D. Dörffel et al.: J Extra Corpor Technol 2025, 57, 89 – 95 91 and wait until more collected blood was available, or fill the missing volume with already collected red blood cell concentrate from the retransfusion bag. After this washing process, the concentrated red blood cells were pumped from the bowl into the retransfusion bag. The hematocrit of the collected blood and the processed autotransfusion blood was directly measured by the cell saver using an optical method. The volume of the collected blood and autotransfusion blood was also indicated by the cell saver for each wash cycle, as well as a total volume in each case. Blood samples before and after washing from a randomly selected bowl were collected and analyzed at the Centre for Transfusion Medicine and Cell Therapy Berlin. A sample was taken from the collected blood during the aspiration process to fill the selected bowl. A set for taking blood samples for quality control measurements( MAT-QM-SET, AB 1522, LivaNova Ò) was sterilely attached between the reservoir and the centrifuge bowl and a 50 mL perfusor syringe was connected to the three-way walve. 50 mL of blood was drawn into the perfusor syringe, and the syringe remained connected. A serum tube was then attached to the three-way valve and filled during the filling process. Afterward, 50 mL of blood from the perfusor syringe was returned to the system. A sample was also required from the autotransfusion blood. This was taken during the emptying process after the bowl had been processed. A three-way valve was fitted to the supply tube between the retransfusion bag and the bowl, and the sample in the serum tube was transferred into the retransfusion bag during the emptying process of the bowl. The hematocrit was determined in the laboratory using a Sysmex XS-800i Ò hematology device. To determine the elimination rate, the total protein in the samples of collected blood and processed autotransfusion blood was analyzed using the Olympus AU480 Ò. The elimination rate of total protein was calculated based on the following equation( PB = processed blood; CB = collected blood):
Elimination rate of total protein ð % Þ 8
Total protein PB Volume PB 1 � Hematocrit 9
PB
>< 100
>=
¼ 100 �
Total protein CB Volume CB 1 � Hematocrit
100:
CB
>: >;
100
Statistics
The statistical analysis was carried out using the statistical program SPSS version 29.0.1.1 for macOS( IBM SPSS Statistics Ò). Data on hematocrit values in autotransfusion blood and total protein elimination were analyzed separately, but the same statistical methods were used. The descriptive statistics include the number of cases, the minimum, the maximum, the arithmetic means, and the median. The frequency distribution of the values was visualized using histograms. Further tests were carried out to generate hypotheses in an exploratory setting. Scatterplots and Spearman’ s correlation were used to further investigate the correlation between the age of the patients and the hematocrit or total protein elimination. The influence of gender on hematocrit and total protein
Figure 1. Histogram of the distribution of the hematocrits. The black vertical line in the histogram represents the hematocrit target value of 50 %.
elimination was examined using the Phi coefficient. The influence of surgical diagnosis on hematocrit and total protein elimination was investigated using Cramer-V and the Kruskal- Wallis tests. P-values less than 0.05 were considered statistically significant.
Results Enrolment
In 234 of 238 cases, the hematocrit of the processed blood was reported by the cell saver. Of these 234 cases, a diagnosis of surgery was given in 200 cases. In the remaining cases, no surgical diagnosis was entered in the SAP Ò hospital information system. A total of over 50 different surgical diagnoses were included in this study. The most common surgical diagnoses for which cell salvage was used were aortic aneurysm( 49 cases), scoliosis( 16 cases), infection and inflammatory reaction due to joint arthroplasty( 12 cases), and atherosclerosis of the limb arteries( 8 cases). Total protein elimination was reported in 179 of 238 cases. Of these 179 cases, the volumes of collected and processed blood checked on the cell savers matched the data in the hospital information system in 130 cases. This is relevant because the total protein elimination was determined based on the volumes specified in the hospital information system. If the volumes did not match, it must be assumed that an error had occurred during manual data transfer, which is why these cases cannot be included in the quality control. In 110 of these cases, a surgical diagnosis was given.
Hematocrit
of the processed blood
The distribution of hematocrit values in the processed blood is shown in the histogram( shown in Fig. 1). The hematocrit values were between 18 % and 70 %. The arithmetic mean was 56.41 %, and the median was 58.00 %, both of which are within the prescribed target value of 50 %. The hematocrit values were not normally distributed. The 25th percentile was at a hematocrit value of 52 %, the 50th percentile at 58 %, and the 75th percentile at 64 %. Of the 234 cases, 52 did not meet the hematocrit target value, corresponding to approximately 22 %.