Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

International Forum: Proper Time for Permanent Prosthetic Fitting

Magnus Lilja, CPO
Tommy Oberg, MD, PhD

ABSTRACT

The authors examined postoperative volume changes of residual limbs to identify the proper time for definitive transtibial prosthetic fitting (the point in time at which the residual-limb volume has stabilized). The number of stockings in the socket was used as a criterion for good or bad prosthetic fitting. Residual-limb volume changes of 11 transtibial amputees were followed for approximately 160 days. The volume was determined by laser scanning with the CAPOD® system.

The authors found the volume changes could be characterized by a simple power function. The range of volume reduction after 160 days was 17 percent to 35 percent. Using a two-stocking criterion (10-percent reduction of volume), the criterion was fulfilled by all patients at 100 days postamputation; with a one-stocking criterion (5-percent reduction of volume), the criterion was fulfilled by 120 days postamputation. From these data, the authors recommend fitting with a temporary prosthesis as early as possible, and the change to a definitive prosthesis should occur four months postoperatively.

Introduction

Early fitting of a prosthesis after transtibial amputation is of vital importance for the future rehabilitation of the patient. On the other hand, early fitting with a definitive prosthesis will be jeopardized by rapid volume changes in the early postoperative phase. Thus, it is important to identify the time when the residual-limb volume has stabilized and the volume changes are small enough not to interfere with the prosthetic fitting. In this study, the authors have examined the volume changes postamputation and have tried to find a criterion defining when to provide a definitive prosthesis.

Since early rehabilitation is important, the patient cannot wait until volume changes have stabilized to receive a prosthesis. One possibility is to provide the patient with an inexpensive temporary prosthesis; this method has been reported as successful in a number of studies (1-6). The advantages of early prosthetic fitting include accomplishing a more normal gait re-education, achieving a more independent life, undertaking more active physical training and gaining psychological advantages such as better acceptance of the amputation and restoration of the body image (1,7-10). Early prosthetic fitting also will hasten the maturation of the residual limb as well as the adaptation of the residual limb form to the definitive socket (1,8). However, since the temporary prosthesis has several limitations, it is important to find the proper time for fitting the patient with the definitive prosthesis.

Most transtibial prostheses are based on the patellar-tendon-bearing (PTB) concept (11), and, today, new materials such as silicone frequently are used (12). The idea of total contact with the pressure distributed over the whole residual-limb surface currently is widely accepted. Stiffness has been described as the most important property of the coupling between the residual limb and the prosthesis (13). In fact, high stiffness improves proprioception and prosthetic function. Thus, stiffness of the coupling is one of the technical goals in prosthetic fitting.

A prerequisite for total contact and high stiffness is a constant residual-limb volume. Incongruency between the residual limb and the socket often is compensated for by one or two (and sometimes even more) stockings. The price of such a compensation, however, is a loss of stiffness of the coupling and increased mobility of the residual limb inside the socket. Consequently, a large number of stockings must be regarded as a negative factor.

The authors have hypothesized a reduction of postoperative residual-limb volume that can be approximately described by a negative exponential curve. This reduction may be attributed to two main factors: a rapid phase due to reduction of edema and a slow phase due to atrophy of the soft tissues. After a major amputation, a severe traumatic edema and hematoma in the residual limb may result, which can be reduced by bandaging and physical activity (14-16). The residual-limb volume decreases gradually during the early maturation process due to reduction of the postoperative edema.

In addition, the process of muscle atrophy starts immediately postoperation and continues parallel to the reduction of the edema (17). The amount of this atrophy is correlated to the physical activity, postoperative bandaging and prosthetic fit (18). Renstrom et al. (19) found an atrophy of the quadriceps muscle after transtibial amputation, which was mainly confined to the vastus lateralis muscle. The mean muscle fiber area was reduced to 74 percent compared with the nonamputated leg.

Several attempts have been made to determine the volume of residual limbs. Fernie et al. (20) used an instrument with a water tank; other authors also have used a water immersion technique or a tape measure to determine residual-limb volume (1,21,22). No evaluation of measurement errors was performed in these studies. Fernie and Holliday (23) showed a reduction of volume of about 5 percent during the first 100 to 200 days in transtibial amputees. Golbranson et al. (22) showed, using a linear regression technique, a reduction of the residual-limb volume of 0.011 percent per day.

With the introduction of the CAD/CAM technique, a new method for volume determinations was devised. Data can be fed into a computer via a digitizer or scanner, and

a two- or three-dimensional model can be constructed. The CAD software also has an option for calculation of volumes of the model (24). However, very little is known about the validity and reliability of such measurements. Does the new technique introduce new errors of measurement, and of what magnitude?

