The mean percent error for measuring actual hemoglobin loss using the device was within 12%.  Representative images using different lighting conditions including measures of brightness (Bv). (Image source: Anesthesia & Analgesia)
The mean percent error for measuring actual hemoglobin loss using the device was within 12%. Representative images using different lighting conditions including measures of brightness (Bv). (Image source: Anesthesia & Analgesia)

Blood is lost during a surgical procedure.  The surgeon wipes it up with sponges, suctions it, or lets it fall on the floor.  We, the anesthesiologist, must then determine how much blood is lost.  This is an inaccurate science at best!  Sponges can be weighed and the dry weight can be subtracted, though it’s assumed that the hemoglobin concentration does not vary.  Dr. Gerhardt Konig, Department of Anesthesiology, University of Pittsburgh School of Medicine, UPMC Department of Anesthesiology, Pittsburgh, Pennsylvania, and colleagues tested the accuracy of a new device, the Triton SystemTM, that can measure hemoglobin mass absorbed by surgical sponges from an image using computer vision system, implemented on an iPad, and machine-learning algorithms implemented on a remote cloud server. This Triton SystemTM was tested in the laboratory in a report in this month’s issue of Anesthesia & Analgesia in the article titled “In Vitro Evaluation of a Novel System for Monitoring Surgical Hemoglobin Loss.”

Blood samples from outdated units of human packed red blood cells and plasma of known hemoglobin concentrations and volumes were reconstituted and applied to laparotomy sponges.  Differing amounts of saline were also added in order to account for varying saturation levels due to hemodilution and/or irrigation use.  The sponges were then scanned using the Triton System™. Image capturing was performed under three different ambient lighting conditions in order to test for changes in accuracy due to the presence of different lighting conditions in different operating rooms.

Hemoglobin analysis as measured by the system correlated well with actual hemoglobin value.  Measures of hemoglobin loss utilizing the Triton System™ displayed a mean percent error of 12%. Values with the device were consistent across the three different ambient light conditions.  Accuracy increased as more sponges were used.

For this particular study, banked blood was used. The tests were conducted in a controlled laboratory setting. However, conditions in a laboratory are well controlled. How does the device work in the hectic environment of a busy operating room handling cases with significant blood loss? Simply put, does it work in the OR? That is the subject of the next post in AA2Day.