Validation of an algorithm to predict reulceration in amputation patients with diabetes

International Wound Journal ISSN 1742-481 ORIGINAL ARTICLE Validation of an algorithm to predict reulceration in amputation patients with diabetes Raúl J Molines-Barroso 1, José L Lázaro-Martínez 1, Francisco J Álvaro-Afonso 1, Irene Sanz-Corbalán 1, José L García-Klepzig 2 & Javier Aragón-Sánchez 3 1 Unidad de Pie Diabético, Facultad de Medicina, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 2 Unidad de Medicina Interna, Hospital Clínico San Carlos, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 3 Unidad de Pie Diabético, Hospital La Paloma, Las Palmas de Gran Canaria, Spain Key words Algorithm; Amputation; Disease-free survival; Ulcer Correspondence to RJ Molines-Barroso Unidad de Pie Diabético Universidad Complutense de Madrid Edificio Facultad de Medicina Pabellón 1. Avda Complutense s/n, 284 Madrid Spain E-mail: rauljmolines@hotmail.com doi: 1.1111/iwj.12639 Molines-Barroso RJ, Lázaro-Martínez JL, Álvaro-Afonso FJ, Sanz-Corbalán I, García-Klepzig JL, Aragón-Sánchez J. Validation of an algorithm to predict reulceration in amputation patients with diabetes. Int Wound J 217; 14:523 528 Abstract The aim of this article was to assess the ability to predict reulceration in people with diabetes and a history of minor amputation according to the formula proposed by Miller et al. A retrospective study was performed on 156 consecutive records of patients with a recent history of simple or multiple forefoot amputation. The sample was divided according to Miller s formula into patients at low risk of reulceration and those at high risk; those were further divided into two subgroups according to whether or not the first segment of the forefoot had been amputated. Forty-eight (47 1%) individuals suffered forefoot reulceration, showing a median reulceration-free survival time of 8 months [interquartile range (IR) 3 6 14 8]. Nephropathy (P =.5) and Miller s formula (P =.28) were risk factors for reulceration-free survival time in the univariate analysis. The pattern relating to the first segment amputated [hazard ratio (HR) 2 853; P = 4; 95% confidence interval (CI) 1 391 5 849] and nephropathy (HR 2 468; P =.4; 95% CI 1.328 4.587) showed a significant hazard ratio in the multivariate Cox model. Participants with first segment amputation and one other amputation showed an association with the probability of reulceration in comparison with any other specific type of minor amputation. Introduction Diabetic foot ulcers are common in individuals with diabetes mellitus (DM). One third of these ulcers is complicated by infection and peripheral arterial disease, which may be extremity or life-threatening (1). Complicated ulcers result in the need for hospitalisation, surgical intervention and medical treatment to manage the ulcerations, increasing the severity of foot disease (1,2) and the cost to the patient and the health care system (3). Considering that recurrence of foot ulcers takes place in more than half of patients after 3 years, patients quality of life will be impaired, and thus, the global burden will dramatically increase (4). Transfer lesions syndrome is a common situation that occurs after minor amputation (5). In this circumstance, the therapeutic approach prevails according to the clinical or individual criteria, and multiple amputations may be performed; therefore, standard protocols are necessary because the foot could result in an inability to walk or to fit into footwear. Both single and multiple toe amputation lead to deformities and alter foot function, increasing the risk of development of new ulceration (5,6). Although location and level of amputation have been also related to high probability of reulceration (7 1), until recently (11), there has Key Message transfer lesions are a common situation after minor amputation that may be extremity or life-threatening, increasing the health care cost this is an assessment of reulceration risk using Miller s formula in 156 consecutive records of participants with diabetes and a history of amputation one amputation more than first segment amputation shows the highest risk of reulceration between patients with previous forefoot amputations 216 Medicalhelplines.com Inc and John Wiley & Sons Ltd 523

