Abstract
Introduction
Portable ultrasound enhances medical assessments in austere environments but is often limited by volume, weight, and availability of commercial gels. This study evaluates the stability and image quality of powder-based ultrasound gel substitutes.
Methods
Eight alternative gel formulations were compared with commercial gel across simulated environments: tundra, desert, altitude, and hospital control. Gels were assessed daily for color, viscosity, odor, homogeneity, and tensile strength. A stability score was calculated and analyzed using ANOVA with Tukey's HSD. Ultrasound images of the internal jugular vein, lung, hepatorenal recess, and quadriceps tendon were obtained using SonoSite and Butterfly devices. Blinded ultrasound-trained faculty graded the image quality (1–10), with t-tests comparing alternatives to control (p < 0.05 significant).
Results
All alternatives degraded over 1 week. ANOVA confirmed significant effects of gel type (p < 0.0001) and time (p < 0.0001), with Tukey's HSD identifying dry commercial and xanthine as most stable.
Conclusion
Sonosite, cassava (p = 0.05), and dry commercial (p = 0.02) outperformed the control at the internal jugular, while tapioca (p < 0.00), cornstarch (p = 0.01), and water (p < 0.00) performed worse at the hepatorenal. At the quadriceps, cassava (p = 0.01), cold glucomannan (p < 0.00), guar gum (p = 0.01), cornstarch (p = 0.04), and water (p < 0.00) performed worse. Using Butterfly, cold glucomannan (p = 0.02), guar gum (p = 0.01), tapioca (p = 0.02), cornstarch (p = 0.01), and water (p < 0.01) performed worse at the hepatorenal, while at the quadriceps only water (p = 0.02) performed worse.
Keywords
Introduction
Since its development in the mid-20th century, ultrasonic imaging technology has continually evolved, becoming increasingly portable, compact, and durable. 1 These advancements have enabled ultrasound to be adapted for austere environments, where it surpasses magnetic resonance imaging (MRI), computed tomography (CT), and x-ray in terms of accessibility, portability, cost, and maintenance requirements. 2
Recent innovations have further enhanced ultrasound device sturdiness with the adoption of solid-state transducers in place of fixed-array crystals, alongside improvements in battery life and miniaturization. 2 This progress has led to the commercial availability of affordable, pocket-sized ultrasound machines. A systematic review comparing handheld and high-end ultrasound systems found strong overall agreement across multiple applications, including emergency-focused assessment with sonography for trauma (FAST); vascular, obstetric, gynecologic, and intra-abdominal solid organ imaging; and procedural guidance. 3 As a result, handheld ultrasound devices have become widely adopted, especially in remote locations, combat zones, and low-resource settings.4,5
Despite these technological advancements, the portability of ultrasound gel remains a limiting factor for its use in austere environments. Ultrasound probes rely on piezoelectric materials to transmit sound waves, interpreting their reflections to generate images based on tissue echogenicity and density.1,6 A coupling agent is required to ensure adequate wave transmission, and commercial ultrasound gels are formulated to enhance image clarity, smoothness, and usability. 6 However, preconstituted ultrasound gel is constrained by its liquid form, making it bulky and difficult to transport, in addition to challenges related to cost, shipping, and waste disposal in austere environments.
An ideal ultrasound gel alternative would be economical, locally available, shelf-stable, and easy to produce. Prior studies have explored powder-based gels derived from cassava, xanthine gum, cornstarch, glucomannan, and guar gum.7–11 However, a comprehensive comparison of their image quality, cost-effectiveness, and shelf life remains unexplored. This study aims to evaluate several sustainable ultrasound gel substitutes based on viscosity, color, image quality, and usability while assessing their degradation in environmental conditions mimicking austere settings.
