Local biological factors that influence amputations in diabetic patients.

Abstract: Lower extremity amputation is one of the worst complications of diabetes, as it usually has a life expectancy that is below that of most cancers. Four local biologic factors--ulcer, ischemia, neuropathy, and infection--have been shown to be related to amputation. These factors interact with one another, such that neuropathy has been shown to cause ulcers and ischemia and to prevent the healing of ulcers. In addition, ischemia and neuropathy are independent risk factors for infection. More coordinated efforts are needed to create better grading schemes and therapeutic protocols.

Key Words: diabetic amputations, ischemia, neuropathy, ulceration, infection

**********

Leg amputation is a dreaded complication of diabetes. Amputation rates per 100,000 per year vary from 43.9 for Native Americans to 2.8 for citizens of Spain, with diabetes being associated with 25 to 90% of all amputations. (1) For above-ankle amputations, the cost has been estimated at $66,215 per patient, of which 77% is postprocedural. (2) Some 6 to 30% of amputees undergo contra-lateral amputation within one to three years of the initial procedure. (3) Of greatest concern is mortality: a recent study gave diabetic patients a 69.4% one-year and a 30.9% five-year survival, significantly worse than that provided for nondiabetic patients. (4) This review explores local biologic factors that impact the incidence of diabetic amputations. Unless otherwise stated, diabetes references adult-onset diabetes. Individual factors will be outlined, after which their interactions will be discussed. Therapeutic interventions, such as appropriate footwear, systemic risk factors (such as hemoglobin A1C levels), and socioeconomic status will not be specifically addressed. In assessing individual factors, unadjusted odds ratios are presented: the explanation of multiple regression analyses to explain the import of individual factors, while of value, would triple the length of the initial discourse and duplicate analyses of interactions of these factors provided in the last section. Although references are provided as proof of a connection, it should be recognized that there is great variation from population to population, a fact best seen by examining a recent article from Africa that shows extensive variation in the prevalence of disorders related to the diabetic foot. (5) P values are not displayed; results, unless otherwise stated, were found to be significant at P < 0.05, with correction by Bonferroni's method for multiple assessments whenever univariate odds ratios were disclosed. The Table lists all odds ratios discussed in this paper.

Factors in Isolation

Ulceration

Foot ulcers have long been accepted as the most important antecedent to amputation. Of 80 diabetic amputations, (6) fifty were preceded by the sequence 1) minor trauma, 2) ulceration, and 3) wound healing failure. Among a superbly studied group of Caribbean patients, 125 of 205 who underwent amputation versus only 2 of 194 who had not undergone amputation had had an ulcer; ulcers increased the amputation risk 150.0 times. (7)

Among diabetic persons, foot ulcers are common. The Behavioral Risk Factor Surveillance System found 11.8% of diabetic adults had had a foot ulcer. Those who were neither married nor cohabitating with someone and those with longer disease duration more often had ulcers. Older patients and African-Americans less often had foot ulcers. (8) By the same token, once an ulcer has developed, the evidence is not strong that diabetes imposes an immense risk of an amputation beyond that of ischemia: in a study of patients with ischemic related ulcers, 29 of 119 patients with diabetes and 8 of 50 patients without diabetes developed an amputation; the odds ratio of 1.7 is neither extraordinary nor statistically significant. (9)

Two toe conditions increase ulcer risk. First, toe deformities can induce ulcers: in one study of 63 diabetic patients, callus yielded a relative risk of ulcer of 11.0; for prior ulcer, the relative risk was 56.8. (10) Second, diseased toenails often ulcerate skin: although the incidence of complications among those with toenail aberrations has not been determined, Rich and Hare (11) clearly document the association of onychomycosis, ulceration, and cellulitis in diabetic patients.

