Share this article on:

A Laboratory Evaluation of the Level 1 Rapid Infuser (H1025) and the Belmont Instrument Fluid Management System (FMS 2000) for Rapid Transfusion

Comunale, Mark E. MD

doi: 10.1213/01.ANE.0000077078.53242.29

In this investigation I compared the heating capabilities, infusion rates, and air management system of the new FMS 2000 and the Level 1 infusion systems. A rollerhead pump was placed at the position of the IV bag spikes on the Level 1 disposable set and used to match the flow of the rollerhead pump of the FMS 2000. Infusate temperature was measured at 250 and 500 mL/min by using packed red blood cells (PRBC) at the infusate supply and distal end of the disposable tubing. Tubing lengths were equal for both systems. Devices were set at maximum flow while temperatures were measured as described previously. Flow was measured with a graduated cylinder and stopwatch. Air management was evaluated with a 10-mL air bolus injected proximal to the heat exchanger at maximum flow. Air passing through the system was collected and measured. At 250 mL/min, both devices provided infusate at ≥37°C. At 500 mL/min, only the FMS 2000 provided PRBC at ≥37°C. PRBC infusate temperature decreased over time with the Level 1 device. The FMS 2000 delivered PRBC at its maximum flow rate of 500 mL/min. The Level 1 attained a maximum flow of 575 mL/min of PRBC with a temperature of 32°C. Air (10 mL) was recovered from the distal tubing of the Level 1. No air was recovered from the tubing of the FMS 2000. In this laboratory investigation, the FMS 2000 performance was superior with respect to providing PRBC at physiologic temperatures (at 500 mL/min) and with respect to air detection and elimination compared with current widely used pressure-infuser technology.

IMPLICATIONS: The FMS 2000 and Level 1 H1025 warming capabilities were similar at metered flows of 500 mL/min. At more rapid flow rates, which are achieved by the H1025, fluid warming decreases dramatically. The air-detection and -elimination system of the FMS 2000 is superior to that of the H1025.

Department of Anesthesia and Critical Care, St. Louis University Medical Center, St. Louis University School of Medicine, St. Louis, Missouri

Supported in part by a grant from Belmont Instrument Corporation, Billerica, MA (Manufacturer of the FMS 2000).

Accepted for publication April 30, 2003.

The author was a product development consultant to Belmont Instrument Corporation during the development of the FMS 2000.

Presented in part at the New York Post Graduate Assembly in Anesthesiology, New York, NY, December, 1999.

Address correspondence and reprint requests to Mark E. Comunale, MD, Department of Anesthesia and Critical Care, St. Louis University Medical Center, St. Louis University School of Medicine, 3635 Vista Ave. at Grand Blvd., St. Louis, MO 63110-0250. Address e-mail to

The rapid transfusion of large volumes of room-temperature crystalloid and/or refrigerated blood, as occurs in resuscitation during hemorrhage, can lead to hypothermia and its attendant sequelae (1–4). Several fluid-warming devices are available that heat fluid by countercurrent exchange, heating plates or heating elements, and water baths (5,6). These devices may have limited capabilities in providing warmed crystalloid and/or blood because of insufficient heater block surface area for properly heating at rapid flows or insufficient heating capability of water-bath systems or countercurrent systems at rapid flow rates (7). Also, these devices have limited, if any, capability to detect and remove air during operation.

A new device on the market, the FMS 2000 (Belmont Instrument Corp., Billerica, MA), is designed to heat fluid by using magnetic induction. This device also includes a semiocclusive rollerhead pump that provides rapid and accurate infusion rates. This article compares the heating capabilities, infusion rates, and air management systems of the Level 1 rapid infusion (SIMS Level 1, Inc., Rockland, MA) system and the FMS 2000.

