---

Retrospective analysis of the prevalence of bilateral amputations in lower limbs
(Portuguese PDF version)

Cézar Ferreira Leite,¹ Airton Delduque Frankini,² Eduardo B. DeDavid,¹ João Haffner³

1. Vascular Surgeon. Resident Physician, Vascular Surgery Service, Hospital Nossa Senhora da Conceição - HNSC (2000-2002), Porto Alegre, RS, Brazil.
2. Ph.D. Professor of the Discipline of Vascular Surgery, Fundação Faculdade Federal de Ciências Médicas de Porto Alegre. Head of the Vascular Surgery Service, HNSC (1988-2002), Porto Alegre, RS, Brazil.
3. Fellow of the Vascular Surgery Service, HNSC (2001), Porto Alegre, RS, Brazil.

Correspondence:
Airton Delduque Frankini
Rua Quintino Bocaiúva, 1290/502
CEP 90440-050 - Porto Alegre, RS
Brazil
Phone: +55 (51) 3222.2716
E-mail: frankini@terra.com.br

---

ABSTRACT

Objective: To identify the prevalence of bilateral amputations in lower limbs in a service of Vascular Surgery, as well as to identify their main features associated.

Patients and Method: Retrospectively, records of 288 patients who underwent major amputations in lower limbs performed by the Vascular Surgery Service at the Hospital Nossa Senhora da Conceição, from January/2000 to April/2001, were reviewed. Two hundred and eighty eight amputees were divided into two groups: 225 (78.1%) with unilateral amputation (group I) and 63 (21.9%) with bilateral amputation (group II). Of these, 24 were performed at the same period of our study, an overall figure of 312 major amputations. Data detailed by age, sex, associated diseases, clinical complaints (rest pain, trophic lesions or diabetic foot complications), type and level of amputation and mortality rate were collected. The chi-square (Χ²) test was used for the statistical analysis, and significant P value was considered to be < 0.05.

Results: Among the 312 amputations, 238 (76.3%) were above-knee and 74 (23.7%) were below-knee, with below-knee/above-knee ratio of 0.31. There was no statistical difference between the groups I and II with relation to age, associated diseases, type of amputation and mortality. Trophic lesion was the most common complaint in both groups (68.4%). The rate of global primary amputation was 72.9% and the presence of diabetic foot was more common in group I (17.3%) than in group II (1.6%) (P < 0.05). Among the causes of amputation in group II, we had 62 patients (98.4%) with critical ischemia and just one with diabetic foot. In group II, we had 49 amputations (77.8%) of bilateral thigh; seven (11.1%) thigh-leg and seven (11.1%) bilateral leg. The non-diabetic patients showed a higher rate of bilateral thigh amputations (96.4%; P < 0.05). The rates of mortality and primary amputation, in bilateral amputations within an interval of 30 days, were 50% and 100%, respectively (P < 0.05).

Conclusions: The prevalence of biamputees was 20.2% in our service. The mortality rate was 50% when both amputations were performed in an interval less than 30 days. We believe this is an important prognostic factor in the development of patients with bilateral amputation.

Key-words: amputation, arteriosclerosis, diabetes mellitus, postoperative complications.
Palavras-chave: amputação, aterosclerose, diabetes melito, complicações pós-operatórias.

J Vasc Br 2004;3(3):206-13

---

Peripheral revascularization is the most effective method of managing critical limb ischemia, although it does not interfere in the natural development of the atherosclerotic disease. Some known risk factors for the peripheral obstructive arterial disease development are: advanced age, smoking and diabetes mellitus,^1-3 and it is well known that gangrene incidence is higher among diabetic subjects.^4,5 Gangrene leads to the majority of amputation procedures, even in patent bypasses.⁵ On the other hand, the observation of patients treated for critical ischemia revealed that 35% undergo amputation, 20% die and 45% remain alive and are not amputated.⁶

The success of peripheral revascularization procedures and the possibility of performing more distal surgeries^7,8 has been reducing the number of amputation procedures.^9,10 Unfortunately, it is not always possible to salvage the patient's limb, and the extremity is lost. Besides, there is a special group of patients who is not frequently mentioned in the literature, mainly in Brazil, although there is an increasing number of them in public hospitals: the bilateral amputees. Concerned with these patients, we developed the present study, with the goal of finding the real prevalence of major bilateral amputations of lower limbs in a reference center in Rio Grande do Sul state, as well as identifying its most important features associated.

