Experimental evaluation of infusion methods and administration speeds in confined spaces during urban search and rescue operations

1 Introduction

On-site rescue and medical cooperation are recognized as necessary in urban search and rescue (USAR) missions during disaster relief operations. Confined space medicine is required, especially in patients with crush syndrome, where adequate fluid infusion before decompression is critical. Occasionally, unique infusion methods must be used in rescue operation settings where there is insufficient height difference between the ceiling and the ground. Rescue personnel and medical personnel must comprehend each other's technical procedures.

However, strong evidence for infusion therapy in confined space (CS) is lacking. The International Search and Rescue Advisory Group recommends an initial fluid load of 1–1.5 L/h in the first 2 h for adults to avoid hemodynamic collapse (1). Another review on crush syndrome suggests an isotonic saline rate of 1,000 ml/h (10–15 ml/kg/h) while the victim is trapped under rubble (2). The Japan disaster relief (JDR) Rescue Team Medical Unit advises completing at least 2,000 ml before decompression and continuing at 500–1,000 ml/h afterward (3).

Notably, even in times of disaster, the team operation must follow the laws of the affected country where the team belongs. During JDR operations, Japanese law currently prohibits the use of mechanical pumps by JDR paramedics. In addition, no infusion pump that is suitable for CS is currently approved in Japan. Therefore, any method other than spontaneous infusion must be carried out by nurses or medical doctors. These personnel are not extensively trained in CS operations as rescuers. Aiming for the reduction of time under CS decreases operational risks.

Knowledge regarding infusion methods and speed in CS is limited. Few studies exist on the speed of peripheral venous infusion (4) or venous pressure (5). The lack of data on field-adapted infusion therapy— especially in low clearance environments—has affected field rescue plans and safe medical care. The primary research question of this study is which infusion methods optimize speed and safety in confined spaces, where gravitational perfusion is ineffective and rescuer exposure must be minimized.

To simulate real-life conditions affecting infusion speed, venous pressure was replicated, and an intravenous infusion model was established based on experiments on healthy participants. An exploratory study was conducted to assess infusion rates. This study aimed to provide data on the infusion methods and administration speeds in CS required for on-site rescue and medical cooperation during USAR missions. This study was approved by the Ethical Review Board of the Aichi Medical University School of Medicine (Approval Number: 2022–018). Consent for the use of each device and data publication was obtained from each company.

2 Materials and methods

2.1 Human peripheral venous pressure and infusion speed

Healthy male participants aged 20–29 years were recruited from students of Aichi Medical University School of Medicine. Individuals with medical complications—especially cardiovascular disease, upper limb vascular or lymphatic disease, or dialysis shunts—those who were severely underweight or obese (body mass index [BMI] < 18 kg/m² or >30 kg/m²), or anyone deemed inappropriate by the investigators were excluded. Finally, 20 participants were enrolled in the study.

The procedures and risks were explained to all participants, and written informed consent was obtained. Information on age, height, weight, medical history, amount of water consumed within 12 h before the experiment, and the time of last fluid intake. Vital signs, central venous diameter and respiratory variation, peripheral venous diameter at the forearm, and blood flow velocity were measured by pulse Doppler method with ultrasonography (MicroMaxx^{^®} Ultrasound System; Sonosite, Inc., Bothell, WA, USA).

A peripheral forearm vein was cannulated with a 20 G needle (Surshield Surflo II^{^®} flush-type 20 G × 1, 1/4 inch; Terumo Corporation, Tokyo, Japan). Peripheral venous pressure was measured using a water manometer in the supine position. Infusion route for adults 125 cm (TI-U350; Terumo Corporation) and extension tube 50 cm (TS-WR1735L; Terumo Corporation) were connected with a total height of 90 cm. The time to administer 500 ml of Lactate Ringer Solution (SOLULACT^{^®} Infusion; Terumo Corporation) at full speed was measured. Results were statistically analyzed to assess the effect of age, height, weight, medical history, fluid intake 12 h prior, and the last time of water intake on infusion rate.

2.2 Establishing the infusion model using a simulator arm

Based on the human experiment results, we established an experimental model with the same height and tube length. The differences between the infusion line's lower end and the top of the cylinder was set to 19.8 cm on a vascular access simulator arm (V-Line^{^®}; Kyoto Kagaku Co., Ltd., Kyoto, Japan), as shown in Figure 1.

