Reusable versus disposable forced air warming

Abstract

BACKGROUND

Perioperative hypothermia increases the risk of surgical complications making effective patient warming essential. Disposable forced air warming blankets, while effective, contribute significantly to healthcare waste and environmental impact. Reusable alternatives may offer similar efficacy while reducing environmental footprint, but their thermal performance requires rigorous evaluation.

OBJECTIVE

To assess the safety and effectiveness of reusable textile blankets as an alternative to disposable forced air warming blankets, as a first step toward developing more sustainable perioperative practices.

DESIGN

We conducted an in-vitro, prospective, head-to-head comparison of heating performance in full-body and upper body designs, using three blanket types in a simulated operating-room scenario. Blanket types included disposable single use (DSU), light reusable cotton (LRC) and heavy reusable cotton (HRC). A single forced-air warmer was set to 38 °C for all tests, which simulated both normothermic (37 °C) and hypothermic (25 °C) patient conditions, measuring surface temperatures at six anatomically relevant points across the simulated body.

SETTING

Laboratory-based simulation in a controlled environment simulating operating room conditions with ambient temperature maintained at 20 ± 0.5 °C.

PARTICIPANTS

A standardised simulated patient thermal model set to either normothermic (37 °C) or hypothermic (25 °C) conditions.

MAIN OUTCOME MEASURE

Surface temperature and temperature homogeneity were measured at six anatomically relevant points across the simulated body.

RESULTS

The heavy cotton blanket provided the highest and most homogenous surface temperature in full-body designs (36.7 ± 0.6 °C) (DSU = 35.4 ± 1.1 °C; LRC = 35.7 ± 1.7 °C) (P < 0.05) and upper body designs (35.5 ± 0.5 °C) (DSU = 35.4 ± 0.5 °C; LRC = 34.9 ± 2.4 °C) (P < 0.05). Unsafe temperature elevations were not recorded for any of the test configurations. The statistical analysis revealed significant differences between blanket types (P < 0.05), with heavy reusable cotton consistently outperforming both disposable single-use and light-reusable cotton options in terms of temperature distribution.

CONCLUSION

Carefully designed reusable blankets may provide a safe and environmentally friendly alternative to disposable forced air warming blankets, potentially reducing costs and environmental impact. Additional research examining infection control, durability through multiple laundering cycles and comprehensive life cycle analysis would further validate these promising findings.

KEY POINTS

• Reusable heavy cotton blankets demonstrated superior warming performance and more homogeneous temperature distribution compared to disposable alternatives in forced air warming systems.

• Implementation of reusable warming blankets may represent a significant opportunity for developing green operating rooms by reducing medical waste while maintaining or improving patient care standards.

• The study found no safety concerns with reusable alternatives, with no unsafe temperature elevations recorded in any experimental configurations.

Introduction

Inadvertent perioperative hypothermia is a common issue, affecting up to 70% of surgical patients if preventive measures are not implemented. It may cause a wide range of complications, including patient discomfort, increased incidence of bleeding and blood transfusion¹ and elevated risk of infection and cardiovascular morbidity,² significantly affecting healthcare costs.³ Maintaining normothermia is a combined challenge that minimises radiation, conduction, evaporation and convection heat losses and by active warming through conductive (warming mattresses and garments) and convective (forced air warmer) devices.²

Forced air warmers reduce the risk of hypothermia and are readily available, easy to apply and operate.² As a result, many consider their use as standard practice for all surgical procedures lasting longer than 20 min. These devices circulate warm air through specially designed blankets, evenly distributing heat over the patient's body. The clinical effectiveness of forced air warming systems relies heavily on the blanket design, which must ensure a consistent and uniform temperature distribution across the contact surface.⁴

Most commercially available forced air warming blankets are disposable and designed for single use. Although practical, their frequent usage creates significant medical waste and can substantially increase healthcare costs. It is estimated that approximately 20 million forced air warming blankets are used globally each year, contributing to the growing environmental impact of healthcare systems.⁵

Plastic pollution from healthcare is a growing concern for environmental and human health. Disposable medical devices contribute significantly to healthcare-related plastic waste, with operating rooms (OR) generating approximately 30% of hospital waste,⁶ with 50–80% of this OR waste is used prior to the sterile phase is initiated, therefore, the majority of OR waste is clean, uncontaminated and much of it can be recycled.^7,8 While transitioning from disposable to reusable devices potentially offers environmental benefits, such transitions must first establish equivalent or superior clinical performance. The eco-design of care involves choosing the less environmentally damaging of two equally effective medical technologies. Therefore, comparing the thermal performance of reusable versus disposable warming blankets represents a critical first step before comprehensive sustainability assessments can be justified.