Systems using digitizers for measuring hand casts have been described by several authors (25,26). Such systems heavily depend on the quality and accuracy of the cast.

Other systems, such as those described in References 27-29, are based on noncontacting laser scanning. The accuracy of one such system, the CAPOD® systema, for determination of volumes has been described in a previous study by Lilja and Öberg (30). The random error, represented by the coefficient of variation, was 0.5 percent, with a linear systematic error of 2.5 percent.

This study was intended to describe postoperative volume changes of transtibial residual limbs during the first 160 days and, more specifically, to test the hypothesis that volume reduction follows a negative exponential function. Furthermore, using the volume data, the authors intended to define a criterion for determining the proper time to fit a definitive prosthesis.

Materials and Methods

Subjects

Eleven transtibial amputees (three men and eight women) with a mean age of 76.8 (range 65 to 87) years were followed for approximately 160 days to study volume changes of their residual limbs. Inclusion criteria were 1) amputation caused by peripheral vascular disease (PVD) with (six subjects) or without (five subjects) diabetes, and 2) healthy enough to undergo an examination with laser scanning. The volumes of the residual limbs were evaluated once a week during the first eight weeks, every second week during the next six weeks and once a month for the remainder of the study. The first residual-limb evaluation occurred during the first three weeks after the amputation. The subjects' body weight was measured at every occasion for volume determination.

Except for the volume determination, the patients were treated according to standard routines at the clinic, i.e., fitted with a temporary prosthesis and rehabilitated with physical training, gait re-education, prosthesis training and occupational therapy. The residual limb was wrapped with an elastic bandage as soon as the surgical dressing was removed. The elastic bandage was removed 10 minutes prior to the volume determination. The exact time for delivery of the definitive prosthesis was recorded (see Table A) .

Laser Scanning

Volume determination was performed using the CAPOD system (28). The scanning procedure followed the CAPOD system's manual. A thin nylon stockingb was put on the residual limb to increase the reflection of the laser beam. A thin black line was marked at the tibial plateau as a reference for the calculation of the volume. The time for scanning was about 10 seconds.

After scanning, a manual procedure was used for axial centering of the residual-limb model. The tibial plateau indicated the proximal level for the volume calculations, and the end of the residual limb represented the distal level. The residual-limb volume was calculated by an automatic procedure included in the CAPOD software.

Curve-Fitting Procedure

All individual volume data were tested with a commercial software package, CURVEFIT®^c, for least-square-fitting of the data to 25 predetermined equations. The coefficient of determination (r²) was used as a measure of the fitting of the model to the data (31). The equation showing the highest coefficient of determination (r²) for most of the cases was selected.

Curve fitting was performed with both non-normalized and normalized data. Normalization was performed by division of all volume data with the calculated value for the 160th day, and the second normalization was performed by division of all volume data with calculated volume for the fifth day postamputation. However, not all patients could be examined at the fifth and 160th days; for these subjects, expected volumes for these days were calculated by the individually fitted equation. Using such a normalization procedure, all subjects had a relative volume of 100 percent.

Criterion for Bad Fitting

Putting stockings on the residual limb to fill the socket is a standard procedure to compensate for bad fitting of the socket to the residual limb. Normally, Swedish prosthetists use one or two stockings, then a new socket must be made. For this reason, the extra volumes added to the residual limb by one and two stockings, respectively, were calculated using a CAPOD system scanner. A normal plaster-of-paris residual-limb model with length = (112) 2 breadth was used (32).

The model then was covered with an isolation rubber bagd, and the volume was determined with the CAPOD scanner. The model then was covered with one or two terry cloth stockingse, and an isolation rubber bag was applied. The air was sucked out with a depression of 700 mbar, maximum of the apparatus, simulating the pressure inside the prosthetic socket during load, and a new volume determination was carried out. This procedure was performed 10 times for each stocking, and the mean volumes of one and two stockings, respectively, were used as criteria for bad fitting. The difference between the two measurement values was used to evaluate the volume of one and two stockings, respectively.

Statistical Analysis

Means, range and coefficient of determination were calculated according to standard procedures (33).

Results

The body weights of the amputees were relatively stable during the study. All patients, except for one who did not change body weight, increased in weight during the study. The maximum change was 5 kg (8 percent of body weight), and the mean weight change was 2.1 (SD=1.4) kg.