Back to Top | Article Outline


A Level 1 infusion system (Model H1025) and a Belmont Instrument FMS 2000 were obtained from their respective manufacturers. A schematic of each device is shown in Figure 1. The Level 1 system consists of a disposable countercurrent heat exchanger (Model D-300) through which water warmed to 41.5°C in a tank at the base of the unit is circulated by a pump. This countercurrent warming device is combined with automated pneumatic pressure infusers. The Belmont Instrument FMS 2000 system warms via electromagnetic induction. The disposable tubing set has a set of surgical stainless-steel rings embedded in a plastic housing through which blood or other fluid flows. The disposable tubing is mounted on the machine in such a way that the ring is placed over an electromagnet. Infrared heat sensors on the inflow and on the outflow side of the heat exchanger measure inflow and outflow fluid temperature. Temperature information is monitored by the FMS 2000 computer, which then controls power to the electromagnet and thus heats the coil depending on the inflow and outflow fluid temperatures. A semiocclusive rollerhead pump drives fluid through the system up to a maximum rate of 500 mL/min. Pressure sensors within the system monitor line pressure and automatically reduce pump speed in the event of partial occlusion. Maximum line pressure is set at 400 mm Hg. In the event of total occlusion, the pump will stop, and the system sounds an alarm when the line pressure exceeds 400 mm Hg or when the rate of pressure increase exceeds a preset value of 40 mm Hg/mL pumped.

Figure 1

Figure 1

The two ultrasonic air detectors monitor for air in the fluid path (Fig. 1A). The detectors discriminate between the presence of air and water by the acoustical differences of liquid and air. The first ultrasonic air detector is the “fluid out” detector, located near the input to the system. This air detector monitors the pump tubing for air that can enter from the fluid bag. The sensitivity of the “fluid out” detector is 0.8 mL of air. The second sensor is the “air detect” detector, located just above the valve. This air detector monitors for air that may enter the patient line. The sensitivity of the “air detect” sensor is 0.1 mL of air. The sensitivity of the ultrasonic air detectors does not differ with blood and saline, because they are both liquids.

Whenever air is detected, the system immediately stops pumping and heating and the valve closes off flow to the patient. An audible alarm is sounded. If it is a “fluid out” alarm, the user is asked to spike more fluid and press the reprime key on the screen. The user cannot bypass the automated reprime cycle, which removes air from the main fluid circuit and into the reservoir chamber, where there is a hydrophobic air vent. With the patient line closed off by the valve, the pump will push 100 mL of fluid at 500 mL/min to clear the air into the recirculate line and back up into the reservoir chamber. Only after this process has been completed will the “infuse” screen return and allow the operator to continue infusion. If the alarm was “air detection” by the second air detector, the user is asked to open the door and tap the air bubbles that are normally lodged in the sensor into the pressure chamber, which doubles as an air trap. The automated reprime cycle must again follow before the operator can resume infusion.

Heating fluids causes normal out-gassing of air. To prevent this out-gassed air from tripping the “air detect” alarm regularly, the system performs a self-purging cycle every 500 mL of fluid infused. The out-gassed air normally exits the heat exchanger and is collected in the pressure chamber/air trap. The pressure chamber has an approximate capacity of 30 mL. While the system is infusing, the valve will momentarily switch from the infuse position to the close position to force the air in the pressure chamber into the recirculate line and back into the reservoir chamber to be vented. Under fast infusion rates, the duration of the self-purging cycle is a few seconds. If the infusion rate is slow, the system automatically ramps up the recirculate flow rate to complete the process.

Outdated packed red blood cells (PRBC) were obtained from American Red Cross for this protocol.

This investigation was performed in two parts. In Part 1, a rollerhead pump (Master Flex Model 7021-24; Cole Parmer, Chicago, IL) was placed at the position of the IV bag spikes on the Level 1 disposable tubing. This pump was used to adjust flow to match the flow produced by the rollerhead pump of the FMS 2000. Disposable tubing sets on both units were primed according to manufacturers’ instructions. The temperature of the infusate was measured for at least 1 min of flow via a rapid-response thermistor (Keithley Instruments, Cleveland, OH) at two points: at the infusate supply (by insertion of temp probe into the supply bag) and at the distal end of the disposable tubing.

The tubing length from each heat exchanger to the distal end of the disposable tubing was adjusted such that the lengths were equal for both the FMS 2000 and the Level 1. Temperatures were measured at flows of 250 and 500 mL/min for both the FMS 2000 and the Level 1 with saline and then undiluted PRBC. All measurements were performed once because of the limited availability of PRBC.