This study evaluated the safety and efficacy of reusable textile blankets as alternatives to disposable forced air warming blankets. Determining performance equivalence or superiority is a necessary first step in the eco-design process towards reducing plastic waste and contributing to more sustainable healthcare practices.

Methods

Ethical approval

The Institutional Review Board waived the requirement for ethical approval, because this investigation was conducted in a laboratory setting without involving human or animal subjects.

Experimental design

We performed an in vitro, prospective, head-to-head comparison of the heating performance of two blanket designs (full-body model 610 and upper-body model 502, Bair Hugger, 3M, USA) in three different blanket materials in a simulated operating room scenario:

• (1)Disposable single use (DSU): the original, single-use commercial product.
• (2)Light reusable cotton (LRC): light cotton fabric (hospital bed sheet, approximately 120 gsm) specially cut and sewn to the same size, shape and design as a disposable blanket.
• (3)Heavy reusable cotton (HRC): heavy cotton fabric (surgical drape, approximately 210 g) specially cut and sewn to the same size, shape, and design as a disposable blanket.

Testing environment and equipment

All tests were conducted using a single forced-air warmer (Temperature Management Unit Model 775, Bair Hugger, 3M, USA) set to 38 °C. The ambient temperature was maintained at 20 ± 0.5 °C using the hospital's central air conditioning system and was continuously monitored at 30 s intervals throughout the experiments, that took place in 2022 to 2023.

Simulated patient model

To create a standardised representation of human thermal characteristics, we developed a patient simulation model consisting of

• (1)Two rolled wool blankets (to represent body mass and provide thermal insulation properties)
• (2)A water-filled thermal blanket (Model 122922, Paragon Medical, USA) connected to a cardiopulmonary bypass cooler system (Stöckert Heater-Cooler System 3T, Sorin Group USA, Inc., Arvada, Colorado, USA)

This model allowed us to simulate both normothermic (37 °C) and hypothermic (25 °C) conditions. The patient simulation model was constructed to approximate the dimensions of an average adult (height, 175 cm; shoulder width, 45 cm and hip width, 35 cm). Two rolled wool blankets (each 150 × 200 cm when unfolded) were tightly rolled to create a cylindrical core with a diameter of approximately 20 cm to simulate the thermal mass of a human torso. This core was then wrapped with a water-filled thermal blanket (Model 122922, Paragon Medical, USA) to provide uniform temperature control across the entire surface. The completed model presented a contact surface area of approximately 0.95 m², comparable to the exposed surface area of an average adult patient during surgery. Before beginning each experiment, the test system was set to the target temperature and equilibrated for 4 h to ensure even temperature distribution throughout the model.

Temperatures of 37 °C and 25 °C were selected to represent clinically relevant scenarios. The 37 °C setting simulates a normothermic patient requiring temperature maintenance during surgery, which is the most common clinical objective of forced air warming.^9,10 The 25 °C setting represents significant hypothermia, which can occur during major surgical procedures with large exposed body cavities, substantial fluid resuscitation or in trauma patients.¹¹ This lower temperature provides a more challenging thermal environment for evaluating blanket performance, as more significant heat transfer is required to warm the patient model, potentially highlighting differences in blanket efficacy that might not be apparent under less demanding conditions.

Temperature measurement protocol

The surface temperature was measured for 4 h for each experimental setup, with recordings taken at 30 s intervals at six anatomically relevant points distributed over the simulated body:

• (1)Upper chest (two measurement points)
• (2)Mid-abdomen (two measurement points)
• (3)Thigh (one measurement point)
• (4)Lower leg (one measurement point)

The measurement locations were selected to correspond to the areas of clinical significance for patient warming during surgery. Six temperature sensors (USB-501, Measurement Computing Corporation, China) with an accuracy of ±0.5 °C were securely attached to the model's surface at consistent locations across all experimental conditions, and a seventh sensor monitored the room temperature. Data were directly transferred to a laptop computer for analysis.

