Clinical Chemistry
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 49: 792-799, 2003; 10.1373/49.5.792
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hirsch, J.
Right arrow Articles by Hempelmann, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hirsch, J.
Right arrow Articles by Hempelmann, G.
Related Collections
Right arrow Other Areas of Clinical Chemistry
Right arrow Hematology
(Clinical Chemistry. 2003;49:792-799.)
© 2003 American Association for Clinical Chemistry, Inc.

Indicators of Erythrocyte Damage after Microwave Warming of Packed Red Blood Cells

Jan Hirsch1,a, Axel Menzebach1, Ingeborg Dorothea Welters1, Gerald Volker Dietrich1, Norbert Katz2 and Gunter Hempelmann1

1 Department of Anesthesiology and Intensive Care Medicine and
2 Institute of Clinical Chemistry, University Hospital, 35385 Giessen, Germany.

aAddress correspondence to this author at: Department of Anesthesiology and Intensive Care Medicine, Universitätsklinikum Giessen, Rudolf Buchheim Strasse 7, 35385 Giessen, Germany. Fax 49-641-99-199-60; e-mail Jan.Hirsch{at}chiru.med.uni-giessen.de.


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results
 Discussion
 References
 
Background: Localized overheating of packed red blood cells (PRBCs) after microwave warming with consequent damage to erythrocytes has been reported. We therefore compared possible cellular markers of erythrocyte damage, as measured by flow cytometry, with laboratory indicators of hemolysis to evaluate the effects of microwave warming on PRBCs.

Methods: PRBC samples were warmed to room temperature or to 37, 42, 47, 52, or 57 °C in a water bath. Flow cytometry was performed after fluorescein labeling using antibodies to spectrin, Ca2+-ATPase, and Na+-K+-ATPase. The forward-to-sideward scatter (FSC/SSC) ratio and antibody binding were evaluated. Plasma free hemoglobin (FHb) and {alpha}-hydroxybutyrate dehydrogenase (HBDH) were measured immediately after heating and after 48 h. In addition, all measurements were made before and after the heating of PRBCs to 35 °C by a microwave blood warmer.

Results: Analysis of 15 000 erythrocytes showed a decrease in the FSC/SSC ratio and antibody binding above 47 °C [at 37 °C, median (SD) of 94.2 (7.4) with 0.07 (0.05)% fluorescein-positive; at 52 °C, median (SD) of 177.0 (19.0) with 18.5 (6.4)% positively gated; P <0.001]. FHb [room temperature, 0.3 (0.2) g/L] was increased 2-fold at 37 and 42 °C, 4-fold at 47 °C, and 25-fold at 52 °C. HBDH increased in parallel. Hemolysis markers showed an additional twofold increase 48 h after heating to 42 and 47 °C. Microwave heating to 35 °C did not produce significant changes of any marker.

Conclusions: All markers of cellular damage were altered after heating to >47 °C, and a substantial part of hemolysis was delayed. The methodology can be used for future testing of other blood warming devices.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results
 Discussion
 References
 
Transfusions of red blood cells are necessary to increase the oxygen-carrying capacity in patients after major trauma or during and after surgery. Units of packed red blood cells (PRBCs) 1 are generally stored at 4 ± 2 °C according to German regulations (1). Cold storage is considered to be interrupted if the temperature of the blood increases above 8 °C. The US Food and Drug Administration regulation recommends 1–6 °C (2). For patients requiring large volumes of PRBCs for transfusion, units must be warmed immediately before transfusion to prevent hypothermia. Especially in emergency situations requiring massive transfusions, this need for warming can lead to considerable delay.

The use of microwaves to warm blood products was propagated soon after the introduction of microwave ovens for other purposes in the mid-1950s and came into regular use in the 1970s. The devices provided shorter heating times than did conventional heating in a water bath, but several reports of complications from overheating of blood products (3)(4)(5) led to widespread abolishment of microwave blood warmers. Considerable debate remains regarding the use of these devices.

The primary concern with microwave blood warmers is the development of hemolysis as a result of general overheating and local hot spots within the blood units. Several studies have evaluated the concentrations of hemolysis markers, such as free hemoglobin (FHb) and lactate dehydrogenase (LDH) in PRBCs and whole blood after microwave heating (4)(5)(6)(7). Recent evaluations showed no clinically relevant difference between blood heated by microwave blood warmers and blood heated by other devices with respect to these markers (6)(7). Consequently, the use of microwave blood warmers is currently increasing.

One shortcoming of these studies is that either hemolysis was measured immediately after heat exposure or the authors did not provide the time of measurement. One experimental study with chicken erythrocytes, however, demonstrated a substantial delayed hemolysis after heat exposure (8). Heating to 51.5 °C for 180 min caused 13% hemolysis measured immediately after heating but almost 100% hemolysis measured after 24 h.

