Can latent heat safely warm blood? – in vitro testing of a portable prototype blood warmer
© McEwen and Roxby; licensee BioMed Central Ltd. 2007
Received: 06 February 2007
Accepted: 01 August 2007
Published: 01 August 2007
Trauma/retrieval patients are often in shock and hypothermic. Treatment of such patients usually involves restoring their blood volume with transfusion of blood (stored at 2°C – 6°C) and/or crystalloids or colloids (stored at ambient temperature). Rapid infusion of these cold fluids can worsen or even induce hypothermia in these patients. Warming of intravenous fluids at accident sites has traditionally been difficult due to a lack of suitable portable fluid warmers that are not dependent on mains electrical or battery power. If latent heat, the heat released when a liquid solidifies (an inherently temperature limiting process) can warm intravenous fluids, portable devices without a reliance on electrical energy could be used to reduce the incidence of hypothermia in trauma patients.
Rapid infusion of red cells into patients was timed to sample typical clinical flow rates.
An approved dry heat blood warmer was compared with a prototype blood warmer using a supercooled liquid latent heat storage material, to warm red cells whilst monitoring inlet and outlet temperatures. To determine the effect of warming on red cell integrity compared to the normal storage lesion of blood, extracellular concentrations of potassium, lactate dehydrogenase and haemoglobin were measured in blood which had been warmed after storage at 2°C – 6°C for 1 to 42 days.
A prototype latent heat fluid warmer consistently warmed red cells from approximately 4°C to approximately 35°C at typical clinical flow rates. Warming of stored blood with latent heat did not affect red cell integrity more than the approved dry heat blood warmer.
Using latent heat as an energy source can satisfactorily warm cold blood or other intravenous fluids to near body temperature, without any adverse affects.
Victims of major trauma are considered to be at risk of hypothermia which often results in deleterious effects (including coagulopathy, cardiac arrhythmias, peripheral vasoconstriction, metabolic acidosis, compensatory increased oxygen requirements during rewarming, and impaired immune response) leading to an associated increased mortality and morbidity [1, 2]. Up to 66% of trauma patients are reported to be hypothermic at the time of hospital admission . Those patients with a core temperature lower than 34°C suffer a mortality rate up to 35% higher than euthermic patients [3–6].
At trauma sites it is common to transfuse red cells that have been stored and transported at 4°C and/or colloid/crystalloid fluids at ambient temperature. This administration of cold intravenous fluids has been identified as one of the main factors that contribute to the high incidence of hypothermia in the trauma population .
To prevent falls of temperature, in hospitals there are several mains powered devices available for warming fluids before infusion into patients. However at retrieval or accident sites electrical energy is generally only available in the form of batteries.
An alternative method of portable energy storage is in the form of latent heat in supercooled liquids. A liquid is supercooled when it remains in the liquid state at temperatures below its nominal solidification temperature. Many substances can exist as supercooled liquids, eg sodium acetate trihydrate (used in "heat packs") normally freezes at approximately 55°C, but can remain in the liquid state at temperatures below 10°C if it is kept free from impurities in a sealed container. When a supercooled liquid solidifies, it releases the latent heat of solidification and its temperature quickly approaches the nominal solidification temperature of the substance. This provides an ideal situation for warming intravenous fluids, providing the desired warming temperature corresponds to the solidification temperature of a latent heat storage material.
A fluid warmer based on latent heat could be a relatively simple and low-cost device. In its most basic form it may consist of some latent heat storage material, some IV tubing and an external casing.
This work aimed to determine if latent heat can be used to warm blood from its storage temperature (2°C – 6°C) to normal body temperature, this article describes the development of a latent heat fluid warmer and its in vitro testing.
Prototype fluid warmers, consisting of (Gambro) infusion extension lines immersed in a latent heat storage material, with flexible or rigid external casings, were constructed. The latent heat storage material had a phase change temperature of 42°C, and was kept free from impurities to enable it to be supercooled. The prototype used to warm blood contained 0.85 litres of latent heat storage material.
Red cell units
Fresh red cells (RC) stored in Optisol®, that were unsuitable for transfusion because they did not meet pre-determined criteria (such as the donor had visited a malarial risk area within the previous 4 months), were supplied by the Australian Red Cross Blood Service for in vitro tests. All these RC units were marked "Component for discard". These RC had a shelf life of 42 days from the day of collection and were kept under standard storage conditions (2°C – 6°C) in validated temperature monitored cold rooms until used for testing. The average volume of these RC units was 275 mL. RC with a range of storage times (1 to 42 days) were tested in the latent heat fluid warmer.