The importance of pressure cannot be overstated. The above study of callus and ulcer also showed a direct relationship of pressure and ulcer. (10) Height is one of the best correlates of pressure upon the feet; a population based study in Canada demonstrated that, for every 10 cm of added height, the risk of amputation increased by a factor of 1.8. (12) This is important because it helps show that offloading is of critical importance in the healing of ulcers. (13)

Peripheral Artery Disease

Diabetics face a much greater risk of peripheral artery disease (PAD) than do nondiabetics. The 1999 to 2000 National Health and Nutrition Examination (NHNE) survey evaluated 2,873 persons over 40 years old, 419 of whom had diabetes. The index used for PAD was the ankle brachial index (ABI), the ratio of the systolic blood pressure in the ankles to that in the right arm; an ABI <0.9 was deemed evidence of PAD. Whereas 9.5% of those with diabetes had PAD, only 4.0% of those without diabetes did. (14)

The influence PAD has upon amputation in diabetes is easily shown. The study of Carribeans (7) found a low ABI in 71 of 166 diabetic patients who had had amputations versus only 23 of 142 diabetic patients who had not undergone amputation; the odds ratio of 3.9 increased to 17.9 when other PAD related measures were included. Of 209 Swiss amputees, almost half had received a diagnosis of diabetes; of 229 admissions for amputation, only 13 (5.7%) were not for arterial conditions, including three for trauma, four for tumor, and three for osteomyelitis complicating an open fracture. (15) All 32 patients said to have undergone amputation principally for diabetic foot ulcer had certain (29/32) or likely (3/32) severe PAD.

Neuropathy

Peripheral neuropathy (PN) commonly afflicts diabetic patients. There is an array of quantitative means to assess neuropathy, but two are most commonly used. One may test the ability to recognize the presence of a 10 g Semmes-Weinstein nylon monofilament being bent against the foot. (14) Alternatively, one might, at greater expense of time, but also with better accuracy, use an electronic tuning fork to determine the vibration perception threshold (VPT), the lowest voltage at which vibration can be detected. (16) The former test is recommended because of its simplicity and speed; it should be part of the routine assessment of diabetic patients by physicians in general.

The NHNE, using the monofilament, found that, whereas 28.5% of diabetic adults over 40-years-old had PN, only 13.3% of nondiabetic adults did. (14) Whereas 111 of 118 Caribbean diabetic patients who had undergone amputation had monofilament insensitivity, only 64 in 193 diabetic patients without amputations did, yielding an odds ratio of 31.9. (7) Among Brazilian diabetic patients, the manner of whose evaluation for neuropathy was not specifically delineated, 97 of 107 patients with lower extremity amputations versus 138 of 232 patients without amputation, had neuropathy, yielding an odds ratio of 6.0. (17) The association of neuropathy and amputation is quite clear, but will be better established when the association of neuropathy and ulceration is explored, as it is the principle mechanism by which the pathogenesis of amputation is said to expresses itself.

Infection

Understanding of infection in isolation as a factor in diabetic amputations is hampered by three factors. First, it is assumed in virtually all studies that infection complicates ulceration, meaning no modern study can be found that assesses infection independent of ulceration. Second, amputation is often accepted as being indicated when osteomyelitis is present, even though it may not be needed in all cases, (18) meaning assessments of infection as a cause of amputations are in some ways more summaries of practice patterns than they are discussions of biologic antecedents. Third, there exists no uniform consensus on the classification of foot infections. (19)

Notwithstanding the above caveats, the import of infection in regards to amputation can be established. In the Caribbean study of diabetic patients, whereas 54 of 203 amputees had had an infection, only 2 of 194 nonamputees did, yielding an odds ratio of 34.5. (7) An Australian study of 799 amputations in 661 diabetic patients found 621 (78%) were performed for ischemia, 53 (7%) were performed for ischemia and infection, and 121 (15%) were performed for infection alone; of importance, the fraction presenting with sepsis alone increased as time progressed. (20) There is not a good scientific proof of what constitutes infection, but a consensus has developed that infection is present when a wound has purulent secretions or at least two of the following signs: pain, redness, tenderness, induration, warmth, lymphangitis, foul smell, or gas formation. (21) Bacterial culture is of little assistance in discriminating infected from noninfected wounds when osteomyelitis is absent. From each of 60 infected diabetic foot wounds, two samples were taken, one a swab before debridement and the other a deep tissue specimen taken at the end of debridement (22); in 90% of the infections not involving bone (36/40), all organisms present in the deep specimen were seen in the superficial swab that many might have attributed to contamination. That identifying the organism is insufficient is also shown by a study of 84 patients with uninfected and infected foot ulcers with both types of diabetes; there were no differences identified in the proportions of isolates of any particular type of organisms, although more of the infected ulcers with S aureus were methicillin-resistant (MRSA) than the uninfected ulcers. (23) For antibiotic guidance in the case of osteomyelitis, biopsy is often recommended, but a recent study shows that needle biopsy is a very promising technique. (24)