PATIENTS AND METHODS

Records of patients submitted to major amputations (transfemoral and transtibial) in the Vascular Surgery Service of HNSC, from 1 January 2000 to 30 Abril 2001, were retrospectively reviewed. Data were obtained from the database of the Service of Medical Records and Statistics (Serviço de Arquivo Médico e Estatística) of the hospital. Procedures of transfemoral and transtibial amputation which were carried out in the period established for the study were selected. Inclusion criteria considered were: patients whose the major indications for amputation were, according to the literature:^11-14 1) chronic arterial obstructive disease with critical ischemia (rest pain and/or trophic lesion); 2) diabetic foot, neuropathy and infected lesion in the lower extremity; and 3) patients with acute arterial lesion and unsalvageable limb. Exclusion criteria were: 1) amputation resultant from trauma, tumor and osteomyelitis, and 2) leg or thigh stump reamputation.

Firstly, the patients were distributed into two groups: group I (unilateral amputees) and group II (bilateral amputees). We defined bilateral amputees as patients who had undergone amputation of a second member within the period of study, regardless of when the first amputation was performed.

Data collected for analysis were age, presence of associated diseases (systemic arterial hypertension; diabetes mellitus; chronic renal insufficiency; stroke; severe ischemic cardiopathy with acute myocardial infarction; and smoking), type of amputation (primary or secondary) and level of amputation (leg or thigh).

We tried to identify in the clinical records of the inpatient period, if there were rest pain and/or trophic lesion in patients with severe chronic ischemia, what the extension of the infectious process was in patients with diabetic neuropathy and if the foot with acute arterial ischemia was unsalvageable. Moreover, we checked the types of previous revascularization surgeries carried out and its condition at the moment of the secondary amputation (patent, thrombosed or with infection). The level of amputation was determined according to the patient's clinical conditions and to the extension of the arterial obstructive disease, which is assessed through a physical examination and an angiographic study.

The statistical analysis was performed by using the chi-square test (Χ²). The difference was considered statistically significant when P < 0.05 (confidence interval of 95%).

RESULTS

Between January 1, 2000, and April 30, 2001, 1,606 surgical procedures were carried out in the Service of Vascular Surgery of HNSC, among them, 342 were major amputations (21.3%). Nineteen amputations resultant from trauma, tumor and osteomyelitis, and 11 reamputations were excluded, performing a total of 312 amputations in 288 patients, once 24 were submitted to bilateral amputation within the period set for this study. Two hundred and thirty-eight amputations were above knee (AKA), or transfemoral (76.3%), and 74 below knee (BKA), or transtibial (23.7%). One hundred and seventy-seven patients were male (61.5%) and 111 were female (38.5%).

Regarding the distribution within groups, group I had 225 patients with unilateral amputation (78.1%) and group II had 63 patients with bilateral amputation (21.9%) (see Table 1). The mean age was almost the same in both groups, and there was not a statistically significant difference when the main associated diseases were compared (diabetes mellitus, systemic arterial hypertension, smoking, stroke, acute myocardial infarction and nephropathy). In both groups, the rate of primary amputation was high (over 70%), and the mortality rate within the 30 postoperative days was 22.7% for group I and 25.4% for group II. There were not statistically significant differences between the two groups, once they showed similar characteristics.