Experimental protocol

Each of the three blanket types (DSU, LRC and HRC) in each of the two designs (full body and upper body) were tested three times under both normothermic (37 °C) and hypothermic (25 °C) conditions, yielding a total of 36 experiments, 144 h of recording, and 10 368 individual temperature measurements.

The test sequence was randomised to minimise any potential order effects, and all equipments were calibrated according to the manufacturer's specifications before the study began. All equipment was allowed to return to ambient temperature between the experiments, and fresh blankets were used for each test.

Statistical analysis

Sample size justification

The sample size determination was based on previous similar studies of forced air warming systems,¹² indicating that with three groups (blanket types), a sample size of 18 measurements per group would achieve 80% power to detect an effect size of f = 0.40 (medium–large) at α = 0.05. With six measurement points per experiment and three repetitions under each condition, our design yielded 18 measurements per blanket type in four test configurations (full/upper body × 25 °C/37 °C), meeting the minimum required sample size. Continuous recording over 4 h (480 measurements per sensor per experiment) also provided robust data for assessing the temperature stability over time.

Analysis

Results are reported as mean ± standard deviation, unless specified otherwise. We performed a one-way ANOVA to compare outcomes across blanket types, with Tukey's honestly significant difference (HSD) post hoc test for pairwise comparisons between groups. Statistical significance was defined as P less than 0.05. Effect sizes were calculated using η² (eta-squared) to quantify the proportion of variance in temperature explained by the blanket-type differences. All analyses were performed using SPSS v25 (IBM Corp., Armonk, New York, USA).

Results

All 144 recordings were successfully completed. No significant changes in temperature over time were observed in any of the six setups after the initial stabilisation, regardless of the model temperature, blanket design or material.

No unsafe temperature elevations (>40 °C) were measured in any of the experiments, confirming the basic safety profile of all three blanket types when used with the tested forced-air warming system.

The results are reported as mean ± standard deviation unless specified otherwise (Table 1, Fig. 1).

Table 1.

Mean, and SD and one-way ANOVA results comparing three blanket types (single-use, heavy fabric, light fabric) across four conditions

	Single use (n = 18)		Heavy reusable (n = 18)		Light reusable (n = 104)
	M	SD	M	SD	M	SD	F
Full-body blankets, 37 °C	35.48	1.11	36.79	0.62	35.78	1.72	5.484a
Full-body blankets, 25 °C	28.53	1.77	30.39	1.45	29.43	3.12	3.110
Upper body blankets, 37 °C	35.53	0.48	36.22	0.54	34.93	2.36	3.714a
Upper body blankets, 25 °C	28.94	1.16	30.31	0.90	28.96	3.39	2.468

^aP < 0.05.

Fig. 1.

The schematic illustrations of the assigned temperature sensor's location over the simulated full body.

Full-body blanket design at 37 °C

A one-way ANOVA was performed to compare the mean temperature of the three different blankets in the full-body design with the model temperature set to 37 °C (Table 2). The ANOVA revealed a statistically significant difference in temperature means [F(2,51) = 5.484, P = 0.007, η² = 0.18].

Table 2.

One-way ANOVA results comparing three blanket types (single-use, heavy fabric, light fabric) across four conditions

Condition	Source	Sum of squares	df	Mean square	F	P	η 2
Full-body blankets, 37 °C	Between groups	16.75	2	8.372	5.484	0.007*	0.18
	Within groups	77.87	51	1.527
	Total	94.61	53
Full-body blankets, 25 °C	Between groups	30.94	2	15.472	3.110	0.053	0.11
	Within groups	253.71	51	4.975
	Total	284.66	53
Upper body blankets, 37 °C	Between groups	15.15	2	7.576	3.714	0.031*	0.13
	Within groups	104.04	51	2.040
	Total	119.19	53
Upper body blankets, 25 °C	Between groups	22.41	2	11.203	2.468	0.095	0.09
	Within groups	231.47	51	4.539
	Total	253.87	53

η² values were calculated as SSBetween/SSTotal.

^*P < 0.05.