Validation of an amputation algorithm been no proposal of a simple algorithm that embraces all variables. Miller et al. (11) distinguished between patients who have undergone one or more forefoot amputations, those who had an appropriate foot function and a low likelihood of reulceration and those who will probably develop a new ulcer and will undergo revisional surgery. In the algorithm, single toe or ray amputation is considered a low risk for reulceration. Multiple toe or ray amputations are potentially considered at great risk of revisional surgery in accordance with the following condition: individuals who have undergone any ablative forefoot procedure that exceeded a hallux plus one additional toe resection or a lesser toe plus two additional lesser toe resections; multiple amputations different from these conditions are considered at low risk. To date, we are unaware of any investigation that has validated the framework proposed by Miller; therefore, we proposed the following objective: to assess the ability to predict reulceration according to the formula proposed by Miller in people with diabetes and a history of amputation. Materials and methods We retrospectively analysed data records from 663 consecutive patients who underwent surgery to treat a diabetic foot ulcer from January 28 to December 213. Of those, 156 underwent amputation surgery (Table 1). Patients underwent longitudinal follow-up after surgery in the Diabetic Foot Unit of the Complutense University of Madrid to analyse the transfer of ulcers onto the ipsilateral forefoot. Records were selected when subjects met the following criteria: individuals over 18 years of age, with diagnosis of type 1 or type 2 DM, who underwent amputation on a forefoot and had at least 12 months of reliable longitudinal follow-up information. Records of individuals were excluded because of the following factors: Charcot s osteoarthropathy, rheumatoid arthritis, a non-healing surgical site, severe infections (sign of systemic inflammatory response syndrome) according to the Infectious Diseases Society of America (12), patients with critical limb ischaemia (ankle pressure less than 7mmHg or toe systolic pressure less than 5mmHg) (13) and those who did not have an autonomic gait. Those participants who need revascularisation treatment in the perioperative period to heal the ulcer were excluded. Longitudinal follow-up All procedures were classified as emergent class according to Armstrong and Frykberg s classification of diabetic foot surgery (14). Sensorimotor neuropathy was diagnosed using the Semmes-Weinstein 5 7/1g monofilament and a biothesiometer (both from Novalab Ibérica, Madrid, Spain). Patients who did not feel either of the two tests were diagnosed with neuropathy (15). Peripheral arterial disease was considered when both distal pulses were absent, the ankle brachial index was below 9 and/or transcutaneous oxygen pressure (TcPo2) was below 3 mmhg (Radiometer Medical, Copenhagen, Denmark) (16). R. J. Molines-Barroso et al. Table 1 Demographic and clinical data of subjects who underwent amputation (N = 156) Patients Male/female n (%) 125 (8)/31 (19) Mean age (years) ± SD 62 41 ± 11 68 Body mass index (kg/m 2 ) 27 61 ± 4 31 Type 1/type 2 diabetes mellitus n (%) 25 (16)/131 (84) Mean duration diabetes mellitus (years) ± SD 18 37 ± 12 48 Glycated haemoglobin mmol/mol ± SD (% ± SD) 57 ± 17 (7 4 ± 1 6) Retinopathy n (%) 83 (53) Nephropathy n (%) 55 (35) Neuropathy n (%) 134 (85) Previous amputation n (%) 96 (62) Peripheral vascular disease n (%) 32 (21) An experienced podiatrist prescribed therapeutic footwear that consisted of either an extra-depth shoe or custom insoles, according to foot biomechanical characteristics and plantar pressure measurements (Rsscan International, Olen, Belgium), after the ulcer was completely healed (17). Patients were subjected to periodic footwear checks and forefoot examination for new ulcers, according to the recommendations of the International Working Group on the Diabetic Foot (IWGDF) (17). Outcome measures Reulceration that occurred during the follow-up period was the main outcome measure assessed in the study. Reulceration was defined as a new full-thickness lesion of the skin, that is, a new wound penetrating through the dermis at the foot, without reference to the time it had been present (18). For participants who suffered reulceration, reulceration-free survival time was considered to be the time between the date of surgical intervention and the date when reulceration took place. In those patients who did not suffer reulceration, reulceration-free survival time was considered to be the time between the date of surgical intervention and the date of the last contact with the participant during the follow-up. When