Frequent versus infrequent partial doffing in transtibial prosthesis users as a means to stabilize residual limb fluid volume

Abstract

Background:

Daily residual limb fluid volume fluctuation is thought to be a driving factor behind prosthetic socket fit deterioration among people with lower limb amputation. Expanding the number and type of strategies available to manage daily residual limb fluid volume changes may improve socket fit.

Methods:

To investigate the effects of partial-doff frequency on residual limb fluid volume, participants completed two controlled laboratory sessions that differed in the number of times participants sat and partially doffed their prosthesis. Transtibial prosthesis users were recruited, fitted with a custom limb fluid volume monitoring device and a socket release-relock mechanism, and asked to complete a series of low- and high-activity ambulation bouts that mimicked daily prosthesis use.

Findings:

Mean limb fluid volume measures from the anterior compartment of the residual limb were significantly different between the one- and three-partial doff experimental protocols (−5.8 % and −2.1 % fluid volume, respectively, P < 0.01). The posterior compartment limb fluid volume, locking pin depth, locking pin magnitude of movement, anterior distal limb position depth, and anterior distal limb position total movement did not differ significantly between the two conditions.

Interpretation:

Partial doffing may offer prosthesis users an alternative means to accommodate limb fluid volume changes. More frequent partial doffing appears to better mitigate limb fluid volume loss when compared to less frequent partial doffing.

1. Introduction

Fit of the prosthetic socket on a residual limb is one of the most important factors in success for people with lower limb amputation (Legro et al., 1999; Turner and McGregor, 2020). Socket fit can deteriorate over time due to factors such as usage (Sanders et al., 2014; Sanders and Fatone, 2011), sweating in elevated temperatures (Davies et al., 2020), and diet. One reason for usage-based fit degradation is residual limb shape changes due to diurnal residual limb fluid volume fluctuations (Sanders et al., 2012a). Shape changes may alter the distribution of stress and motion between the limb and socket, thereby affecting fit.

One method for accommodating residual limb volume change is to add or remove prosthetic socks, and clinicians regularly advise patients use this strategy when they experience a deterioration in fit (D’Silva et al., 2014). Yet sock thickness can have a detrimental relationship with residual limb fluid volume. By reducing the socket volume available for the residual limb, limb fluid may be driven out, creating a chain of events that further increase fluid volume loss with each subsequent sock donned (Cagle et al., 2016; Sanders et al., 2012b) (Fig. 1 A). This relationship may be exacerbated by the vascular health of the patient. Poor vascular health is common among people with limb loss.

Fig. 1.

A,B. Figures illustrating residual limb fluid volume change over time for different accommodation strategies. (A) Adding Sock Ply: The addition of prosthetic socks reduces the volume within the socket available for the residual limb, driving additional residual limb fluid volume out. A new window of acceptable socket volume is created that is smaller than the initial volume. (B) Temporary Partial Doffing allows the user’s residual limb to regain volume and maintain the “original acceptable window of socket volume.”

A tether release-relock mechanism (Gurrey et al., 2021) that allows participants to partially doff their prosthesis may not only be more convenient than sock changes but may also limit residual limb volume fluctuation by allowing fluid to return to the limb (Fig. 1B) (Sanders et al., 2016). In a sensitivity analysis on participants with transtibial amputation repeating twelve cycles of walking and partial-doff resting over a 2–3 h testing protocol, Lanahan et al. found that both 4-min and 10-min partial-doff durations significantly reduced limb fluid volume loss over no release (P < 0.04) (Lanahan et al., 2023).

In the present study, two partial doff frequencies were compared to help establish doff frequency guidelines in routine clinical care. Sensors measuring pin depth and limb motion were integrated into participants’ sockets to gain insight into how doff frequency affected movement of the limb in the socket. The hypotheses evaluated were that, compared to one partial doff, three partial doffs reduced residual limb fluid volume loss, downward limb displacement, anterior distal pistoning, and socket discomfort over the 2-h testing protocol.