Methods
This study was submitted to the institutional review board and deemed exempt. A literature review was conducted to identify previously studied ultrasound gel formulations. Among the articles reviewed, gels made from glucomannan, guar gum, cassava flour, xanthan gum, tapioca powder, and cornstarch were selected, and their constituent ingredients were obtained. Commercial ultrasound gel (Aquasonic®) and dry commercial gel (IMV Imaging) were also included for comparison. During preliminary testing, all gels were prepared and subjectively evaluated for usability and viscosity. Recipes were then adjusted to match the subjective viscosity of commercial ultrasound gel (Table 1). The cost per 500 mL of gel was calculated in U.S. dollars, based on the price of locally available ingredients (costs may vary depending on regional supply and ingredient pricing). While most alternative gels were comparable in price, Table 1 highlights the higher cost of commercially available gel.
Comparison of Utilized Ultrasound Gel Recipes in Cost per 500 mL.
Simulated Environments
The tundra, desert, and altitude environments were simulated using insulated terrariums and compared to a hospital control environment. Each terrarium was equipped with a thermometer and humidity gauge to monitor conditions. The tundra environment was maintained at sea level, with an average temperature of 5 °C (41°F) and 40% humidity. Similarly, the desert environment, also at sea level, had an average temperature of 40 °C (104°F) and 26% humidity. The altitude environment was constructed in Golden, Colorado, at an elevation of 2408 m, with an average temperature of 18 °C (65°F) and 26% humidity. Lastly, the hospital control environment, maintained at sea level, had an average temperature of 24 °C (76°F) and 60% humidity (Table 2). This setup ensured controlled comparisons of gel stability across distinct environmental conditions. A tropical environment was not included, as we were unable to create a terrarium with appropriate humidity despite the use of air humidifiers within the constructed tropical terrarium.
Average Temperature and Humidity of Environments During the Experimental Window.
Degradation Testing
After initial preparation, all gels were tested for color, relative viscosity, odor, homogeneity, and tensile strength. Color was assessed on a binary scale, with gels classified as either unchanged from baseline or exhibiting a detectable change. Malodor was subjectively evaluated and recorded as present or absent. Homogeneity was visually inspected without agitation and recorded as either complete or separated. Viscosity was measured indirectly using a “droplet spread test,” in which a 1 mL gel droplet was released from a height of 1 cm onto standardized paper. The average diameter of the droplet was recorded 30 s later, calculated from 2 perpendicular measurements—the longest and shortest diameters—since the droplets were not perfectly circular. Tensile strength was tested using a “string test,” in which a 1 mL gel sample was placed between an investigator's gloved fingers, which were then slowly pulled apart until the gel strands broke. The maximum distance between the fingers at the moment of breakage was recorded in millimeters. Both of these investigator-devised methods were selected for their feasibility and reproducibility in resource-limited settings where access to specialized laboratory equipment is not possible.
Initial gel characteristics were recorded 2 h after creation, designated as the day 0 measurement. Afterward, the gels were divided into 90 cc sealed plastic containers and placed in 4 different environments designed to simulate climatic conditions commonly encountered in austere settings. These environments were used to assess how long-term degradation varied with different environmental exposures. Gel characteristics were subsequently measured every 24 h for 7 days following the Day 0 measurement to track changes over time. Notably, only measurements from Day 1 through Day 6 were obtained in the altitude environment. This limitation arose from a misunderstanding in data collection procedures, resulting in the omission of these time points. Efforts were made to ensure consistency in measurement protocols for the remaining data points.
A stability score was developed using a 0 to 5 point-based system (0 indicating no stability, 5 indicating complete stability), with each of the 5 variables scoring a maximum of 1 point. Binary variables (color, malodor, and homogeneity) were each assigned 1 point for no change and 0 points for any detected change. Continuous numerical variables (viscosity and tensile strength) received a fractional score based on their percentage change from the control. The total points for all variables were averaged over the experimental period to calculate a daily stability score.
Image Quality Testing
Image quality testing was performed in the control hospital climate. Gels were tested for image quality on day 0 (freshly made). Each gel was applied to the same human subject participant, and images of a marked location on the quadriceps tendon, internal jugular vein, lung apex, and hepatorenal pouch were obtained and cataloged. Gel was cleaned off the subject's skin with soap and water between each gel application. Two ultrasound fellowship-trained faculty members acquired and cataloged the images using two ultrasound machines (conventional cart-based Sonosite M Turbo and hand-held Butterfly iQ). The images were assigned a number for reference use. Seven ultrasound faculty, blinded to the ultrasound gel type, were then asked to grade the previously obtained still images based on perceived quality on a scale of 1–10 for each image.