Furthermore, global markers are of limited value for the detection of hot spots as a result of localized overheating in a microwave oven. It has been shown that characteristic changes exist in overheated blood with respect to cell size and hemoglobin distribution (9). These changes are similar to the changes in the blood of patients with severe burns (10)(11). The increasing availability of flow cytometry has made this method a powerful and rapid means to examine changes in large numbers of individual cells. To our knowledge, there have been no previous studies describing cellular changes of erythrocytes after warming of PRBCs.


   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results
 Discussion
 References
 
Samples of PRBCs were warmed to different temperatures from room temperature to 57 °C in a water bath. Room temperature was chosen as the control value because PRBCs that are not actively warmed will, in clinical practice, be transfused after reaching ambient temperature to avoid discomfort and hypothermia of the patient. The concentrations of FHb and the isoenzymes LDH1 and LDH2 [also known as {alpha}-hydroxybutyrate dehydrogenase (HBDH)] were measured immediately after heating and after 48 h. Flow cytometry was used to characterize cells and to evaluate changes in erythrocyte antigen patterns (Na+-K+-ATPase, Ca2+-ATPase, and spectrin) with respect to thermal changes. After we established a detection assay for general and localized overheating, we performed measurements before and after the heating of PRBCs with a commercially available microwave blood warmer that is in regular use in our department.

PRBCs that had been discarded by the blood bank of our institution because of age or undocumented interruption of cool storage were used for the experiments. The units were stored in a refrigerator certified for the storage of PRBCs before and between the measurements. Before the experiments, the units were checked visually for signs of clot formation or hemolysis. Samples were heated by immersing 10-mL samples in closed plastic tubes in a calibrated water bath (Haake DC 10; Haake GmbH) for 60 min. The temperature of the samples was cross-checked with a calibrated thermometer (Testo® 110; Testo GmbH & Co; accuracy ±0.2 °C).

markers of overheating
Samples from 12 units were heated to room temperature (~20 °C), 37, 42, 47, 52, and 57 °C. Samples for the measurement of hemolysis markers (FHb and HBDH) and for the characterization of cell changes by flow cytometry were taken from the same plastic tubes.

hemolysis markers
For preparation of cell-free supernatant, an aliquot of the PRBCs was centrifuged at 3000g for 10 min. We measured {alpha}-HBDH activity in the supernatant according to the recommendation of the German Society for Clinical Chemistry (12), using a commercially available assay (Roche Diagnostics) on a Hitachi 917 clinical chemistry analyzer (Hitachi Ltd.). FHb was measured by the formation of cyanhemoglobin (13), using a commercially available assay (Roche Diagnostics) on a Hitachi 917 analyzer.

Because the addition of lidocaine had led to substantial changes in the heat-induced concentration changes of the hemolysis markers in previous measurements (data not shown), these measurements were performed without lidocaine.

flow cytometry measurements
For flow cytometry measurements, we used a Becton Dickinson FACSCaliburTM flow cytometer that was calibrated using fluorescein-labeled chicken erythrocytes. For each measurement, 15 000 cells (events) were evaluated. In preliminary measurements, two regions of interest of the forward-to-sideward side scatter (FSC/SSC) ratio were defined (Fig. 2 ). In region 1 (R1), a large proportion of erythrocytes could be found at room temperature. Without heat exposure, the cells in this region could not be labeled with the fluorescent antibodies. In region 2 (R2), a substantial number of cells could be found after heat treatment (>=52 °C); the cells in this region showed fluorescence after tagging with fluorescein. The analysis of cell changes was performed using these regions ("gating"). The gate settings were not changed during the experiments. To compare measurements we used the percentage of the gated events that could be assigned to these regions (%) and the median of the fluorescence.



View larger version (50K):
[in this window]
[in a new window]
  		Figure 2. Changes in the FSC/SSC ratio with heating.

At room temperature (RT; A), a large proportion of the 15 000 counted erythrocytes were in R1. The remainder had no specific pattern of distribution. Only a very low percentage of events was present in R2. This pattern was virtually unchanged after heating to 37 °C (not shown) and 42 °C (B). At 47 °C, we observed an increased amount of particles with distinctly lower FSC and SSC signals, which could not be attributed to a specific region (C). Most erythrocytes could still be found in the same regions. In the samples that had been heated to 52 °C (D), a large proportion of erythrocytes showed a distinct reduction in the FSC signal, whereas the SSC signal was constant.