2. Methods

2.1. Participants

All study procedures were approved by a University of Washington Institutional Review Board (Study 00006874); written informed consent was obtained from all participants prior to enrollment. Participants who underwent a transtibial amputation at least 18 months prior, had a residual limb length of at least 9.0 cm, reported wearing a definitive prosthesis for at least 7 h/wk, had the ability to walk for ≥2-min bouts, and were capable of sitting, standing, and navigating a step of at least 5.0 cm were recruited from our local registry. The 9.0 cm residual limb length was necessary for bioimpedance monitoring of residual limb fluid volume. Those with presence of skin breakdown or who used a walking aide were excluded.

2.2. Investigational prosthesis fabrication and sensor calibration

Prior to the first test session, participants visited the lab so that a digital scan of their traditional prosthesis could be taken. A high-accuracy coordinate measurement machine (FaroArm Platinum, FARO Technologies) was used to trace the inner socket shape, and software was used to construct a digital point cloud surface (GeoMagic, Oqton; OMEGA Tracer CAD, WillowWood). The surface was then used to create an investigational socket which matched the shape of the participant’s traditional socket. All investigational sockets were fitted with an energy storage and return prosthetic foot, either the participant’s regular foot or a Proteor RUSH foot (LoPro).

Three inductive sensor antennae were embedded in the socket wall during the lamination process using methods previously described (Ballesteros et al., 2024). The sensors were designed to measure the distance between the socket and liner (Fig. 2 A). The embedded sensors were instrumented with thermistors to allow for thermal compensation during post-processing. An investigational liner, manufactured by a prosthetics company under contract (Alpha Silicone, WillowWood), was also fitted to each participant. The investigational liners had a trace amount of iron powder in the outer part of the elastomer next to the fabric backing that served as a magnetically-permeable target for the inductive distance sensors in the socket (Weathersby et al., 2019). A data logger, as described by Mertens et al. (Mertens et al., 2022), was fastened to the outside of the socket to store data collected from the socket sensors and other sensors described below.

Fig. 2.

A-C. investigational socket and TARPIN system. (A) An embedded distance sensor is indicated with the red circle. (B) The resistive position sensor within the RevoFit dial located on the posterior distal socket measures panel tightening and loosening. (C) The motor-driven ratcheting TARPIN system is removed from the socket and shown at the left. The TARPIN is controlled by buttons on the posterior distal aspect (don) and the lateral aspect (doff) (left-legged) of the distal socket enclosure. Vectran 400 cabling extended from the spool on the TARPIN to the pin lock receptacle in the liner. The data acquisition/controller printed circuit board is not shown in this image but mounts within the distal socket enclosure.

The investigational sockets were fabricated with three adjustable panels within the socket wall. A cable-ratcheting dial (RevoFit, Click Medical) was embedded in the posterior aspect of the socket, and cabling (Vectran 400, TwinLine, LLC) was routed through the panels so that the single cable passed through each panel twice (Fig. 2B). The ratcheting dial allowed the panels to be released to execute partial doffing. A resistive position sensor (RDC803101A, Alps Alpine) and custom lamination housing were placed in the ratcheting dial to measure the cable length so that the change in socket volume could be quantified during each walking bout in a session (Gurrey et al., 2021). Changes in cable length were converted to percentage change in socket volume relative to the original socket volume (panels flush) taken from the scan of each participant’s socket using a calculation described by Vamos et al. (Vamos et al., 2020). Each of the panels had a piece of foam (6.35-mm thick Plastazote^™) affixed to the inside surface so that socket size could be reduced and not just enlarged using the ratcheting dial. The foam also helped to facilitate an equalized panel pressure on the residual limb.

The sockets were designed to include an enclosure distal to the shuttle lock (Icelock, Össur) that allowed for a custom Tether-Aided Release/Relock Pin system (TARPIN) to be installed (Fig. 2C). The TARPIN, as previously described (Gurrey et al., 2021), is an electromechanical system that aids the donning and doffing process of a pin-lock suspension transtibial socket. A tether (Vectran 400) was run from a motor-driven ratcheting spool mechanism within the TARPIN to a locking pin fastened to the liner. The pin was a custom smooth pin (34.0-mm long, 7.8-mm diameter) with a central hole (2.0-mm diameter). One push of the TARPIN’s ‘don’ button on the posterior aspect of the enclosure caused the tether to draw the socket onto the limb. One push of the ‘doff’ button on the medial aspect for a participant with a left-legged amputation, lateral aspect for a participant with a right-legged amputation, allowed the user to back-drive the motor and partially doff the socket to a stop at 5.0 cm. A second push of the ‘doff button,’ executed only at the end of the session protocol, allowed the user to fully doff the prosthesis. A custom sensor was seated between the shuttle lock body and the TARPIN system to measure the depth of the locking pin in the distal end of the socket (Bennett et al., 2021). In total the TARPIN unit weighed about 450 g.