Statistics
Degradation testing: This analysis employed a two-factor ANOVA to assess the effects of gel type and time on stability, followed by Tukey's Honestly Significant Difference (HSD) post-hoc test to identify significant pairwise differences between gels.
Image quality testing: Mean image quality scores were analyzed using T-tests to determine if there were significant differences in the perceived quality of gel images at the various anatomic sites. Significance for all statistical tests was accepted at p < 0.05.
Results and Discussion
Daily average stability scores across all environments were calculated and are presented in Table 3, with trends in stability visualized in Figure 1. While the commercial gel exhibited no degradation, all alternative gels showed a gradual decline in stability scores over time. Although many gels demonstrated comparable stability, tapioca experienced a sharp and rapid decline (Figure 1).
Stability Score Over Time for Each Gel Type. Score Range 0 (No Stability) to 5 (Full Stability).
Daily Average Stability Scores for Each Gel Type.
Score range 0 (no stability) to 5 (full stability).
A two-factor ANOVA without replication (Table 4) was performed to assess the independent effects of gel type and time (days) on stability scores. This test was chosen because the study involved multiple gel formulations measured over multiple time points, allowing us to determine (1) whether different gels exhibit significantly different stability trends and (2) whether stability changes significantly over time. The ANOVA analysis indicated that gel type had a highly significant effect on stability (F(8, 40) = 65.59, p < 0.0001), confirming that different formulations degrade at different rates. Additionally, time was a significant factor (F(5, 40) = 14.29, p < 0.0001), demonstrating a progressive loss of stability across all alternative gels. Additionally, the error term (SS) was noted to be small.
Two-Factor Without Replication ANOVA Testing of the Variables of Gel Type and Time on the Stability of Each Gel.
Tukey's Honestly Significant Difference (HSD) test (Table 5) was conducted as a post-hoc analysis following the two-factor ANOVA to determine which specific gel formulations differed significantly from each other in terms of stability. While ANOVA confirmed that gel type had a significant effect on stability (p < 0.0001), it did not specify which gels were significantly different from one another. Tukey's HSD allowed for pairwise comparisons while controlling for multiple comparisons, ensuring that significant differences between gels were identified without inflating the risk of Type I errors.
Tukey's HSD Analysis of ANOVA Gel Stability Data.
The null hypothesis of equal stability between gel formulations was rejected in all 36 pairwise comparisons (p < 0.05 for all comparisons). Commercial gel was significantly more stable than all alternative gels (p < 0.0001 for all comparisons). Among the alternatives, dry commercial and xanthine exhibited the highest stability scores and were significantly more stable than cassava, cold glucomannan, and hot glucomannan. In ranked order, the most stable gel was the commercial gel (control), followed by the dry commercial, xanthine, cassava, guar gum, hot glucomannan, cold glucomannan, cornstarch, and lastly tapioca.
Image Quality
SonoSite: Mean image quality scores and corresponding p-values for images collected with the conventional cart-based SonoSite are reported in Table 6. At the internal jugular vein, cassava and dry commercial had significantly increased mean image quality scores (p = 0.05 and p = 0.02, respectively). The rest of the reconstituted gels at the internal jugular vein produced images that were noninferior to the control gel. At the hepatorenal recess, tapioca, corn starch, and water produced images that were significantly inferior to control (p < 0.00, p = 0.01, and p < 0.00, respectively), while the rest of the reconstituted gels produced mean image quality scores that were noninferior to control. At the lung, there were no significant differences between the mean image quality scores of the reconstituted ultrasound gels and the control. At the quadriceps tendon, cassava, cold glucomannan, guar gum, corn starch, and water all produced mean image quality scores that were significantly less than control (p = 0.01, p < 0.00, p = 0.01, p = 0.04, p < 0.00, respectively).
Comparison of Mean Image Quality Scores Among Reconstituted Ultrasound Gels at Various Body Sites Using SonoSite.
IJ, internal jugular; HR, hepatorenal recess; QT, quadriceps tendon; Avg., average; P, P value.