For flow cytometry analysis, the PRBCs were diluted 1:1000 in 1x phosphate-buffered saline. The cells were washed three times by centrifugation at 9000g for 10 min and resuspension in phosphate-buffered saline. The detection antibodies were then added, and the cells were incubated for 60 min at room temperature. The primary antibodies were mouse monoclonal anti-human Ca2+-ATPase IgG (Maine Biotechnology Inc.; MAP 457P), mouse anti-human Na+-K+-ATPase ({alpha}-subunit) IgG (cat. no. A-276; Sigma-Aldrich), and mouse monoclonal anti-spectrin ({alpha} & ß) IgG (cat. no. S3396; Sigma-Aldrich).

After three washing steps, the cells were incubated for 60 min with a fluorescein-conjugated goat anti-mouse IgG F(ab') antibody (Rockland Inc.) in a dark environment. The cells were then washed, and flow cytometry was performed. Measurements were generally done in parallel with room temperature samples as a negative and blood heated to 52 °C as a positive control.

Previous experiments had shown a higher sensitivity of erythrocytes to heat challenge with changes at temperatures <47 °C after incubation with lidocaine (data not shown). We therefore repeated the flow cytometry measurements after adding 200 g/L lidocaine to the samples (final concentration of lidocaine, 15 mmol/L). Gating for these experiments was the same as described above.

sensitivity for local hot spots
We examined the sensitivity of flow cytometry and the assays for markers of hemolysis to detect local hot spots by heating samples from the 12 PRBC units to 52 °C. We then diluted the heated blood stepwise (1:1 to 1:64) in blood stored at room temperature from the same unit. In a 320-mL PRBC unit, a local hot spot of 1:64 of the unit would equal ~5 mL. Laboratory and flow cytometry measurements were performed as described above. Because the results for the different antibodies had been very similar, these experiments were performed with the anti-spectrin antibody only.

evaluation of microwave heating
The transfusio-therm® 2000 microwave blood warmer (Zeipel Medical GmbH) is currently being used in several hospitals in Europe and Japan. The device has holders for three bags of fresh-frozen plasma or PRBCs. The holders are equipped with metal shields to protect the thinner peripheral regions of the blood units from excess doses of microwave radiation. During the heating process, the holders rotate continuously within the device. Two separate temperature probes with separate regulatory circuits are installed in each holder.

We used 30 units of PRBCs for these experiments. Before heating, the units were gently mixed and opened at the filling tubes. After mixing, 10 mL of blood was removed from all units for control measurements. Lidocaine (200 g/L) was injected into 15 of the units to a final concentration of 15 mmol/L. The bags were then sealed with an electrical welder. The units were heated until the cutoff temperature of the device at 35 °C. For comparisons, samples from each unit without lidocaine were left at room temperature and heated to 52 °C in a water bath as described above. Laboratory and flow cytometry measurements were then performed as described above. Samples from the units with lidocaine were heated to 37 and 47 °C in a water bath; for these samples, only flow cytometry measurements were performed.

statistical analysis
The data are generally expressed as the mean (SD). Statistical analysis was performed with the software package SPSS 10.0® (SPSS GmbH) on a Windows® PC. Gaussian distribution was tested using the Kolmogorov–Smirnov goodness-of-fit test. All data sets were compared with the t-test for paired samples. P <0.05 was considered statistically significant. Because some of the datasets used for correlation analysis did not follow a gaussian distribution, correlation analysis was performed with the Spearman rank correlation. Curve fitting was performed with the curve-fitting algorithms provided in the SPSS software package.


   Results
Top
 Abstract
 Introduction
 Material and Methods
 Results
Discussion
 References
 
markers of overheating
We evaluated 12 PRBC units with a mean (SD) hematocrit of 55.9 (4.9)%, hemoglobin of 174 (16) g/L, and red cell count of 6.0 (0.5) x 106/µL. The mean (SD) age of the units at the time of measurement was 39.3 (9.8) days.

Hemolysis markers.
FHb and the HBDH values were higher than the reference values in humans [FHb, <0.3 g/L; HBDH, <140 U/L (14)] in the room temperature samples. FHb and HBDH showed a parallel and exponential increase with heat (see Fig. 1 and Table 1 of the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol49/issue5/).

Both FHb and HBDH increased further 48 h after warming to room temperature (P <0.001 compared with measurements immediately after heating for all temperatures). After heating to 37 °C, the proportion of delayed increase in FHb and HBDH concentrations was similar to that after warming to room temperature. In the samples that were heated to 42 °C, the FHb concentrations were more than twice as high as the concentrations measured immediately after heating, whereas the HBDH was increased ~1.5-fold. Forty-eight hours after heating to 47 °C, there was a threefold increase compared with the measurements immediately after heating for both FHb and HBDH.