In Part 2, each device was evaluated as used in clinical practice. To evaluate temperature and flow rates, each device was set up, primed according to the manufacturer’s instructions, and allowed to function at its maximum rate of flow while temperatures at the infusate supply and distal end of the disposable tubing were measured as described previously. PRBC were used as the infusate. Infusate bags were placed in the Level 1 pressurized infusing system, and pressure was set at 300 mm Hg per the manufacturer’s recommendations. The FMS 2000 was set at its maximum flow rate (500 mL/min). The temperature of the infusate supply and of the infusate at the distal end of the disposable tubing was measured for at least 1 min of flow, as described previously. Infusate flow (mL/min) was measured with a graduated cylinder and stop watch. All measurements were performed only once because of the limited availability of PRBC.

The air management system of each device was evaluated by injecting a 10-mL bolus of air (over 1 s) at a site proximal to the heat exchanger in each device while the device ran at maximum flow rate. An air trap was created to capture any air not removed by the air-elimination system of each device by using a graduated cylinder filled with water and inverted in a basin containing water. The distal end of the disposable tubing was placed underwater and directly under the graduated cylinder to contain any air bubbles passing through the tubing. Any air passing through the system (and into a patient, in a clinical setting) was thus captured and measured.

Back to Top | Article Outline


The PRBC infusate supply temperature was 8°C. At metered flows of 250 mL/min, both the Level 1 and the FMS 2000 provided infusate at physiologic temperatures (≥37°C) (Fig. 2). At the faster flow rate tested (500 mL/min), only the FMS 2000 provided infusate at physiologic temperatures (Fig. 4). Additionally, infusate temperature decreased over time with the Level 1 device (Fig. 4).

Figure 2

Figure 2

Figure 4

Figure 4

The infusate supply temperature was 8°C. When each device was allowed to run at maximum flow rates, the FMS 2000 delivered PRBC at physiologic temperature at a maximum flow rate of 500 mL/min (Fig. 3). The Level 1 attained a maximum flow rate of 575 mL/min and an average infusate temperature of 32°C ± 0.2°C (Fig. 3).

Figure 3

Figure 3

When a 10-mL bolus of air was injected into each unit proximal to the heat exchanger, the following results were obtained: a bolus of air was observed passing through the Level 1 air filter and into the distal portion of the disposable tubing. This air was captured as it displaced fluid in the inverted graduated cylinder and had a volume of approximately 10 mL. This is the volume of air that would have passed into a patient in a clinical setting. In the FMS 2000, infusate flow to the distal portion of the disposable tubing was interrupted while air was purged from the disposable tubing. No air was observed in the distal portion of the disposable tubing or in the graduated cylinder.

Back to Top | Article Outline


Slow flows (≤250 mL/min) deliver fluid by both units at physiologic temperatures. However, at more rapid flows (500 mL/min), the FMS 2000 appears superior to the Level 1 in providing PRBC at physiologic temperatures. The air-elimination system of the FMS 2000 was superior to that of the Level 1 and was completely effective in preventing air from entering the patient’s supply line.

Fluid warming has made a major contribution to the maintenance of normothermia (8). Typically these devices are used in patients who are expected to be undergoing a surgical procedure of longer than two hours or for whom large intravascular volume fluid replacement is anticipated. Since the mid 1980s, a number of fluid-warming devices have been available, each of which strives to deliver fluids at 37°C. Most of the devices available today use either water baths or heating plates to warm the fluids before they are administered to the patient (9). Control of the flow is accomplished through the use of pressure infusers and roller clamps.

