Hemolysis risk after packed red blood cells transfusion with infusion pumps*

ABSTRACT Objective: To evaluate the hemolysis biomarkers of packed red blood cells transfused by two different linear peristaltic infusion pumps at two infusion rates. Method: An experimental and randomized study was designed simulating the clinical practice of transfusion. Two linear peristaltic infusion pumps from different manufactures were studied in triplicate at 100 mL/h and 300mL/h infusion rates. The chosen hemolysis biomarkers were total hemoglobin, free hemoglobin, hematocrit, potassium and degree of hemolysis. They were analyzed before and after each infusion. Results: Potassium showed statistically significant variations in all scenarios of the experiment (P<0.010). In a separated analysis, potassium increased mainly at 300mL/h rate (P=0.021) and free hemoglobin had significant variation when comparing infusion pumps from different manufacturers (P=0.026). Although hematocrit, total hemoglobin and degree of hemolysis had increased after infusion, no statistically significance variations were identified. Conclusions: Hemolysis risk induced by a linear peristaltic infusion pump was identified by an increase in free hemoglobin and potassium markers. As the potassium biomarker is often increased in aged packed red blood cells, we do not recommend using them in this scenario. Additional studies should be performed about other markers and using larger samples in order to reinforce the transfusion practice in nursing.


Introduction
Packed red blood cell (RBC) transfusion is often indicated as supportive care in different clinical scenarios of anemia as a purpose of increasing the oxygen-carrying capacity of erythrocytes (1)(2) . Hemolysis is one of the clinical complications from blood transfusion therapy and it is related to functional reduction of hemoglobin. Red cell damage may release potassium and free hemoglobin to the patient plasma, possibly causing clinical complications in renal and cardiovascular systems (3)(4) .
Packed RBC is often delivered by gravitational infusion sets and in some scenarios with additional aid of positive or negative external pressures sets (5)(6) . However, the safety of these pressure devices is not clear, once there are several factors that can influence erythrocyte integrity (3,6) .
Infusion pumps (IPs) are among the most commonly used devices for controlling fluid infusion.
Several models of IPs are available in the market with different propulsion mechanisms, such as peristaltic, shuttle, piston cassette, syringe and diaphragm (6)(7) .

Electronic infusion devices have brought innovation
to the intravenous therapy through rigorous fluid controlling, security in pressure and air alarms; however, the concern with the safety of this equipment in blood transfusions remains (3)(4)8) . The IPs mechanism can also be influenced by the infusion rate, age of RBC and preservative solution, different types of intravenous sets, needle gauge and in-line filters (7)(8) .
The existing evidence does not precisely define the result of the mechanical actions of infusion devices over the red blood cells. Also, there are few and conflicting studies about the impact of using IPs on the integrity of RBCs (8) .
However, other studies with linear peristaltic IP showed low risk of hemolysis and concluded that the evaluated IPs were suitable for transfusion when considering limit of the degree of hemolysis at 0.8%, recommended by Brazilian, European and Canadian regulatory agencies (13)(14)(15)(16) .
Infusion rate is also a variable mentioned in different studies, and it may have effect on the integrity of RBCs. Researchers have observed higher degree of hemolysis when there are higher RBC infusion rates in peristaltic IPs (17)(18) .  software, version 20.0. Variables were pre-tested for equal standard deviations and were estimated to follow a Gaussian distribution according to the Kolmogorov-Smirnov test. The distribution of these outcomes was examined using histograms. Normally distributed data were presented as mean ± standard deviation (SD), whereas non-normally distributed data were presented as medians and interquartile range (IQR).

Hemolysis markers were assessed by repeated measures analysis of variance test (ANOVA) and the
Bonferroni post hoc test to discover which specific means differed one from another. A P value less than 0.05 (P < 0.05) was considered statistically significant.