Volume Changes

Individual raw volumes are shown in Figure 1 . Despite considerable individual differences, the same pattern is seen: A rapid decrease of the volume occurs in the early postoperative phase and then reduction slows. The volumes, normalized relative to the fifth day, are shown in Figure 2 . The total range of the volume changes is 17 to 35 percent. The relative volumes, using the 160th day postamputation as a reference, are shown in Figure 3 . The following equation shows the best fit of the data for most of the patients:

y = a × x^b    (a power function)

        a = 1558 (SD=363)
        b = 10,099 (SD=0,028)

where y is the residual-limb volume, x is time after amputation, and a and b are constants. The coefficients of determination (r²) for all individual data range between 0.81 and 0.96, with a mean of 0.90.

Volume of Stockings

The mean volume of one stocking, when measured on a plaster-of-paris residual-limb model with a volume of 1192.1 ml, was 5.2 percent. The mean volume of two stockings was 9.4 percent of the residual-limb volume.

Criterion for Good/Bad Fitting

The number of patients fulfilling the one- and two-stocking compensation (5 percent and 10 percent, respectively) criteria is shown in Figure 3 . If the two-stocking criterion is adopted, nine of 11 patients fulfill the criterion after 80 days; but if the one-stocking criterion is used, it takes 120 days until the criterion is fulfilled.

Discussion

Early fitting of a prosthesis after transtibial amputation will be jeopardized by rapid volume changes during the rehabilitation phase so it is important to identify the time when the residual-limb volume has stabilized and the volume changes are small enough not to interfere with the prosthetic fitting. The results of the present study indicate a volume reduction for all patients following a power function, y = a × x^b. While the number of patients included in the study is relatively small, as can be seen in Figure 2 , most volumes do follow the same pattern. These results suggest a proper time for definitive prosthetic fitting.

All patients included in the study were amputated due to PVD--six subjects with diabetes and five subjects without diabetes. This inclusion criterion was used since these diagnoses are the main indications for amputation in Scandinavia. Further, the age distribution of the patients in the study roughly corresponds to that reported in other Scandinavian studies (34-36). None of the patients included in the study underwent diuretic treatment.

Golbrandson et al. (22) studied the volume changes over a period of almost 340 days, and Fernie and Holliday (23) studied their patients over a 600-day period. In the present study, the patients were followed for about 160 days. The reason for this time limit can easily be seen in Figure 1 , which shows the residual-limb volume stabilizes about 100 days postamputation. The body weights for all patients in the present study were relatively stable during the whole study. The maximum body weight fluctuation was 5 kg (see Table A) . Fernie and Holliday (23) found the body weights of amputees started to increase after about 200 days, and concluded this increased body weight could affect the residual-limb volume. However, the volume fluctuations between 200 and 600 days postamputation do not influence the first fitting of the definitive prosthesis. In the present study, the subjects increased their body weight, with a mean of 2.1 kg.

All patients included in the study had residual-limb bandaging during the rehabilitation, and the wrap was removed only during the gait re-education and exercise. While most authors agree wrapping the residual limb greatly influences final residual-limb form (15,16), Golbrandson et al. (22) did not find such an influence. To standardize the scanning procedure, the wrap was removed about 10 minutes prior to the scanning. Before the volume determination, a thin white nylon stocking was pulled onto the residual limb to increase the reflection from the laser beam. The nylon stocking was used in all evaluations and therefore did not influence the volume determinations.

The accuracy and precision of the CAPOD system has been evaluated in an earlier study. The random error represented by the coefficient of variation was relatively small (0.5 percent), and the linear systematic error was 2.5 percent (30). Such a linear systematic error can easily be corrected for; therefore, it did not influence the present study. In the authors' opinion, the random error is too small to have any essential influence on the volume changes presented in this study.

The authors hypothesized the volume reduction of the residual limb can be described by a negative exponential function. To verify this hypothesis, all data from the volume determination were tested in 25 different mathematical functions (31). As a degree of the relation between the volume data and the different functions, the coefficient of determination (r²) was used. The function y = a × x^b seemed to be the simplest function with a high coefficient of determination (mean 0.90; range 0.81-0.96). This value of r2 shows that at least 81 percent of the volume can be explained by the mathematical model y = a × x^b (33). This procedure permits a description of the volume changes in residual limbs during the first 160 days after the amputation.