A study by Patel et al. (7), which compared both conventional and newer types of fluid-warming devices, demonstrated that the efficiency of many warmers is reduced over a range of flows and that in some cases, temperature readouts may be misleading. For example, the authors reported that at moderate flow rates (20 mL/min), the measured outward temperature of the infusate through the Flotem® warmer (Datacem, Indianapolis, IN) was 33.1°C. Despite the fact that the temperature display indicated that the heater plates were set at 37°C, the temperature at the distal end of the 84-cm patient line was 30.8°C. Similarly, the Hot Line® (SIMS Level 1) device was capable of delivering normothermic infusate at flow rates up to 25 mL/min; however, with maximum flow used for pressure infusion (218 mL/min), the temperature of the infusate was 28.9°C. This device is specifically designed for warming at slow flow rates, and the manufacturer does not recommend pressure infusion. However, the current study with the Level 1 rapid infusion system Model H1025 yielded surprisingly similar results. When flow was controlled through the Level 1 by using the rollerpump, a flow of 500 mL/min produced fluid at a near-physiologic temperature for blood (36.7°C). However, a decline in infusate temperature was observed over the period of flow (Fig. 3). When the Level 1 system was pressurized, as in clinical practice, flow rates increased to 575 mL/min. At this flow rate, temperature decreased dramatically to 32.1°C ± 0.4°C. This is consistent with previous laboratory data examining countercurrent heat exchangers (6) and is an important finding, because the rate of fluid infusion with the Level 1 rapid infuser cannot be accurately controlled. Thus, when the system is pressurized at the manufacturer’s recommended 300 mm Hg and maximum flow is allowed to occur, a decline in infusate temperature results. Practitioners should therefore be aware that although rapid-volume infusion is occurring, infusate temperature will be substantially reduced, creating the potential for hypothermia.

While effective fluid-warming technology has significantly reduced the morbidity and costs of medical care directly associated with surgical hypothermia (10,11), there are important safety concerns related to the use of some currently available devices, most notably the potential risk of air infusion. Intraoperative air embolus can be catastrophic. The lethal volume of air when it is injected rapidly (as during rapid-volume administration) is 7.5 mL/kg in dogs (12). In humans, the volume of gas tolerated is not known. However, case reports of accidental venous air injection suggest that a volume of 100–300 mL can be fatal (13–16).

Ten milliliters of air was chosen as a bolus in this experiment because it represents only a fraction of the air embolus size that is thought to be lethal. Also, small amounts of air can result in significant morbidity (17) if the air passes through a patent foramen ovale, as occurs in approximately 25% of the population (18). Because IV fluid bags contain approximately 60 mL of air per 1000 mL of crystalloid (19), careful de-airing is an essential step in avoiding delivery of air to the patient.

There is a high likelihood that air remaining in an emptied IV fluid bag may be delivered to the surgical patient with the use of fluid-warming technology that uses a compressor type of infusion system to drive the contents of the bag through the heating unit. In particular, the absence of an air detector or “fluid out” warning feature on such devices increases the possibility of inadvertent delivery of air, a risk that may be increased when rapid, large-volume fluid resuscitation is necessary (20). Smith (9) recently acknowledged the existence of four cases of massive air embolus during use of pressurized infusions. A review of the ASA closed claims project database (Karen L. Posner, ASA Committee on Professional Liability, personal communication, 2001) for air emboli caused by IV infusions revealed seven claims for air embolism from IV infusions. Three of these claims involved an infusion device, with settlements of up to US$1.6 million per case. The results of this investigation confirm previous findings that the use of the FMS 2000 prevents the infusion of air (21,22).

Due to the lack of air-detection technology used on most fluid-warming devices, air infusion remains a potentially hazardous situation. Pressurized infusion devices require constant vigilance by a member of the anesthesia team. Furthermore, a member of the anesthesia team must “de-air” the IV bags before placing them into the compressors. Air filters provided with these systems are designed only to remove out-gassed air at the rate of 14 mL/min (20) and therefore are ineffective in managing a large air bolus that may be injected with empty or partially “de-aired” IV bags. Additionally, air-elimination filters often fail in their venting function when the hydrophobic membrane becomes wet. 1 Indeed, an advisory notice mailing from SIMS Level 1 reminded users of steps to be taken to avoid introducing air into the patient (23).

Fluid infusers and warmers can play a major role in preventing perioperative hypothermia; however, the use of pressure infusers with uncontrollable flow rates and limited air management systems can be a patient safety hazard because of the potential for infusion of air and ineffective fluid warming at rapid flow rates. In this laboratory investigation, the FMS 2000 performance was superior to that of the Level 1 H1025 with respect to providing PRBC at physiologic temperatures at rapid (500 mL/min) flow rates and with respect to air detection and elimination.

