Amputees and Their Prostheses

Amputees and Their Prostheses

Elizabeth J Davies. M.A. *
Barbara R. Friz, M.S. *
Frank W. Clippinger, M.D. *

Information on 8,698 amputations was collected during a period of approximately two years, ending June 30, 1967. This information was extracted from case-record forms provided by 44 prosthetics facilities in 30 states. The case-record form used was initially developed and standardized by the Conference of Prosthetists of the American Orthotic and Prosthetic Association. Its purpose was to encourage prosthetists in the accurate recording of pertinent information relating to the amputee and his prosthesis. Duplicate copies of the case-record forms were submitted to the Committee on Prosthetic-Orthotic Education (CPOE), National Research Council, in order that significant data could be identified and reported.

"The Facility Case Record Study: A Preliminary Report" and "Children with Amputations", both reporting findings emerging from this study, have been published previously.

Data analyzed in the study included those related to age, sex, level and cause of amputations, reamputations, stump length and contractures, work status of amputees, referrals, months to delivery of prosthesis, age of replaced prosthesis and reason for replacement, components most frequently prescribed for upper- and lower-extremity prostheses, and source of payment for prostheses.

Methods

Each of the 44 facilities submitted case record forms on amputees as they were seen. Three forms were utilized, one for the amputee's medical history, one for the lower-extremity prosthesis, and one for the upper-extremity prosthesis. In cases where the meaning of the data was uncertain, follow-up forms were sent to the prosthetics facilities to clarify or add to the information provided.

A coding system was devised, and information was transferred from the case-record forms to coding sheets and then to IBM cards and magnetic tape. Selection of pertinent data for retrieval was determined by an ad hoc group and the staff of CPOE.

In order to make comparisons between different areas of the country, the states represented in the study were arbitrarily grouped into five geographical regions Fig. 1.

Subjects

The study included 8,323 amputees with a total of 8,698 amputations. Statistics in this study refer only to patients fitted with a prosthesis; amputees not fitted are not included. Table 1 indicates the types of cases included in the study.

Amputees or amputations being fitted for the first time were considered "new" cases. Amputees or amputations being fitted with replacement prostheses were considered to be "old" cases. There was a total of 4,034 "new" amputations and 4,664 "old" amputations Table 2. Amputations in males accounted for 6,848 amputations, and amputations in females, 1,850-a ratio of 3.7:1.

Findings

Aage of Amputees

Table 3 shows the age of amputees fitted in prosthetics facilities during the two years covered by this study. The incidence of amputations for males peaked in the fifth decade; for females, the peak was reached in the seventh decade. Forty-eight per cent of the amputees were 51 years of age or older, 30 per cent were over 61 years, and 12 per cent were over 71 years. The fact that 23 per cent of the amputees were fitted with either a new or a replacement prosthesis after 65 years of age has Medicare implications. (It should be noted that Medicare was in effect during only the second year of data collection.)

Level of Amputations

Amputations of the lower extremity accounted for 86 per cent of the total number of amputations Table 4. Of these, 53 per cent were at the below-knee level. In the upper extremity, 57 per cent of the amputations were at the below-elbow level.

There was no significant difference in the incidence of left- and right- side amputation in either the upper or lower extremities. A total of 4,386 left-limb and 4,312 right-limb amputations was reported. The right upper extremity was involved slightly more than the left, 605 to 573, and the left lower extremity fractionally more than the right, 3,813 to 3,707.

Cause of Amputation

Causes of amputation were considered in four categories: congenital, tumor, trauma, and disease. Cases of infection, gangrene, or osteomyelitis resulting from trauma were classified under "trauma." Cases of trauma associated with vascular disease were classified under "disease."

Causes of amputation were analyzed by age group and level. Of the 8,698 amputations reported in this study, the cause was known for 8,487 cases; both cause and age were known for 8,394 cases. Fifty per cent of all amputations were caused by trauma, 37.3 per cent by disease, 8.4 per cent were of congenital origin, and 4.3 per cent were due to tumor. Table 5 shows the relative incidence of amputation by cause and level.

In Fig. 2 the total number of amputations by cause of amputation and age is indicated. Amputees most frequently fitted or returning for replacement in the first ten years of life were those with congenital limb deficiencies. Amputations for trauma led all other categories fitted or returning for replacement between the ages of 11 through 50. In the third, fourth, and fifth decades, this group accounted for 76 per cent, 82 per cent, and 72 per cent, respectively, of all cases fitted or returning. Of those fitted in the sixth decade of life, the incidence was almost equally distributed between traumatic amputations and amputations due to disease. After age 60, the latter group led all other categories by a ratio of more than 2:1.

"New" Cases by Cause

Analysis of all amputations entered in the study gives an overview of the type of amputee being seen and fitted in prosthetics facilities, as reported above. Analysis of those being fitted for the first time, however, provides a picture of persons amputated during the two-year period of data collection and gives a better current indication of cause related to age, sex, and level of amputation.

It is probable that the statistics on age are slightly distorted, since age was reported as of the time of fitting. Age at the time of amputation, therefore, would be less, and to a variable degree.

In the group of "new" amputees, cause was reported for 3,963 cases, and both cause and age for 3,920. Fig. 3 indicates the incidence of amputation by age. Of the "new" cases, 60.2 per cent of amputations were caused by disease, 29.1 per cent by trauma, 5.9 per cent by tumor, and 4.8 per cent were of congenital origin.

The predominance of trauma as the cause of amputation in the overall amputee population of the study Fig. 2 is in striking contrast to the predominance of disease as a cause of amputation when only new patients are considered Fig. 3. In the overall picture, the ratio of trauma to disease is 1.3:1, whereas in new patients the ratio is reversed, and disease as a cause of amputation outnumbers trauma 2:1.