2.3. Protocol

Upon arriving for the first test session, participants sat with their traditional socket donned for 10 min while the research prosthetist collected demographic information. The participant was fitted with the investigational prosthesis which required walking and standing for short durations while the prosthetist adjusted the alignment and sock thickness, if necessary.

Electrodes were placed on the participant’s residual limb to measure the limb fluid volume via bioimpedance analysis in the anterior and posterior tissue compartments, as previously reported (Hinrichs et al., 2019; Youngblood et al., 2019). The time it took to place bioimpedance electrodes allowed the user to reach a homeostatic condition prior to the start of the structured protocol. The protocol had five groups of activities: pre-conditioning; three cycles of low- and high-activity bouts; and one intermediate activity bout (pseudo-low activity). During these activities, participants were asked to complete various ambulatory tasks such as treadmill walking, standing, sitting, and walking in a Figure-of-8 pattern (Fig. 3).

Fig. 3.

Study protocol. Participants were asked to complete two protocols with 1- versus 3-partial doffs in the span of two hours. Fluid volume and socket fit data were analyzed after the pre-conditioning bout was completed (t = 0 min). Bouts of low and high activity were alternated, and partial doffing was performed after bouts of high activity. Relative Socket Comfort Scores (RSCR) were recorded at the start and end of the analyzed session and each time a partial doff was performed. A Figure-of-8 walk was performed at the conclusion of the session as a functional mobility test.

The two test sessions, 1-doff and 3-doff protocols, were completed on two days approximately 1–2 weeks apart, conducted in a randomized order. In the 1-doff experimental condition, participants performed a single, aided, 5-min partial doff approximately halfway into a 2-h structured protocol, while in the 3-doff experimental condition, participants performed three, aided, 5-min partial doffs over the course of the 2-h structured protocol (Fig. 3). Socket Comfort Score (SCS) and Relative Socket Comfort Rating (RSCR) (Youngblood et al., 2022) were recorded during the protocol and the ordinal RSCR was converted to a numeric scale for reporting in this publication. The ratings compared to the previous walking segment were: a little better (+1), better (+2), the same (0), a little worse (−1), and worse (−2). Number of revolutions in each Figure-of-8 activity were counted. Participants were not allowed to add socks during the test session.

Research staff performed the aided partial doff and re-don process to ensure a consistent procedure was used across all participant test sessions. To partially doff the prosthesis, the panels were released using the ratcheting dial, and the doff button was pressed once to allow the limb to displace 5.0 cm out of the socket. While partially doffed, the participant’s limb was placed so that it rested gently in the socket. All partial doffs were 5 min in duration. To don the socket, the don button was pressed and the socket guided back onto the limb. Once the socket was donned, the participant stood, and the don button was pressed again to draw in any remaining slack. If panels had been released to accommodate partial doffing, they were returned to the prior position once the participant donned the socket, using a live output of the instrumented dial encoder for feedback.

2.4. Data reduction

Data stored to the data logger was downloaded and processed using scripts written in MATLAB (MathWorks) and Python. Data from the socket sensors, locking pin, ratcheted-dial sensor, and TARPIN motor encoder were converted into units of distance, and bioimpedance data were converted into units of percent fluid volume change as a percentage of the reference fluid volume. The reference fluid volume was the mean of the stance phase minima from all steps within the walking bout at the end of pre-conditioning, i.e., during the 3rd walk of the protocol (Fig. 3). The mean of the stance phase minimum percent fluid volume from all steps within each walking bout was calculated and plotted over time.

The mean pin position during each walking bout was calculated from the stance phase minima from all steps in the bout. Cumulative pin movement was the difference in pin position from the 10th step of the walk minus that from the reference (mean pin position from the 3rd walking bout). The magnitude of pin movement (MPM) was the sum of the absolute difference in pin position from one walking bout to the next (i.e., how much the pin moved, up or down, from one walking bout to the next) from the 3rd to 10th bout. These seven absolute value differences were summed to compute cumulative pin movement.