Table 1 - Comparison between unilateral amputees (group I) and bilateral amputees (group II) as for associated diseases and type of amputation

Variable	Unilateral (I) (n = 225)	Bilateral (II) (n = 63)	P
Mean age	65 years	64 years
Diabetes mellitus	136 (60.4%)	35 (55.5%)	0.58
SAH *	108 (48%)	39 (61.9%)	0.07
Smoking	91 (40.4%)	28 (44.4%)	0.67
Stroke	50 (22.2%)	15 (23.8%)	0.92
MI ‡	23 (10.2%)	9 (14.3%)	0.49
IRC §	43 (19.1%)	7 (11.1%)	0.19
Primary amputation	163 (72.4%)	47 (74.6%)	0.85
Secondary amputation	62 (27.5%)	16 (25.4%)	0.85
Death	51 (22.7%)	16 (25.4%)	0.77

* Systemic arterial hypertension; ‡ Myocardial infarction; § Chronic renal insufficiency.

The most frequent indication for amputation in both groups was the trophic lesion in 197/288 patients (68.4%), characterized by gangrene or ulcer in toe and/or forefoot, with or without infection. Rest pain was also frequent and was similarly distributed between groups, with no statistical significance when incidences were compared. The indication for amputation because of infection in the diabetic foot was most common in unilateral amputations, 39 cases (17.3%); different from what happened to bilateral amputation, with only one case (1.6%) and statistical significance of P = 0.002* when groups were compared (see Table 2).

Table 2 - Clinical signs determinant of amputation in unilateral (group I) and bilateral (group II) procedures

Clinical complaint	Unilateral (I) (n = 225)	Bilateral (II) (n = 63)	P
Trophic lesion	149 (66.2%)	48 (76.2%)	0.17
Rest pain	37 (16.4%)	14 (22.2%)	0.38
Diabetic foot	39 (17.3%)	01 (1.6%)	0.002 *
Gangrene	65 (28.9%)	24 (38.1%)	0.21
Gangrene + infection	62 (27.5%)	18 (28.6%)	1
Ulcer	16 (7.1%)	04 (6.3%)	1
Ulcer + infection	06 (2.7%)	02 (3.2%)	0.68

* Statistical significance.

In group II (bilateral amputees), the mean age was 64 years (42-91 years). Among causes for amputation there were: 62 patients with critical ischemia and one patient with diabetic foot and neuropathy (with the presence of distal pulse). Among the 48 patients with trophic lesion in group II (Table 2), 31 (64.5%) entered the hospital with trophic lesion in the foot, and 17 (35.5%) had lesions that were already involving the leg. The rate of primary amputation was 74.6%; and 25.4% were previously submitted to revascularization (Table 1). Previous procedures carried out in these 16 patients (25.4%) were four aorto-bifemoral bypasses, three iliac-femoral bypasses, one profundoplasty, six femoro-popliteal bypasses and two femoro-distal bypasses. At the moment of the secondary amputation, seven bypasses were patent (43.7%) and nine were thrombosed (56.3%). Three cases presented infection in the prosthesis (18.7%). On the other hand, five patients underwent simultaneous amputations (7.9%). In six amputations (9.5%), there was infection in the immediate post-operative stump and stump ischemia in three cases (4.8%).

Regarding the level of amputation in group II, there were 49 cases (77.8%) of bilateral thigh, seven thigh and contralateral leg amputations (11.1%) and seven bilateral leg amputations (11.1%). Non-diabetic patients had a higher rate of thigh bilateral amputation (27/28 - 96.4%). In the group of diabetic patients (22/34) the rate was 64.7% (P = 0.003*). Table 3 shows the higher prevalence of diabetic bilateral amputees within 1 and 3 years: 12/34 patients (35.3%) compared to 3/28 (10.7%) non-diabetic patients, with P equals 0.05.

