Secondary Logo

Journal Logo


Volume Management

Smart Variable Geometry Socket (SVGS) Technology for Lower-Limb Prostheses

Greenwald, Richard M. PhD; Dean, Robert C. ScD; Board, Wayne J. MS

Author Information
JPO Journal of Prosthetics and Orthotics: July 2003 - Volume 15 - Issue 3 - p 107-112
  • Free


The lower-limb prosthetic socket is effectively a fixed volume into which the residual limb is placed. An increase or decrease in the volume of the residual limb results in a socket that is either too tight or too loose, respectively. Numerous factors contribute to prosthesis fit, including socket design, normal daily volume fluctuations, and long-term changes in limb volume. Poor socket fit remains the primary concern of lower limb amputees with regard to their prostheses.1

Daily variations in limb volume play a critical role in prosthesis fit. Of great importance is the maintenance of a precision socket fit. The precision fit of most suction sockets is lost within a short period. This occurs, in part, because of physiological volume changes of the residual limb resulting from normal ambulation and especially from the high compressive pressures exerted upon the limb. Volumetric changes occur hourly, daily, and monthly.

There are limited data in the literature that quantify volume fluctuation that occurs in the residuum over the course of the day. Residual limb volume changes of between −11% and +7% support the argument that the socket fit changes throughout the day.2 A volume increase of +3% to 5% may cause significant difficulty for an amputee donning his/her prosthesis.3 The amount of daily volume fluctuation varies greatly among individual amputees and is a function of prosthesis fit, activity level, ambient conditions, body composition, dietary habits, and for women, monthly cycles.

Current methods used to compensate for daily volume changes in the limb include multiple socks, pads, and inflatable air bladders. Socks and pads represent discrete levels of change in socket volume and do not continuously match volume changes occurring in the limb. Pneumatic systems, using inflatable bladders inserted in the socket to manage volume fluctuations, suffer from inadequate support of the limb during walking due to under-inflation, localized tissue compression due to over-inflation; high compliance from compression of the air inside the bladder causes excessive pumping of the residual limb in and out of the socket.4 The clinical use of pneumatic inserts is limited.

A novel technology, dynamic variable geometry fitting system (patent pending) has been employed in a lower-limb suction socket application to compensate for daily limb volume fluctuations automatically and continuously over the course of the day. The smart variable geometry socket (SVGS) system developed for this application overcomes limitations of existing socket volume management methods by dynamically changing the volume of the socket to compensate for changes in volume of the residual limb. This technical note describes the theory of operation and method of use of the SVGS system for clinical use.


Stability and comfort of the prosthesis are likely to be related to the quality of the fit at the interface between the limb and the socket. Volume changes in the lower limb over the course of the day result in a mismatch in the volume of the limb and the volume of the constant volume socket. Volume loss results in a pistoning or pumping motion between the tissue and the socket, which can decrease retention security during ambulation and cause skin lesions and, with suction sockets, loss of suction.5

To prevent pistoning, a method of maintaining a consistent and comfortable level of socket fit is required that is independent of individual intrinsic factors (hormone levels, water retention, short-term weight change) and extrinsic factors (air temperature, activity level). On the other hand, skin tissue cannot be subjected to constant levels of high pressure (>8 kPa or ˜1 psig) for any sustained period of time without tissue necrosis.6 The desired level of socket tightness (i.e., comfortable and stable fit) during various activities is usually a personal preference for the amputee; it may change for different activities, such as office work and athletics.

The SVGS system (Figure 1) has three main components: a fluid reservoir, a mechanical control circuit (Figure 2), and multiple discrete bladders located inside the socket. Power for the system is provided entirely by the amputee—no external power source is needed. During normal ambulation, a natural pumping cycle occurs within a lower-limb suction socket: suction is created during the swing phase followed by compression during the stance phase (Figure 3). Fluid, typically water, is drawn into the bladders from the reservoir during the swing phase, then distributed among the bladders during stance. Gravity and the dynamic forces on the swinging prosthesis control the level of suction, so no vacuum control is required. Various check valves (CV) control the flow direction within the system. The bladders inflate until they reach a maximum stance pressure, set by the prosthetist using the adjustable stance pressure regulator (PRS) to a level that is comfortable for the amputee within the limits of the pressure regulator, between 28 and 105 kPa (4 and 15 psig).