Pistoning at the anterior-distal site during each walking bout was calculated as the mean of the maximum minus minimum anterior distal position from all steps in a bout. Cumulative pistoning was the difference in pistoning from the 10th bout of walking minus that from the reference (3rd bout of walking). The magnitude of anterior-distal pistoning was the sum of the absolute difference in pistoning from one walking bout to the next (i.e., how much the pistoning changed from one walking bout to the next) from the 3rd to 10th bout. These seven absolute value differences were summed to compute cumulative pistoning.

Shapiro-Wilk tests were performed to evaluate each of the metrics described for a normal distribution across the cohort of patients in the study. A repeated measures ANOVA with Bonferroni corrected pairwise-comparisons was conducted for the limb fluid volume data because it had four measurements of like unit, while Student t-tests were done for individual comparisons of the Pin Depth-Cumulative, Pin Depth-Magnitude, and Anterior Distal Max-to-Min-Cumulative. A Wilcoxon-rank sum comparison was done for the final individual comparison, the Anterior Distal Max-to-Min-Magnitude, since the data had a non-parametric distribution.

3. Results

3.1. Demographics

Thirteen participants were recruited and consented. Nine males and two females completed all study procedures. One participant did not complete the study because he was not available within the study time frame, and one participant withdrew before starting the structured protocols. All participants who completed the study procedures were at a Medicare Functional Classification Level 2 or higher (Medicare Region C Durable Medical Equipment Prosthetics Orthotic Supplier (DMEPOS) Manual, 2005) as determined by the research prosthetist. Their median age was 58 years (range 37–76) and time since amputation was 16 years (range 4–54). Additional demographic data is listed in Table 1.

Table 1.

Participant demographics.

Participant	Age (Years)	Height (cm)	Time Since Amputation (Years)	Limb Length (cm)	Comfortable Walking Speed (MPH)	Sex	Weight Average (kg)	BMI	Traditional Prosthesis Suspension Type	Prosthesis weight (kg)	Health Notes
1	61	188.0	36	18.5	1.8	M	82.3	23.5	Locking Pin	1.8	None
2	37	177.8	15	18.0	1.5	M	111.7	35.7	Locking Pin + REVO	2.4	None HBP; Upper limb
3	58	182.9	30	11.5	1.6	M	104.3	32.8	Locking Pin	2.0	amputation
4	57	177.8	4	16.2	1.3	M	117.2	37.4	Locking Pin	2.7	DM; Smoker
5	39	160.0	15	14.0	2.2	F	68.9	27.5	Locking Pin	0.9	None
6	48	172.7	7	15.0	1.8	F	83.5	28.1	Vacuum	2.3	None
7	75	185.0	46	12.3	2.9	M	89.0	26.2	Locking Pin	2.2	HBP; Seizures
8	47	175.0	7	12.5	1.9	M	87.4	28.8	Locking Pin	2.2	HBP
9	73	182.9	54	15.0	2.3	M	96.8	29.1	Locking Pin	2.7	HBP DM, HBP, ABI Abnormal;
10	61	193.0	16	21.1	2.0	M	128.3	34.9	Locking Pin	2.5	Smoker
11	76	185.4	52	21.8	2.5	M	95.5	28.0	Locking Pin	2.4	HBP

3.2. Socket volume within a session

The change in socket volume due to panel tightening and loosening between bouts was calculated from the resistive position sensor data. Data from the first fifteen sessions demonstrated that the change in socket volume within a session ranged from a 0.60 % volume reduction to a 0.07 % volume enlargement. Since prior research demonstrated that a 1.00 % socket volume change was the threshold necessary to create a clinically detectable change in socket fit (Sanders et al., 2012c), the socket volume changes recorded in this study were deemed not likely to affect the analysis of interest. Thus, these data were not collected during the remaining sessions.