This kind of modeling has not earlier been described in the prosthetic literature. Fernie and Holliday (23) found some shrinking of the residual limb similar to what was found in the present study, but they did not try to fit their data into a mathematical function. In the rehabilitation process and the prosthetic fitting process, no scientific criteria have been used to decide when to fit the transtibial amputee with a definitive prosthesis. It may seem natural to provide patients with a definitive prosthesis as soon as the wound has healed, i.e., two to three weeks postamputation. Early fitting with a definitive prosthesis will, however, result in a bad fitting due to volume reduction of the residual limb, causing a need for a new socket. From an economic point of view, a high failure rate is unacceptable. A late fitting with a prosthesis, on the other hand, will delay the rehabilitation of the patient.

One alternative is to provide the patient with an inexpensive temporary prosthesis in the early postoperative phase and then switch to a more expensive definitive prosthesis later, when the residual-limb volume has stabilized. Identifying the proper time for this definitive prosthesis fitting requires establishing criteria for good and bad fitting. These criteria depend on several different factors and variables (13).

In the present study, the discussion of such criteria was limited to the number of stockings used to obtain an acceptable fitting of the prosthesis. Most prosthetists in Sweden consider one or two stockings acceptable. Consequently, bad fitting occurs when the patient needs three or more stockings to be able to use the prosthesis. The number of stockings used is related to the stiffness of the coupling between residual limb and socket. One of the aims of prosthetic fitting is to have a high stiffness, thereby reducing the existing movement between the skeleton and the socket (13,37,38).

It is possible to estimate the volume of one and two stockings. The authors determined the volume of one and two stockings by using an isolation rubber bag and air evacuation. Fernie and Holliday (23) estimated the volume of two stockings through mathematical calculations. Their results correspond with the results of this study: One stocking represents about 5 percent of total residual limb volume. The stocking volume expressed in percentages relates to the residual-limb volume. In the present study, the volume of the residual-limb model was 1,192 ml.

In the attempts to define the proper time for definitive prosthesis fitting, all residual-limb volumes represented by a power function were normalized related to day 160 postamputation (see Figure 3) . The number of patients who do not fulfill the criterion represents the failure rate; thus, after 40 days, only three out of 11 patients fulfill the 10-percent criterion (two stockings), and none fulfill the 5-percent criterion (one stocking).

After 100 days, all patients fulfill the 10-percent criterion, and, after 120 days, all meet the 5-percent criterion. Today, most definitive prostheses are provided about three months or more postamputation. This practice is in good accordance with the findings of the present study.

Conclusion

The authors examined postoperative volume changes of residual limbs and found postoperative residual-limb volume can be described with a power function that closely agrees with the authors' primary assumptions. Furthermore, the changes of residual-limb volume could be related to a criterion for good prosthesis fitting. Using this criterion, the authors found the proper time for definitive prosthesis fitting to occur at about 120 days. This finding agrees fairly well with the practice applied by many prosthetists over the world. Based on these findings, the authors recommend a fitting with a temporary prosthesis as early as possible and a change to a definitive prosthesis four months postoperatively.

References:

1. Liedberg E, Hommerberg H, Persson BM. Tolerance of early walking with total contact among below-knee amputees--a randomized test. Pros Orth Int 1983;7:91-5.
2. Engstrom B, Van de Ven C. In: Physiotherapy for amputees. London: Churchill Livingstone, 1993.
3. Redhead RG, Davis BC, Robinsson KP, Vitali M. Postamputation pneumatic walking aid. Br J Surg 1978;65:611-2.
4. HÖlter W, Echterhoff M, BlÖmer A, Verfürden H. The management of amputations of the leg using a new rigid foam plaster for prosthetic fitting. Int Orthop 1980;4:73-7.
5. Mooney V, McClellan B, Cummings D, Smith P. Early fitting of the below-knee amputee. Orthop 1985;8:199-202.
6. Folsom D, King T, Rubin JR. Lower-extremity amputation with immediate postoperative prosthetic placement. Am J Surg 1992;164:320-2.
7. Kitowski VJ, Appel MF, Haslam T. Prosthetic fitting immediately after below-knee amputation. South Med J 1975;68:739-42.
8. Kraker D, Pinzur MS, Daley R, Osterman H. Early postsurgical prosthetic fitting with a prefabricated plastic limb. Orthop 1986;9:989-92.
9. Angel JC. Principles of prosthetic prescription in the early month after lower-limb amputation. In: Murdoch G, Donovan RG, eds. Amputation surgery and lower-limb prosthetics. Oxford: Blackwell Scientific Publications, 1988:1,63-70.
10. Ham R, Cotton L. In: Limb amputation, from aetiology to rehabilitation. London: Chapman & Hall, 1991.
11. Radcliffe CW, Foort J. The patellar-tendon-bearing below-knee prosthesis. Berkeley and San Francisco: University of California, 1961.
12. Gottschalk FA. Prosthetic components in clinical practice. J Rehab Sci 1994; 7:55-6.
13. Klasson B. In: Appreciation of prosthetic socket fitting from basic engineering principles. Glasgow: University of Strathclyde, 1995.
14. Levy SW. Disabling skin reactions associated with stump edema. Int J Dermatol 1977;16:122-5.
15. Manella KJ. Comparing the effectiveness of elastic bandages and shrinker socks for lower-extremity amputees. Phys Ther 1981;61:334-7.
16. MacLean N, Fick GH. The effect of semirigid dressings on below-knee amputations. Phys Ther 1994;74:668-73.
17. Baumgartner R, Langlotz M. Amputee stump radiology. Pros Orth Int 1980;4:97-100.
18. Levy SW. Skin problems of the leg amputee. Pros Orth Int 1980;4:37-44.
19. RenstrÖm P, Grimby G, Morelli B, Palmertz B. Thigh muscle atrophy in below-knee amputees. Scand J Rehab Med Suppl 1983;9:150-62.
20. Fernie GR, Holliday PJ, Lobb RJ. An instrument for monitoring stump edema and shrinkage in amputees. Pros Orth Int 1978;2:69-72.
21. Renstrom P, Goldie IF. In: The below-knee amputee. GÖteborg: Department of Orthopaedic Surgery II, Sahlgren's Hospital, University of GÖteborg, Thesis, 1981.
22. Golbrandson FL, Wirta RW, Kuncir EJ, Lieber RL, Oishi C. Volume changes occurring in postoperative below-knee residual limbs. J Rehab Res Dev 1988; 25:11-8.
23. Fernie GR, Holliday PJ. Volume fluctuations in the residual limbs of lower-limb amputees. Arch Phys Med Rehab 1982;63: 162-5.
24. Lilja M, Öberg T, Karsznia A. The applications of CAD/CAM technique in prosthetics and orthotics. Biomechanics seminar, ISSN 1100-2247, 1993; 7:157-62.
25. Klasson B. Computer-aided design, computer-aided manufacture and other computer aids in prosthetics and orthotics. Pros Orth Int 1985;9:3-11.
26. Brussel A. CAD/CAM in prothetik und orthetik--stand der technik. (CAD/CAM in Prosthetics and Orthotics--state of the art). Orthopädie Technik 1991; 7:487-95.
27. Fernie GR, Griggs G, Bartlett S, Lunau K. Shape sensing for computer-aided below-knee prosthetic socket design. Pros and Orth Int 1985;9:12-6.
28. Öberg K, Kofman J, Karlsson A, Lindstrom B, Sigblad G. The CAPOD® system--a Scandinavian CAD/CAM system for prosthetic sockets. JPO 1989;1:139-48.
29. Engsberg JR, Clynch GS, Lee AG, Allan JS, Harder JA. A CAD/CAM method for custom below-knee sockets. Pros Orth Int 1992;16:183-8.
30. Lilja M, Öberg T. Volumetric determinations with CAD/ CAM in prosthetics and orthotics: errors of measurement. J Rehab Res Dev 1995;32:141-8.
31. Kolb WM. Curve fitting for programmable calculators. IMTEC, P.O. Box 1402, Bowie, MD 20716.
32. Persson BM, Liedberg E. A clinical standard of stump measurement and classification in lower-limb amputees. Pros Orth Int 1983;7:17-24.
33. Armitage P, Berry G. Statistical methods in medical research. 2nd ed. Oxford: Blackwell, 1987.
34. Ebskov LB. Lower-limb amputation for vascular insufficiency. Int J Rehab Res 1991;14:59-64.
35. Eneroth M, Persson BM. Amputation for occlusive arterial disease. A prospective multicentre study of 177 amputees. Int Orthop 1992;16:383-7.
36. Alaranta H, Alaranta R, Pohjolainen T, Kärkkäinen M. Lower-limb amputees in Southern Finland. Pros Orth Int 1995;19:155-8.
37. Eriksson U, James U. Roentgenological study of certain stump-socket relationship in above-knee amputees with special regard to tissue proportions, socket fit and attachment stability. Upsala J Med Sci 1973;78:203-14.
38. Lilja M, Johansson T, Öberg T. Movement of the tibial end in a PTB prosthesis socket: a sagittal X-ray study of the PTB prosthesis. Pros Orth Int 1993;17:21-6.