Figure 1.:
Rendering of SVGS applied to a transfemoral suction socket. The placement and size of the bladders and the placement of the reservoir below the socket are for illustrative purposes only and represent one possible combination for use.
Figure 2.:
Prototype control system photo (left) and schematic (right). CV, check valve; F, particle filter; NV, night valve; PRi, ischemic pressure regulator; PRs, stance pressure regulator; R, flow resistor; RES, reservoir; FP, fluid fill port; B, bladder.
Figure 3.:
Sample data demonstrating intrasocket pressure distribution during a single stride measured by discrete pressure transducers placed through the wall of a suction socket during gait for a single transfemoral amputee (top) and transtibial amputee (bottom). Pressure measurement location key: A, anterior, D, distal, L, lateral, M, medial, P, posterior, and X, proximal.

The volume of fluid drawn from the reservoir and pumped into the bladders to achieve the maximum stance pressure is a function of the instantaneous volume mismatch between the socket and the residuum. Various sizes, shapes, and placement of the bladders within the socket are possible. The prosthetist selects the number, location, and size of bladders that are considered to best match the limb geometry, musculature, and bony prominences. It is important to note that the tissues are not subjected to a continuous and highpressure level during ambulation because the pressure in the socket is continuously varying between suction and the maximum pressure set by PRs.

Ischemic relief for the residuum tissues is integrated into the system to relieve bladder pressure during stasis below the tissue ischemic limit of 8 kPa (˜1 psig) above ambient. A resistor (R) continuously bleeds liquid from the bladders back into the reservoir at approximately 1/10 of the liquid circulating rate. Because the liquid is essentially incompressible and the socket frame is relatively stiff, a few milliliters of liquid bleeding from the bladders reduces the stance pressure to below the ischemic limit. To prevent loss of suction when not walking, the system drains only enough liquid to drop the pressure below the ischemic limit. This is accomplished by an ischemic pressure regulator (PRi) that maintains residual bladder pressure at approximately 3.5 kPa (0.5 psig). By retaining most of the liquid required for volume management during walking, a relaxed fit of the socket is created while still maintaining a close-fitting socket; only a few steps are required, when walking resumes, to automatically return stance pressure to its set point and the socket fit to its dynamic walking tightness.

A manually operated night-valve (NV) bypasses PRi and R to release all of the fluid from the bladders back into the reservoir to provide maximum socket volume for donning and doffing.

The rationale for using liquid fill rather than air fill is presented below. The intrasocket environment in lower-limb, total-contact suction socket prostheses is characterized by dynamic changes in normal and shear forces applied to the tissues.4,7 Leg motion, muscle contraction, and ambulation create changes in the pressure as the tissue shifts within the socket. The tissue, composed primarily of water, reacts to changes in pressure with a fluid transport mechanism that forces fluid either into or out of the limb. It is normal for residual limb volume to fluctuate by as much as 11% during the day.2

To quantify volume changes in the socket during the course of the day, we measured manually the change in volume of liquid in a reservoir placed external to a socket modified with SVGS between the time of donning the socket and at 4-hr intervals during the course of the day until the socket was removed. Fluid was removed from the reservoir and added to the socket automatically by SVGS as needed to maintain a preset comfortable pressure level for the wearer. For example, the average daily maximum volume variation for a single transfemoral amputee over a 20-day recording was 40 mL in a 1500-mL socket, or approximately 2.7% (Figure 4).

Figure 4.:
Maximum daily volume variation over a 20-day recording period for a transfemoral amputee with a residual limb volume of 1500 mL. Data were generated by measuring manually the change in volume of liquid in a reservoir placed external to a socket modified with SVGS between the time of donning the socket and at 4-hr intervals during the course of the day until the socket was removed.