Thus, the total sample data obviously includes a considerable number of traumatic amputees who lost their limbs at an earlier age and survived to require replacement prostheses. However, the noteworthy finding is that, in the period surveyed, disease-caused amputations were occurring at double the rate of those attributable to trauma.

Congenital. In the 191 reported n males, 86 in females Table 6. Of this number, 137 did not require amputation surgery, while 54 did. This surgery presumably involved the conversion of anomalous limbs to stumps that were more suitable for the fitting of a prosthesis. Eighty-three amputations occurred in the lower extremity, of which 44 were at the below-knee level. Of 108 upper-extremity amputations, 78 were at the below-elbow level. Thirty-two per cent of congenital amputations were not fitted until after 11 years of age.

Tumor. Of 235 "new" amputations caused by tumor, 206 (88 per cent) were of the lower extremity Table 7. There were 120 amputations at the above-knee level, accounting for 58 per cent of the lower-extremity amputations. An additional 27 per cent were at a level higher than above-knee, i.e., hip-disarticulation or hemipelvectomy. Males outnumbered females 130 to 105.

The highest incidence of tumor (66 cases or 29 per cent) occurred in the second decade of life. Within this decade, no particular pattern of incidence is discernible Table 8. These data are somewhat at variance with those reported by Taft and Fishman from a study conducted by the staff of New York University Child Prosthetic Studies. This study, which involved a larger sampling (278 children whose amputations were caused by tumor), showed a gradual increase in incidence beginning about the 6-8 year period and peaking in the 14-16 year group. Unfortunately, the age groupings are slightly different from those of our study, so an exact comparison cannot be made. However, both studies agree that tumor occurs most frequently in the second decade by a wide margin.

Trauma. Of the 1,156 new cases of amputations resulting from trauma, amputations in males accounted for a total of 1,050, and those in females for 106, a ratio of approximately 10:1 Table 9. The highest incidence of trauma-related amputations occurred in the third decade (250 cases), followed closely by that in the fourth decade (216 cases). The number of amputees in these two decades accounted for 41 per cent of all new cases where age was known. The incidence of amputations in females varied only slightly in each decade between the ages of 11 and 60. The incidence of amputations in males exhibited a sharp rise through the second and third decades, and then receded gradually.

In every decade the involvement of the lower extremity exceeded that of the upper. Actually, the lower extremity was involved 1.9 times as often as the upper, 753 times as opposed to 403.

Disease. Sixty per cent (2,381 cases) of all new amputations were caused by disease Fig. 13. Although males outnumbered females by more than 2:1 in this category, the relative percentages of males and females in each age group were closely parallel, e.g., 980 or 61 per cent of males were over the age of 61 years, while 464 or 62 per cent of females were also over the age of 61. After 40 years of age, a sharp rise in the incidence of amputations caused by disease was noticeable. Approximately one-third of the amputations occurred in the seventh decade. Eighty-five per cent of all new amputees in the disease category were over the age of 51 years, and 49 per cent were in the Medicare age group.

In disease-caused "new" amputations, involvement of the lower extremity greatly exceeded that of the upper, the ratio being 73:1.

Comparison with Amputee Census

The Glattly study, reported in 1964 and commonly referred to as the "Amputee Census," included only "new" amputees. It is of interest to compare the findings of that study with the present one. Findings of our study relating to the sex and age of new amputees and the cause, side, and level of amputations closely parallel the findings of the Glattly study. Comparative data of the two studies are depicted in Fig. 4, Fig. 5, Fig. 6, and Fig. 7, and Table 11.

In our study, newly fitted amputees 51 years of age and older accounted for 60.2 per cent of the total, as compared with 58.8 per cent in the Amputee Census Fig. 4. In both studies, the highest incidence of amputation was in the seventh decade. Because many geriatric amputees are not fitted with prostheses, the incidence of amputation in the older age groups would presumably be even higher if statistics on nonfitted amputees were included.

In both studies, male amputees exceeded female amputees by approximately three to one Fig. 5.

The distribution of right- and left-side amputations was almost equal in both studies, and lower-extremity amputations still accounted for about 85 per cent of all new fittings Table 11. In Fig. 6 a higher incidence of below-knee amputations and a lower incidence of above-knee amputations were evident in the more recent study. Among new patients in this study, there was a total of 3,254 above-and below-knee amputations. Of these, 50.9 per cent were above-knee.

The relative incidence of trauma as a cause of amputation decreased by four per cent from the Glattly to the present study, and the incidence by cause in other categories increased, but by relatively small amounts Fig. 7.

Original Level of Amputation for Disease Correlated with Geographical Area and Age

The original level of amputation for disease was examined for 2,242 new cases whose amputations were at either the above- or below-knee level. Comparisons were made between below- and above-knee as the choice of amputation level in each of the five geographical areas Table 12. Below-knee appeared to be the site of choice in less than half the total number of cases. The South led the other geographical areas in percentage of amputations at the below-knee level (54 per cent), followed in order by the Midwest (51 per cent), New England (48 per cent), East Central (46 per cent), and the West (45 per cent).

A look at the site of the original disease-related amputation for new patients 41 years of age and above revealed some interesting statistics Table 13. In the fifth decade, below-knee was selected in preference to above-knee in 58 per cent of the cases. This percentage gradually decreased over the next two decades to a low of 43 per cent in the seventh decade. After the seventh decade, there was an increase to 47 per cent in the eighth decade and to 50 per cent after the eighth decade. For all new amputations for disease in patients 41 years of age and above, above-knee was selected in 52 per cent of the cases, below-knee in 48 per cent.

The lack of a consistent pattern in these data is intriguing. A progressive decrease in the proportion of below-knee amputations with increase in age might logically be anticipated. Surgeons, for example, might wish to be more sure of obtaining healing in older patients and elect to amputate at the above-knee level. However, other factors than age of patient obviously enter into the selection of amputation level.