Antibiotic guidance by bacterial culture is important: in one study, MRSA was isolated in 30% of 63 diabetic patients with infected foot ulcers, a fraction that had doubled in just three years. (25) The diagnosis of osteomyelitis itself is extremely complicated and controversial (18); the once regnant probe-to-bone test was recently shown in an assessment of 81 patients with a total of 104 foot ulcers to have a sensitivity of only 38%, a specificity of 91%, a positive predictive value of only 53%, and a negative predictive value of 85%. (26)

Interactions of Factors

Interactions with Ulceration

The loss of sensation produced by neuropathy increases the ulcer risk from minor, everyday trauma to the foot. Two groups of diabetic Australians without active foot lesions were followed for two years (27); 34 of 250 patients with PN, as defined by VPT >30, versus 3 of 22 patients without PN developed ulcers, yielding an odds ratio of 11.5. An interesting article (28) used a complex statistical model to incorporate data from prior studies to estimate the risk of ulceration from PN, defined as VPT >25; for 10,000 Monte Carlo simulations, 4,722 ulcers per year for patients with PN and only 1,547 ulcers per year for patients without PN were estimated. Similar studies led Singh et al (29) to recommend the Semmes-Weinstein monofilament test be administered to all diabetic patients. The article is recommended to the reader, both for the clarity with which it is written and for its frank presentation of the nonuniformity in risk categories and recommendations that exist in regards to the prevention of foot ulcers in patients with diabetes.

PAD is itself well-known to produce ulcers, but its primary demonstrated role in diabetic foot ulcers is to prevent healing and encourage amputation. Among a study of patients with ischemic ulcers, 70% of whom had diabetes, (9) multiple regression showed that, although diabetes and neuropathy were not shown to be related to the subsequent risk of amputation, an ABI of >0.5 was shown to yield a significant decrease in the risk of subsequent amputation; comparing the 12 of 39 patients who had an ABI [less than or equal to]0.5 and underwent amputation at 18 months to the 8 of 51 patients who had an ABI >0.5 who did yield an odds ratio of 2.4. Among patients with diabetic foot ulcers, another study (30) showed that whereas 96 of 120 patients with intact pedal pulses healed, only 127 of 223 with reduced or absent pulses did, yielding an odds ratio of 3.0.

Interactions with Infection

Because ulceration and infection are intertwined, definite statements about the effects of PAD and PN upon infection independent of ulceration are difficult to make. Among 374 Indian diabetic patients with foot infections, who were followed for six months, both PAD, defined as ABI <0.8, and PN, a VPT >25 V, were studied. (31) Whereas 49 of 76 patients with PAD developed a recurrent infection, only 149 of 298 did; the odds ratio of 1.8 is not statistically-significant when all other univariate analyses are taken into account. Whereas 134 of 201 patients with PN developed a recurrent infection, only 64 of 173 without PN did, yielding an odds ratio of 3.4.

On the other hand, more definite results were seen when 68 diabetic patients with infected feet were compared with 44 diabetic patients without infections. (32) Fifty nine of the 68 patients with an infection and only 16 of the 44 patients without infection had PAD, when defined as the absence of a peripheral pulse, yielding an odds ratio of 11.5. Fifty of the 68 patients with an infection and only 12 of the 44 patients without an infection had PN, when defined as a VPT >25 V, yielding an odds ratio of 7.4. The results held after multivariate analysis with a very wide array of covariates.

Conclusions

Ulceration is the best established precursor for an amputation, such that many articles that assess various therapeutic options for diabetic feet give almost the same weight to ulceration that they do to amputation. Peripheral artery disease, well-known as a cause of amputation in its own right, is associated with an increased risk of amputations in diabetic patients and also with an increased likelihood of amputations among patients with foot ulcers. Diabetic neuropathy, by decreasing the sensitivity to trauma, increases the risk of ulceration and amputation; it is of such importance that all persons caring for diabetic patients should routinely assess these persons' ability to detect a standardized 10 g Semmes-Weinstein nylon monofilament bent against their feet. Finally, infection may be said to serve as both a consequence of infection and a cause of amputation, but definition of its role is hampered by the absence of universally-accepted, scientifically-validated grading systems and criteria for therapeutic intervention.

References

1. Global Lower Extremity Amputation Study Group. Epidemiology of lower extremity amputations in centres in Europe, North America, and East Asia. Br J Surg 2000;87:328-337.