3.3. Limb fluid volume

Example traces illustrating temporal changes in the data are shown in Appendix A. Shapiro-Wilk test results suggested that the cumulative fluid volume changes in the anterior compartment during the 1-doff protocol, the anterior compartment during the 3-doff protocol, the posterior compartment during the 1-doff protocol, and the posterior compartment during the 3-doff protocol were normally distributed (Appendix B). Results from a repeated measures ANOVA showed that one or more of the observed means was different than the rest. The mean fluid volume changes for the anterior compartment for the 1-doff and 3-doff protocol were −5.8 % and −2.1 % respectively (Table 2). For the posterior compartment we observed −5.0 % and −3.0 % mean fluid volume changes for the 1- and 3-doff protocol, respectively. The Bonferroni corrected pairwise comparisons suggested that there was a meaningful difference between the 1- and 3-doff protocol limb fluid volume changes in the anterior compartment (P < 0.01) but not the posterior compartment (P = 0.48). The other differences were statistically not significant.

Table 2.

Means (Standard Deviations) for each metric and results from statistical tests.

Metric (unit)	1-Doff (St.Dev.)	3 Doff (St.Dev.)	P-Value (test performed)
Anterior Compartment Fluid Volume Cumulative Change (%)	−5.8 (4.2)	−2.1 (2.6)	P < 0.01 Repeated Measures ANOVA with Bonferroni Adjustment P = 0.48
Posterior Compartment Fluid Volume Cumulative Change (%)	−5.0 (4.3)	−3.0 (3.0)	Repeated Measures ANOVA with Bonferroni Adjustment
Pin Depth Cumulative (mm)	−1.8 (0.56)	−1.3 (0.42)	P = 0.50 Paired t-test
Pin Depth Magnitude (mm)	2.6 (2.1)	3.4 (2.4)	P = 0.36 Paired t-test
Anterior Distal Max-to-Min Cumulative Change (mm)	−0.35 (0.40)	−0.31 (0.37)	P = 0.82 Paired t-test
Anterior Distal Max-to-Min Magnitude (mm)	0.51*	0.83*	P = 0.22 Wilcoxon-rank sum
End of Session Figure-of-Eight Walk (Revolutions)	13.2 (1.2)	13.3 (2.0)	P = 0.73 Paired t-test

3.4. Pin depth

Shapiro-Wilk test results also suggested that the cumulative change in pin depth for the 1- and 3-doff and the magnitude of pin movement (MPM) for the 1- and 3-doff were normally distributed. Mean cumulative pin displacement for the 1- and 3-doff protocols were −1.8 and −1.3 mm respectively, and mean MPM for the 1- and 3-doff protocols were 2.6 and 3.4 mm, respectively. Statistical testing showed that neither differences in cumulative pin movement nor MPM were statistically significant (P = 0.50 and 0.36, respectively), suggesting that there were comparable pin displacements in both conditions (Table 2).

3.5. Anterior-distal site pistoning

Shapiro-Wilk test results suggested that anterior cumulative anterior-distal pistoning was normally distributed for the 1- and 3-doff condition. A right skewed distribution was found for the magnitude of pistoning changes in the 1-doff condition but normally distributed in the 3-doff condition. A paired t-test was run for the 1- and 3-doff cumulative change in pistoning where the means were −0.35 and −0.31 mm, respectively. A Wilcoxon-rank sum test was performed on the non-normal magnitude of pistoning data for the 1- and 3-doff protocols, where the medians were 0.51 and 0.83 mm, respectively. Differences in cumulative pistoning and magnitude of pistoning were not statistically significant (P = 0.82 and 0.22, respectively), suggesting that pistoning was comparable in both conditions (Table 2).

3.6. RSCR

Of the 22 test sessions, seven had a change in RSCR of 3 points or greater (Appendix B). Based on the participant comments recorded in the session notes, the five sessions that had decreases in comfort were primarily due to the socket becoming looser over time (three from the 1-doff protocol, and two from the 3-doff protocol). Reasons for RSCR score increases in two of the sessions (both from the 1-doff protocol) could not be determined from the session notes.

4. Discussion

Results from this study demonstrate a meaningful difference in limb fluid volume change between the 1- and 3-partial doff sessions and are similar to prior studies of temporary prosthetic socket doffing (Lanahan et al., 2023; Youngblood et al., 2022). One previous investigation (Lanahan et al., 2023) reported that partial doffing for both 4-min and 10-min durations reduced fluid volume loss compared to sitting with the socket fully donned, but there was no significant difference between durations of 4 min and 10 min. A second study on participants wearing pin suspension reported that participants experienced less posterior fluid volume loss when they intermittently fully-doffed their prosthesis than when they did not doff (Youngblood et al., 2022).