Specific Causes of Traumatic Amputations

Trauma was listed as the primary or precipitating cause of 4,306 amputations ("old" and "new" cases). As noted earlier, some of this number were classified in categories other than trauma, since trauma was not considered the primary cause of amputation; hence, the number 4,306 exceeds the number of cases actually coded in the trauma category. Of these 4,306 instances where trauma was mentioned, there were 392 cases where the type of trauma was unknown, so, for purposes of this analysis, reference will be to the 3,914 cases where type was known.

Fig. 8 summarizes the causes of traumatic amputations. In this category, men were affected ten times as frequently as women: 3,561 to 353. In males, cars, industrial accidents, and war each accounted for approximately 20 per cent of the cases. On the other hand, automobiles were by far the outstanding cause of traumatic amputations in women (49 per cent), with no other cause approaching this in frequency. It is noteworthy that the ratio of male to female automobile-caused amputations was in the order of 4:1, in contrast to the 10:1 overall ratio. Since it is not known whether these female victims were predominantly drivers or riders, the full significance of these data is not clear.

Table 14 relates cause of trauma to sex, side, and level of amputation. Involvement of the right upper extremity in males was greater than the left. This preponderance was especially evident in farm and industrial accidents and is doubtless related to handedness. In car accidents, the left upper extremity was involved significantly more than the right for both males and females, 62 per cent as compared with 38 per cent. One can speculate that this incidence might be attributable to the fact that many motorists ride with the left elbow extending beyond an open window. In the small sample of train accidents, the involvement of the left upper extremity in males was also considerably greater than the right but, because of the small number, this probably was without significance.

The left lower limb was involved slightly more than the right in males, and the right and left limbs almost equally in females.

Table 15 compares causes cited for "new" traumatic amputations in males with those given for "old" traumatic amputations. Twenty-six per cent of the amputations of "old" cases were due to war injuries, whereas only 2 per cent of the new cases were due to this cause. At the time of this study, the Vietnam War had not yet exerted its full impact. The greatest increase in trauma-caused amputations was seen in the industrial-accident category. Industrial accidents caused 29 per cent of the "new" traumatic amputations, but only 15 per cent of the "old" amputations. Elimination of war cases from the total number avoids distortion of the data due to the preponderance of old war injuries, and thus presents a somewhat truer comparative picture of other traumatic causes. With war injuries eliminated, industrial accidents accounted for 29 per cent of the "new" amputations and 20 per cent of the "old" amputations, which still reflects an increased incidence of amputations caused by industrial accidents. Industrial accidents exceeded all other categories as the cause of amputation in new patients.

Reamputations of the Lower Extremity

Reamputations were studied in relation to cause, original level of amputation, and present level. Level was reported for 396 reamputations of the lower extremity. Some members of this group had second reamputations, but for the purposes of this study, only the original and present level of amputation were considered. An attempt was made to exclude simple revisions that involved no shortening of bone.

In reviewing the figures presented here, it should be remembered, again, that only those patients fitted with prostheses at the time of the study are considered. Despite this limitation, analysis of the available data is thought-provoking. Of 396 reamputations reported, 189 were in the disease-related category involving a total of 3,122 cases Table 16, and 182 were in the trauma-caused group with 3,387 total cases Table 17. Thus, reamputations in the first group ran a shade over 6 per cent, those in the second group a shade under 6 per cent. Stated in reverse, approximately 94 per cent of the cases in both groups did not require re-amputation. The statistics for specific levels are also quite fascinating. In disease-related below-knee amputations, approximately 6 per cent required reampu-tation versus approximately 5 per cent in the like trauma group. In the above-knee group, the comparative proportions are 1 per cent versus 0.6 per cent. At the Syme's level, comparative figures are 25 per cent versus 28 per cent, and for partial feet 96 per cent versus 25 per cent. The reasons for the sharp increase in reampu-tations at the last two levels are worthy of further study. It would also be of interest to know whether partial foot amputations, for example, were or were not successfully performed on many patients who were never fitted with prostheses.

For the 189 (48 per cent) reamputations due to disease, Table 16 gives the final as compared to the original level. Of 93 below-knee amputations requiring ream-putation, 22 (24 per cent) remained in the same segment, 67 (72 per cent) were converted to an above-knee level, 3 to a knee-disarticulation, and 1 to a hip-disarticula-tion level. Of the 15 original above-knee amputations, 9 were reamputated in the same segment and 6 became hip disarticulations.

Of the 11 Syme's reamputations reported, 2 were reamputated to an above-knee level and 9 to a below-knee level. Of the 67 reamputations at the partial foot level, 22 were converted to an above-knee, 41 to below-knee, and 4 to a Syme's level.

Causes of reamputation for patients in the disease category were indicated for 181 of the 189 reamputations. In some instances, two causes of reamputation were cited. In each instance where a cause was mentioned, it was counted as contributing to the reamputation. The total number of contributing causes to reamputation in the disease category therefore was 192 Table 18. "Recurrence of the original cause of amputation" accounted for almost half (48 per cent) of the reasons cited for reamputations. This generalized response is interpreted as meaning a continuance of the original vascular problem responsible for the initial amputation. Specific causes cited were a nonhealing wound (18 per cent), gangrene (12 per cent), infection (5 per cent) stump breakdown (3 per cent), and "other" (14 per cent).

Most reamputations in the disease category occurred very shortly after the original surgery, 49 per cent occurring in less than 1 1/2 months, and 60 per cent occurring in less than 2 1/2 months. Eighty-two per cent occurred in the first year following the amputation.