Among the three blankets, the HRC blanket showed the highest mean temperature (36.79 ± 0.62 °C) compared to DSU (35.48 ± 1.11 °C) and LRC (35.78 ± 1.72 °C). In Tukey's post hoc analysis, we found significant differences between the HRC blanket and DSU blanket (mean difference = 1.30 °C, P = 0.007) and between the HRC blanket and LRC blanket (mean difference = 1 °C, P = 0.047).

Full-body blanket design at 25 °C

For the full-body design with the model temperature set to 25 °C, ANOVA revealed a nearly statistically significant difference in temperature means [F(2,51) = 3.110, P = 0.053, η² = 0.11] (Table 2). Although the overall ANOVA was marginally nonsignificant, we conducted planned post hoc comparisons, given the exploratory nature of this study.

The HRC blanket again showed the highest mean temperature (30.39 ± 1.45 °C) compared to DSU (28.53 ± 1.77 °C) and LRC (29.43 ± 3.12 °C). Although this P-value slightly exceeded the conventional significance threshold of 0.05, we proceeded with planned post hoc comparisons for several reasons: the P value was very close to the threshold, suggesting a likely meaningful effect that may have reached conventional significance with a larger sample; the effect size (η² = 0.11) indicates a medium practical effect, irrespective of the P value; and the exploratory nature of this study warrants the examination of potential patterns across all testing conditions. Consistent with our transparent reporting approach, we acknowledge that these comparisons should be appropriately interpreted. Therefore, we conducted post hoc comparisons, which revealed a significant difference between the HRC blanket and DSU blanket (mean difference = 1.85 °C, P = 0.041), but no significant difference between the HRC and LRC blankets (mean difference = 0.96 °C, P = 0.408) (Table 2).

Upper body blanket design at 37 °C

For the upper body design with the model temperature set to 37 °C, ANOVA showed a statistically significant difference in temperature means (F(2,51) = 3.714, P = 0.031, η² = 0.13) (Table 2).

The HRC blanket demonstrated the highest mean temperature (36.22 ± 0.54 °C) compared to DSU (35.53 ± 0.48 °C) and LRC (34.93 ± 2.36 °C). Tukey's post hoc test identified a significant difference between the HRC blanket and LRC blanket (mean difference = 1.30 °C, P = 0.024), but no significant difference between the HRC blanket and DSU blanket (mean difference = 0.69°C, P = 0.324) (Table 3).

Table 3.

Post hoc comparisons (Tukey's honestly significant difference test)

Condition	Comparison	Mean difference	Standard error	P
Full-body blankets, 37 °C	Single-use vs. heavy	−1.30*	0.41	0.007*
	Single-use vs. light	−0.30	0.41	0.751
	Heavy vs. light	1.00*	0.41	0.047*
Full-body blankets, 25 °C	Single-use vs. heavy	−1.85*	0.74	0.041*
	Single-use vs. light	−0.90	0.74	0.456
	Heavy vs. light	0.96	0.74	0.408
Upper body blankets, 37 °C	Single-use vs. heavy	−0.69	0.48	0.324
	Single-use vs. light	0.61	0.48	0.416
	Heavy vs. light	1.30*	0.48	0.024*
Upper body blankets, 25 °C	Single-use vs. heavy	−1.37	0.71	0.139
	Single-use vs. light	−0.02	0.71	1.000
	Heavy vs. light	1.36	0.71	0.145

CI, confidence interval.

^*P < 0.05.

Upper body blanket design at 25 °C

For the upper body design with the model temperature set to 25 °C, ANOVA did not show a statistically significant difference in temperature means [F(2,51) = 2.468, P = 0.095, η² = 0.09].

Although the HRC blanket showed the highest mean (30.31 ± 0.90 °C) compared to DSU (28.94 ± 1.16 °C) and LRC (28.96 ± 3.39 °C), the differences did not reach statistical significance.

Temperature homogeneity

In addition to the mean temperatures, we analysed the standard deviation of the temperature measurements to assess temperature homogeneity (Table 2).

The HRC blanket consistently showed the lowest standard deviation in surface temperatures (full-body 37 °C : 0.62 °C; upper body 37 °C : 0.54 °C) and (full-body 25 °C : 1.45 °C; upper body 25 °C :  0.90 °C), indicating a more homogeneous temperature distribution compared with both DSU and LRC blankets across most test configurations.