A simple argument demonstrates the appropriateness of liquid fill. As the residuum shrinks in volume, it loses liquid to the lymph and vascular systems. Because the soft tissue contains no gas pockets and is itself about 90% liquid, it behaves physically like a liquid. To maintain physiological conditions in the socket, the shrinkage void should be back-filled with liquid. Then the compliance, or compressibility of the substance (i.e., residuum plus fill material) within the socket remains essentially constant despite residuum shrinkage. In contrast, a gas used to fill the void has approximately 20,000 times the compressibility of water (at 100 kPa and 30°C). A gas-filled bladder will expand and contract in volume in a fashion drastically different from that of the soft tissue of the residuum. The proximal-distal (P-D) compliance will be large, engendering excessive pumping (i.e., pistoning).

The difference in P-D compliance between liquid and gas fill in a socket can be determined mathematically using the laws that govern the response of gas and liquid to pressure changes. For example, the exemplar data of intrasocket pressure data from a single stride for both a transfemoral (TF) and transtibial (TT) amputee illustrated in Figure 3, are transformed, using the perfect-gas laws, into socket-volume variations for both liquidand air-filled bladders for that stride cycle (Figure 5). Theoretical results are shown for air- and liquid-filled bladders for typical TF and TT variable-volume sockets (1800- and 1000-mL volume, respectively) with maximum bladder volume/socket volume at 10% (180 and 100 mL, respectively). The theoretical P-D compliance, which represents the amount of pistoning that would occur in the socket, was calculated from this same data by dividing volume change by the mean cross-sectional area of the socket (approximately 1.8 × 10−2 m2 for TF and 7.9 × 10−3 m2 for TT). Note in Figure 5 the large difference in volume change required between conventional air-filled and water-filled bladders. The peak TT bladder change in volume for this stride is about −65% for air-filled and −1% for liquid-filled bladders. These convert to P-D pumping strokes for this TT socket (1000 mL) of 83 mm for air fill and 1.3 mm for liquid fill. The TT amputee would probably lose a suction-retention prosthesis under these conditions. With liquid-fill, the socket would remain tight and the pumping stroke (1.3 mm) very small.

Figure 5.:
Theoretical socket-volume variations during a single stride for liquid- and air-filled bladders calculated from the exemplar anterior lateral bladder data shown in Figure 3 using the perfect-gas laws. Transfemoral amputee walking on a hard floor with an air-filled bladder (top left) and a liquid-filled bladder (bottom left). Transtibial amputee walking on hard floor with an air-filled bladder (top right) and a liquid-filled bladder (bottom right).

Raising the pressure in gas-filled bladders might be thought to reduce compliance. The bulk modulus (i.e., compressibility) of gas (dp/dv) varies as pk+1/k where k = ratio of specific heats, so increasing the bladder inflation pressure by a factor of 4, to approximately 300 kPa (43.5 psig), would increase bladder stiffness by 3.25-fold. Then the maximum bladder volume change required for air-filled TT bladders would decrease to −20%, which is still excessive for suction retention. Additionally, such a high pressure would greatly exceed the ischemic limit of the tissue and might cause tissue damage.

These calculations serve to corroborate both the experimental findings of recently published research on the mechanical properties of commercially available air-filled bladder systems4 and the lack of clinical satisfaction with these devices.


During the course of 10 years of research on the concept of a variable volume socket, with significant research and development support from the National Institutes of Health and National Science Foundation Small Business Innovation Research program, several alternative methods for activating liquid-filled bladders were considered. We attempted to develop a manually operated unitary pump that would be integrated with an external reservoir and internal bladder to form an element that the prosthetist could insert through a 15-mm hole in the socket frame. The concept was that the amputee could inject or extract liquid from the bladder by merely pressing a pump valve or relief valve, respectively. But building a pump valve that would not leak within our desired pressure range specifications between −40 kPa (−6 psig) and 133 kPa (19.3 psig) proved very difficult. Further, we required that manual operating force of the pump be < 0.5 N (2 lbf) for ready actuation by the arthritic hand. The manual pumping system proved to be difficult to construct within desirable functional and cost specifications.

Second, we considered an electrically actuated and controlled pumping system (Figure 6). This electrical system was not intended to be cost effective nor practical for clinical use. The system was heavy, noisy, bulky, and relied on battery power; it did not meet any practical functional or cost effective measures required for widespread clinical use. Experience with it confirmed its lack of robustness. Responsive gels in variable volume liners were also evaluated and tested. They proved too constrained by the operating pressures required.