In the category of traumatic amputations, levels for 182 reamputations of the lower extremity were reported. Of the 114 amputations at the below-knee level requiring reamputation, 57 per cent (65 amputations) remained at the below-knee level, a percentage considerably higher than was the case for reamputations due to disease. Forty-five amputations were converted to above-knee levels and 4 were converted to knee disarticulations. There were 29 Syme's reamputations, of which 23 were converted to below-knee, 3 to above-knee, and 3 remained at the Syme's level. Of the 22 partial foot reamputations, 14 were converted to below-knee levels, 7 to Syme's and 1 to above-knee.

Causes of reamputation were known for 157 of the trauma cases. As with reamputations in the disease category, every instance where a cause was mentioned was counted. There were 165 contributing causes to reamputations Table 19. In 71 instances (43 per cent), "other" was coded as the cause of reamputation. Included in the "other" category were causes that could not be readily classified, such as "stump not satisfactory for prosthesis," "shorten bone and remove neuroma," "painful stump." The median number of months between amputation and reamputation was six.

There were 16 reamputations for congenital amputees and 6 for patients whose amputations were caused by tumor. Three of the latter were reamputated because of recurrence of the tumor. Reported reasons for reamputations in congenital amputees were too diverse for classification, except that 4 reamputations were because of bony overgrowth.

Table 20 summarizes the total number of reamputations for each level and includes the percentage of reamputations converted to a higher segment or remaining in the same segment.

Bony overgrowth was cited eight times as a reason for reamputation: four tibial overgrowths, two fibular overgrowths, and two not specified. All of these reamputa-tions were performed on children, with the exception of one on a 27-year-old amputee. While not implicit in the data, it is conceivable that this 27-year-old had had bony overgrowth for a long time prior to reamputation (his first amputation occurred at age 10).

Stump Length and Contractures

There were 2,602 above-knee amputations for which the presence or absence of contractures of the hip was reported. Of this group, 1,345 had either no flexion contracture or a contracture of less than 5 deg, and are not included in this analysis, other than the notation that they comprised over half of the group reported. Stumps with 5+ deg of contracture ranged in length from 2 - 2 1/2 inches to 14 - 15 1/2 inches. Three stumps had flexion contractures of more than 60 deg. Hip-flexion contractures were greatest in the very short stump. The average contracture at the above-knee level fell in the 5-9 deg range.

There were 3,781 below-knee amputations for which the presence or absence of knee contractures was reported. Of this number, only 12 per cent were reported as having contractures of 5 deg or more. In general, the shorter the stump, the more severe the contracture. Considering only those cases reporting contractures of 5 deg or more, stumps averaging more than 7 1/2 in. in length had average contractures of between 5 and 9 deg; for stumps between 4 and 7 1/2 in. long, contractures averaged between 10 and 14 deg; and for stumps 3 1/2 in. and less in length, contractures averaged 15 to 19 deg. The average contracture, excluding those of less than 5 deg, was 10-14 deg. Three stumps had contractures of 60 deg or more.

Work Status

The work status of "old" male amputees between the ages of 21 and 64, with 2,694 amputations, was reported. "New" amputees were not studied, since the majority of the group had not yet had time to return to employment. Eighty-four per cent of the "old" amputees in the cited age group were employed, the highest employment rate (89 per cent) occurring in the 41- to 50-year-old age group Fig. 9. In each of the age groups studied, a higher rate of employment was reported for upper-extremity than for lower-extremity amputees. It should be noted here that only 6.4 per cent of amputees between the ages of 21 and 64 were reported as not being gainfully employed. The remainder of the group (9.3 per cent) were students, retired, or fell into some other category. This percentage of unemployment is a little higher than that reported for the national average for the years 1965, 1966, and 1967 (4.5, 3.8, and 3.8 per cent respectively).

The rate of employment in relation to each upper- and lower-extremity amputation level appears in Fig. 10 and Fig. 11.

Work status was reported for 383 female amputees between the ages of 21 and 64. Of this number, 200 were housewives, 148 were gainfully employed, and only 18 were not gainfully employed. Seventeen had either retired or reported their work status in some other category.

Referrals

The majority (58 per cent) of cases fitted at prosthetics facilities were referred by amputee clinics; 26 per cent were referred by physicians; 16 per cent were not referred. Of the "new" cases, 5 per cent were not referred to prosthetics facilities by either a clinic or physician, as contrasted to the 26 per cent of the "old" cases not so referred.

Months to Delivery of Prostheses

For "new" amputations, the time from amputation (or from birth for congenital amputees not requiring surgery) to date of delivery of the prosthesis was analyzed by level and cause for the five geographical regions Table 21. The median period to delivery for all prostheses was 6 months. Comparing geographical areas, the median was 5 months for New England, the Midwest and West, 6 months for the South, and 7 months for the East Central region. Of the 3,588 prostheses with times to delivery reported, 71 were delivered in 1 month or less, 67 were not delivered for 99 months or longer. Thirty-seven of the latter were for congenital amputations not requiring surgery, i.e., 37 children were not fitted with their first

prosthesis until after the age of eight years, three months. A comparison of time to delivery by levels indicated that the median time lapse was 5 months for the below-knee prosthesis and 6 months for all other levels. Time to delivery of prostheses ranged from a median of 4 months for below-knee prostheses in the New England area and the West to a median of 10 months for below-elbow prostheses in the East Central region. These data will provide a basis for later comparisons in areas where programs of immediate and early prosthetic fitting have been instituted.

Data on months to delivery were analyzed by cause of amputation and related to geographical regions Table 22. The shortest median length of time for delivery was 3 months for congenital amputees who had had surgery. The longest time was for congenital amputations without surgery, where the median was 31 to 36 months; however, it should be recognized here that this median also represents the median age of congenital amputees not requiring surgery who were being fitted for the first time. Median time to delivery for amputations caused by tumor was 4 months; by trauma, 5 months; and by disease, 6 months.