Infusion flow rates
The infusion rate of red cells under normal clinical conditions (not warmed with latent heat) was sampled by timing infusions into patients in the Intensive and Critical Care Unit of Flinders Medical Centre. Red cells were delivered intravenously via 14, 16 or 20 gauge needles inserted into the patient's right cubital fossa or right hand, whilst being pressurised at 250 mmHg. All decisions affecting patient care (choice of needle gauge, insertion site, infusion pressure etc) were made by the staff of the Intensive and Critical Care Unit, and the investigators were merely observers.
Latent heat prototype testing
Latent heat was used to warm 67 units of RC in 25 tests; 12 tests were conducted using 1 unit of RC, 11 tests were conducted using 4-units of RC, one test was conducted using 5 units of RC and the final test was conducted using 6-units of RC. RC were gently massaged by hand prior to testing, to mix the cells with clear fluid that had separated during storage.
A similar test protocol was used with a Cobe SpectraTherm (Lakewood, CO) dry heat mains powered fluid warmer (regulated drum temperature of 36.7°C). The dry heat blood warmer was used to warm 6 units of RC, with one unit being warmed in each test.
In addition to the RC samples withdrawn during fluid warmer testing, 10 mL samples were extracted from 4 RC units kept in storage (2°C – 6°C) for their lifetimes (42 days), on storage days 2, 3, 6, 8, 9, 10, 13, 16, 22, 30, 34, 37 and 42. These samples were centrifuged and supernatant extracted (as described below) while they were cold.
Plasma potassium (K), lactate dehydrogenase (LD) and haemoglobin (Hb) levels were determined to assess red cell damage and level of haemolysis occurring either during normal storage conditions (2°C – 6°C) or following warming through the approved dry heat blood warmer or the prototype.
All samples were centrifuged at an accelerating force of 2050 g for 4 minutes, after which time 500 μL of supernatant was removed and analyte concentrations were determined:
Potassium (mmol/L) was measured with a Roche modular ISE1800 analyser.
Lactate dehydrogenase (units/L) and a serum index, equivalent to haemoglobin concentration (mg/dl) were determined with a Roche modular P800 analyser.
Infusion flow rates
Rapid infusion flow rates
No of infusions
Average flow rate (mL/min)
Fluid warmer testing
The output of the latent heat prototype fluid warmer did decrease over time (as the release of latent heat decreased).
It was found that there was usually considerable supernatant in RC samples extracted before warming, and there was much less supernatant in samples extracted from the outlet mid way through RC units and even less supernatant in samples extracted from the outlet at the end of RC units. This was probably an indication that cells had settled towards the bottom of RC units, so the first portion expressed from RC units was the leanest in cells and the last portion expressed was the richest in cells.
Means and standard deviations of extracellular potassium, haemoglobin and lactate dehydrogenase concentrations in RC units warmed with latent heat (n = 28) or the dry heat blood warmer (n = 6) between day 7 & 25 of storage, with initial haemoglobin concentration between 70 & 700 mg/dL
Sample from inlet line (before warming)
Sample from outlet line (midway through RC unit)
Sample from outlet line (at RC unit end)
Lactate Dehydrogenase (units/L)
Flow rates of ambient temperature saline (pressurised at 300 mmHg) through IV delivery systems, including a fluid warmer and 14 gauge cannula have been found to be between 225 and 446 mL/minute, depending upon the type of fluid warmer used . The flow rates sampled in the present study in actual clinical situations were significantly lower. Reasons for the lower flow rate in the present study include: 250 mmHg used instead of 300 mmHg, the viscosity of blood is much higher than that of saline and flow restrictions due to venous access sites.
Red cells have previously been heated to 44.7°C for 30 minutes with no damage reported , and it has also been shown that incubation at 45°C for up to 1 hour does not cause significant haemolysis . Therefore damage due to heating red cells to 42°C with the latent heat storage material in the present study was not expected. Since latent heat is only released at or below the solidification temperature there could be no possibility of overheating due to using the latent heat of solidification as an energy source for warming blood.
Plasma potassium, lactate dehydrogenase and haemoglobin were used in this study to indicate RC damage, because they have previously been shown to be accurate markers of haemolysis .
In general extracellular concentrations of these analytes were higher in samples after warming than before warming in the present study. This may indicate damage as a result of warming, but could also be a function of sampling. As the amount of supernatant was greatest in the first (unwarmed) samples and least in the last (warmed) samples from RC units, the amount of cells in extracted samples was least in the first samples and greatest in the last samples from the RC units. The amount of extracellular potassium, haemoglobin and lactate dehydrogenase could be expected to be related to the amount of cells in the samples. Therefore the concentration of these analytes could be expected to be lowest in the first ("pre") samples extracted from RC units and highest in the last ("post") samples extracted from RC units whether units were warmed or not.