These findings cumulatively suggest that intermittent doffing frequency may have a greater impact on limb fluid volume than the duration that the prosthesis is partially doffed. In the previous two studies and the present study, there was reduced residual limb fluid volume loss when participants fully or partially doffed their socket compared to non-doffed rest. In the 4-min duration study (Lanahan et al., 2023), the participants completed 12 partial doffs in a 90-min experimental procedure (once every 7.5 min), which is five times higher than the doffing frequency presented in this study (once every 40 min). Though both studies suggested that partially doffing the prosthesis for 4–5 min during a seated rest can offer benefit, differences between the protocols for the two studies do not allow a quantitative relationship between the extent of fluid volume benefit and partial doff frequency increase to be calculated.

The rate of anterior limb fluid volume change was −2.9 %/hour for the 1-doff protocol and −1.1 %/hour for the 3-doff protocol (beginning minus end percent fluid volume divided by time). The difference in rate (1.8 %/hour difference, higher loss for 1-doff) is greater than that reported in investigations comparing the first two hours vs. the last two hours of 6-h activity protocol (0.9 %/hour difference (higher loss for the first two hours)) and comparing elevated vacuum vs. suction (0.1 %/hour difference (higher loss for suction)) (Youngblood et al., 2019; Youngblood et al., 2020). It is less than that comparing high activity vs. low activity (2.4 %/hour difference (higher loss for low activity)) and comparing oversized and nominal-sized sockets in the anterior-distal region (2.2 %/hour difference (higher loss for oversized socket)) (Sanders et al., 2017; Youngblood et al., 2019).

The present study was conducted with clinically applicable rates of partial doffing in mind, once every 2 h and once every 40 min. We observed a statistically significant reduction in limb fluid volume loss with a higher doff frequency, once every 40 min. Partial doffing may offer prosthesis users an alternative means to accommodate limb fluid volume change compared to the traditional practice of changing sock ply. Clinicians may ask their patients to aim for one to two, 5-min, partial or full doff rests during each hour of prosthesis use in an attempt to better stabilize residual limb fluid volume, and consequently better maintain socket fit.

The statistically significant change in the anterior compartment but not the posterior compartment may reflect the local tissue content. There is less soft tissue over the underlying bone anteriorly than posteriorly, which will induce greater tissue stress because bone is much stiffer than residual limb soft tissue and stresses tend to concentrate at the mismatch in materials interface (Traa et al., 2019; Vannah and Childress, 1996). In addition, the locking pin maintains a restrictive proximal-distal force that may accentuate skin stresses during swing phase when the socket is pulled distally away from the residual limb. The skin stresses may be reduced posteriorly compared with anteriorly because the soft tissue is thicker, thereby allowing more lateral displacement between the bone and socket.

Changes in anterior residual limb fluid volume were also observed in prior studies on transtibial prosthesis users testing sockets of different sizes. In one study, limb fluid volume change was measured in participants before and after they had their socket replaced or modified by their regular prosthetist. Participants who rated their new socket at least two ESCS_ave points higher than their old socket lost significantly less fluid volume in the anterior region (P = 0.002) than participants who rated their new socket less than 2 ESCS_ave (Morgan et al., 2022) points higher. There were no significant changes in the posterior region fluid volume, however (P = 0.957) (Vamos et al., 2025). In another study assessing limb fluid volume change using sockets of two different sizes (one socket was the same size as the participants practitioner-prescribed socket and a second had an increase in socket volume by about 6 %), it was found that most of the participants exhibited a strong dependence of limb fluid volume loss at the anterior and anterior-distal regions, with greater percentage fluid volume loss/h in the larger socket than in the smaller socket. The anterior effect sizes were large (−0.593 and −0.580, respectively) but the posterior effect sizes were small (−0.093 and 0.206) (Sanders et al., 2017).

While meaningful changes in limb depth have been observed in controlled laboratory studies testing different sock thicknesses (Coburn et al., 2022; Krout et al., 2023; Rich et al., 2023), changes in limb depth were not observed for the 1- and 3-partial doff protocols studied here. Pin downward motion over the 2-h protocol was typically about 1.0 mm total which is small, within the range of the distance between notches in a notched locking pin (3.0 mm).