Age of Replaced Prostheses and Reasons for Replacement

The average age of replaced prostheses for all patients was 6.1 years. For children up to 21 years of age, it was 2.5 years, and for adults, 6.7 years.

Comparisons of the ages of replaced prostheses for above- and below-elbow and above- and below-knee amputees in relation to the age of the patient (by decade) are shown in Table 23. In almost every instance, the "life" of the prosthesis increased with the age of the patient. The average life of above-elbow prostheses for 124 amputations was 9.2 years. The range was from 2.5 years for the child through the age of 10 years to 16.7 years for amputees over the age of 61. The average age of below-elbow prostheses for 349 amputations was 6.5 years, ranging from 2.5 years for the child through age 10, to 10.3 years for amputees over age 51. The average age of above-knee prostheses for 1,269 amputations was 6.2 years, with a range from 2.2 years for the child in the first decade, to 8.1 years for amputees over age 71. The below-knee prosthesis had the shortest life, averaging 5.8 years for 2,201 amputations, and ranging from an average of 1.7 years for the child through age 10, to 8.6 years for amputees over 71 years of age.

In comparing ages of replaced prostheses by cause of amputation and the sex of the amputee, it is found that prostheses for congenital amputees had the shortest life, averaging 3.5 years, and prostheses for traumatic amputees had the longest life, averaging 6.8 years Table 24. The growth rate of children in the congenital group undoubtedly accounts for the more frequent replacements of prostheses evident here. Replacement of prostheses for patients in the disease category occurred, on average, every 5 years, and there was very little difference between replacements for males and females. The life of prostheses for tumor patients also averaged 5 years; however, prostheses for males in this category needed more frequent replacement, lasting 4.5 years as compared with an average 5.6 years for females.

It is interesting to note that the age of replaced prostheses for males averaged 6.2 years, and that of females 5.4 years. The large number of males in the trauma category may account for this difference, inasmuch as the average life of prostheses in this category is longer than in others.

Table 25 indicates the reason for replacement of prostheses. The majority of prostheses were replaced because they were worn out. "Worn out" was listed as the sole or contributing cause of replacing a prosthesis in 58 per cent of the cases. It was the leading reason for replacing prostheses of persons whose amputations were caused by tumor (50 per cent), trauma (67 per cent), and disease (44 per cent). As would be expected, the primary reason for replacing prostheses of congenital amputees was that the prosthesis was "outgrown." In 52 per cent of replacements for congenital amputees, the prosthesis was outgrown; in 33 per cent of the cases it was worn out.

"Unsatisfactory" was cited as the reason for replacement in four per cent of the cases. However, it should be noted that although the "unsatisfactory" category was meant to include only those cases in which problems arose relating to fabrication or patient tolerance, it was often cited for other reasons which rendered the prosthesis unsatisfactory. Had this item been interpreted correctly, the percentage undoubtedly would have been lower.

The average age of all "worn out" prostheses that were replaced was 7.6 years Table 26. This exceeds the average age of prostheses replaced for any reason (6.1 years) by a year and a half. This higher age undoubtedly reflects the longer life of the prostheses of traumatic amputees reported above, since "worn out" was the sole or contributing factor for 67 per cent of the replacements in the trauma category. Additionally, the lower average age of all the replaced prostheses was affected by the inclusion of children's prostheses, which had shorter lives.

Ccomponents for Upper-Extremity Prostheses

The components most frequently used for upper-extremity prostheses at the above- and below-elbow levels are depicted in Fig. 12a,Fig. 12b. The voluntary-opening hook was used with 87 per cent (201 instances) of the above-elbow prostheses and 90 per cent (517 instances) of below-elbow prostheses. The preference for this type of hook was reflected in all areas except the West, which showed a preference for the voluntary-closing hook with below-elbow prostheses. New England was the only area that did not prescribe the voluntary-closing hook at all.

The hand-type terminal device was utilized to a limited extent, being prescribed 309 times as opposed to the hook-type device which was prescribed 806 times. Many amputees for whom hooks were prescribed were also equipped with hands. Where hand-type devices were reported, the voluntary opening hand was prescribed for above-elbow prostheses 40 per cent of the time (36 cases) and for below-elbow prostheses 36 per cent of the time (79 cases). Both the East Central and Midwest areas preferred voluntary-closing hands for use with above-elbow prostheses. The East Central and Western areas preferred voluntary-closing hands for below-elbow prostheses. New England showed a preference for the passive hand with the below-elbow prosthesis.

The simple friction wrist unit was overwhelmingly preferred to quick-change types in all geographical areas, being used with 83 per cent of above-elbow and 85 per cent of below-elbow prostheses.

Although the triceps pad was used with 56 per cent of the below-elbow prostheses, its use ranged from 35 per cent in the South to 94 per cent in the New England area. The South preferred the half cuff. Plastic laminate was the cuff material of choice in 61 per cent of the total cases, although the East Central and Western areas preferred leather to the extent of 54 per cent and 55 per cent respectively.

The double-wall socket was used in 89 per cent of the above-elbow and 77 per cent of the below-elbow prostheses. Pre-flexed sockets, some of which also had double walls, were used in 11 per cent of the below-elbow prostheses. Sixty-one per cent of the preflexed sockets were utilized by children.

In 98 per cent of the upper-extremity prostheses, the sockets were made of plastic.

The elbow unit with internal lock was the item of choice for above-elbow prostheses in all geographical areas, being used in 78 per cent of all fittings. Seventeen per cent of all elbow units had spring-flexion assists. Sixty-four per cent of the elbow hinges used in below-elbow prostheses were flexible, the range being from 44 per cent in the West to 92 per cent in New England. The Midwest showed almost equal preference for the single-pivot (47 per cent) and the flexible hinge (50 per cent).

Dual-control systems were used in 80 per cent of above-elbow and single control in 96 per cent of the below-elbow prostheses.