Table 3 - Comparison between diabetic and non-diabetic patients submitted to bilateral amputation (group II)

Variable	Diabetic (n = 34*)	Non-diabetic (n = 28)	P
Rest pain	5 (14.7%)	9 (32.1%)	0.16
Trophic lesion	29 (85.3%)	19 (67.9%)	0.27
Lesion + infection	15 (44.1%)	5 (17.9%)	0.06
Foot lesion	20/29 (69.0%)	11/19 (57.9%)	0.63
Leg lesion	9/29 (31%)	8/19 (42.1%)	0.63
Primary amputation	26 (76.5%)	20 (71.4%)	0.97
Secondary amputation	8 (23.5%)	8 (28.6%)	0.82
Bilateral thigh	22 (64.7%)	27 (96.4%)	0.003 *
Thigh-leg	7 (20.6%)	0	-
Bilateral leg	5 (14.7%)	1 (3.6%)	0.21
< 30 days	8 (23.5%)	10 (35.7%)	0.39
30 days - 1 year	5 (14.7%)	8 (28.6%)	0.28
1 - 3 years	12 (35.3%)	3 (10.7%)	0.05
> 3 years	6 (17.6%)	5 (17.9%)	1
Indeterminate	3 (8.8%)	2 (7.1%)	1
Death	9 (26.5%)	7 (25%)	0.82

* Diabetic foot excluded for comparison reasons.

There was a higher number of amputations and a death rate of 50% (P = 0.04*) in the group of patients who underwent bilateral amputations in less than 30 days (see Table 4). Although risk factors as acute myocardial infarction, stroke and nephropathy had a higher incidence in these patients, the main causes of death were respiratory problems (12.5%) and sepsis (12.5%). The sub-group with an interval of 30 days to 1 year between amputations (13 patients) had a higher rate of secondary amputations (six patients or 46.1%), resultant from a higher rate of bypass thrombosis (five patients or 38.5%; P = 0.04*). Among bilateral amputees, 31 patients (49.2%) underwent contralateral amputation before 1 year, and 18 of them (28.6%) before 30 days (Table 4).

Table 4 - Interval between bilateral amputations (group II)

Variable	< 30 days (n = 18)%	30 d - 1 year (n = 13) %	1 - 3 years (n = 15) %	> 3 years (n = 11) %	Indeterminate (n = 6) %	Total (n = 63) %	P
Primary amputation	13 (72.2)	7 (53.8)	12 (80)	9 (81.8)	6 (100)	47(74.6)	0.37
Secondary Amputation	5 (27.8)	6 (46.1)	3 (20)	2 (18.2)	0	16 (25.4)	0.37
Patent bypass	2 (11.1)	1 (7.7)	2 (13.3)	2 (18.2%	0	7 (11.1)	0.88
Occluded derivation	3 (16.7)	5 (38.5)	1 (6.7)	0	0	9 (14.3)	0.04*
Bilateral thigh	17 (94.4)	10 (76.9)	10 (66.7)	8 (72.7)	4 (66.7)	49 (77.8)	0.23
Thigh-leg	0	2 (15.4)	2 (13.3)	2 (18.2)	1 (16.7)	7 (11.1)	0.35
Bilateral leg	1 (5.5)	1 (7.7)	3 (20)	1 (9.1)	1 (16.7)	7 (11.1)	0.56
Death	9 (50)	1 (7.7)	3 (20)	2 (18.2)	1 (16.7)	16 (25.4)	0.04*

In an analysis of living and death patients within the period of 30 days, four from the nine who died had a previous myocardial infarction (44.4%), what did not happen with living patients. Besides, all patients who died had been submitted to primary amputation, while only four living patients (44.4%) had been submitted to primary amputation (P = 0.02*) (see Table 5).

Table 5 - Comparison between death and living patients with an interval of bilateral amputation (group II) < 30 days

Variable	Dead (n = 9)	Living (n = 9)	Total (n = 18)	P
Rest pain	1 (11.1%)	0	1 (5.5%)	1
Trophic lesion	8 (88.9%)	9 (100%)	17 (94.4%)	1
Lesion + infection	4 (44.4%)	1 (11.1%)	5 (27.8%)	0.29
Foot lesion	5 (55.5%)	4 (44.4%)	9 (50%)	1
Leg lesion	3 (33.3%)	5 (55.5%)	8 (44.4%)	0.63
Primary amputation	9 (100%)	4 (44.4%)	13 (72.2%)	0.02*
Secondary amputation	0	5 (55.5%)	5 (27.8%)	0.02*

For several reasons, none of the bilateral amputees were able to use prosthesis.