Figure 6.:
Electrically actuated and controlled pumping system

Experience with these various potential solutions for a continuously adjusting mechanism to react to limb-volume changes in the socket environment led us to reconsider the intrasocket pressure-time curves that occur during the gait cycle. We realized that this curve describes a pumping cycle, and that the pumping actuator was already on board; that is, the amputee could supply the required power for the system just by walking. Additionally, we realized that the smart features of the system need not be electronic. This led to development and construction of the system presented in this technical note. The most attractive system is an automatic (i.e., smart) system that can maintain proper-fit pressure magnitude and distribution in the socket and continuously during ambulation and long-term sitting.


We believe that the retention or coupling efficacy of most lowerlimb amputees' prostheses is compromised by uncompensated diurnal, soft-tissue volume variations. The socket is the mechanical interface between a lower-limb prosthesis and the amputee's body. This interface is responsible for the mechanics behind motion and pressure in the amputee's socket. If a socket fit is unsatisfactory, it can cause sores, localized and distributed pain, ambulatory difficulty, and insecurity. Poor fit remains the primary complaint of lower-limb amputees with regard to their prostheses.1 The issues of confidence and a sense of security are of great concern to amputees, particularly for older wearers. They fear that the prosthesis will fail to perform consistently and safely. This fear puts the amputee constantly on his/her guard against prosthesis malfunction, which itself may cause additional instability and inconsistent performance that can result in a fall. Falls are the older amputee's greatest safety hazard and a leading cause of morbidity and mortality (from secondary consequences of a fall).

Our solution to this generic challenge is to employ a dynamic, continuous and automatic varying volume socket to maintain: 1) good security and safety through enhanced retention, 2) low P-D compliance (i.e., low pumping or pistoning), 3) comfort, and 4) the amputee's confidence in the function of the prosthesis. We are currently completing a multisite clinical trial of 20 transfemoral and 20 transtibial subjects to statistically determine the effects of SVGS on socket comfort, gait symmetry, and walking performance.


Many other individuals have contributed to and supported our work, including our colleagues: David J. Loney of WillowBrook Prosthetics & Orthotics, Lebanon, NH; Peter Couture and Robert Emerson of Next Step Orthotics & Prosthetics, Inc., Manchester, NH; Michael B. Mayor, MD, Hanover, NH; Algis Maciunas of Wethersfield-Hanger Prosthetics & Orthotics, Inc., Wethersfield, CT; and Louis J. Haberman of Progressive Restorations, Inc., Denville, NJ. We acknowledge and appreciate the patience and eager cooperation of our α-prototype test team.


1. Legro MW, Reiber G, del Aguila M, Ajax MJ, Boone DA, Larsen JA, Smith DG, Sangeorzan B. Issues of importance reported by persons with lower limb amputations and prostheses. J Rehabil Res Dev. 1999;36:155-163.
2. Board WJ, Street GM, Caspers C. A comparison of trans-tibial amputee suction and vacuum socket conditions. Prosthet Orthot Int. 2001;25:202-209.
3. Fernie GR, Holliday PJ. Volume fluctuations in the residual limbs of lower limb amputees. Arch Phys Med Rehabil. 1982;63: 162-165.
4. Sanders JE, Cassisi DV. Mechanical performance of inflatable inserts used in limb-prosthetics. J Rehabil Res Dev. 2001;38: 365-374.
5. Mak AFT, Zhang M, Boone DA. State-of-the-art research in lowerlimb prosthetic biomechanics-socket interface. J Rehabil Res Dev. 2001;38:161-173.
6. Daly CH, Chimoskey JE, Holloway GA, Kennedy D. The effect of pressure loading on the blood flow rate in human skin. In: Kenedi RM, Cowden JM, Scales JY, eds. Bedsore Biomechanics. London: Macmillan; 1976:69-77.
7. Zhang M, Turner-Smith AR, Tanner A, Roberts VC. Clinical investigations of the pressure and shear stress on the transtibial residuum with prosthesis. Med Eng Phys. 1998;20: 188-198.

Lower-limb prosthesis; volume management; socket fit; socket retention

© 2003 American Academy of Orthotists & Prosthetists