Eighty-three per cent of the harnesses for above-elbow prostheses were of the figure-eight type, the majority of this group (55 per cent) being equipped with the Northwestern University harness ring. The East Central area and the West showed a preference for the figure-eight harness without the ring. Of the 14 cases with reported type of harness in the West, none used the ring with the figure-eight. The South used the ring to the greatest extent for above-elbow prostheses.

Ninety-two per cent of the below-elbow harness were of the figure-eight type, 59 per cent of these being equipped with rings. The East Central, South, and Midwest areas showed greatest preference for the ring figure-eight harness; the New England and Western areas used the figure-eight harness without the ring almost as often as with it.

Components for Lower-Extremity Prostheses

Components most frequently used for above- and below-knee prostheses appear in Fig. 13a,Fig. 13b. The various geographical areas showed more consistency in prescription of lower-extremity than upper-extremity components. In most instances, only the percentage varied, not the type of component.

The SACH foot was prescribed for 55 per cent of the above-knee and 73 per cent of the below-knee prostheses. In area comparisons, the South showed the greatest usage of the SACH foot, and the Midwest the lowest. For the above-knee prosthesis, prescription of the SACH foot rose from 76 per cent in the first to 83 per cent in the second decade, and then gradually declined with advancing amputee age. In the below-knee group, the SACH foot was prescribed 96 per cent of the time for children under 10 years of age; the percentage declined steadily to a low of 56 per cent in the eighth decade, then rose to 63 per cent for the group of amputees 81 years of age and over.

Wood was used as the shank material in 95 per cent of the above-knee and in 90 per cent of the below-knee prostheses.

The most frequently used knee component for above-knee prostheses was the single axis, with friction being used in 74 per cent of the fittings. Twelve per cent of the knees were single axis with manual locks. Eight per cent of the knees were hydraulic, with the West showing the greatest preference (17 per cent) and the Midwest the least (4 per cent). In instances where metal joints were reported for below-knee prostheses, the lap joint was specified in 48 per cent of the cases and the clevis joint in 22 per cent. The type of joint was not specified in 30 per cent of the cases.

For above-knee amputees, the quadrilateral socket was used in 85 per cent of the prostheses. It was the overwhelming choice in each of the geographical areas. The socket of choice for below-knee amputations was the patellar-tendon-bearing. Preference for this socket averaged 58 per cent, the South and West showing greatest utilization, 79 per cent and 82 per cent respectively, and the New England and Midwest areas the least utilization, 44 per cent and 47 per cent respectively.

Wood was used most often for above-knee sockets, averaging 57 per cent, although the South showed a preference for plastic, using it for 55 per cent of all sockets. Below-knee sockets were most often (55 per cent) fabricated in plastic. New England showed a preference for leather sockets, and the Midwest preferred wood (41 per cent) to either plastic or leather.

The pelvic belt was the preferred method of suspension (56 per cent) for above-knee prostheses. Only in the West did the use of suction, either alone or in combination with other suspension, exceed the use of the pelvic belt. In correlating methods of suspension with age, it was noteworthy that during the second, third, and fourth decades, suction alone was preferred to all other types of suspension. In all other decades, the pelvic belt was preferred.

In considering types of suspension reported for all below-knee prostheses, the knee cuff alone was the choice of suspension in 36 per cent of the cases. It was least used in the Midwest (22 per cent). The South and West utilized the knee cuff alone most frequently (55 per cent). When type of suspension for the patellar-tendon-bearing prosthesis is analyzed by age group, it is found that, while the knee cuff alone was used for 62 per cent of all

the prostheses, greatest usage occurred in the second decade (73 per cent) and next greatest in the third decade (71 per cent). Least use of the knee cuff alone occurred in the very young child (48 per cent), but the inclusion of cases where a waist belt was used in conjunction with the knee cuff raised this percentage to 68.

Sources of Payment

Table 27, Table 28, and Table 29 indicate the sources of payment for prostheses. More than one source was sometimes listed, in which case they are reported under "combinations of the above "or" "other". Medicare had been in operation only one year prior to the conclusion of this study and presumably would rank considerably higher as a source of payment at the present time. As mentioned earlier, over 23 per cent of the amputees in this study were in the Medicare age bracket.

Source of payment was given for 8,631 prostheses Table 27. The greatest contributors to defraying the costs of prostheses were State Bureaus of Vocational Rehabilitation (22.5 per cent) and the patient himself (22.8 per cent). Next in order were the Veterans Administration (14.3 per cent), welfare (10.8 per cent) and insurance (9.9 per cent).

The Children's Bureau paid for 46.5 per cent of the prostheses for children up to the age of 21. Through the wage-earning years, 21 to 64, State Bureaus of Vocational Rehabilitation paid for 31.9 per cent of the prostheses, the amputee for 24.3 per cent, and the Veterans Administration for 19.3 per cent. During the retirement years, 65 and over, the amputee alone paid for 29.9 per cent of the prostheses, Social Security and Medicare for 19.5 per cent, and welfare for 15.3 per cent.

A further analysis of sources of payment relating to the wage-earning years yields some interesting facts Table 28. The Veterans Administration paid for 30 per cent of replacement prostheses, but only 10 per cent of new prostheses. This statistic doubtless reflects the continuing supply of prostheses to veterans of World War II and the Korean War and a decreased number of fresh cases. More "new" male amputees were supported by insurance or compensation than "old" male amputees, 24 per cent as opposed to 8 per cent. This may reflect the policy of some insurance companies to pay for the first prosthesis only. On the other hand, it may indicate an increase in opportunity for insuring oneself against disability and a greater awareness of the values of health insurance. In comparing source of payment for males and females in this age group, one notices the higher level of support by the amputees themselves and the Bureaus of Vocational Rehabilitation for the female group, and also the very low percentage of females supported by insurance or compensation.