Flow cytometry measurements.
In preliminary measurements, two regions of the FSC/SSC ratio were defined and were used throughout the measurements without changes (see Fig. 2Up ). At room temperature, there was only spontaneous fluorescence in R1 in all 12 samples. This was the case for Ca2+-ATPase, Na+-K+-ATPase, and spectrin antibodies [see Table 2 of the online Data Supplement (available at http://www.clinchem.org/content/vol49/issue5/)]. A large proportion of the 15 000 counted erythrocytes could be found in R1 (Fig. 2AUp ). The remainder of the erythrocytes showed no specific pattern of distribution of the FSC/SSC ratio. We found only a very low percentage of events in R2. This pattern was identical for the samples that had been stored at 37 °C. Above 42 °C, we observed significant fluorescence in R1 with all three antibodies (Fig. 3 and Table 2 of the online Data Supplement). The pattern of FSC/SSC distribution was significantly changed after heating to 47 °C and above (Fig. 2B and Table 2 of the online Data Supplement).

In a second step, we added 15 mmol/L lidocaine to samples from the same 12 PRBC units. After heating to 37 °C, the median fluorescence was increased and the proportion of erythrocytes in R1 was decreased [Table 3 of the online Data Supplement (available at http://www.clinchem.org/content/vol49/issue5/)] compared with the samples without lidocaine. A very low proportion of the cells were found in R2. After heating to 42 °C, we observed a trend toward a higher median fluorescence in R1, which was statistically significant for Ca2+-ATPase and Na+-K+-ATPase antibodies. There were also substantially more events in R2 than at 37 °C. After heating to 47 °C, we found a significant increase in the median fluorescence in R1 with all antibodies. The proportion of cells in R1 was significantly lower, and the proportion of the events in R2 increased significantly.

To further evaluate the changes with increasing temperature, we heated samples to room temperature, 37, 39, 41, 43, 45, and 47 °C after the addition of 15 mmol/L lidocaine. The erythrocytes were labeled with the anti-spectrin antibody [Table 4 of the online Data Supplement (available at http://www.clinchem.org/content/vol49/issue5/)]. We observed a gradual increase in the median fluorescence in R1. Compared with the values at room temperature, the differences were statistically significant at temperatures >43 °C. At the same time there was a stepwise increase of the proportion of events in R2 with statistically significant differences at temperatures >43 °C.

sensitivity for local hot spots
Hemolysis markers.
Blood from 12 PRBC units was heated to room temperature and to 52 °C. The blood heated to 52 °C was then diluted stepwise (1:1 to 1:64) in the blood warmed to room temperature. The concentrations of FHb and HBDH were measured at room temperature, at 52 °C, and in the dilutions 1:8, 1:32 and 1:64 [Table 5 of the online Data Supplement (available at http://www.clinchem.org/content/vol49/issue5/)]. Even at the 1:64 dilution, both FHb and HBDH were still significantly increased compared with the room temperature samples. At this dilution, the FHb and HBDH concentrations in some of the samples were in the range of some of the room temperature samples. We observed a stepwise decrease in hemolysis markers with significant correlations between the dilution factor and both FHb (r = 0.41; P = 0.001; n = 60) and HBDH (r = 0.91; P <0.001; n = 60).

Flow cytometry measurements.
Samples diluted 1:2, 1:4, 1:8, 1:16, 1:32, and 1:64 were labeled with anti-spectrin antibodies. The median fluorescence was significantly increased in R1 in the samples that had been heated to 52 °C [Table 6 of the online Data Supplement (available at http://www.clinchem.org/content/vol49/issue5/)]. The proportion of events in R1 and R2 was significantly different after heating to 52 °C compared with room temperature. The number of events in R2 decreased in parallel with the dilution factor. At a dilution of 1:64, there was still a significant difference compared with the samples that were stored at room temperature, but the absolute number of events in some of the samples was within the same range as in some of the room temperature samples.

The dilution factor and the median fluorescence were significantly correlated (R1, r = 0.62; P <0.001; n = 96; R2, r = -0.31; P = 0.002; n = 96), as were the dilution factor and the percentage of events in the regions (R1, r = -0.61; P <0.001; n = 96; R2, r = 0.97; P <0.001; n = 96). For R2, the relationship between the dilution factor and the percentage of events could be described by a quadratic function (Fig. 4 ).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 4. Percentage of cells in R2 for different dilutions of heated in nonheated blood.