2.3 Infusion experiment with various methods

Using the established infusion model, the administration speed was measured using various infusion methods, as shown in Figure 2. Measurements were performed seven times for each method.

2.3.1 Spontaneous dripping

The infusion route for adults (125 cm) and extension tube (50 cm) were connected 90 cm high (same as in the human experiment).

2.3.2 Manually pressurized bag + re-pressure

Inspiron^{^®} pressurized bag for 1,000 ml (IN9,000K; MC Medical Inc., Ponce, Puerto Rico), followed by re-pressure after 10 min (because infusion speed decreases by half in 10 min).

2.3.3 Tube extension + pumping

Infusion route for adults 125 cm and extension tube 50 × 10 cm = 625 cm connected with 50 cm height + pumping (repeating manual bolus with 20 ml syringe).

2.3.4 Droper

Automated pressure device with continuous pressure by spring-loaded tray (Droper^{^®}; Promovet SARL, Givet, France).

2.3.5 AutoCuff

Automated pressure device with continuous pressure by battery (AutoCuff^{^®}; Aztech Co., Ltd., Tokyo, Japan).

2.3.6 Terumo

Automated pressure device with temporary pressure by battery (PG pressure Q series^{^®} pressurized bag + electric pump PE-PR40DP PE-PR40DPB; Terumo Corporation) followed by re-pressure after 10 min.

2.3.7 Nipro

Automated pressure device with temporary pressure by battery (Nipro pressure pump EN pressurized bag CS^{^®}; Nipro Corporation, Osaka, Japan) followed by re-pressure after 10 min.

2.4 Statistical analysis

Values are presented as median (interquartile range). Data were analyzed using the Mann–Whitney U-test between two groups, and the Kruskal–Wallis-test with Dunn's multiple comparison test between more than three groups. Spearman's rank correlation coefficient was used for the correlation analysis. Statistical analyses were performed using GraphPad Prism 9 (GraphPad Inc., San Diego, CA, USA), with significance set at P < 0.05.

3 Results

3.1 Human peripheral venous pressure and infusion speed

The median time to administer 500 ml lactated Ringer's solution in 20 participants was 16 min 8 s (14 min 33 s−17 min 33 s). Peripheral venous pressure was 21.5 (16.5–26.1) cm H₂O, and flow velocity was 3.85 (0.96–5.34) cm/s. Infusion time was significantly negatively correlated with flow velocity (P < 0.05) and weakly positively correlated with BMI (P = 0.07) (Figure 3). No correlations were found with other parameters (venous pressure, fluid intake, time of last drinking, venous diameter, and central venous respiratory variation).

3.2 Establishing infusion model using a simulator arm

The results of preliminary measurements on seven simulators showed that there was no statistically significant difference between the 500 ml infusion time of 15 min 27 s (14 min 50 s−16 min 10 s) and that of human infusion, and the model was considered adequate (Figure 4).

3.3 Infusion experiment with various methods

The time to administer 1 L of infusion solution was measured using the model; results are shown in Table 1 and Figure 5. Compared to spontaneous dripping, tube extension + pumping (P < 0.0001), pressurized bag + re-pressurization after 10 min, and automated devices, such as AutoCuff (P < 0.001) and Nipro (P < 0.05), had significantly shorter administration times. Other automated pressure devices, such as Droper and Terumo, tended to have shorter administration times; however, the difference was not significant.

Table 1. Time required to administer 1 L of infusion solution via each administration method.

4 Discussion

In USAR operations during disaster relief, rescue and medical cooperation are critical to save lives. One such common case is crush syndrome, caused by muscle compression from heavy debris, especially during earthquakes. This can result in serious conditions such as hypovolemic shock or kidney injury after decompression (2). In Japan, where earthquakes often occur, many victims suffered from crush syndrome during the 1995 Hanshin-Awaji earthquake (6). Several reports highlighted treatments such as infusions and dialysis (7); with early massive infusion is especially important for preventing renal failure and avoiding dialysis (8, 9). Usuda et al. emphasized the importance of fluid infusion before decompression (10).