2. Apelqvist J, Ragnarson Tennvall G, Persson U, et al. Diabetic foot ulcers in a multi-disciplinary setting: an economic analysis of primary healing and healing with amputation. J Intern Med 1994;235:463-471.

3. Kanade RV, van Deursen RWM, Price P, et al. Risk of planta ulceration in diabetic patients with single-leg amputation. Clin Biomech 2006;21:306-313.

4. Aulivola B, Hile CH, Hamden AD, et al. Major lower extremity amputation: outcome of a modern series. Arch Surg 2004;139:395-399.

5. Abbas ZG, Archibald LK. Epidemiology of the diabetic foot in Africa. Med Sci Monit 2005;11:RA262-RA270.

6. Pecoraro RE, Reiber GE, Burgess EM. Pathways to diabetic limb amputation: basis for prevention. Diabetes Care 1990;13:513-521.

7. Hennis AJ, Fraser HS, Jonnalagadda R, et al. Explanations for the high risk of diabetes-related amputation in a Caribbean population of Black African descent and potential for prevention. Diabetes Care 2004;27:2636-2641.

8. History of foot ulcer among persons with diabetes: United States, 2000-2002. MMWR Morb Mortal Wkly Rep 2003;53:1098-1102.

9. Marston WA, Davies SW, Armstrong B, et al. Natural history of limbs with arterial insufficiency and chronic ulceration treated without revascularization. J Vase Surg 2006;44:108-114.

10. Murray HJ, Young MJ, Hollis S, et al. The association between callus formation, high pressures and neuropathy in diabetic foot ulceration. Diabet Med 1996;13:979-982.

11. Rich P, Hare A. Onychomycosis in a special patient population: focus on the diabetic. Int J Dermatol 1999;38(suppl 2):17-19.

12. Tseng CH. Prevalence of lower-extremity amputation among patients with diabetes mellitus: is height a factor? CMAJ 2006:174:319-323.

13. Wu SC, Crews RT, Armstrong DG. The pivotal role of offloading in the management of neuropathic foot ulceration. Curr Diab Rep 2005;5:423-429.

14. Gregg EW, Sorlie P, Paulose-Ram R, et al. Prevalence of lower-extremity disease in the U.S. adult population [greater than or equal to]40 years of age with and without diabetes. Diabetes Care 2004;27:1591-1597.

15. Carmona GA, Hoffmeyer P, Herrmann FR, et al. Major limb amputations in the elderly observed over ten years: the role of diabetes and peripheral arterial disease. Diabetes Metab 2005;31:449-454.

16. Shearer A, Scuffham P. Gordois A, et al. Predicted costs and outcomes from reduced vibration detection in people with diabetes in the US. Diabetes Care 2003;26:2305-2310.

17. Gamba MA, Gotlieb SL, Bergamaschi DP, et al. Lower extremity amputations in diabetic patients: a case-control study. Rev Saude Publica 2004;38:399-404.

18. Jeffocoate WJ, Lipsky BA. Controversies in diagnosing and managing osteomyelitis of the foot in diabetes. Clin Infect Dis 2004;39:S115-S122.

19. Lipsky BA, Berendt AR, Deery HG, et al. Diagnosis and treatment of diabetic foot infections. Plast Reconstr Surg 2006:117(7 suppl):212S-238S.

20. Adam DJ, Raptis S, Fitridge RA. Trends in the presentation and surgical management of the acute diabetic foot. Eur J Vasc Endovasc Surg 2006;31:151-156.

21. Lipsky BA. A report from the international consensus on diagnosing and treating the infected diabetic foot. Diabetes Metab Res Rev 2004;20(Suppl 1):S68-S77.

22. Slater RA, Lazarovitch T, Boldur I, et al. Swab cultures accurately identify bacterial pathogens in diabetic wounds not involving bone. Diabet Med 2004;21:705-709.

23. Tentolouris N, Petrikkos G, Vallioanou N, et al. Prevalence of methicillin-resistant Staphylococcus aureus in infected and uninfected diabetic-foot ulcers. Clin Microbiol Infect 2006;12:186-189.

24. Kessler L, Piemont Y, Ortega F, et al. Comparison of microbiological results of needle puncture vs superficial swab in infected diabetic foot ulcer with osteromyelitis. Diabet Med 2006;23:99-102.