DISCUSSION

When a limb is revascularized, the surgeon has always in mind the goal of salvaging it, however, this may not be always achieved. Besides, many extremities undergo amputation because of the late search for a medical service, what characterizes the primary amputations. In the present review, we have approached only amputation procedures related to the peripheral obstructive disease, acute ischemia and diabetes mellitus, focusing specially on bilateral amputees.

The prevalence of bilateral amputees in our service, in the period of 16 months, was 20.2%, a total of 63 in 312 amputations carried out due to chronic or acute arterial obstructive disease, and infection in the diabetic foot. In a recent national publication, Spichler et al.¹⁵ found that, of 4,673 patients amputated due to peripheral obstructive arterial disease or diabetes mellitus complications, 3.1% underwent bilateral amputations. The literature suggests that the range of contralateral limb loss varies from 15 to 33% in 5 years.^16,17 Dawson et al.¹⁸ say that the progression of the arterial occlusive disease leads to amputation of a contralateral limb at a rate of about 10% a year. During our study, 63 patients were bilateral amputees, of these, 31 underwent major contralateral amputation within 1 year, a 9.9% rate. Inderbitzi et al.¹⁹ noticed that 25% of second limb amputations were carried out in the first year of follow-up, 50% within the second, and 75% within the third year.

Around two thirds of patients who entered our study were diabetic (171/288). This fact has not significantly altered both the bilateral amputation rate (55.5 versus 44.4%) and the mortality rate in comparison with patients without diabetes (26.5 versus 25.0%) (see Table 3). Table 2 shows another fact worthy of notice: in group II there was only one patient with palpable distal pulse, whilst in group I, 39 patients (17.3%) were in that situation. The patient in group II underwent bilateral amputation because of diabetic foot.

This may be attributed to the fact that the patient had already lost a limb and, as there was not an atherosclerotic obstructive component as worsening factor, the patient was more careful with the foot, taking preventive measures and consequently reducing the probability of loosing the only remaining limb. Malone et al.¹⁶ performed a randomized prospective study, by assessing diabetic patients with ulcers, infection, previous amputation and high risk lesions. Two hundred and three patients were divided into two groups: "educated" and "non-educated" about preventive measures with the diabetic foot. The outcomes did not show any difference as for medical approach, risk factors and incidence of infection. However, the presence of ulcers and amputation was three times higher in the group which had not received adequate information as for basic care with the injured foot: 26/177 versus 8/177 for ulceration (P < 0.05) and 21/177 versus 7/177 for amputation (P < 0.05).¹⁶

Another review of literature shows that the ratio between BKA/AKA varies from 1 to 2.1.^20,21 In our study, the ratio between BKA(74)/AKA(238) was quite reduced (0.31), with 76.3% prevalence of AKA. This may be because of the severe situation of patients at the moment of surgery: most of them were in an advanced age, with high surgical risk, low perspective of healing, sometimes confined to bed. In these cases, the surgeon wants to achieve primary healing, with low stump complications. This picture is also reflected in the high mortality rate recorded for patients submitted to bilateral amputation (25.4%), as well as in the primary amputation rate (74.6%) (Table 4). In these patients, if death causes would be analyzed, we would see the high incidence of respiratory problems (12.7%), such as acute lung edema and pulmonary thromboembolic disease, as well as sepsis, which most of times is also associated with respiratory complications (atelectasy, pneumonia). All these facts show, partly, the severe situation of patients who died.

Although there was not statistical significance, there was a trend of patients who died within an interval of contralateral amputation under 30 days to be diabetic (5/9 or 55.5%) and to have trophic lesions with infection (4/9 or 44.4%) (See table 5). These complications are significantly higher in patients with AKA, as observed by Huston et al.²² The incidence of pneumonia and sepsis reached 60% of cases; pulmonary and cardiovascular complications are, beyond doubt, the biggest problems for elder patients submitted to major amputations.