a participant was missed during the follow-up, they were phoned to evaluate new ulcerations, and the data of ulceration or the last phone contact were noted. Identification of final amputation pattern Radiological records were abstracted to identify the amputation end stage for each participant. A toe was considered amputated when more than 75% of the total toe was resected. A ray was considered amputated when a toe was completely removed and the upper metatarsal level rather than the metatarsal head was resected. Radiological records were not analysed in this study when they were not available or if the amputation pattern was unclear. The sample was divided into those patients who presented an amputation end stage different from Miller s formula (group 1) and those who presented an amputation end stage according to the formula proposed by Miller et al. (group 2). In group 2, participants considered at high risk of ulceration were divided 524 216 Medicalhelplines.com Inc and John Wiley & Sons Ltd

R. J. Molines-Barroso et al. Validation of an amputation algorithm Figure 1 Algorithm for classifying the end stage of amputation according to Miller s formula. into two subgroups according to whether or not the hallux had been amputated (patterns 1 and 2, respectively). Pattern 1 was defined as those individuals who had a history of hallux or first ray amputation and, in addition, one or more amputations involving segments 2 5, either toe or lesser ray amputation. Pattern 2 included participants with a history of lesser toe amputation plus two digital/ray amputations (Figure 1). Statistical analysis Statistical analysis was performed using SPSS for Macintosh, version 19. (SPSS, Inc. Chicago, IL). Reulceration was analysed in terms of frequency and reulceration-free survival time. A Mantel-Haenszel Chi-square test (χ 2 MH) was used to compare amputation patterns with reulceration frequency. Univariate analysis was applied to evaluate survival differences between groups by using the log-rank test. The Cox survival model adjusted for confounding variables was applied to evaluate the covariates that showed proportional hazards over reulceration-free survival time, and were expressed as a hazard ratio. P < 5 was accepted as statistically significant with a confidence interval of 95%. This retrospective study based on chart review did not require approval by an ethics committee as all procedures were part of our standard practice, and we did not alter or exceed our usual scope of medical care. Results Patient population and treatment One hundred and eighteen individuals met the inclusion criteria. Four patients whose radiological records were unclear and three patients who could not be contacted by phone were excluded from the analysis. Seventy subjects (63 1%) were classified into group 1 and 41 individuals (36 9%) into group 2. In group 2, 25 (22 5%) individuals showed pattern 1 and 25 (22 5%) showed pattern 2 after undergoing surgery; for nine individuals, both patterns coincided in the same foot. Individuals who had both patterns were eliminated from the analysis to ensure the association of each pattern with the risk of reulceration; therefore, 12 participants were analysed in the study (Figure 2). There were no differences in clinical or demographic variables between groups or between subgroups (Tables 2 and 3, respectively). Reulceration took place in 48 (47 1%) individuals. The reulceration frequency was 29 (41 4%) events in group 1, while 19 (59 4%) events occurred in subjects belonging to group 2. Eleven individuals (68 8%) suffered reulceration in subgroup 1, while eight individuals (5 %) suffered it in subgroup 2. Survival analysis Eight subjects were contacted by phone to assess the development of new ulcers, and six patients died during the follow-up period. Patients were evaluated for reulceration during a median follow-up period of 18 months (IR 8 2 36 6 months). Individuals who suffered reulceration (47 6%) showed a median reulceration-free survival time of 8 months (IR 3 6 14 8 months). Reulceration-free individuals were followed up for a median time of 3 3 months (IR 18 8 44 6 months). Miller s formula (11 2; IR 4 6 22 1 months) was a risk factor for reulceration-free survival time (P = 28) compared to group 1 (21 4; IR 12 37 7 months). In group 2, subjects who had pattern 1 (8 6 months; IR 2 9 15 1) showed an association with reulceration-free survival time (P = 4), while those who displayed pattern 2 (13 9 months; IR 5 9 33 8) did not show (P = 841) an association (Figure 3). Additionally, nephropathy (P = 5) was one more variable associated with reulceration-free survival time in the univariate analysis (Figure 4). 216 Medicalhelplines.com Inc and John Wiley & Sons Ltd 525