Discussion

This experimental comparative study demonstrates that carefully designed reusable forced air warming blankets can match or exceed the thermal performance of disposable alternatives. The HRC blanket consistently provided superior warming efficacy and temperature homogeneity across both full-body and upper body designs compared to both DSY and LRC options.

Efficacy of reusable versus disposable blankets

Many factors influence the efficacy of forced-air warming systems, including blanket design, material properties, and airflow dynamics.^4,9,13 The ideal forced-air warming blanket provides an optimal heat distribution with minimal temperature variance across the patient's body surface. Performance evaluation typically focuses on temperature homogeneity and mean surface temperature, because these factors directly influence clinical effectiveness.^14,15

Our findings demonstrate that HRC provided superior performance in terms of both mean temperature and temperature distribution homogeneity. The HRC blanket achieved statistically higher mean temperatures than the disposable alternative in two of the four test configurations and consistently showed better temperature uniformity across all configurations. This aligns with previous research by Bräuer et al.,¹⁰ who identified material properties as critical determinants of forced-air warming efficiency.

The superior performance of the HRC blanket can be explained by several principles of thermodynamic and material science. Firstly, the higher density fabric (~210 vs. ~120 gsm) provides an enhanced thermal mass, which helps maintain temperature stability by resisting rapid fluctuations.¹⁶ Secondly, the specific weave structure of heavier cotton fabrics creates an optimal balance between insulation and air permeability.¹⁶ This balance is crucial for forced-air warming systems, which require sufficient porosity to allow warmed air circulation while maintaining sufficient thermal resistance to prevent excessive heat loss.¹⁷ Thirdly, cotton fibers have inherent hygroscopic properties that moderate humidity, potentially creating a more stable microenvironment at the blanket–patient interface.¹⁸

Comparing our results with those of previous studies, Sessler et al. demonstrated that maintaining temperature homogeneity is as important as achieving high mean temperatures to prevent perioperative hypothermia. The superior homogeneity of our HRC blanket (as evidenced by lower standard deviations across measurement points) suggests that it may provide more consistent patient warming, potentially reducing ‘cold spots’ that could compromise clinical outcomes.

Thermal performance and patient safety

Therefore, temperature safety is essential for patient warming. No unsafe temperature elevations (>40 °C) were observed during our experiments, suggesting that all the three blanket types maintained acceptable safety profiles under controlled conditions. This is consistent with the well established safety records of properly functioning forced-air warming systems in clinical practice.^19,20

The superior temperature homogeneity of the HRC blanket may offer additional safety benefits beyond those measured in our study. As Hopf ²¹ noted, excessive localised heating can potentially cause thermal discomfort or contribute to developing pressure injuries in vulnerable patients in extreme cases. A more uniform temperature distribution observed with the HRC blanket might minimise such risks, although clinical studies are needed to confirm this hypothesis.

Recent meta-analyses have confirmed that maintaining normothermia reduces surgical site infections by 64% and cardiac complications by 55%.^22,23 The improved thermal performance of the HRC blanket observed in our study could potentially translate to better clinical outcomes, particularly for more prolonged procedures or vulnerable patient populations, where maintaining normothermia is challenging.²⁴

Sustainability and economic implications

Healthcare systems are increasingly focusing on sustainable practices, with the environmental impact of medical waste receiving particular attention.^25,26 Disposable medical textiles, including forced air warming blankets, contribute significantly to healthcare waste streams. McGain et al. ²⁷ estimated that operating rooms generate approximately 30% of hospital waste, with single-use products being major contributors.

Quantitative life cycle assessments comparing reusable and disposable surgical textiles have demonstrated substantial environmental benefits for reusable systems. Overcash (5) found that reusable surgical textiles produce 65% less energy consumption, 73% less greenhouse gas emissions, and 93% less solid waste by weight than their disposable counterparts. Similarly, Vozzola et al. ²⁸ demonstrated that reusable surgical gowns have a 50–75% lower global warming potential than disposable alternatives, depending on the specific products compared.

From an economic perspective, long-term savings often offset initial investments in reusable systems. Conrardy et al. ²⁹ conducted a comprehensive cost analysis of reusable and disposable surgical textiles. They found that reusable systems yielded net savings of 39–50% when accounting for all costs, including purchase, laundering, sterilisation and disposal. By applying these findings to forced air warming blankets, we can project potential cost savings.