Although there was considerable variation in RC units before warming in this study, when the biochemical test results for units of similar age and initial condition were compared for the latent heat prototype and the dry heat blood warmer similar results were obtained for both warming methods. Overall the results of the biochemical tests for the latent heat fluid warmer, showed no signs of increased cellular damage compared to red cells passed through the clinically approved dry heat blood warmer.
It should be noted that the severity of the storage lesion of blood found in this study may not be an indication of typical storage lesion, because the RC units used in this study had been declared unsuitable for transfusion.
The further development of the use of latent heat as a blood warming modality will depend upon the willingness of medical device manufacturers to explore this field.
The feasibility of warming blood and other intravenous fluids with a latent heat storage material has been demonstrated. A low-cost portable non-electrically powered fluid warmer using latent heat storage may be used to significantly warm cold blood or intravenous fluids to near body temperature with minimal risk of cellular damage and haemolysis.
The big advantage of warming IV fluids using the latent heat of fusion of a solidifying liquid is that heat is released at the solidification temperature, which is inherently temperature limiting. Consequently the risk of overheating fluids with a latent heat IV fluid warmer is negligible. Testing reported in this paper found that the warm blood outlet temperature from a latent heat fluid warmer did not exceed the phase change temperature of the latent heat storage material, even during times of flow cessation.
latent heat prototype blood warmer
dry heat blood warmer
acceleration due to gravity
sample extracted from blood warmer inlet line (before warming)
sample extracted from blood warmer outlet line midway through RC unit (warmed)
sample extracted from blood warmer outlet line at the end of the RC unit (warmed)
number of RC units tested
The authors would like to thank Anne-Louise Smith (of Flinders Biomedical Engineering and Flinders university) for collating information regarding hypothermia from the literature and for making suggestions regarding the drafting of the manuscript, and John Robson and Robin Woolford (of Flinders Biomedical Engineering) who provided resources used in this study.
The authors would also like to thank the reviewers of this paper, J. Paton and M. Gyongyossy-Issa, for their comments and suggestions.
- Smith CE: Prevention and treatment of hypothermia in trauma patients. ITACCS. 2004, Spring 2004: 68-80.Google Scholar
- Herron DM, Grabowy R, Connolly R, Schwaitzberg SD: The limits of bloodwarming: maximally heating blood with an inline microwave bloodwarmer. J Trauma. 1997, 43 (2): 219-226.View ArticlePubMedGoogle Scholar
- Luna GK, Maier RV, Pavlin EG, Anardi D, Copass MK, Oreskovich MR: Incidence and effect of hypothermia in seriously injured patients. J Trauma. 1987, 27 (9): 1014-1018.View ArticlePubMedGoogle Scholar
- Gregory JS, Flancbaum L, Townsend MC, Cloutier CT, Jonasson O: Incidence and timing of hypothermia in trauma patients undergoing operations. J Trauma. 1991, 31 (6): 795-798.View ArticlePubMedGoogle Scholar
- Jurkovich GJ, Greiser WB, Luterman A, Curreri PW: Hypothermia in trauma victims: an ominous predictor of survival. J Trauma. 1987, 27 (9): 1019-1024.View ArticlePubMedGoogle Scholar
- Steinemann S, Shackford SR, Davis JW: Implications of admission hypothermia in trauma patients. J Trauma. 1990, 30 (2): 200-202.View ArticlePubMedGoogle Scholar
- Turner M, Hodzovic I, Mapleson WW: Simulated clinical evaluation of four fluid warming devices*. Anaesthesia. 2006, 61 (6): 571-575. 10.1111/j.1365-2044.2006.04589.x.View ArticlePubMedGoogle Scholar
- Eastlund T, Van Duren A, Clay ME: Effect of heat on stored red cells during non-flow conditions in a blood-warming device. Vox Sang. 1999, 76 (4): 216-219. 10.1159/000031054.View ArticlePubMedGoogle Scholar
- Chalmers C, Russell WJ: When does blood haemolyse? A temperature study. Br J Anaesth. 1974, 46 (10): 742-746. 10.1093/bja/46.10.742.View ArticlePubMedGoogle Scholar
- Holzman S, Connolly RJ, Schwaitzberg SD: The effect of in-line microwave energy on blood: a potential modality for blood warming. J Trauma. 1992, 33 (1): 89-93; discussion 93-4.View ArticlePubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-227X/7/8/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.