Maintaining sample integrity and producing a homogenous sample are key to consistent and reliable results. However, manual mixing has shown to provide inconsistent results.* That was the conclusion of a study conducted at Adventist Midwest Health, part of AMITA Health, a four-hospital system serving the western Chicago suburbs.
The blood gas analyzers at Adventist Midwest Health are managed by the respiratory departments at each of the hospitals. The respiratory departments serve all the other units, providing a number of tests, including total hemoglobin (tHb) measurements for surgery. Adventist currently deploys a number of Radiometer ABL blood gas analyzers, including the ABL800 FLEX, the ABL90 FLEX, and the ABL80 FLEX CO-OX.
The idea behind the study came from Lou Ann Veldt, Respiratory Therapy Manager at Adventist La Grange Memorial Hospital, when she came across an odd tHb result from post-surgery labs for a patient. Veldt noticed that the CBC from the lab showed a tHb of 8.0 g/dL, the same as the pre-surgery value. At the same time, the ABL90 FLEX was showing a result of 16.0 g/dL as part of the post- surgery labs. This baffled Veldt, as the method comparisons that were originally run on the ABL90 FLEX perfectly correlated to the lab. Veldt’s first thought was that "something must have changed with the instrument."
To test the hypothesis of a potential issue with the tHb measurement on the ABL90 FLEX, Veldt performed a precision study with the central lab method using the same syringes that were used in the hospital, a basic ABG syringe from Portex. Veldt took 20 samples, mixed them thoroughly, ran them on the ABL90 FLEX and sent them to the lab for a comparison value. Veldt’s method comparison showed perfect correlation with the lab. So what contributed to the discrepancy?
Veldt performed a second method comparison with one change: she asked her colleagues to mix and run the samples on the ABL90 FLEX as they normally would. She observed each of her colleagues and noticed that each had their own technique. After 20 samples were run on the ABL90 FLEX, she sent the samples to the lab. Unlike her previous method comparison, there was a large variability in the correlation with the lab results. Could the observed discrepancy be due to the variability in mixing?
A third method comparison was performed, this time switching out the Portex syringes with Radiometer safePICO syringes. This way, automatic mixing could be incorporated, thus standardizing the mixing of the sample. Again, Veldt asked her colleagues to run 20 samples, except this time she asked them to use the automatic mixer on the ABL90 FLEX prior to running the sample. This time, when compared to the lab, the results correlated perfectly.
This was a tremendous "a-ha" moment for Veldt. In her 32 years as a respiratory therapist, she had data showing that something as simple as improper mixing could have a huge impact on the clinical decision for any given patient.
According to Veldt, incorporating Radiometer's automated mixing ultimately not only reduced preanalytical errors and secure sample integrity; it helped to enhance care at Adventist.
*Grenache, David G., et al. Integrated and automatic mixing of whole blood: An evaluation of a novel blood gas analyzer. Clinica Chimica Acta. 375 (2007) 153 - 157