25. Dang CN, Prasad YD, Boulton AJ, et al. Methicillin-resistant Staphylococcus aureus in the diabetic foot clinic: a worsening problem. Diabet Med 2003;20:159-161.

26. Shone A, Burnside J, Chipchase S, et al. Probing the validity of the probe-to-bone test in the diagnosis of osteomyelitis of the foot in diabetes. Diabetes Care 2006;29:945.

27. McGill M, Molyneaux L, Yue DK. Which diabetic patients should receive podiatry care? Int Med J 2005;35:451-456.

28. Shearer A, Scuffham P, Gordois A, et al. Predicted costs and outcomes from reduced vibration detection in people with diabetes in the US. Diabetes Care 2003;26:2305-2310.

29. Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers in patients with diabetes. JAMA 2005;293:217-228.

30. Pound N, Chipchase S, Treece K, et al. Ulcer-free survival following management of foot ulcers in diabetes. Diabet Med 2005;22:1306-1309.

31. Vijay V, Narasimham DVL, Seena R, et al. Clinical profile of diabetic foot infections in south India: a retrospective study. Diabet Med 2000;17:215-218.

32. Peters EJ, Lavery LA, Armstrong DG. Diabetic lower extremity infection: influence of physical, psychological, and social factors. J Diabetes Complications 2005;19:107-112.

Mitchell S. Wachtel, MD, and Eldo E. Frezza, MD, MBA, FACS 

From the Departments of Pathology and General Surgery, Texas Tech University Health Sciences Center, Lubbock, TX.

Reprint requests to Eldo E. Frezza, MD, MBA, Professor, Chief, Division of General Surgery, Department of Surgery, Texas Tech University Health Sciences Center, Lubbock, TX 79430. Email: [e-mail]

Neither author received financial support to create this article. Neither author has a proprietary or commercial interest in any drug, device, or equipment mentioned in this article.

Accepted October 16, 2006

RELATED ARTICLE: Key Points

* Most studies cite ulceration as the single most important local biologic factor affecting the risk of amputation.

* Neuropathy is the most important cause of ulceration.

* Ischemia is an important risk factor for patients with ulcers to undergo amputations; it is the most important principle cause of amputation in some studies.

* Infection, universally accepted as a consequence of ulceration, is often the direct cause of amputation.

Table. Odds ratios described in this paper

Fraction with ulcer who underwent amputation: fraction         150.0
  without ulcer who underwent amputation (7)
Fraction with ulcer & diabetes who underwent amputation:         1.7 (a)
  fraction with ulcer without diabetes who underwent
  amputation (9)
Fraction with callus who develop ulcers: fraction without       11.0
  callus who developed ulcers (10)
Fraction with prior ulcer who develop second ulcer: fraction    56.8
  without prior ulcer who develop ulcer (10)
Fraction with ulcer & 10 cm of added height who develop          1.8
  ulcer: fraction with ulcer without 10 cm of added height
  who develop ulcer (12)
Fraction with peripheral artery disease (PAD) who receive       17.9
  amputation: fraction without PAD who receive amputation (7)
Fraction with monofilament insensitivity who receive            31.9
  amputation: fraction without monofilament insensitivity who
  receive amputation (7)
Fraction with peripheral neuropathy (PN) who receive             6.0
  amputation: fraction without PN who receive amputation (17)
Fraction with infection who receive amputation: fraction        34.5
  without infection who receive amputation (7)
Fraction with PN who develop ulcer: fraction without PN who     11.5
  develop ulcer (27)
Fraction with ulcer & ankle brachial index (ABI)                 2.4
  [less than or equal to]0.5 who receive amputation:
  fraction with ulcer & ABI >0.5 who receive amputation (9)
Fraction with ulcer & reduced/absent pedal pulses who did not    3.0
  heal: fraction with ulcer & intact pedal pulses who did not
  heal (30)
Fraction with ABI [less than or equal to]0.8 who developed       1.8 (a)
  recurrent infection: fraction with ABI >0.8 who developed
  recurrent infection (31)
Fraction with PN who developed recurrent infection: fraction     3.4
  without PN who developed recurrent infection (31)
Fraction with infection who lacked pedal pulses: fraction       11.5
  without infection who lacked pedal pulses (32)
Fraction with infection who had peripheral neuropathy:           7.4
  fraction without infection who had peripheral
  neuropathy (32)

(a) Not statistically significant.