Blood from 12 units of PRBCs was either left at room temperature (RT) or heated to 52 °C in a water bath. The heated blood was diluted stepwise in the room temperature blood. The 1:64 dilution of heated blood showed a significant change in the percentage of cells in R2 (P <0.001). The relationship between the dilution factor d and the percentage of cells in R2 could be described by the quadratic function: F(d) = 0.12 + (9.83d) + (9.68d2); R2 = 0.93; P <0.0001. More detailed results are provided in Table 6 of the online Data Supplement (http://www.clinchem.org/content/vol49/issue5/).

evaluation of microwave heating
FHb and HBDH.
Twelve units of PRBCs [mean (SD) age, 32.5 (8.3) days; weight, 343.8 (29.2) g] were used. Three more units with substantial hemolysis by visual inspection [age, 36.3 (4.0) days; weight, 336.3 (18.0) g] were evaluated separately. No unit showed signs of overheating on visual examination after microwave warming.

At room temperature, the mean (SD) FHb was 0.5 (0.5) g/L (minimum–maximum, 0.03–1.39 g/L). After warming to 35 °C with the microwave blood warmer, FHb was 0.5 (0.6) g/L (0.04–1.99 g/L). The difference between the groups was not statistically significant. The three hemolytic samples had a mean (SD) FHb of 3.3 (2.7) g/L (1.63–6.36 g/L); after heating to 35 °C, the mean FHb was not significantly different: 2.6 (1.5) g/L (1.69–4.27 g/L).

Samples warmed to room temperature had a mean (SD) HBDH concentration of 531 (358) U/L (minimum–maximum, 106-1040 U/L). After heating to 35 °C with the microwave device, the mean (SD) concentration was 526 (384) U/L (135–1427 U/L) with no statistically significant difference. The three hemolytic units had a mean (SD) HBDH concentration of 2585 (1797) U/L (1496–4660 U/L) with no significant difference after heating to 35 °C [2160 (1107) U/L; 1270–3400 U/L].

Flow cytometry.
We compared the percentage of events in R2 and the median fluorescence in R1 by flow cytometry.

In samples without lidocaine, at room temperature, 0.2 (0.04)% (minimum–maximum, 0.11–0.23%) of the cells in the 12 PRBC units that did not have visual evidence of hemolysis were in R2; the median (SD) fluorescence in R1 was 126.1 (28.5), with a minimum of 105 and a maximum of 187. In the three hemolytic units, 0.1 (0.04)% of the events were in R2 (minimum–maximum, 0.11–0.19%). The median (SD) fluorescence in R1 was 139.3 (40.5), with a minimum of 113 and a maximum of 186.

After heating to 52 °C in a water bath, the percentage of events in R2 increased to 14.8 (6.4)% (1.28–25.21%; P <0.001). The median (SD) fluorescence in R1 was also increased [189.1 (27.8); minimum–maximum, 156–236]. The three hemolytic samples had a higher percentage of events in R2 [22.8 (5.8)%; 17.88–29.14%] and a higher median (SD) fluorescence in R1 [211.3 (14.6); minimum–maximum, 195–223] than the other samples.

After heating to 35 °C with the microwave blood warmer, there were no significant changes in the mean (SD) percentage of events in R2 [0.2 (0.05)%; minimum–maximum, 0.08–0.25%] or the median (SD) fluorescence in R1 [124.6 (29.6); minimum–maximum, 85–184]. The three hemolytic units had a higher number of events [0.2 (0.1)%; 0.16–0.32%] and a higher median fluorescence [147.3 (30.1); 128–182] than the remainder of the samples.

The portions of the 15 PRBC units to which 15 mmol/L lidocaine had been added had a mean of 0.2 (0.5)% events (minimum–maximum, 0.03–1.95%) in R2 at room temperature. The median fluorescence in R1 was 105.0 (4.4), with a minimum of 96 and a maximum of 114.

Heating to 47 °C in a water bath produced significant increases in the number of events [33.9 (7.7)%; minimum–maximum, 20.51–44.51%; P <0.001] and in the median fluorescence [143.7 (17.1); minimum–maximum, 114–181; P <0.05].

After samples were heated to 37 °C in a water bath, there was no significant increase in events [0.1 (0.1)%; minimum–maximum, 0.04–0.29%] or median fluorescence [107.3 (4.3); minimum–maximum, 100–115] compared with the room temperature samples.

We found no significant differences between the room temperature samples, the samples that had been heated to 37 °C in a water bath, and the samples from the units heated to 35 °C in the microwave device. This was true for the number of events [35 °C, 0.1 (0.2)%; minimum–maximum, 0.02–0.93%] and the median fluorescence [35 °C, 106.4 (4.7); minimum–maximum, 100–117].