In correlating source of support with occupation, only "old" amputees were considered, since in most instances "new" amputees had not yet returned to work at the time the data forms were submitted. Amputees were studied in three categories: those gainfully employed, those not gainfully employed, and those who were students, housewives, or retired Table 29.

Of the 3,055 "old" cases included above, only 187, or 6 per cent, were reported as not being gainfully employed. The Bureaus of Vocational Rehabilitation paid for 35 per cent of the prostheses for the gainfully employed group, the Veterans Administration for 28 per cent, and the amputee for 25 per cent. For the group of amputees not gainfully employed, the Bureaus of Vocational Rehabilitation were the source of payment for 28 per cent of the prostheses, the Veterans Administration for 27 per cent, and welfare for 24 per cent. In the 468 amputations of students, housewives, or retired amputees, 31 per cent of the prostheses were paid for by the amputee, 28 per cent by the Bureaus of Vocational Rehabilitation, and 17 per cent by the Veterans Administration.

Discussion

In recent years, there has been increasing interest in defining the characteristics of the amputee population, and also in providing amputees with functional stumps and prostheses. Much progress has been made in understanding the amputee and his problems, and in the fabrication of improved prosthetic components. This study has sought to document some of the characteristics of the amputee and his prosthesis during a particular period in time-the approximately two years ending June 30, 1967.

Certain characteristics of amputees, namely sex and age, and the cause, side, and site of amputation, were well established in Glattly's study of 12,000 new amputees for whom data were collected over a two-year period, ending in 1963. In the present study of over 8,000 amputees, 4,034 of whom were new, data were likewise collected over a two-year period which ended in 1967, four years later. Unless some catastrophic event had occurred immediately before or during either of the two periods, it would be expected that in large samples such as these, the sex and age of the amputee and side and cause of the amputation would be relatively constant. Such was indeed the case, indicating that the sample in the latest study was a valid cross-section of the amputee population. As noted before, neither the Medicare Act nor the conflict in Vietnam had exerted a significant impact on this study. Although medical advances over a number of years have been largely responsible for the increasing age of the amputee, with a resulting shift from trauma to disease as a predominant cause of amputation, such changes would not be expected to exert a significant difference in as short a period as four years.

In amputations caused by disease, the site of amputation can be influenced by medical judgment at a particular time. In the vast majority of cases where amputation is categorized as disease, the amputees had vascular insufficiency. For this condition, amputation at a level above the knee had been widely advocated for many years because it was felt that this procedure facilitated healing. It has been found, however, that amputation may be performed at a below-knee level, with primary healing occurring in the majority of cases. By preserving the knee joint, amputation at this level greatly enhances the rehabilitation potential of the patient.

Burgess has reported that most below-knee amputations for ischemia heal primarily, and with proper prosthetic care do not break down. Lim reports that 92 per cent of below-knee amputations were successful when a popliteal pulse was present, and 75 per cent were successful when pulse was absent. He also reports a lower mortality rate for below-knee amputees, 16 per cent as opposed to 35 per cent for above-knee amputations. Tracy cites a 90 per cent successful healing rate for below-knee amputations for ischemic gangrene.

Although the increase in the percentage of below-knee amputations in our study, as compared with the Glattly study, is relatively small in view of the potential increase, it is nevertheless an encouraging trend, and it is to be hoped that a dramatic increase will be reflected in future surveys as the results of ongoing educational programs take effect.

Although the incidence of amputations due to trauma appears to have declined, as far as percentage of the total amputee population is concerned, this does not necessarily imply a decrease in the overall incidence of traumatic amputations. Actually, the increasing age of amputees, with its corollary of increasing incidence of amputations due to disease, is certainly partly responsible for the decline in percentage of trauma cases. In the younger age groups, trauma continues as the major cause of amputations. The Public Health Service report published in 1964 shows that "absence of major extremity," classified as an accident "while at work," occurred almost three times as often as amputation caused by "moving motor vehicles." In the present study, the ratio was closer to 1:1 than 3:1, i.e., moving vehicles as a cause of traumatic amputations was almost equal to that of industrial accident. A higher percentage of auto accidents than industrial accidents occurred in the female group, a pattern which is typical of other reported findings. These results may indicate improved safety controls in industry, or may underscore the soaring rate of automobile accidents, or both. The large number of amputations resulting from trauma continues to have strong implication for improved accident-prevention programs and more effective human-factors engineering. The need for greater safety of design, particularly in cars and industry, continues to be great.

It is of interest to note that prosthetic prescription varied among the geographical areas, some areas having a greater tendency than others to incorporate newer prosthetic techniques. It might be expected that the latest prosthetic developments would be incorporated into prosthetic practice in those areas which were near the prosthetic-orthotic educational centers (New York, Chicago, and Los Angeles) or in areas of greatest concentration of prosthetic facilities (California, Pennsylvania, New York, and Illinois), or amputee clinics (New York, Pennsylvania, California, and Texas). With the exception of the West, where newer developments were used in a high percentage of cases, there appeared to be no relationship between the nature of prosthetic services provided and the factors cited above. Both the South and the West showed a more consistent use of newer techniques than did the other areas.

The provision of prosthetic services reported in the study indicates that much improvement is to be desired as far as length of time for delivery of the prosthesis is concerned. The time between the date of amputation (or reamputation) and delivery of the prosthesis was inordinately long, ranging from a median of four months for patients whose amputations were caused by tumor to six months for patients with vascular disease. The provision of temporary prostheses and immediate postsurgical fitting of prostheses would help shorten this time lag.

