose (what I was claiming was from “stress of death”). So think of the common clinical picture to remember the affect of POC glucometers on BG!

So, Lane et al (2) out of University of Georgia wanted to assess if point-of-care glucometers were accurate in hemodiluted and hemoconcentrated canine blood samples. The purpose of their study was to first devise a mathematical correction formula to improve upon the error created by hemodilute and hemoconcentrated samples on POC glucometer measurements and then to evaluate the clinical impact this formula would have in the clinical setting on real patient samples.

In this study, they used the veterinary AlphaTRAK2 glucometer (what VETgirl uses in her clinic!); this specific glucometer has been shown to produce less error than other portable glucometers in canine patients when compared to lab analysis measurements (3).

They utilized blood from 6 healthy, staff-owned dogs. 60 mls of venous blood was collected in syringes with sodium heparin anticoagulant, and the whole blood samples were immediately analyzed for POC glucose, PCV and total protein levels. The samples were then centrifuged to separate the plasma from the RBCs. Plasma and red blood cells were reconstituted in 17 glass tubes to create varying hemoconcentrations ranging from 0% (all plasma) to 40% PCV. The POC and PCV/TS were then analyzed on each of the reconstituted “whole blood” samples.

As expected, the POC glucose reading decreased with increasing PCV and the POC glucose reading increased with a decreasing PCV. The glucometer was unable to read over half of the samples that had PCVs >80%. All the reconstituted whole blood samples of varying hemoconcentrations were then centrifuged, plasma collected, frozen, and run on a laboratory glucose analyzer within 7 days of collection. The difference between the POC glucose measurements and the corresponding lab analyzer glucose measurements were then plotted against each other and a correction formula was calculated. The formula created is as follows:

Corrected POC Glucose = POCglucose (whole blood) + ([1.6 x PCV]-81.3)

Without the correction formula, the mean difference between POC glucose readings and lab analyzed glucose readings was 41 mg/dL (with a range of 62 to 99 mg/dL). Samples with a PCV in the normal reference range (42% to 56%) had less than a 10 mg/dL error when compared to LABglu.

After using the correction formula on all validation samples, the mean difference dropped from 41 mg/dL to 5.4 mg/dL with a maximum difference of 23 mg/dL. The formula was then applied to 30 blood samples available in heparinized blood tubes from actual patients that had been admitted to the teaching hospital. The patient PCV’s ranged from 12% to 72% with total protein measurements of 4.2 and 9.0 g/dL respectively. The mean difference between POCglu and LABglu fell from 29 mg/dL to 5.5 mg/dL with the use of this correction formula.

This study found that the use of the CorrPOCglu formula significantly reduced the discrepancy between the POCglu measurement and the LABglu measurements. This study also attempted to analyze whether the CorrPOCglu would improve upon clinical action. As example, a POCglu measurement for a patient reveals a clinically significant hypoglycemia that would normally trigger medical intervention (administration of dextrose), but when the CorrPOCglu formula is applied to the POCglu measurement, the resultant glucose measurement is actually less dramatic and not within the range to warrant medical intervention. The study’s results from this analysis were not explained in great detail, but it appears that the CorrPOCglu helped to correct the hypoglycemic measurements enough to prevent unnecessary medical interventions from being pursued based on the level of hypoglycemia.

So, what can we take away from this VETgirl podcast?
The study findings supported the previously proven trend between POCglu measurements being falsely elevated in hemodilute samples and falsely lowered in hemoconcentrated samples. Utilizing the corrPOCglu formula will help to limit the exaggeration of false hypoglycemia or false hyperglycemia obtained from whole blood samples that are either hemoconcentrated or hemodiluted respectively.

In conclusion, this study addresses a concern regarding glucose measurements obtained at time of triage when we have not had enough time or blood sample to analyze the patient’s PCV/TS and/or lab analysis of the glucose measurement. Perhaps we should start using this formula within our triage protocols to obtain larger sample sizes for analysis on the clinical usefulness of this formula.
References:
1. Paul AEH, Shiel RE, Juvet F, et. Al. Effect of hematocrit on accuracy of two point-of-care glucometers for use in dogs. Am J of Vet Res 2011;72:1204-1208.
2. Lane SL, Koenig A, Brainard B. Formulation and validation of a predictive model to correct blood glucose concentrations obtained with a veterinary point-of-care glucometer in hemodiluted and hemoconcentrated canine blood samples. J Am Vet Med Assoc 2015; 246(3):307-312.
3. Cohen TA, Nelson RW, Kass PH, et al. Evaluation of six portable blood glucose meters in measuring blood glucose concentration in dogs. J Am Vet Med Assoc 2009;235:276-280.

Abbreviations:
POC: Point of care
POCglu: blood glucose measurement from the point-of-care glucometer
LABglu: plasma glucose measurement from a clinical laboratory biochemical analyzer
CorrPOCglu: blood glucose measurement after applying the correction formula to the point-of-care glucometer measurement

Only VETgirl members can leave comments. Sign In or Join VETgirl now!