   Discussion
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Discussion
References
 
The primary results of our study were as follows: (a) Analysis of the FSC/SSC ratio and fluorescence labeling with anti-spectrin antibodies can be used to identify damage to individual erythrocytes after heating to >52 °C. (b) The detection threshold of this method can be lowered to 47 °C by the addition of 15 mmol/L lidocaine to the samples. (c) Analysis of FHb and HBDH can detect heat damage to PRBC units in the same temperature range. If PRBCs are heated to temperatures of 42 and 47 °C, a considerable part of the hemolysis occurs after 48 h. (d) Both methods can detect the localized overheating of ~5 mL in a 320-mL PRBC unit. (e) We did not observe damage to PRBCs or an increase in hemolysis after heating to 35 °C with a commercially available microwave blood warmer.

We observed labeling of cell membrane antigens with all antibodies only after exposure to temperatures >42 °C. There are several possible explanations for this phenomenon. Destruction of the cell membrane may lead to an influx of antibodies to the inner side of the cell membrane and allow them to bind to previously inaccessible sites. Permeability changes of the cell membrane could also explain the observed changes in FSC, which is correlated to cell size. It has been shown previously that the membranes of erythrocytes make three transitions after heating to 50.0, 56.8, and 63.8 °C, respectively (15). Both spectrin and the antibody-binding site of the {alpha}-subunit of Na+-K+-ATPase (16)(17) are on the cytoplasmic side of the cell membrane, and the anti-Ca2+-ATPase antibody we used is also reported to bind to the Ca2+ translocating channel protein associated with the intracellular membrane Ca2+-Mg2+-ATPase (18), which would support this hypothesis. Another mechanism of action could be heat-induced denaturation of the detected proteins themselves, leading to a change in the tertiary structure of the proteins with subsequent exposure of antigen-binding domains. This theory is supported by the fact that heat-induced changes in the properties of all three proteins can be observed in this temperature range. It has been demonstrated that the unfolding of spectrin occurs at 49.5 °C (19), but the denaturation of spectrin alone seems insufficient to induce thermohemolysis (15). The activities of both Ca2+-ATPase and Na+-K+-ATPase are decreased after heating to 45 °C (20) and 43 °C(21), respectively. This could be the result of protein denaturation leading to a change in conformation and antibody binding, although it has been shown that there is no increase in Lowry-reactive protein fragments after the heating of erythrocytes up to 49 °C (22).

Local anesthetics sensitize cells to hyperthermia (23)(24). It has been shown by differential scanning calorimetry that the incubation of human erythrocyte ghosts and sarcoplasmic reticulum with lidocaine leads to a decrease in the transition temperature of the transmembrane domain of the band III protein and Ca2+-ATPase. This produces a decrease in the denaturation temperature (25). A previous study showed a similar decrease in the critical temperature of fragmentation from 49 to 47 °C with diamide, but not with several other membrane-active agents (26).

We cannot firmly conclude from our data which of these factors are the most important contributors to the observed changes in antibody binding, but the uniformity of antibody binding to all three antigens at temperatures >47 °C and the change in cell size makes it likely that a change in the permeability of the cell membrane to antibodies is an important factor. This is also supported by the results of other investigators, who found a linear increase in the potassium leakage of human erythrocytes between 46 and 54 °C (27) and as well by the reports of massive changes in erythrocyte size and hemoglobin distribution after the heating of blood specimens to temperatures >49 °C and in patients with severe burns (9)(10)(11).

Especially after heating to 42 and 47 °C, there was a considerable further increase in hemolysis after 48 h compared with the measurements immediately after heating. The time between heating and measurement is therefore an important factor for the comparison of hemolysis after blood warming. This should be considered when new blood warming devices are being evaluated.

The results of the flow cytometry and the biochemical assays were significantly correlated with the presence of heat-damaged erythrocytes in the dilution assay. Both assays were capable of detecting heat-damaged (52 °C) samples diluted in untreated samples up to a dilution of 1:64. This would correspond to the localized overheating of 5 mL in a 320-mL unit of PRBCs. At this dilution the absolute values of the hemolysis markers in some samples were similar to the room temperature values of others. The transfusion of units with a lower percentage of overheated cells would probably have little clinical relevance. The methods therefore provide an adequate means to evaluate blood-warming devices for general and localized overheating.

When we used the presented methods to test PRBCs warmed with the microwave device, neither method detected signs of overheating. On the basis of these results, the microwave system can be safely used in the present configuration.

In conclusion, we believe that this methodology can be used to detect both hot spots and general overheating in PRBC units. The sensitivity is sufficient to be useful for the future testing of blood-warming devices. Another potential application for the flow cytometry measurements may be the initial screening of burn victims to assess the risk for future complications, such as kidney failure from hemoglobinemia. The substantial amount of delayed hemolysis that we found based on the measurement of biochemical markers of hemolysis after heat exposure should be taken into account in both recommendations for the temperature set points for the warming of PRBCs and for future testing of blood-warming devices. The exponential increase in hemolysis beyond 42 °C suggests that the temperature set point for blood warmers should allow a sufficient margin of safety for these temperatures.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 1. Increase in FHb of PRBCs at room temperature (RT) and after heating to different temperatures.

