Podcasts

Join Now

Using a point-of-care glucometer to identify septic peritonitis in the dog | VETgirl Veterinary Continuing Education Podcasts

Need Help?

Do you use a AlphaTRAK 2 glucometer in your veterinary clinic? In the veterinary emergency room, many of us utilize the veterinary handheld point-of-care (POC) glucometers to obtain rapid glucose measurements, as it only requires a tiny volume of blood. Not all hospitals have the benefit of having expensive lab analyzers and instead rely on the POC glucometers for glucose measuring. However, it’s important to note that the accuracy of these POC glucometers can be affected by various factors such as the concentration of red blood cells present in the sample (e.g., anemia, hemoconcentration) and various medications. The POC glucometer utilizes a different mechanism by which to measure blood glucose levels than our traditional lab analyzers.

Why is this important in the emergent, critically ill, vomiting dog or cat? Back in 2003, Bonczynski et al compared peritoneal fluid to peripheral blood pH, bicarbonate, glucose, and lactate concentrations, and utilized this information as a diagnostic tool for septic peritonitis in dogs and cats. However, knowing that these POC glucometers can be affected by red blood cell count, it stands to reason that differences in glucose measurement could introduce error in measuring the septic abdomen glucose difference. So, Koenig et al out of University of Georgia (my internmate, and smartest criticalist and internist ever) performed a a study called “Usefulness of whole blood, plasma, peritoneal fluid, and peritoneal fluid supernatant glucose concentrations obtained by a veterinary point-of-care glucometer to identify septic peritonitis in dogs with peritoneal effusion” to evaluate the accuracy of the septic abdominal effusion calculation using the POC glucometer on blood and abdominal effusion, and then devised an adjustment to the calculation, if needed.

In this study, venous blood was collected into lithium heparin collection tubes and analyzed within 5 minutes after collection with the POC glucometer. The whole blood was then spun down and the POC glucometer used to obtain glucose measurements on the plasma. Peritoneal fluid samples were obtained concurrently at the time of venous blood collection into lithium heparin collection tubes and analyzed within 5 minutes after collection with the POC glucometer. The peritoneal effusion was also spun down to create a peritoneal effusion supernatant, and the POC glucometer was used to analyze glucose concentrations in the peritoneal effusion supernatant. Samples of the whole blood and peritoneal fluid were also analyzed for PCV and total protein concentration via refractometry.

Once this data was obtained, the glucose concentration differences for three groups were calculated:

  • whole blood compared to peritoneal effusion (WB-PF)
  • plasma compared to peritoneal effusion (P-PF) and
  • plasma compared to the peritoneal supernatant (P-PFS).

In this study, the cases were retrospectively analyzed to assign the subjects as having septic abdominal effusion by documentation of intracellular bacteria on cytologic examination, positive bacterial culture growth from the abdominal effusion, identification of gastrointestinal perforation at the time of exploratory surgery, or on necropsy. Those that failed to meet these criteria were instead diagnosed as having sterile abdominal effusion. Only 17 of the 39 dogs (43.6%) with abdominal effusion were confirmed as having septic abdominal effusion and 22 were confirmed as having a non-septic abdominal effusion. Four dogs had evidence of intracellular bacteria on cytologic examination of abdominal effusion, but did not have positive growth on bacterial culture media and were hence excluded from this study due to these conflicting results.

So, what’d they find? When the authors evaluated the POC glucose readings on whole blood versus on the corresponding plasma of that sample, they found significant differences in the measurements. On closer inspection, 4 of the samples were anemic, which is known to falsely elevate the glucose reading on whole blood when compared to the glucose reading of its plasma. What Koenig and all found was that the peritoneal effusion did not require being spun prior to running the sample through the POC glucometer to obtain a reliable glucose measurement (so save the step!).

When the authors calculated the difference between the peritoneal effusion and the peripheral blood sugar they found the following: For whole blood samples compared to peritoneal effusion samples (WB-PF), all cases determined by the glucose difference formula to be septic were indeed septic. However, 10 of the total 17 septic cases were incorrectly identified by this formula as being “non-septic.” (In other words, the specificity of this test using whole blood was excellent but the sensitivity was extremely low). A complicating factor in this study is that 5 of these 10 misdiagnosed patients were on antimicrobial therapy for unspecified amounts of time, which may have skewed the diagnosis of sepsis.