In CS, gravity-dependent rapid dripping is ineffective due to height, and techniques such as pressurized bags or pumping to shorten the medical personnel's working time. Experimental studies have been conducted on pressurized infusion and velocity in CS (11, 12). White et al. demonstrated that pressure infusers more than doubled IV flow rates and clearly outperformed all other field methods. Shirokawa et al. demonstrated that infusion using pressure bags or body weight pressure is effective during disasters. Although previous studies have examined pressurized infusion, many used open-air models lacking venous resistance, limiting clinical relevance.

Given the limited knowledge about infusion methods and speed in CS, an exploratory study on the administration speed of various infusion methods was conducted using an infusion model based on healthy human participants. Each method was evaluated based on the concept of victim survival and rescuer safety. Shortening the time spent working in CS can reduce the risk of secondary damage and increase the chance of victims' survival. This is a novel approach in that the infusion model was established based on in vivo experiments.

Data from in vivo experiments showed a weak correlation between infusion time, peripheral venous flow velocity, and BMI—likely reflecting the resistance in the peripheral veins, and suggesting that body size affects infusion speed.

Although Franklin et al. experimented with spontaneous dripping and pressurized bags in vivo (13), our study assessed various infusion methods and found meaningful differences in administration speed.

Significant reductions in the administration time for tube extension + pumping, manually pressurized bags + re-pressure after 10 min, and automated pressure devices (continuous and temporary by battery). These methods should be selected based on effectiveness and the safety field conditions.

However, practical challenges remain with automated devices, including the size, necessity of power supply, environmental durability, training requirements and device portability (Table 2). Although the safety concern requiring re-entry into confined spaces exists, manually pressurized bag + re-pressure after 10 min strikes a balance between safety of pumping, and is actively used by JDR Rescue Team Medical Unit.

Table 2. Advantages and disadvantages of each infusion method.

CS, confined space.

There are some limitations regarding this study. The infusion time may vary depending on anatomical factors not replicable in simulator models, such as venous branches, valves, collapsibility, tissue edema, or patient movement and the victim's physiological state. Such systematic biases (e.g., lack of tissue resistance or dynamic venous collapse) might affect real-world applicability. Besides, it is experimental and performed on a group of young, healthy male participants, who may differ from actual crush syndrome victims who may be hypovolemic, acidotic, or shock. Additionally, other administration methods (e.g., intraosseous infusion) and needle sizes other than 20 G were not evaluated due to cost and device limitations. Finally, larger sample sizes or testing under simulated stress conditions (e.g., vibrations, dust) could further validate the findings.

While our study focused on intravenous infusion methods, international guidelines such as those from INSARAG (2023) and the WHO emphasize intraosseous (IO) access as a critical alternative when IV access is unfeasible (1). IO infusion offers several advantages in confined spaces, including rapid deployment, minimal training requirements, and reliability in hypotensive patients (14). Future studies should compare IV and IO methods in simulated confined space environments to determine optimal protocols for USAR teams. Our exclusion of IO access reflects a limitation of the current study, as it may represent the most viable option in scenarios where IV cannulation is impossible (e.g., complete limb entrapment or severe vasoconstriction).

Despite these limitations, this study is worth presenting as evidence from the JDR rescue team medical unit and has a potential to be integrated into USAR training programsor standardized protocols (e.g., INSARAG's technical guidance notes), especially given the lack of international studies on infusion methods and rates in CS. Further research is required; this study will continue to collect more relevant evidence. Studies involving trauma patients or IO access would require additional ethical considerations (e.g., informed consent in austere environments).

This exploratory study evaluated the infusion speeds using an infusion model based on the results from healthy human participants. based on in vivo experiments novel approach in the infusion model was established. Among the methods tested, the manual pressurized bag with re-pressurization after 10 minutes emerged as a promising balance between speed and practicality. However, the optimal method may vary depending on field conditions, and further validation in real-world USAR scenarios is warranted. Although each method has its advantages and disadvantages in terms of power supply and safety, selection should be based on careful evaluation. Further research is required to determine the optimal method for each rescue scenario.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Kris Spaepen, Vrije University Brussels, Belgium

Reviewed by: Tamorish Kole, University of South Wales, United Kingdom

Jan-Cedric Hansen, Centre d‘Hébergement et d'Accompagnement Gérontologique (CHAG), France