The finding that a relatively high percentage of congenital amputees (32 per cent) were not fitted until after their eleventh birthday is distressing. Since current philosophy is to fit congenital amputees at a very early age, it would be interesting to know the reason for this reported delay. Whether the fault lies with amputee clinics, or with parents who are either reluctant to take their children to clinics or are ignorant of the prosthetic opportunities available to them, is not evident from the present analysis. The implication is that more needs to be done at the educational level. The growth and implementation of dynamic treatment programs would surely result in a much more optimistic picture.

A composite picture of amputees reported in this study would present the following profile:

1. The congenital amputee seen in prosthetic facilities was a male under 10 years of age with involvement at the below-knee level.
2. The amputee whose amputation was caused by tumor was a male between 11 and 20 years of age whose amputation was at the above-knee level.
3. The traumatic amputee was a male now between the ages of 41 and 50 years who had received his amputation between the ages of 21 and 30 years. His amputation was at the below-knee level and was most likely received as a result of a car accident, industrial accident, or war injury.
4. The amputee whose amputation was caused by disease was also a male, between the ages of 61 and 70 years, who was amputated during these same years. His amputation was as likely to be at the above-knee level as at the below-knee level.

Summary

1. This study, which extended over a two-year period ending in June 1967, presents data on 8,323 amputees with 8,698 amputations, all of whom were fitted with prostheses.
2. Of the "new" amputations seen in prosthetic facilities, 60 per cent were caused by disease, 29 per cent by trauma, 6 per cent by tumor, and 5 per cent were of congenital origin.
3. Of all amputations, "new" and "old," being fitted in prosthetic facilities, 50 per cent were caused by trauma, 37.3 per cent were caused by disease, 8.4 per cent were of congenital origin, and 4.3 per cent were caused by tumor.
4. The greatest incidence of disease-caused amputations occurred in the seventh decade, those of trauma in the third decade, and those of tumor in the second decade.
5. Males outnumbered females in every category, the ratio for "new" amputations of males to females being approximately 2:1 for disease, 10:1 for trauma, and 1.2:1 for both congenital causes and tumor.
6. Eighty-six per cent of the total number of amputations were of the lower extremity, with 53 per cent of this group being at the below-knee level.
7. Although automobile accidents were cited as the single greatest cause of all traumatic amputations, war injuries, industrial accidents, and automobile accidents were cited almost equally for male amputees.
8. Forty-eight per cent of all reampu-tations were in the disease category, 60 per cent of these occurring within two and one-half months of the original amputation. The reamputation rate for below-knee amputations caused by disease was not significantly higher than that for trauma-caused amputations-approximately 6 per cent in both instances.
9. Degree of contracture reported at both hip and knee varied inversely with the length of the stump. Excluding contractures of less than 5 deg, the average hip flexion contracture for above-knee amputations was in the 5-9 deg range; the average knee flexion contracture for be-low-knee amputations fell in the 10-14 deg range. Fifty-two per cent of those cases reporting presence or absence of contractures had either no contracture or one of less than 5 deg.
10. Unemployment rate for "old" male amputees between the ages of 21 and 64 was 6.4 per cent, slightly higher than the national average for the years covered by the report.
11. Fifty-eight per cent of patients were referred to prosthetic facilities by amputee clinics, 26 per cent by physicians, and 16 per cent were not referred.
12. The median time from amputation to delivery of a prosthesis was six months, the below-knee prosthesis being delivered in the shortest length of time. Congenital amputees who required surgery received prostheses in a median time of three months postsurgery. Patients in the disease category waited the longest time- six months.
13. Prostheses had an average life of 6.1 years, with the life of the prosthesis increasing with the age of the patient. Below-knee prostheses generally and prostheses for congenital amputees had the shortest life. Prostheses for males lasted longer than those for females. "Worn out" was the primary reason given for replacing a prosthesis.
14. Prosthetic prescription varied in the geographical areas, some regions demonstrating a greater tendency than others to incorporate newer prosthetic techniques. Generally, as the age of the amputee advanced, there was a tendency to use the older types of components, e.g., pelvic hands, articulated ankles.
15. The Children's Bureau was the largest single source of financial support for the purchase of prostheses for children, and the State Bureaus of Vocational Rehabilitation provided the greatest financial support for amputees during the wage-earning years. The Veterans Administration paid for a high percentage of prostheses for males who were in the "old" category. In all, the federal government paid entirely for 48 per cent of all prostheses and provided partial support for another 3 per cent.

Acknowledgments

Grateful appreciation is extended to the 44 facility owners and their staffs who provided the data on which this study is based.

References:

1. Burgess, Ernest M., The below-knee amputation, Bull. Pros. Res., 10-9:19-25, Spring 1968.
2. Davies, E. J., B. R. Friz, and F. W. Clippinger, Jr., Children with amputations, Inter-Clinic Inform. Bull., 9:3:6-19, December 1969.
3. Friz, Barbara R., and Frank W. Clippinger, Jr., The facility case record study: a preliminary report, Orth. and Pros., 23:1:8-17, March 1969.
4. Glattly, H. W., A statistical study of 12,000 new amputees, Southern Med. J., 57:1373-1378, November 1964,
5. Lim, R. C, Jr., et al.. Below-knee amputation for ischemic gangrene, Surg. Gynec. Obstet., 125: 493-501, September 1967.
6. Sarmiento, A., and W. D. Warren, A re-evaluation of lower extremity amputations, Surg. Gynec. Obstet., 129:799-802, October 1969.
7. Taft, C. B., and S. Fishman, Survival and prosthetic fitting of children amputated for malignancy, Inter-Clinic Inform. Bull., 5:5:9-28, February 1966.
8. Tracy, G. D., Below-knee amputation for ischemic gangrene, Pacif. Med. Surg., 74:251-253, September-October 1966.
9. U. S. Department of Health, Education, and Welfare, Public Health Service, Impairments due to injury by class and type of accident, United States, July 1959-June 1961, Washington, D.C., 1964.