*Statistical significance.
Butterfly: Data for images collected with the hand-held Butterfly are reported in Table 7. At the internal jugular vein, there were no significant differences in mean image quality scores between reconstituted gels and the control. At the hepatorenal recess, cold glucomannan, guar gum, tapioca, corn starch, and water all produced images that were significantly inferior to the control (p = 0.02, p = 0.01, p = 0.02, p = 0.01, p < 0.00, respectively). There were no significant differences in images of the quadriceps tendon between reconstituted gels and control, but water produced significantly worse images (p = 0.02).
Comparison of Mean Image Quality Scores Among Reconstituted Ultrasound Gels at Various Body Sites Using Butterfly.
IJ, internal jugular; HR, hepatorenal recess; QT, quadriceps tendon; Avg., average; P, P value
*Statistical significance.
Discussion
Degradation
The ANOVA analysis indicated that both gel type and time had significant effects on stability. These findings suggest that the rate of degradation is formulation-dependent and varies over time. Additionally, the error term (SS) was relatively small, meaning that the variability in the data can be well attributed to the gel types and time rather than random noise.
The null hypothesis was rejected in all comparisons, meaning that every gel exhibited statistically significant stability differences compared to the others. As anticipated, commercial gel was significantly more stable than all alternative gels. Among the alternatives, dry commercial and xanthine exhibited the highest stability scores and were significantly more stable than cassava, cold glucomannan, and hot glucomannan. This suggests that of the alternative ultrasound gels, dry commercial was the most stable, followed by xanthine and cassava. Many wilderness providers are likely to whip up a fresh batch of gel at the time of evaluation without storing gel for longer periods of time, as this would take up more space and weight in one's pack. However, we also recognize the utility of using these powder gel alternatives in remote clinics where availability, cost, and waste avoidance would benefit from reasonable stability when storing for several days.
Image Quality
Using the conventional cart-based SonoSite, several reconstituted ultrasound gel substitutes were found to produce images similar to commercial gel. Interestingly, mean quality scores for internal jugular vein images using cassava and dry commercial gel were significantly higher than the control. In addition, xanthine, hot glucomannan, and dry commercial produced images that were at least equivalent to control across all body sites. Our findings are consistent with prior studies that have found cassava, xanthine, and glucomannan to produce images of comparable quality to commercial gel.7,9–11 At the hepatorenal recess, the image quality using tapioca, corn starch, and water was significantly worse than the control. Gel quality may be more important when evaluating deep structures since they require a lower-frequency probe, which is prone to produce a greater degree of attenuation and speckle interference. 12 The majority of gels at the quadriceps tendon produced mean image quality scores that were significantly worse than the control (Table 6). It is unclear why so many gels produced inferior image quality at this body site. However, the quadriceps tendon is anisotropic in nature, and perhaps the results indicate that higher quality gel helps to reduce the amount of the speckle pattern produced. The results suggest that body sites that are anechoic, such as a fluid-filled vessel, or hyperechoic, as with lung pleural, do well using ultrasound gel substitutes. However, when imaging structures are anisotropic, gel quality appears to matter more.
Using the Butterfly, most ultrasound gel substitutes produced images similar to commercial gel when imaging the internal jugular vein and quadriceps tendon (Table 7). Across body sites, the mean image quality scores appeared lower than SonoSite, which likely reflects a difference in the capabilities of the machines. The literature has shown that hand-held devices can be reliably used to answer certain clinical questions, but that image quality is not yet equivalent to high-end machines. 3 Most gels used with the Butterfly at the hepatorenal recess produced images that were significantly inferior to commercial gel (Table 7). This may again be due to the use of a lower-frequency probe to image deep structures, which already limits image resolution. In a systematic review, five studies evaluating abdominal image quality in hand-held devices found some limitations in assessing the liver and gallbladder. 3 Thus, the poor hepatorenal images are a likely consequence of current technological constraints. Interestingly, all alternative gels, and even water, were not inferior to the control when imaging superficial structures with a hand-held linear transducer. The results suggest that when using hand-held ultrasound machines such as the Butterfly, alternative gels are likely appropriate for use on superficial structures but may limit diagnostic value when evaluating deeper structures.