Validation of an amputation algorithm R. J. Molines-Barroso et al. Figure 2 Flow chart of the process of inclusion of patients in the present study. Table 2 Demographic and clinical data of patients at low (Group 1) and high (Group 2) risk of reulceration as proposed by Miller et al. Table 3 Demographic and clinical data of patients in the group at high risk of reulceration and divided into subgroups according to Miller s formula (N = 12 patients) Non-Miller s formula (Group 1) n = 7 Miller s formula (Group 2) n = 32 P-value (N = 32 patients) Related to first segment (Subgroup 1) n = 16 Related to lesser segment (Subgroup 2) n = 16 P-value Male/female n (%) 59 (84)/11 (16) 23 (72)/9 (28) 143 Mean age (years) ± SD 65.3 ± 12.64 62 28 ± 11 6 254 Body mass index (kg/m 2 ) 28 17 ± 4 17 26 59 ± 3 96 75 Type 1/type 2 diabetes 9 (13)/61 (87) 7 (22)/25 (78) 245 mellitus n (%) Mean duration diabetes mellitus (years) ± SD 16 96 ± 12 27 18 62 ± 12 65 53 Glycated haemoglobin mmol/mol ± SD (% ± SD) 57 ± 17 (7 4 ± 1 6) 55 ± 21 (7 2 ± 1 9) 638 Retinopathy n (%) 3 (43) 17 (53) 334 Nephropathy n (%) 2 (29) 9 (28) 963 Neuropathy n (%) 56 (8) 28 (87) 357 Peripheral vascular disease n (%) 1 (14) 6 (19) 565 Multivariate analysis Miller s formula did not show a hazard risk in the multivariate analysis. Nephropathy and pattern 1 were associated with the reulceration-free survival time in the Cox model adjusted by age, body mass index and duration of DM (Table 4). Discussion Data from this study show that Miller s formula is not indicated for evaluating the risk of reulceration in every patient with diabetes and a history of amputation. However, in a subset of patients, first segment amputation was an independent prognostic factor for predicting risk of reulceration [P = 4; confidence interval (CI) 1 391 5 849; hazard ratio (HR) 2 853]. Male/female n (%) 11 (69)/5 (31) 12 (75)/4 (25) 694 Mean age (years) ± SD 62 31 ± 11 48 62 25 ± 11 8 988 Body mass index 26 63 ± 3 4 26 56 ± 4 56 965 (kg/m 2 ) Type 1/type 2 diabetes 3 (19)/13 (81) 4 (25)/12 (75) 669 mellitus n (%) Mean duration 2 75 ± 14 6 16 5 ± 11 12 948 diabetes mellitus (years) ± SD Glycated haemoglobin 53 ± 14 (7 ± 1 3) 57 ± 27 (7 3 ± 2 4) 643 mmol/mol ± SD (% ± SD) Retinopathy n (%) 8 (5) 9 (56) 723 Nephropathy n (%) 4 (25) 5 (31) 694 Neuropathy n (%) 13 (81) 15 (94) 285 Peripheral vascular disease n (%) 4 (25) 2 (12) 365 Revisional surgery after first ray amputation and resection has been widely investigated (19), and reamputation (2) and reulceration (5) rates of about 7% have been found, depending on the amputation level. Amputation of the hallux and/or first metatarsal leads to changes in the distribution of pressure and changes in the foot structure (21), resulting in the development of new bony deformities (6) and an increased risk of new ulceration. Furthermore, Dalla Paola et al., (22) utilising a sample of patients who had undergone first ray amputation for 16 35 months, showed a low ulcer recurrence and reamputation risk when patients were included in an intensive clinical programme. Similar data were obtained in our study as patients with a single first segment amputation belonging to group 1 526 216 Medicalhelplines.com Inc and John Wiley & Sons Ltd

R. J. Molines-Barroso et al. Validation of an amputation algorithm 1 Non Miller s formula Pattern 1 Pattern 2 1 Non previous nephropathy Nephropathy 8 8 Cumulative survival (%) 6 4 2 Cumulative survival (%) 6 4 2 2 4 6 Follow-up period (months) 8 1 Figure 3 Kaplan Meier curves of reulceration-free survival for pattern 1, pattern 2 and the group different from Miller s formula. (followed for 21 4 months) showed a lower risk of reulceration than those belonging to the Miller s formula group. However, our data suggest that a new amputation in patients with a history of a single first ray amputation will dramatically increase the risk of reulceration. This subset of patients should be included in further investigations for a new conservative or surgical treatment that can avoid reulceration. According to our results, pattern 2 of Miller s could not be used to identify people with a higher risk of reulceration compared to other partial forefoot amputation (P = 841). Multiple lesser ray amputations were not associated with people at a higher risk of reulceration or reamputation