For a medium-sized hospital performing 10 000 surgical procedures annually, the approximate annual cost of disposable forced air warming blankets is $15–25 per unit × 10 000 procedures = $150 000–250 000. Estimated reusable system costs: initial investment ($100–150 per blanket × 300 blankets = $30 000–45 000) plus laundering costs ($3–5 per cycle × ~3500 cycles = $10 500–17 500 = $40 500–62 500 annually. Potential annual savings: $109 500–187 500 (58–73%).²⁹ In line with the economic cost, the estimated plastic weight saved from the DSU in our hospital for 2024 is 2756 tons. The full-body blanket weighs 162 g, while the upper body blanket weighs 146 g. In 2024, we used approximately 10 000 upper body blankets and 8000 full-body blankets. The calculation is as follows: 10 000 × 146 g + 8000 × 162 g = 2 756 000 g.

These projections align with findings from DiConsiglio,³⁰ who documented case studies of hospitals achieving payback periods of 6–18 months after transitioning to reusable surgical textiles. The economic case is further strengthened when considering waste disposal costs, which continue to rise in many regions.³¹

Infection control remains a critical consideration in the evaluation of reusable medical textiles. Modern laundering and quality control processes have largely addressed historical concerns regarding infection risk. Recent systematic reviews by Rutala and Weber³² and Kieser et al. ³³ found no evidence of increased infection rates associated with properly processed reusable surgical textiles compared with disposable alternatives. Indeed, some studies have suggested that disposable nonwoven fabrics may shed more particulate matter than reusable woven textiles, potentially increasing airborne contamination.⁵

Our study demonstrates that reusable cotton blankets maintain, and in some configurations, the thermal performance of disposable alternatives. Thus, the combined environmental, economic, and clinical arguments for transitioning to reusable forced-air warming blankets are compelling.

Limitations and future directions

This study has several limitations that should be addressed in future research. The use of an in vitro thermal model to simulate human body characteristics, while providing standardised and reproducible conditions, may not fully replicate the complex thermal dynamics of human patients, necessitating further clinical validation studies. We did not evaluate the performance of the reusable blankets after multiple laundering cycles, which is essential for determining their long-term efficacy and cost-effectiveness in real-world settings. Future studies should implement a systematic protocol that evaluates reusable blankets after standardized wash cycles (e.g. at 0, 10, 20, 30 and 40 cycles) to assess durability and performance consistency after repeated laundering. These evaluations should include thermal performance testing identical to the current study, material integrity assessment using standardised textile testing methods (tensile strength, air permeability and thickness measurements), microbial sampling to ensure decontamination efficacy, cost analysis accounting for replacement needs, and a Life Cycle Assessment to quantify the actual environmental impacts of reusable vs. disposable warming blankets. Such a comprehensive approach would provide robust data on the long-term viability of reusable warming blankets and strengthen the economic and environmental case for their implementation in clinical practice. A more detailed cost–benefit analysis that includes procurement, processing, laundering and disposal costs would provide more precise estimates of the economic advantages of reusable systems. Additionally, a full life cycle assessment of disposable and reusable forced-air warming blankets would better quantify the potential environmental benefits beyond our current projections and would strengthen the sustainability argument for transitioning to reusable alternatives.

Conclusion

Carefully designed and tested reusable blankets may provide an environmentally friendly alternative to disposable forced-air warming blankets, potentially reducing costs and environmental impacts. The heavy cotton reusable blanket (210 gsm) demonstrated superior warming performance and temperature homogeneity compared to light cotton reusable and disposable alternatives.

Fabric density and air permeability play significant roles in blanket efficiency, with heavier cotton fabrics providing better temperature distribution than lighter materials. These findings suggest that adequately designed reusable forced-air warming blankets could be introduced into clinical practice without compromising patient safety or warming efficacy.

Future research should focus on comprehensive life cycle assessments, clinical validation, durability through multiple laundering cycles, infection control protocols, and comprehensive economic and environmental impact analyses to further support the implementation of sustainable alternatives for perioperative warmingFig. 2 and Fig. 3.

Fig. 2.

Mean surface temperature across measurement points, across the four simulated designs.

Fig. 3.

The schematic illustrations of the assigned temperature sensor's location over the simulated upper body.