The author acknowledges Jill LaRue for her assistance in the preparation of this manuscript.

Back to Top | Article Outline


1 Philip JH. Performance of Level 1 and Microwave Medical Systems air eliminators [abstract]. Anesthesiology 1999;91:A513.
Cited Here...

Back to Top | Article Outline


1. Collins JA. Problems associated with the massive transfusion of stored blood. Surgery 1974; 75: 274–95.
2. Hervey GR. Proceedings: physiological changes encountered in hypothermia. Proc R Soc Med 1973; 66: 1053–8.
3. Boyan CP, Howland WS. Blood temperature: a critical factor in massive transfusion. Anesthesiology 1961; 22: 559–63.
4. Boyan CP. Cold or warmed blood for massive transfusions. Ann Surg 1964; 160: 282–6.
5. Russell WJ. A review of blood warmers for massive transfusion. Anaesth Intensive Care 1974; 2: 109–30.
6. Uhl L, Pacini D, Kruskall MS. A comparative study of blood warmer performance. Anesthesiology 1992; 77: 1022–8.
7. Patel N, Knapke DM, Smith CE, et al. Simulated clinical evaluation of conventional and newer fluid warming devices. Anesth Analg 1996; 82: 517–24.
8. Muth CM, Mainzer B, Peters J. The use of countercurrent heat exchangers diminishes accidental hypothermia during abdominal aortic aneurysm surgery. Acta Anaesthesiol Scand 1996; 40: 1197–201.
9. Smith CE, for the ITACCS Seminar Panel. Principles of fluid warming and trauma. In: Smith CE, Rosenberg AD, Grande CM, eds. Massive transfusion and control of hemorrhage in the trauma patient. Baltimore, MD: International Trauma Anesthesia and Critical Care Society, 1999: 30–4.
10. Bush ML, Hydo LJ, Fischer E, et al. Hypothermia during elective abdominal aortic repair: the high price of avoidable morbidity. J Vasc Surg 1995; 21: 392–400.
11. Kurz A, Sessler DI, Lenhardt R, et al. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization: Study of Wound Infection and Temperature Group. N Engl J Med 1996; 334: 1209–15.
12. Oppenheimer MJ, Durant TM, Lynch P. Body position related to venous air embolism and associated cardiovascular-respiratory changes. Am J Med Sci 1953; 225: 362–73.
13. Yeakel AE. Lethal air embolism from plastic blood storage container. JAMA 1968; 204: 267–9.
14. Martland HS. Air embolism: fatal air embolism due to power insufflators used in gynecologic treatments. Am J Surg 1945; 68: 164–9.
15. Toung TJK, Rossberg MI, Hutchins GM. Volume of air in a lethal venous air embolism. Anesthesiology 2001; 94: 360–1.
16. Palmon SC, Moore LE, Lundberg J, Toung T. Venous air embolism: a review. J Clin Anesth 1997; 9: 251–7.
17. Orebaugh SL. Venous air embolism. Crit Care Med 1992; 20: 1169–77.
18. Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first ten decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1984; 59: 17–20.
19. Hartsmannsgruber WB, Gravenstein N. Very limited air elimination capability of the Level 1 fluid warmer. J Clin Anesth 1997; 9: 233–5.
20. Adhikary GS, Massey SR. Massive air embolism: a case report. J Clin Anesth 1998; 10: 70–2.
21. Smith CE, Kabbara A, Kramer RP, Gill I. A new IV fluid and blood warming system to prevent air embolism and compartment syndrome. Trauma Care 2001; 11: 78–82.
22. Comunale ME. Comparison of the fluid warming and flow capabilities of the Belmont Instrument FMS 2000 and the SIMS Level 1 1000. N Y Postgrad Assembly Anesthesiol 1999; 116: P9030.
23. SIMS Medical Systems. Advisory notice: reminder of label warnings. Rockland, MA: SIMS Medical Systems, 2001.
© 2003 International Anesthesia Research Society