than a single lesser ray amputation in other studies (5,6). Amputations of digits 2 5 have been considered functionally less important for mobility and quality of life (1). However, we believe that a specific type of minor amputation exists in people who have undergone amputations related to lesser segments that increases the risk of reamputation, but according to the findings of our study, it is different to that proposed by Miller. In our experience, surgical interventions on the fifth metatarsal have been associated with the lowest risk of reulceration (5) among lesser metatarsal procedures and are common interventions. The location of lesser ray amputation was not 2 4 6 8 Follow-up period (months) Figure 4 Kaplan Meier curves of reulceration-free survival for nephropathy. considered in the formula proposed by Miller and was therefore not considered in this study either; this could influence the risk of reulceration. The number and level of variables next to the location (6) of lesser segment amputation should be included in further multivariate investigations to identify a more precise lesser segment pattern related to the risk of new ulceration. Reulceration occurred in 41 4 % (29/7) of patients belonging to group 1 and 5% (8/16) of patients in subgroup 2, and these rates should not be underestimated. Every patient with a previous amputation (23) should be aware of this risk of reulceration and be included in an intensive follow-up programme of foot care in diabetic foot units; otherwise, in non-specialised centres, these numbers may increase considerably over time. The limitations of the present study were its retrospective nature, that, at recruitment, many of the patients had undergone previous amputation at other centres and that the duration since first amputation was unreliable. Structured joint deformities (6) resulting from a lack of monitoring and use of therapeutic footwear could lead to differences in reulceration compared to people undergoing their first amputation event. The findings are intended to assist clinicians in identifying which specific type of forefoot amputation results in a patient having a higher probability of suffering reulceration and ultimately undergoing revisional surgery. Although Miller s 1 Table 4 Cox survival model adjusted for confounding variables Variables Adjusted* hazard ratio [95% CI] P-value Adjusted hazard ratio [95% CI] P-value Adjusted hazard ratio [95% CI] P-value Nephropathy 2 28 [1 218 4 266] 1 2 468 [1 328 4 587] 4 2 26 [1 191 4 84] 12 Pattern of amputation 1 796 [ 963 3 348] 65 2 853 [1 391 5 849] 4 895 [ 399 2 1] 788 *Miller s formula, adjusted for age, body mass index, duration of diabetes mellitus and nephropathy. Pattern 1 of Miller s formula, adjusted for age, body mass index, duration of diabetes mellitus and nephropathy. Pattern 2 of Miller s formula, adjusted for age, body mass index, duration of diabetes mellitus and nephropathy. 216 Medicalhelplines.com Inc and John Wiley & Sons Ltd 527

Validation of an amputation algorithm formula is not suitable for every patient with a previous amputation, we believe that it is clinically useful in certain patients who have undergone first ray amputation. Participants with hallux or first ray amputation who have to undergo another amputation will have the highest probability of reulceration in comparison with other specific types of minor amputation. More research is required that includes other risk factors, such as the location of the amputation, to find the specific pattern related to lesser segment amputation. Acknowledgement The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article. References 1. Prompers L, Huijberts M, Apelqvist J, Jude E, Piaggesi A, Bakker K, Edmonds M, Holstein P, Jirkovska A, Mauricio D, Ragnarson Tennvall G, Reike H, Spraul M, Uccioli L, Urbancic V, Van Acker K, van Baal J, van Merode F, Schaper N. High prevalence of ischaemia, infection and serious comorbidity in patients with diabetic foot disease in Europe. Baseline results from the Eurodiale study. Diabetologia 27;5:18 25. 2. Jeffcoate WJ, Harding KG. Diabetic foot ulcers. Lancet 23;361:1545 51. 3. Prompers L, Huijberts M, Schaper N, Apelqvist J, Bakker K, Edmonds M, Holstein P, Jude E, Jirkovska A, Mauricio D, Piaggesi A, Reike H, Spraul M, Van Acker K, Van Baal S, Van Merode F, Uccioli L, Urbancic V, Ragnarson Tennvall G. Resource utilisation and costs associated with the treatment of diabetic foot ulcers. Prospective data from the Eurodiale Study. Diabetologia 28;51: 1826 34. 