When the plasma samples were compared to the peritoneal effusion samples (P-PF), some of the cases classified by this formula as septic were found later to be non-septic though the paper doesn’t specify how many were incorrectly identified as “septic” based on this method alone. Only 2 of the 17 confirmed septic effusions were incorrectly identified as “non-septic”. For the plasma samples compared to peritoneal effusion supernatant samples (P-PFS), only 3 of the 17 confirmed septic effusions were incorrectly identified as “non septic”. 1 of the 3 patients had a prior intestinal surgical dehiscence. It remains unknown how many days post-operatively this cut off definer of glucose in the effusion to peripheral blood glucose will be unreliable in the post-operative period, but it is not recommended to use this glucose comparison in the post-operative period as it may cause false positive results for diagnosing a septic abdomen.2

Most importantly, when the cutoff definer was increased from 20 mg/dL of glucose to 38 mg/dL, the specificity of the P-PF and P-PSF calculations were improved. Another complicating factor was that some of the peritoneal effusion samples were too low to read and were instead assigned a value of 20 mg/dL which is the lowest the glucometer can read instead of reflecting the true glucose level.

So, what can we take away from this VETgirl podcast?
When using a lab analyzer, a glucose difference of > 20 mg/dL between peripheral blood and peritoneal effusion is diagnostic for septic peritonitis. Using a POC glucometer in place of the lab analyzer can provide 100% specificity only if the cutoff is raised to 38 mg/dL. However, use of the POC glucometer, even with the adjusted cutoff of 38 mg/dL results in a lower sensitivity than if the lab analyzer were used with the original cutoff value.

Conclusion
Although the gold standard in identifying a septic abdomen is to have positive bacteria growth on culture of the abdominal effusion sample, this particular way of testing takes too long to be of use in making emergency medical decisions. Instead, by comparing the difference in glucose levels between peripheral blood and the abdominal effusion, we can quickly rule in or out the presence of an septic abdominal effusion. When the glucose of the abdominal effusion is found to be at least 20 mg/dL less than the peripheral blood glucose, the effusion can be diagnosed as a septic abdominal effusion with 100% sensitivity and 100% specificity in canines.1 In this referenced landmark study, the glucose measurements were performed using a lab analyzer, which may or may not be available at all animal hospitals. When calculating the glucose difference between abdominal effusion and peripheral blood for the purpose of diagnosing a septic abdominal effusion, the use of a handheld glucometer in place of the lab analyzers cannot provide reliable results.

This study highlights the fact that using POC glucometers for identifying a septic abdomen may create false negatives and false positives. The authors recommend using plasma samples instead of whole blood for glucose measurements of the peripheral blood. They also recommend that when comparing plasma glucose levels to the effusion or to the effusion supernatant glucose levels to use a broader cutoff range of > 38 mg/dL. The fact that this method creates both false negatives and false positives makes me want to take the extra time to run all samples through a lab analyzer instead of trying to make them work with the POC glucometer. If a lab analyzer is not available, this method may be of clinical use, but based on this study’s findings on the performance of the POC glucometer in this calculation, I would strongly recommend employing additional means of diagnosis to support the findings of this calculation such as performing in-house cytologic evaluation of a spun effusion sample.

References:
1. Bonczynski JJ, Ludwig LL, Barton LJ, et. Al. Comparison of peritoneal fluid and peripheral blood pH, bicarbonate, glucose, and lactate concentration as a diagnostic tool for septic peritonitis in dogs and cats. Vet Surg 2003;32:161-166.
2. Szabo SD, Jermyn K, Neel J, et.al. Evaluation of post celiotomy peritoneal drain fluid volume, cytology, and blood-to-peritoneal fluid lactate and glucose difference in normal dogs. Vet Surg 2011;40:444-449.
3. Koenig A, Verlander LL. Usefulness of whole blood, plasma, peritoneal fluid, and peritoneal fluid supernatant glucose concentrations obtained by a veterinary point-of-care glucometer to identify septic peritonitis in dogs with peritoneal effusion. J Am Vet Med Assoc 2015;247:1027-1032

Abbreviations:
POC Point of care
P-PF Plasma versus peritoneal fluid
P-PFS Plasma versus peritoneal fluid supernatant
WB-PF Whole blood versus peritoneal fluid

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