Limitations
Degradation
The different environmental stressors were simulated as closely as possible using heat lamps, refrigerators, humidifiers, and dehumidifiers, but there were limitations in how closely the environment could be imitated. There were no natural fluctuations in temperature or humidity, as there would be in a natural environment, because the devices used to simulate environments were set to standard settings for the entirety of the gel testing. Furthermore, there was significant difficulty creating an appropriate sealed capsule for appropriate increased/decreased humidity environments. We were unable to create a terrarium with an appropriate humidity corresponding to a tropical environment despite the use of air humidifiers within the constructed tropical terrarium. However, given that the gels were stored in sealed bottles, this humidity variable was likely irrelevant. There was further limitation in the manner in which the gels were reconstituted, as a hand immersion blender was used to maximize homogeneity. While this was the best manner for creating appropriately mixed gels, this hand mixer device would likely be unavailable in most austere environments, and the maximized homogeneity of the gels may have created a falsely optimized gel efficacy. Day 0 and Day 7 measurements were omitted from the altitude environment analysis, which may have reduced the observed differences in gel stability on those days. However, since Tukey's analysis confirmed statistically significant differences among all gels, the omission likely had minimal impact on the overall findings, as the additional data points would have only contributed to the existing variation between gels.
An additional limitation is the creation of the degradation stability score, which was a proprietary system developed by the investigators and has not been standardized or previously validated in the scientific literature. While the assessment of color, malodor, and homogeneity was relatively straightforward, our measures of viscosity and tensile strength relied on investigator-devised surrogate methods (the “droplet spread test” and the “string test,” respectively). To our knowledge, the “string test” has not been previously described in the peer-reviewed literature as a formal method of tensile strength testing. These approaches were chosen for their feasibility and reproducibility in resource-limited and austere settings, though they do not represent standardized laboratory assays. Although these techniques may not capture the full spectrum of rheological or tensile properties, they provided consistent comparative data. Future work should aim to validate these approaches against established rheological and tensile testing methods to strengthen their generalizability.
Image Quality
Only one human model was used to collect images, and it is well known that differences in body habitus can impact the quality of images captured. Also, the tapioca gel would separate quickly after mixing, which made it difficult to reliably apply a homogenous solution to body sites. The images collected using the hand-held Butterfly machine were collected under a portable tent on a sunny day outdoors, possibly making it difficult for the ultrasonographer to adequately assess image quality when capturing the still image. This, however, represents many outdoor environments where imaging may take place in remote areas. Both ultrasound machines used were the most recent models on the market, potentially optimizing the image quality, and the performance of the gel alternatives on older model ultrasound machines may be less reliable.
Availability and Price
Not all of the gel powders listed may be available or affordable in different global markets, depending on the location. However, this study aims to provide a ranking list for comparable options, regardless of the available products in a location.
Conclusion
This study identified several sustainable ultrasound gel substitutes as viable options for use in austere environments. Among the alternative dry gels, dry commercial demonstrated the highest stability over time, while xanthine and cassava also performed well. Additionally, xanthine, hot glucomannan, and dry commercial reliably produced image quality comparable to commercial gel when imaging various anatomical locations, including the internal jugular vein, hepatorenal recess, lung, and quadriceps tendon. When using alternative gels, image quality was notably better with high-end ultrasound machines compared to handheld devices. Ultimately, the availability of ingredients in different austere environments may be the most significant factor in determining which gel to use. This study provides a preliminary assessment of key considerations, including cost, stability, and image quality, to guide alternative powder ultrasound gel selection.
Future research could focus on evaluating these alternative gels in clinical decision-making, rather than image quality alone, to identify a broader range of viable substitutes. This study contributes to the growing body of literature on reconstituted powder-based ultrasound gels in resource-limited settings, offering insights into their stability, usability, and imaging performance.
Presentations
This work was presented as an oral abstract at the Wilderness Medical Society Summer Conference, Snowmass, Colorado, in July 2022.
Footnotes
Acknowledgment
The authors thank Melodie Santodomingo for assistance with statistical analysis.