There is an everlasting debate about the effect of previous revascularization with relation to the amputation level. Crouch et al.²³ report a significant increase of AKA after graft failure. Other authors believe that the failure predisposes the patient to a higher amputation level.^24,25 The present study has not showed any statistically significant correlation between previous revascularization and amputation level. In the group of bilateral amputees (group II), 39 of 47 primary amputations (82.9%), and 14 of 16 secondary amputations (87.5%) were accomplished above knee (Table 4).

With no doubt, the presence of a trophic lesion means a bad prognosis, especially when associated to diabetes mellitus and infection (Table 3), in which the chances for amputation are much higher.¹⁹ In the present study, 31/48 patients (64.6%) had trophic lesion limited to foot, and if only the group of diabetic patients is considered, these numbers raise to 20/29 patients or 69%.

A study with more than 200 patients showed that 75% of patients submitted to amputation had lesions that first appeared in the foot.²⁶ Although it was not a prospective study, the fact that there were trophic lesions, either by gangrene or foot ulcer, seems to be associated with a bad prognosis, more than the rest pain (Table 2). That is the importance of a careful assessment on the functionality of the foot on walking, because the patient should no be submitted to multiple vascular procedures (aggressive vascular reconstructions and myocutaneous flap), without the expected success and unavoidably ending in a higher amputation of the limb.

In our opinion, one of the most important findings in this study was the relation of the interval between amputations and the mortality rate. It can be noticed that half of deaths of bilateral amputees were in an interval less than 30 days between amputations (Table 4). However, a higher percentage of myocardium infarct, stroke and nephropathy should be considered in this group. Comorbidities, such as diabetes mellitus, cardiorespiratory diseases and strokes are strongly related with worst prognostic factors.¹⁵,²⁷ As it is not proved that an interval of less than 30 days between the two amputation procedures in one patient corresponds to a higher risk of death (50% in the present review study), we understand that the accomplishment of prospective studies analyzing the importance of this factor in the prognostic assessment of the bilateral amputees would be required.

As a conclusion, we could say that the features of bilateral and unilateral amputee patients are similar (Table 1), except in cases of amputation resultant from the infection of a diabetic foot with neuropathy. These predominate in the unilateral group (Table 2). The trophic lesion has shown to be the most frequent clinic complaint after the loss of the inferior limb, in both groups (Table 2). Diabetes mellitus has not shown any difference as comorbidity that has some influence on the bilateral amputation (Table 1) or the mortality rate (Table 3). However, there was a higher incidence of death in bilateral amputations in a period less than 30 days (Table 4). In this subgroup we also noticed an association between diabetes mellitus and the presence of trophic lesion in the extremity. In our review, we found a group of patients whose mortality rate reaches 50% when the interval of the contralateral limb loss is inferior than 30 days (Table 4). We believe this is an important prognostic factor in the development of patients with bilateral amputation.

REFERENCES

1. Hiatt WR. Pharmacologic therapy for peripheral arterial disease and claudication. J Vasc Surg 2002;36:1283-91.

2. Pentecost MJ, Criqui MH, Dorros G, et al. Guidelines for peripheral percutaneous transluminal angioplasty of the abdominal aorta and lower extremity vessels. J Vasc Interv Radiol 2003;14(9 Pt 2):S495-515.

3. Burns P, Gough S, Bradbury AW. Management of peripheral arterial disease in primary care. BMJ 2003;326:584-8.

4. Kannel WB. Risk factors for atherosclerotic cardiovascular outcomes in different arterial territories. J Cardiovasc Risk 1994;1:333-9.

5. Frankini AD, Ramos LA, Pezzella MV, Medina M, Belaus A. Revascularização infra-inguinal em isquemia crítica nos diabéticos e não diabéticos: análise de 215 casos. Cir Vasc Angiol 1999;15 (Supl 3):S66.