Results
This study covered the assessment of five markers in the different phases of the experiment conducted in each device (IPA and IPB); and in the rates of 100 mL/h and 300 mL/h selected in each IP.
The outcome variables are regarded in Table 1 that shows the variation of hemolysis biomarkers. The effect of infusion system (C2) and infusion pump (V) on the RBC (C1), regardless of the flow and IP, are summarized in Table 2.
Multiple comparisons were performed with potassium in the three moments of the experiment according to the Bonferroni method, as showed in Figure 1.   These results, shown in Table 2 and Figure 1, indicate the potassium was significantly increased by the IPs.
The level of hemolysis markers of RBC was assessed according to the IPs (IPA and IPB) and infusion rates (100 and 300 mL/h), as shown in Table 3.
Potassium concentration increased regardless on the programed infusion rate, but with a significant increase on experiments at rate of 300mL/h (P=0.021). The other markers did not present statistically significant variation regardless of the infusion rate.
The five selected markers were assessed in the experiment phases according to the IP, as shown in Table 4.

Discussion
The main findings of this study indicate that the most prone to hemolytic events, suggesting that the damage caused to the erythrocytes may occur due to the device mechanism (7-8) .
The RBC quality control is a rigorous phase aimed at assessing the blood component preparation process.
Quality is evaluated through previously set indicators by extrapolating the results of the referred production of the analyzed fraction. Quality control analyses in blood banks must be carried out in at least 1% of the production or in 10 units per month, according to the value representing most of the production. It is worth highlighting that not all blood component units transfused to patients are subjected to quality control assessment (9)(10)15) .
The present study showed that the hematocrit and the total hemoglobin remained within the quality control standard. The degree of hemolysis, according to Brazilian legislation, is set at a maximum of 0.8%, for the last storage day of RBC preserved in CPDA-1 solution (19) . The reference values of potassium and free hemoglobin have not been stablished in literature, so they are not routinely assessed in blood banks. These markers were assessed in the present study according to their alteration during the experiment phases. Although free hemoglobin is not routinely assessed, the analysis thereof is an evaluation stage of degree of hemolysis in some analysis techniques used by blood banks (15) . the whole production. Thus, there is concern with the fact that the lack of analyses in markers throughout the whole production may result in damages to the patient.
The prolonged storage time of RBC could be an additional factor that contributes to increasing hemolysis markers. The intracellular concentration of potassium decreases as the storage time increases, whereas the extracellular concentration grows (16,(20)(21) .
Studies have recommended that blood transfusions in some populations, such as newborns and children, must be performed with recently collected RBCs (at least seven days), in order to preserve the original features of the blood component and to avoid the damaging effects caused by its storage (23)(24) . Potassium  cases and in hypercalcemia cases described in the literature. It is recommended RBC infusion of at most 5mL kg -1 minute -1 to avoid hypercalcemia (22)(23) . Some publications suggest that hemolytic effects are more prone to occur in higher rates (8,17) .
The analyses of the markers according to the infusion pump (Table 4) presented statistically significant increase in the hemolysis markers, such as free hemoglobin in IPA (p=0.022) and potassium in IPB (p=0.026). These changes could be possibly explained by the infusion mechanism of the device.
Other studies corroborate the potassium increase in experiments carried out with infusion pump and relate higher values to longer storage time (11)(12)(13)(14)(15)(16)(17) . A publication described extracellular potassium increase according to the storage time and to the infusion mechanism of the device, regardless of the selected flow (11) . High potassium levels used as hemolysis indicators have been described in studies associated with severe adverse events, such as arrhythmias and cardiocirculatory arrest, in massive and fast blood transfusion performed during emergency situations (21)(22)(23) .
The other markers, such as hematocrit, increased in all the phases of the conducted experiments, but such increase was not statistically significant.
Hematocrit has been associated with the viscosity of the blood components in some publications; thus, the higher the hematocrit value is, the more viscous the blood product (18)(19) . The free hemoglobin marker had increased in all experiment scenarios. These

Conclusion
Hemolysis risk induced by a linear peristaltic infusion pumps was identified in this study by an increase in free hemoglobin and potassium markers.
Potassium appeared to be an important parameter to assess the fragility of the plasmatic membrane of erythrocytes. There was also a significant increase in evolution of the experiments when it was assessed in all scenarios, as well as greater increase predisposition at flow of 300mL/h. As the potassium biomarker is often increased in aged packed red blood cells, we do not recommend using them in this scenario. Additional studies must be conducted on this theme.