FHb values are on a logarithmic (ln) scale. • indicate the concentration immediately after heating, indicate the values after 48 h. After heating to 37 °C, there were no significant changes compared with the room temperature values. After heating to temperatures of 42 °C and above, the FHb concentration increased significantly (P <0.001 compared with room temperature). For temperatures (T) from 37 to 57 °C, the change in FHb could be described by the exponential function: f(T) = 8.6 x 10-5 x e0.21xT (R2 = 0.87; P <0.0001). FHb values 48 h after heating were significantly higher than those immediately after heating (P <0.001). For temperatures from 37 to 47 °C, the increase of FHb after 48 h could be described by the exponential function: f(T) = 1.4 x10-3 x e0.16xT (R2 = 0.53; P <0.0001). The increase in HBDH with temperature immediately after heating and after 48 h also followed an exponential pattern (data not shown). Long horizontal bars indicate the means, and short horizontal bars indicate the SD. More detailed results are provided in Table 1 of the Data Supplement that accompanies the online version of this article (http://www.clinchem.org/content/vol49/issue5/).



View larger version (11K):
[in this window]
[in a new window]
  		Figure 3. Median fluorescence of erythrocytes in region R1 (A) and distribution of FSC/SSC ratio of erythrocytes in region R2 (B) at room temperature (RT) and after heating to different temperatures.

(A), cells were labeled with anti-spectrin antibodies. Heating to 42 °C and above produced significant (P <0.001) antibody binding. The apparently diminished antibody binding and the large SD (short horizontal bars) at 57 °C are caused by the reduction in the number of cells in this region (see Fig. 2 and Table 2 of the online Data Supplement). (B), heating to temperatures >37 °C produced an exponential increase in the number of cells in R2. The increase in the number of cells in R2 with temperature (T) could be described by the function: f(T) = 1.7 x 10-7 x e0.34xT (R2 = 0.88; P <0.0001). More details on these results are provided in Table 2 of the online Data Supplement (http://www.clinchem.org/content/vol49/issue5/).


   Acknowledgments
 
This project was funded by the research foundation of the Department of Anesthesiology and Intensive Care Medicine at Giessen University Hospital. The foundation received an unrestricted donation from Zeipel Medical GmbH (Bad Heiligenstadt, Germany). We also wish to thank the staffs of the blood bank and the laboratories of the Department of Anesthesiology of Giessen University Hospital for their cooperation.


   Footnotes
 
1 Nonstandard abbreviations: PRBC, packed red blood cell; FHb, free hemoglobin; LDH, lactate dehydrogenase; HBDH, {alpha}-hydroxybutyrate dehydrogenase; FSC, forward side scatter; and SSC, sideward side scatter.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Discussion
 References
 