6. TransAtlantic Inter-Society Consensus (TASC). Management of peripheral arterial disease (PAD). Section A: Introduction. J Vasc Surg 2000;31:S1-53.

7. Karlstrom L, Bergqvist D. Effects of vascular surgery on amputation rates and mortality. Eur J Vasc Endovasc Surg 1997;14:273-83.

8. DHSS. Amputation Statistics for England, Wales and Northern Ireland, London: Department of Health and Social Security; 1986.

9. Darling III RC, Chang BB, Paty PSK, Lloyd WE, Leather RP, Shah DM. Choice of peroneal or dorsalis pedis artery bypass for limb salvage. Am J Surg 1995;170:109-12.

10. Frankini AD, Pezzella MV. Revascularização no pé em paciente com isquemia crítica. J Vasc Br 2002;1:193-200.

11. Rutherford RB, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;4:80-94.

12. Rutherford RB, Baker JD, Ernst C, et al. Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg 1997;26:517-38.

13. TransAtlantic Inter-Society Consensus (TASC). Management of peripheral arterial disease (PAD). Section D: Chronic Critical Limb Ischemia. J Vasc Surg 2000; 31(1): S168-S287.

14. TransAtlantic Inter-Society Consensus (TASC). Management of peripheral arterial disease (PAD). Section C: Acute Limb Ischemia. J Vasc Surg 2000; 31(1 Suppl): S135-S67.

15. Spichler D, Miranda Jr F, Spichler ES, Franco LJ. Amputações maiores de membros inferiores por doença arterial e diabetes melito no município do Rio de Janeiro. J Vasc Br 2004;3:111-22.

16. Malone JM, Synder M, Anderson GG, Bernhard VM, Holloway GA, Bunt TJ. Prevention of amputation by diabetic education. Am J Surg 1989;158:520-4.

17. Whitehouse FW, Jurgensen C, Block MA. The later life of diabetic amputee. Another look at fate of the second leg. Diabetes 1968;17:520-1.

18. Dawson I, Keller BP, Brand R, Pesch-Batenburg J, Hajo van Bockel J. Late outcomes of limb loss after failed infrainguinal bypass. J Vasc Surg 1995;21:613-22.

19. Inderbitzi R, Büttiker M, Pfluger D, Nachbur B. The fate of bilateral lower limb amputees in end-stage vascular disease. Eur J Vasc Surg 1992;6(3):321-6.

20. Mooney V, Wagner FW, Waddell J, Ackerson T. The below the knee amputation for vascular disease. J Bone Joint Surg Am 1976;58(3):365-8.

21. Rush DS, Huston CC, Bivins BA, Hyde GL. Operative and late mortality rates of above-knee and below-knee amputation. Am Surg 1981;47:36-9.

22. Huston CC, Bivins BA, Ernst CB, Criffen WO Jr. Morbid implications of above-knee amputations. Report of a series and review of the literature. Arch Surg 1980;115:165-7.

23. Crouch FM, Robicsek F, Hanley EM, Lawhorn RL. Vascular Surgery: Possible adverse effect on extent of subsequent lower limb amputation. South Med J 1992;85(12):1190-2.

24. Epstein SB, Worth MH, Ferzli G. Level of amputation following failed vascular reconstruction for lower limb ischemia. Curr Surg 1989;46:185-92.

25. Bloom RJ, Stevick CA. Amputation level and distal bypass salvage of the limb. Surg Gynecol Obstr 1988;166:01-5.

26. Isakov E, Budoragin N, Shenrav S, Mendelevich I, Korzets A, Susak Z. Anatomic sites of foot lesions resulting in amputation among diabetics and non-diabetics. Am J Physical Med Rehabil 1995;74:130-3.

27. Inderbitzi R, Buettiker M, Enzler M. The long-term mobility and mortality of patients with peripheral arterial disease following bilateral amputation. Eur J Vasc Endovasc Surg 2003;26:59-64.