4. Boulton AJ, Vileikyte L, Ragnarson-Tennvall G, Apelqvist J. The global burden of diabetic foot disease. Lancet 25;366:1719 24. 5. Molines-Barroso RJ, Lázaro-Martínez JL, Aragón-Sánchez J, García-Morales E, Beneit-Montesinos JV, Álvaro-Afonso FJ. Analysis of transfer lesions in patients who underwent surgery for diabetic foot ulcers located on the plantar aspect of the metatarsal heads. Diabet Med 213;3:973 6. 6. Quebedeaux TL, Lavery LA, Lavery DC. The development of foot deformities and ulcers after great toe amputation in diabetes. Diabetes Care 1996;19:165 7. 7. Chu YJ, Li XW, Wang PH, Xu J, Sun HJ, Ding M, Jiao J, Ji XY, Feng SH. Clinical outcomes of toe amputation in patients with type 2 diabetes in Tianjin, China. Int Wound J 216;13(2):175 81. 8. Greteman B, Dale S. Digital amputations in neuropathic feet. JAm Podiatr Med Assoc 199;8:12 6. 9. Izumi Y, Satterfield K, Lee S, Harkless LB. Risk of reamputation in diabetic patients stratified by limb and level of amputation: a 1-year observation. Diabetes Care 26;29:566 7. 1. Pickwell K, Siersma V, Kars M, Apelqvist J, Bakker K, Edmonds M, Holstein P, Jirkovská A, Jude E, Mauricio D, Piaggesi A, Ragnarson Tennvall G, Reike H, Spraul M, Uccioli L, Urbancic V, R. J. Molines-Barroso et al. van Acker K, van Baal J, Schaper N. Predictors of lower-extremity amputation in patients with an infected diabetic foot ulcer. Diabetes Care 215;38:852 7. 11. Miller J, Zhubrak M, Giovinco N, Mills J, Armstrong D. The Too Few Toes principle: a formula for limb-sparing low-level amputation planning. Wound Med 214;4:37 41. 12. Lipsky BA, Berendt AR, Cornia PB, Pile JC, Peters EJ, Armstrong DG, Deery HG, Embil JM, Joseph WS, Karchmer AW, Pinzur MS, Senneville E; Infectious Diseases Society of America. 212 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. Clin Infect Dis 212;54:132 73. 13. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG; TASC II Working Group, Bell K, Caporusso J, Durand-Zaleski I, Komori K, Lammer J, Liapis C, Novo S, Razavi M, Robbs J, Schaper N, Shigematsu H, Sapoval M, White C, White J, Clement D, Creager M, Jaff M, Mohler E 3rd, Rutherford RB, Sheehan P, Sillesen H, Rosenfield K. Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). Eur J Vasc Endovasc Surg 27;33:1 75. 14. Armstrong DG, Frykberg RG. Classifying diabetic foot surgery: toward a rational definition. Diabet Med 23;2:329 31. 15. Boulton AJ, Armstrong DG, Albert SF, Frykberg RG, Hellman R, Kirkman MS, Lavery LA, Lemaster JW, Mills JL Sr, Mueller MJ, Sheehan P, Wukich DK; American Diabetes Association; American Association of Clinical Endocrinologists. Comprehensive foot examination and risk assessment: a report of the task force of the foot care interest group of the American Diabetes Association, with endorsement by the American Association of Clinical Endocrinologists. Diabetes Care 28;31:1679 85. 16. Schaper NC, Andros G, Apelqvist J, Bakker K, Lammer J, Lepantalo M, Mills JL, Reekers J, Shearman CP, Zierler RE, Hinchliffe RJ; International Working Group on Diabetic foot. Specific guidelines for the diagnosis and treatment of peripheral arterial disease in a patient with diabetes and ulceration of the foot 211. Diabetes Metab Res Rev 212;28:236 7. 17. Bakker K, Apelqvist J, Schaper NC. Practical guidelines on the management and prevention of the diabetic foot 211. Diabetes Metab Res Rev 212;28:225 31. 18. Schaper NC. Diabetic foot ulcer classification system for research purposes: a progress report on criteria for including patients in research studies. Diabetes Metab Res Rev 24;2:9 5. 19. Borkosky SL, Roukis TS. Incidence of re-amputation following partial first ray amputation associated with diabetes mellitus and peripheral sensory neuropathy: a systematic review. Diabetic Foot & Ankle 212;3:12169. DOI: 1.342/dfa.v3i.12169. 2. Murdoch DP, Armstrong DG, Dacus JB, Laughlin TJ, Morgan CB, Lavery LA. The natural history of great toe amputations. J Foot Ankle Surg 1997;36:24 8. 21. Mann RA, Poppen NK, O Konski M. Amputation of the great toe. A clinical and biomechanical study. Clin Orthop Relat Res 1988;226:192 25. 22. Dalla Paola L, Faglia E, Caminiti M, Clerici G, Ninkovic S, Deanesi V. Ulcer recurrence following first ray amputation in diabetic patients: a cohort prospective study. Diabetes Care 23;26:1874 8. 23. Lavery LA, Armstrong DG, Vela SA, Quebedeaux TL, Fleischli JG. Practical criteria for screening patients at high risk for diabetic foot ulceration. Arch Intern Med 1998;158:157 62. 528 216 Medicalhelplines.com Inc and John Wiley & Sons Ltd