Part of the challenge in interpretation of distal pin depth data may be the complex relationship between limb volume and limb depth. A limb that gains volume proximally may sit higher in the socket and not allow the distal part of the limb to travel deeper into the socket during ambulation. Conversely, a limb may gain volume distally and not be constrained by a proximal part of the socket shape, allowing the limb to travel more distally in the socket. In this case, an increase in limb fluid volume is measured but the opposite trend in the pin depth sensor data is observed (i.e., the pin would move distally). This example illustrates the complex, individualized nature of how a residual limb interacts with a socket and explains why limb depth may not always correlate to limb volume.

A similar argument could be made for anterior distal motion. Depending on where the limb fluid was gained or lost, the distal limb may be tighter or looser in the sagittal plane. The variability in where the limb fluid volume change occurred may explain day-to-day differences in distal limb motion observed among suction and elevated vacuum users in a recent take-home study (Coburn et al., 2022).

This investigation used the RSCR, a modified socket comfort scale, to assess relative changes in socket comfort from one walking bout to the next. The majority (15 of 22) of sessions did not show notable changes in socket comfort from one bout to the next, which is consistent with other recent studies that examined the effect of elevated vacuum on relative socket comfort over the course of an experimental trial (Hafner et al., 2016; Youngblood et al., 2022). Participants in the present study only reported a meaningful (i.e., 2 or more point) change in RSCR in 32 % of the sessions (9 % were an increase and 23 % were a decrease in RSCR). This study was not powered to investigate limb fluid volume or socket fit data as it relates to changes in socket comfort and thus no detailed analysis of participants’ comfort ratings was conducted.

Conceivably, changes in RSCR may relate to the “acceptable window of socket fit” as alluded to in Fig. 1. There is no evidence that any clinical metric reliably indicates when a user reaches the edge of the acceptable fit window (Morgan et al., 2022). In this investigation we have introduced within-socket metrics that can be used to evaluate fit, but additional studies are needed to properly investigate the relationship between objective socket fit measures like those studied here and prosthesis users’ perceived socket comfort.

The primary limitations in applying the study findings to clinical practice were the weight of the investigational sockets compared with participants’ traditional sockets and the mechanics of how the TARPIN pin functioned compared with a traditional pin. The TARPIN unit weighed about 450 g, greater than a standard pin-lock system (~37 g) but comparable to some electronic elevated vacuum systems (~225 to 475 g). The heavier weight may have caused the socket to feel different, affecting participants’ performance and self-report ratings of comfort, and thus the influence of the altered suspension system must be considered when drawing comparisons to clinical observations. In addition, the TARPIN system pulled the pin into the socket, unlike traditional pins that are pushed deeper into the socket by the user. It is unknown if this difference affected pin depths compared to users’ traditional sockets. The sample size of this study was small (n = 11), possibly leading to high inter-subject variability. Though this is not uncommon in prosthetics research, this is a limitation of the evaluation performed.

5. Conclusion

This study demonstrates that a frequency of 1.5 partial doffs per hour significantly reduces anterior residual limb fluid volume loss in transtibial prosthesis users compared to 0.5 partial doffs per hour without compromising limb stability or function. These findings support and extend previous research suggesting that doffing frequency may be a more critical factor than doffing duration in mitigating fluid volume loss. While limb depth metrics did not differ between conditions, likely due to the complex and individualized nature of socket-limb interactions, residual limb volume changes were consistent with prior work. Importantly, the reduction in fluid volume loss did not consistently translate to perceived comfort improvements, highlighting the multifactorial nature of socket fit. Although the investigational socket’s weight and suspension mechanics limit direct clinical translation, the study underscores the potential clinical utility of intermittent partial doffing as a strategy to manage diurnal limb volume changes. Future work should further investigate patient-specific responses and explore how objective measures of socket fit relate to user comfort in real-world settings.

Supplementary Material

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.clinbiomech.2025.106537.

Abbreviations:

ANOVA

Analysis of variance

MPM

Magnitude of pin movement

RSCR

Relative socket comfort rating

SCS

Socket comfort score

TARPIN

Tether-aided release-relock pin system