  1. Richtlinien zur Gewinnung von Blut und Blutbestandteilen und zur Anwendung von Blutprodukten (Hämotherapie). Bundesgesundheitsbl-Gesundheitsforsch-Gesundheitsschutz 2000;43:555–89..
  2. . US Food and Drug Administration. Guide to inspections of blood banks 1994;9:14 US Food and Drug Administration, Office of Regulatory Affairs Washington. .
  3. Arens JF, Leonard GL. Danger of overwarming blood by microwave. JAMA 1971;218:1045-1046.[Abstract/Free Full Text]
  4. McCullough J, Polesky HF, Nelson C, Hoff T. Iatrogenic hemolysis: a complication of blood warmed by a microwave device. Anesth Analg 1972;51:102-106.[Free Full Text]
  5. Staples PJ, Griner PF. Extracorporeal hemolysis of blood in a microwave blood warmer. N Engl J Med 1971;285:317-319.
  6. Roth Henschker H, Holzberg E, Oppitz KH, Lehmann C. [Warming resuspended erythrocyte concentrates with a new microwave device. In vitro results and clinical experiences in comparison with continuous flow warming]. Anaesthesist 1988;37:321-330.[Medline] [Order article via Infotrieve]
  7. Pappas CG, Paddock H, Goyette P, Grabowy R, Connolly RJ, Schwaitzberg SD. In-line microwave blood warming of in-date human packed red blood cells. Crit Care Med 1995;23:1243-1250.[Medline] [Order article via Infotrieve]
  8. Lee SW, Ducoff HS. Effects of hyperthermia on avian red blood cells. Int J Hyperthermia 1993;9:709-722.[Medline] [Order article via Infotrieve]
  9. Bain BJ, Liesner R. Pseudopyropoikilocytosis: a striking artefact. J Clin Pathol 1996;49:772-773.[Abstract/Free Full Text]
  10. Zarkowsky HS, Mohandas N, Speaker CB, Shohet SB. A congenital haemolytic anaemia with thermal sensitivity of the erythrocyte membrane. Br J Haematol 1975;29:537-543.[Web of Science][Medline] [Order article via Infotrieve]
  11. Navari RM, Sheehy TW, McLean BK, Sutton FD. The peripheral blood smear in heat stroke. An aid to diagnosis. Ala J Med Sci 1983;20:137-140.[Medline] [Order article via Infotrieve]
  12. Recommendations of the German Society for Clinical Chemistry. Standardisation of methods for the estimation of enzyme activities in biological fluids. Experimental basis for the optimized standard conditions. Z Klin Chem Klin Biochem 1972;10:281-291.[Medline] [Order article via Infotrieve]
  13. van Kampen EJ, Zijlstra WG. Determination of hemoglobin and its derivatives. Adv Clin Chem 1965;8:141-187.[Medline] [Order article via Infotrieve]
  14. MSD Manual der Diagnostik und Therapie. 5th German Edition 1993:3408 Urban & Schwarzenberg Munich-Vienna-Baltimore. .
  15. Lepock JR, Frey HE, Bayne H, Markus J. Relationship of hyperthermia-induced hemolysis of human erythrocytes to the thermal denaturation of membrane proteins. Biochim Biophys Acta 1989;980:191-201.[Medline] [Order article via Infotrieve]
  16. Ball WJ, Schwartz A, Lessard JL. Isolation and characterization of monoclonal antibodies to (Na+ + K+)-ATPase. Biochim Biophys Acta 1982;719:413-423.[Medline] [Order article via Infotrieve]
  17. Blanco G, Mercer RW. Isozymes of the Na-K-ATPase: heterogeneity in structure, diversity in function. Am J Physiol 1998;275:F633-F650.
  18. Hack N, Authi KS, Crawford N. Introduction of antibody (PL/IM 430) to a 100 kDa protein into permeabilised platelets inhibits intracellular sequestration of Ca2+. Biosci Rep 1988;8:379-388.[CrossRef][Medline] [Order article via Infotrieve]
  19. Brandts JF, Erickson L, Lysko K, Schwartz AT, Taverna RD. Calorimetric studies of the structural transitions of the human erythrocyte membrane. The involvement of spectrin in the A transition. Biochemistry 1977;16:3450-3454.[CrossRef][Medline] [Order article via Infotrieve]
  20. Kosk Kosicka D, Wawrzynow A, Roszczynska G. Stabilizing and destabilizing effects on plasma membrane Ca(2+)-ATPase activity. Mol Cell Biochem 1994;139:1-9.[Medline] [Order article via Infotrieve]
  21. Grzelinska E, Bartosz G, Leyko W, Chapman IV. Effect of hyperthermia and ionizing radiation on the erythrocyte membrane. Int J Radiat Biol Relat Stud Phys Chem Med 1982;42:45-55.[Medline] [Order article via Infotrieve]
  22. Bartosz G, Gaczynska M, Retelewska W, Grzelinska E, Rosin J. Hyperthermia, unlike ionizing radiation and chemical oxidative stress, does not stimulate proteolysis in erythrocytes. Int J Biochem 1990;22:25-30.[Medline] [Order article via Infotrieve]
  23. Yatvin MB. The influence of membrane lipid composition and procaine on hyperthermic death of cells. Int J Radiat Biol Relat Stud Phys Chem Med 1977;32:513-521.[Medline] [Order article via Infotrieve]
  24. Yau TM. Procaine-mediated modification of membranes and of the response to X irradiation and hyperthermia in mammalian cells. Radiat Res 1979;80:523-541.[CrossRef][Medline] [Order article via Infotrieve]
  25. Senisterra GA, Lepock JR. Thermal destabilization of transmembrane proteins by local anaesthetics. Int J Hyperthermia 2000;16:1-17.[CrossRef][Medline] [Order article via Infotrieve]
  26. Zarkowsky HS. Membrane-active agents and heat-induced erythrocyte fragmentation. Br J Haematol 1982;50:361-365.[Web of Science][Medline] [Order article via Infotrieve]
  27. Prinsze C, Tijssen K, Dubbelman TM, Van Steveninck J. Potentiation of hyperthermia-induced haemolysis of human erythrocytes by photodynamic treatment. Evidence for the involvement of the anion transporter in this synergistic interaction. Biochem J 1991;277:183-188.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hirsch, J.
Right arrow Articles by Hempelmann, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hirsch, J.
Right arrow Articles by Hempelmann, G.
Related Collections
Right arrow Other Areas of Clinical Chemistry
Right arrow Hematology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS