Effect of Limb Preservation Status and Body Mass Index on the Survival of Patients with Limb-threatening Diabetic Foot Ulcers Cheng-Wei Lin, Brend Ray-Sea Hsu, Jir-Shiong Tsai, Hui-Mei Yang, Jr-Rung Lin, Chia-Hung Lin, Chung-Huei Huang, Shih-Yuan Hung, Yu-Yao Huang PII: DOI: Reference:

S1056-8727(16)30623-7 doi: 10.1016/j.jdiacomp.2016.09.011 JDC 6855

To appear in:

Journal of Diabetes and Its Complications

Received date: Revised date: Accepted date:

30 March 2016 22 September 2016 24 September 2016

Please cite this article as: Lin, C.-W., Hsu, B.R.-S., Tsai, J.-S., Yang, H.-M., Lin, J.-R., Lin, C.-H., Huang, C.-H., Hung, S.-Y. & Huang, Y.-Y., Effect of Limb Preservation Status and Body Mass Index on the Survival of Patients with Limbthreatening Diabetic Foot Ulcers, Journal of Diabetes and Its Complications (2016), doi: 10.1016/j.jdiacomp.2016.09.011

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Effect of Limb Preservation Status and Body Mass Index on the Survival of

T

Patients with Limb-threatening Diabetic Foot Ulcers

RI P

Cheng-Wei Lin, M.D.a; Brend Ray-Sea Hsu, M.D., Ph.D.a; Jir-Shiong Tsai, M.D.,

SC

F.A.C.P.b; Hui-Mei Yang, R.N.a; Jr-Rung Lin, Ph.D.c; Chia-Hung Lin, M.D.a;

a

MA NU

Chung-Huei Huang, M.D.a; Shih-Yuan Hung, M.D.a; Yu-Yao Huang, M.D., Ph.D.a, d*

Division of Endocrinology and Metabolism, Department of Internal Medicine, Chang

Gung Memorial Hospital, Chang Gung University, Taoyuan City, Taiwan

Clinical Informatics and Medical Statistics Research Center, Chang Gung University,

PT

c

Sun Yat-Sen Cancer Center, Taipei City, Taiwan

ED

b

Department of Medical Nutrition Therapy, Chang Gung Memorial Hospital, Chang

AC

d

CE

Taoyuan City, Taiwan

Gung University, Taoyuan City, Taiwan *Correspondence and reprint requests to: Yu-Yao Huang, Division of Endocrinology and Metabolism, Chang Gung Memorial Hospital. 5, Fuxing St., Guishan Dist., Taoyuan City 333, Taiwan Tel.: +886-3-3281200 (ext. 8826) Fax: +886-3-3288257 E-mail: [email protected]

1

ACCEPTED MANUSCRIPT Abstract

T

Aims

RI P

To evaluate the effect of limb preservation status and body mass index (BMI) on

SC

the survival of patients with diabetic foot ulcers (DFUs). Methods

MA NU

A total of 1346 patients treated for limb-threatening DFUs at a major diabetic foot center in Taiwan from 2002 to 2009 were tracked until December 2012. The patients were classified into three groups: limb-preserved (n=858), minor

ED

lower-extremity amputation (LEA) (n=249), and major LEA (n=239). Clinical data

PT

during treatment were used for survival analysis.

CE

Results

AC

With 729 deaths, the median survival time (MST) was 6.14 (95% CI 5.63－6.65) years. Major LEA and BMI were two independent factors associated with mortality after adjusting for age, diabetic duration, HbA1c level, comorbidities and peripheral artery diseases. The mortality hazard ratios for the minor and major LEA groups were 0.92 (95% CI 0.74－1.16) and 1.34 (95% CI 1.07－1.68), respectively, to the reference group (limb-preserved). After stratifying BMI into four categories (underweight, normal weight, overweight and obesity, according to the Taiwanese definition), the MSTs for each

2

ACCEPTED MANUSCRIPT category were 2.57, 5.24, 7.47 and 7.85 years, respectively (P for trend < 0.01). This

T

“obesity paradox” was not observed in the major LEA group (P for trend 0.25). For

RI P

patients with LEA, the obesity patients had lower MST than those in overweight

SC

category (7.97 and 8.84 in minor and 3.25 and 5.42 in major LEA, respectively). Conclusions

MA NU

For the patients treated for DFUs, major － but not minor － LEA was associated with poor survival compared with the limb-preserved group. The MST had positive correlation with BMI levels for patients with limb-preserved and minor LEA,

PT

ED

but not for those with major LEA.

CE

Keywords

AC

Diabetic foot ulcers, Survival, BMI, Lower-extremity amputation

3

ACCEPTED MANUSCRIPT Main Text

T

1. Introduction

RI P

Recent studies have reported poor long-term outcomes for patients with diabetic

SC

foot ulcers (DFUs) (Iversen et al., 2009; Moulik et al., 2003), with a 5-year survival rate of around 55% (Morbach et al., 2012; Moulik et al., 2003). Although factors such

MA NU

as age (Morbach et al., 2012; Tseng et al., 2008), renal function (Lavery et al., 2010; Morbach et al., 2012), and co-existing peripheral artery disease (Morbach et al., 2012) have been reported to be associated with mortality, the impact of factors such as limb

ED

preservation status and body mass index (BMI) have not been fully elucidated.

PT

A higher mortality rate has been reported in diabetic subjects with major

CE

lower-extremity amputation (LEA) (above the ankle) than minor LEA (below the

AC

ankle) (Hambleton et al., 2009; Izumi et al., 2009). The current study is an investigation of the long-term risk of mortality for patients with limb-threatening DFUs, with specific analysis of limb status after in-hospital treatment. Obesity is a known risk factor associated with diabetes mellitus (Colditz et al., 1995), hypertension (Huang et al., 1998), cardiovascular disease (Kim et al., 2000) and death (Adams et al., 2006; Flegal et al., 2005). Nevertheless, the so-called “obesity paradox”, which suggests obese people have a greater chance of survival than people with normal weight, has recently been reported for certain chronic

4

ACCEPTED MANUSCRIPT diseases (Lavie et al., 2015). These included chronic obstructive pulmonary disease

T

(Blum et al., 2011), stroke (Andersen&Olsen, 2013), hemodialysis (Kalantar-Zadeh et

RI P

al., 2012), diabetes (Carnethon et al., 2012), and peripheral artery disease (Kumakura

SC

et al., 2010). BMI may also play a role in the survival of patients with DFUs, although this has yet to be verified.

MA NU

The aim of this study was to investigate the survival and associated risk factors for mortality of patients with limb-threatening DFUs. Limb preservation status after

PT

2. Patients and Methods

ED

treatment and BMI level were the two main factors of interest in this study.

CE

2.1 Patients and hospital care

AC

In this study, we retrospectively analyzed the medical chart notes of patients with type 2 diabetes admitted to Chang Gung Memorial Hospital (a 3,700-bed teaching and referring hospital) for limb-threatening foot ulcers at our Diabetic Foot Center between January 2002 and December 2009. A limb-threatening foot ulcer is defined as an ulcer accompanied with limb-threatening infection (Frykberg et al., 2006) or critical limb ischemia (Norgren et al., 2007). The limb-threatening infection was prescribed if at least one of the following conditions was present: greater than 2 cm of cellulitis around an ulcer, lymphangiitis, soft tissue necrosis, gangrene, and

5

ACCEPTED MANUSCRIPT osteomyelitis (Frykberg et al., 2006). This foot center was certified by the

T

International Diabetes Federation-West Pacific Region. The Institutional Review

RI P

Board of Chang Gung Memorial Hospital approved this study (no. 104-1401B).

SC

All patients were admitted for comprehensive foot care by a multidisciplinary team. This team included diabetologists, cardiologists, and vascular, plastic, and

MA NU

orthopedic surgeons. The patients received prompt empiric systemic antibiotic treatment, wound debridement and dressing. Surgical interventions such as debridement, drainage, digital or foot amputation or revascularization were scheduled

ED

after the investigative diabetic team reached a consensus (Hung et al., 2015; Sun et al.,

PT

2012).

CE

A total of 1346 type 2 diabetic patients were identified who completed treatment

AC

for limb-threatening DFUs. All were discharged with a clean ulcer or healed wound (surgical debridement or amputated wound included) which could be managed at home. The mean duration of hospitalization was 30.62 ± 23.65 days. 2.2 Data collection Survival time was calculated from the discharge date to December 31, 2012, or death. Causes and the date of death were obtained from the National Death Registry of the Taiwan National Health Insurance Database and classified in accordance with the ninth revision of the International Classification of Diseases (Huang et al., 2012;

6

ACCEPTED MANUSCRIPT Li et al., 2012).

T

Demographic information was recorded from the patients’ first visit at admission

RI P

and used for survival analysis. Limb preservation status at discharge was stratified

SC

into three groups: limb-preserved (i.e., no amputation performed, n=858), minor LEA (i.e., amputation performed including digital amputation or tarsal-metatarsal

MA NU

amputation, as long as it did not involve the ankle area, n=249), and major LEA (i.e., amputation performed above the ankle joint, n=239). The medical records included the patient’s age, gender, diabetes duration, body weight/height, HbA1c level and

ED

medical history such as hypertension, history of major adverse cardiac events

PT

(MACEs; including coronary artery disease and cerebrovascular accidents) and

CE

dialysis. BMI was determined on the first day of admission and calculated as the

AC

weight in kilograms divided by the square of the height in meters (kg/m2). Smokers were classified as currently smoking if they smoked at least one cigarette per day. The estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease Study equation: 175 x serum creatinine (exp[−1.154]) × age (exp[0.203]) × (0.742 if female). Renal status was categorized as follows: normal or mild chronic kidney disease (eGFR ≥ 60 ml/min); moderate to severe chronic kidney disease (eGFR < 60 ml/min); and dialysis according to the National Kidney Disease Outcomes Quality Initiative (NKDOQI) guidelines. Ankle brachial index (ABI) was

7

ACCEPTED MANUSCRIPT obtained by measuring blood pressure at the ankle and in the arm while the patient

T

was at rest. Peripheral artery disease (PAD) was defined as either an abnormal ABI (≤

RI P

0.9 or > 1.4 according to the 2011 American Heart Association guidelines) or

SC

evidence of arterial occlusion by CT angiography or Doppler waveform (Rooke et al., 2011; Tsai et al., 2013).

MA NU

2.3 Statistics

Age, duration of diabetes, BMI and HbA1c level were used as continuous variables, and gender, smoking status, and co-morbidities were used as categorical

ED

variables. Comparisons between the three limb preservation groups were performed

PT

using one-way ANOVA for continuous variables and Pearson’s chi-square test for

CE

categorical variables. The ability of each variable to predict mortality (hazard ratio)

AC

was calculated using univariate Cox regression analysis. The significant risk factors in the univariate analysis were then entered into a multivariate Cox regression model with adjustments for covariates to identify independent risk factors. The information for those individuals who were alive at the end of the study was collected as right-censored data and the Kaplan－Meier method was therefore used to calculate the median survival time (MST). All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS for Windows, version 19.0, Armonk, NY: IBM Corp.) data analysis software.

8

ACCEPTED MANUSCRIPT

T

3. Results

RI P

3.1 Demographic data

SC

Seven hundred and twenty-nine of the 1346 patients died during the study period. The 5-year survival rate was 57% with a MST of 6.14 (95% CI 5.63－6.65) years and

MA NU

a crude mortality rate of 114.5 per 1000 patient-years. The observation period of this study was up to 10 years (median: 4.33 years, range: 0.03－10.47). The demographic and clinical data of the study population are presented in Table 1. The mean age of the

ED

patients was 63.84 ± 12.51 years, and 55.5% of them were male. The mean BMI was

PT

24.53 ± 4.24 kg/m2, which is considered to be overweight according to the Taiwanese

CE

definition (underweight: < 18.5, normal weight: 18.5－23.9, overweight: 24－26.9,

AC

and obesity: ≥ 27 kg/m2) (Hwang et al., 2006). Sixty percent of the subjects had underlying hypertension, 24% had MACEs, and 12% were undergoing chronic dialysis for end-stage renal disease. Thirty-three percent of the patients (n=440) had PAD, of whom 71 (16%) received revascularization interventions (62 endovascular treatment and 9 bypass grafting). The patients with major LEA tended to be older (mean 66.08 years) with a lower BMI (mean 23.56 kg/m2), more underlying co-morbidities (72% hypertension, 35% MACEs, 27% dialysis), and to have PAD (49%) when compared with the

9

ACCEPTED MANUSCRIPT limb-preserved group or the minor LEA group.

T

3.2 Survival analysis

RI P

All variables listed in Table 1 except for gender and smoking status were entered

SC

into the multivariate Cox regression model. Age, chronic kidney disease or dialysis, MACEs, PAD, major LEA and BMI were independent risk factors associated with

MA NU

long-term mortality (Table 2). The patients with PAD had a higher risk of mortality [adjusted HR 1.31 (95% CI 1.09－1.56)]. The MSTs for the patients with PAD who did and did not receive interventions were 3.74 and 3.97 years, respectively, and the

ED

difference did not reach statistical significance (P value 0.69).

PT

Of note, there was no increase in mortality between the limb-preserved and

CE

minor LEA groups (Table 2 and Figure 1). When the data were compared with respect

AC

to the limb-preserved group, the adjusted mortality hazard ratios for minor and major LEAs were 0.92 (95% CI 0.74－1.16) and 1.34 (95% CI 1.07－1.68), respectively (Table 2). There was a negative correlation between BMI and mortality [adjusted HR 0.95 (95% CI 0.93－0.97)] (Table 2). 3.3 Survival analysis between limb preservation status and BMI categories After stratifying BMI into four categories of BMI classification (Hwang et al., 2006), the MSTs were 2.57 (95% CI 1.94－3.2), 5.24 (95% CI 4.66－5.82), 7.47 (95% CI 6.41－8.53) and 7.85 (95% CI 6.57－9.13) years, respectively, in the

10

ACCEPTED MANUSCRIPT underweight, normal weight, overweight and obesity groups, with a significant

T

increasing trend (P for trend < 0.01) (Table 3). When the data of the normal weight

RI P

patients were used as reference, the adjusted mortality hazard ratios for the

SC

underweight, overweight, and obesity groups were 1.77 (95% CI 1.25－2.51), 0.78 (95% CI 0.63－0.96), and 0.68 (95% CI 0.54－0.86), respectively (Table 3).

MA NU

Similar trends were observed in the limb-preserved and minor LEA groups (Figure 2), with both P values for trend < 0.01 (Table 3), but not in the major LEA group (P value 0.25). The longest MST for the minor and major LEA groups was in

ED

the overweight patients (8.84 and 5.42 years, respectively) (Figure 2 and Table 3). The

PT

subjects with LEA in the obesity group had a decrease in MST (8.84 to 7.97 years in

CE

minor and 5.42 to 3.25 years in major LEA).

AC

3.4 Causes of death

The most common cause of mortality in the 729 subjects who died was: documented cardiovascular disease (18.6% overall; coronary heart disease 11.8%; cerebrovascular disease 6.7%; and peripheral vascular disease 0.01%), followed by kidney complications (8.4%), infectious diseases (8.4%), and malignancy (8%). Approximately 44% of those who died had an uncertain diagnosis.

4. Discussion

11

ACCEPTED MANUSCRIPT In addition to age (Tseng et al., 2008), concurrent MACEs (Morbach et al., 2012),

T

chronic kidney disease (Lavery et al., 2010), and PAD (Morbach et al., 2012), major

RI P

LEA status and BMI were major factors affecting survival in the patients with DFUs

SC

in this study. The 5-year survival rate of the study population was 57%, which is similar to previous studies (Morbach et al., 2012; Moulik et al., 2003).

MA NU

Several studies on survival rates of amputees have shown a poor prognosis in patients with a higher LEA level (Hambleton et al., 2009; Izumi et al., 2009; Lavery et al., 2010). Compared to the patients with minor LEA in this study, those with major

ED

LEA had a 33% increased risk of mortality, which is consistent with the hazard ratios

PT

reported in previous studies (range 1.1－2.43) (Hambleton et al., 2009; Izumi et al.,

CE

2009; Lavery et al., 2010; Tseng et al., 2008). Difficulty in ambulation following

AC

LEAs may increase the risk of mortality. That is, limited mobility (Davies&Datta, 2003), bed-ridden status (Inderbitzi et al., 2003), and higher risk of falls (Pauley et al., 2006) may all contribute to poor survival. As a higher amputation level may result in poor ambulation (Suckow et al., 2012), patients with major LEA have a poorer survival than those with minor LEA. Furthermore, a major LEA itself may increase the risk of cardiovascular disease because of behavioral changes, psychosocial stress, and increases in inflammation and insulin resistance (Naschitz&Lenger, 2008). Nevertheless, it is difficult to determine the direction of causality. In this study, the

12

ACCEPTED MANUSCRIPT patients with major LEA had more underlying diseases (Table 1). The underlying

T

disease may have led to the DFU and the amputation. Therefore, amputation may be

RI P

involved in the progression to death. For example, patients with major LEA had more

SC

severe PAD, which is an indicator of more severe generalized atherosclerosis resulting in a higher mortality rate (Leng et al., 1996). Of note, this study is the first to show no

MA NU

differences in survival between patients with preserved limbs and those with minor LEA in those with DFUs. Sometimes sacrificing part of a limb is inevitable when treating patients with limb-threatening DFUs, such as in those with co-existing severe

ED

infections or ischemia. Our results suggest that the level of LEA should be kept to a

PT

minimum as far as possible, as this not only improves the quality of life but also leads

CE

to better long-term survival.

AC

The relationship between BMI and mortality usually manifests as a U-shape in both Western and Eastern populations, indicating the highest mortality rates in patients who are either underweight or obese (Adams et al., 2006; Chen et al., 2012; Flegal et al., 2005; Tamakoshi et al., 2010). The phenomenon of obesity paradox suggests a better chance of survival in obese subjects. Several hypotheses have been postulated to explain the obesity paradox (Hainer&Aldhoon-Hainerova, 2013). Patients with a higher BMI may be receiving optimal medication for hypertension or hyperlipidemia (Schenkeveld et al., 2012), and they may have more muscle mass

13

ACCEPTED MANUSCRIPT (Oreopoulos et al., 2010; Wannamethee et al., 2007) or better nutritional status

T

(Casas-Vara et al., 2012) and hence more strength to handle a severe illness. A higher

RI P

BMI may also be associated with decreased thromboxane production (Graziani et al.,

SC

2011) or increased production of soluble tumor necrosis factor receptor (Feldman et al., 2000), both of which may contribute to survival from cardiovascular disease. To

MA NU

the best of our knowledge, this study is the first to provide evidence of the obesity paradox with regards to survival for patients with DFUs. Although a higher BMI means better survival in this study population,

ED

differences were found between the three groups of limb status based on BMI

PT

categories. For patients with LEA, the highest MST was found for the overweight

CE

subjects. A lower MST for the patients with obesity was observed more prominent for

AC

the patients with major LEA than for those with minor LEA. Although it is difficult to determine the direct etiology, obesity is known to affect the musculoskeletal system, including osteoarthritis, osteoporosis, low back pain, and gait disturbance (Anandacoomarasamy et al., 2008). Therefore, it is possible that obesity in patients with LEA further hampers their mobility due to the burden on their musculoskeletal system. Such immobility may then lead to organ dysfunction including cardiovascular, respiratory, musculoskeletal, and neuropsychological changes, and ultimately mortality (Brown et al., 2004; Harper&Lyles, 1988; Siu et al., 2006). However, the

14

ACCEPTED MANUSCRIPT causal relationship was difficult to identify in our study, and further studies are

T

needed to verify our hypothesis.

RI P

This study is limited by the single center and retrospective design. The

SC

limb-threatening DFUs of our patients may not be generalized to patients with chronic foot ulcers. In addition, data on mobility and ambulation ability were not available in

MA NU

these patients after the critical foot event. Moreover, certain clinical features such as waist circumference, body fat and nutritional status were not available. The uncertain diagnosis in the records of mortality from the registry database makes further analysis

ED

of the causal relationship to mortality difficult. Further studies may be needed to

PT

verify the range of weight and muscle strength and physical performance in patients

AC

CE

with DFUs and different limb status.

Conclusions

In summary, limb status at treatment and BMI were independent factors for survival in patients with DFUs. Since major － but not minor － LEA was associated with poor survival, aggressive treatment to reduce the need for a major LEA may not only be beneficial for limb preservation, but also for the long-term prognosis. The median survival time increased with increasing BMI for patients with limb-preserved and minor LEA, but not for those with major LEA.

15

ACCEPTED MANUSCRIPT

T

Acknowledgment

RI P

The authors thank the staff and participants of this study for their important

SC

contributions. We also thank Professor Jr-Rung Lin for assistance with statistics. This research was supported by a grant (CMRPG3D0332) from Chang Gung Memorial

MA NU

Hospital to YY Huang. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No other potential conflicts of interest relevant to this article were reported.

ED

YY Huang contributed to the study design. CW Lin contributed to the statistical

PT

analysis and wrote the manuscript. BRS Hsu and JS Tsai reviewed and edited the

CE

manuscript. HM Yang, CH Lin, CH Huang, and SY Hung collected the data and

AC

contributed to the discussion. YY Huang gave the final approval of the version to be published.

16

ACCEPTED MANUSCRIPT References

T

Adams, K.F., Schatzkin, A., Harris, T.B., Kipnis, V., Mouw, T., Ballard-Barbash, R,

RI P

Hollenbeck, A., and Leitzmann, M.F. (2006). Overweight, obesity, and mortality in a

SC

large prospective cohort of persons 50 to 71 years old. N Engl J Med 355, 763-778. Anandacoomarasamy, A., Caterson, I., Sambrook, P., Fransen, M., and March, L.

MA NU

(2008). The impact of obesity on the musculoskeletal system. Int J Obes (Lond) 32, 211-222.

Andersen, K.K., and Olsen, T.S. (2015). The obesity paradox in stroke: Lower

PT

Int J Stroke 10, 99-104.

ED

mortality and lower risk of readmission for recurrent stroke in obese stroke patients.

CE

Blum, A., Simsolo, C., Sirchan, R., and Haiek, S. (2011). "Obesity paradox" in

AC

chronic obstructive pulmonary disease. Isr Med Assoc J 13, 672-675. Brown, C.J., Friedkin, R.J., and Inouye, S.K. (2004). Prevalence and outcomes of low mobility in hospitalized older patients. J Am Geriatr Soc 52, 1263-1270. Carnethon, M.R., De Chavez, P.J., Biggs, M.L., Lewis, C.E., Pankow, J.S., Bertoni, A.G., Golden, SH., Liu, K., Mukamal, K.J., Campbell-Jenkins, B., and Dyer, A.R. (2012). Association of weight status with mortality in adults with incident diabetes. JAMA 308, 581-590. Casas-Vara, A., Santolaria, F., Fernandez-Bereciartua, A., Gonzalez-Reimers, E.,

17

ACCEPTED MANUSCRIPT Garcia-Ochoa, A., and Martinez-Riera, A. (2012). The obesity paradox in elderly

T

patients with heart failure: analysis of nutritional status. Nutrition 28, 616-622.

RI P

Chen, Z., Yang, G., Offer, A., Zhou, M., Smith, M., Peto, R., Ge, H., Yang, L., and

SC

Whitlock, G. (2012). Body mass index and mortality in China: a 15-year prospective study of 220000 men. Int J Epidemiol 41, 472-481.

MA NU

Colditz, G.A., Willett, W.C., Rotnitzky, A., and Manson, J.E. (1995). Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med 122, 481-486 Davies, B., and Datta, D. (2003). Mobility outcome following unilateral lower limb

ED

amputation. Prosthet Orthot Int 27, 186-190.

PT

Feldman, A.M., Combes, A., Wagner, D., Kadakomi, T., Kubota, T., Li, Y.Y., and

CE

McTiernan, C. (2000). The role of tumor necrosis factor in the pathophysiology of

AC

heart failure. J Am Coll Cardiol 35, 537-544. Flegal, K.M., Graubard, B.I., Williamson, D.F., and Gail, M.H. (2005). Excess deaths associated with underweight, overweight, and obesity. JAMA 293, 1861-1867. Frykberg, R.G., Zgonis, T., Armstrong, D.G., Driver, V.R., Giurini, J.M., Kravitz, S.R., Landsman, A.S., Lavery, L.A., Moore, J.C., Schuberth, J.M., Wukich, D.K., Andersen, C., and Vanore, J.V. (2006). Diabetic foot disorders: a clinical practice guideline (2006 revision). J Foot Ankle Surg 45, 1-66. Graziani, F., Biasucci, L.M., Cialdella, P., Liuzzo, G., Giubilato, S., Della Bona, R.,

18

ACCEPTED MANUSCRIPT Pulcinelli, F.M., Iaconelli, A., Mingrone, G., and Crea, F. (2011). Thromboxane

T

production in morbidly obese subjects. Am J Cardiol 107, 1656-1661.

RI P

Hainer, V., and Aldhoon-Hainerova, I. (2013). Obesity paradox does exist. Diabetes

SC

Care 36, S276-281.

Hambleton, I.R., Jonnalagadda, R., Davis, C.R., Fraser, H.S., Chaturvedi, N., and

MA NU

Hennis, A.J. (2009). All-cause mortality after diabetes-related amputation in Barbados: a prospective case-control study. Diabetes Care 32, 306-307. Harper, C.M., and Lyles, Y.M. (1988). Physiology and complications of bed rest. J

ED

Am Geriatr Soc 36, 1047-1054.

PT

Huang, Y.Y., Lin, K.D., Jiang, Y.D., Chang, C.H., Chung, C.H., Chuang, L.M., Tai,

CE

T.Y., Ho, L.T., and Shin, S.J. (2012). Diabetes-related kidney, eye, and foot disease in

AC

Taiwan: an analysis of the nationwide data for 2000-2009. J Formos Med Assoc 111, 637-644.

Huang, Z., Willett, W.C., Manson, J.E., Rosner, B., Stampfer, M.J., Speizer, F.E., and Colditz, G.A. (1998). Body weight, weight change, and risk for hypertension in women. Ann Intern Med 128, 81-88. Hung, S.Y., Huang, Y.Y., Hsu, L.A., Chen, C.C., Yang, H.M., Sun, J.H., Lin, C.W., Wang, C.C. (2015). Treatment for diabetic foot ulcers complicated by major cardiac events. Can J Diabetes 39. 183-187.

19

ACCEPTED MANUSCRIPT Hwang, L.C., Bai, C.H., and Chen, C.J. (2006). Prevalence of obesity and metabolic

T

syndrome in Taiwan. J Formos Med Assoc 105, 626-635.

RI P

Inderbitzi, R., Buettiker, M., and Enzler, M. (2003). The long-term mobility and

Eur J Vasc Endovasc Surg 26, 59-64.

SC

mortality of patients with peripheral arterial disease following bilateral amputation.

MA NU

Iversen, M.M., Tell, G.S., Riise, T., Hanestad, B.R., Ostbye, T., Graue, M., and Midthjell, K. (2009). History of foot ulcer increases mortality among individuals with diabetes: ten-year follow-up of the Nord-Trondelag Health Study, Norway. Diabetes

ED

Care 32, 2193-2199.

PT

Izumi, Y., Satterfield, K., Lee, S., Harkless, L.B., and Lavery, L.A. (2009). Mortality

AC

126-131.

CE

of first-time amputees in diabetics: a 10-year observation. Diabetes Res Clin Pract 83,

Kalantar-Zadeh, K., Streja, E., Molnar, M.Z., Lukowsky, L.R., Krishnan, M., Kovesdy, C.P., and Greenland, S. (2012). Mortality prediction by surrogates of body composition: an examination of the obesity paradox in hemodialysis patients using composite ranking score analysis. Am J Epidemiol 175, 793-803. Kim, K.S., Owen, W.L., Williams, D., and Adams-Campbell, L.L. (2000). A comparison between BMI and Conicity index on predicting coronary heart disease: the Framingham Heart Study. Ann Epidemiol 10, 424-431.

20

ACCEPTED MANUSCRIPT Kumakura, H., Kanai, H., Aizaki, M., Mitsui, K., Araki, Y., Kasama, S., Iwasaki, T.,

T

and Ichikawa, S. (2010). The influence of the obesity paradox and chronic kidney

RI P

disease on long-term survival in a Japanese cohort with peripheral arterial disease. J

SC

Vasc Surg 52, 110-117.

Lavery, L.A., Hunt, N.A., Ndip, A., Lavery, D.C., Van Houtum, W., and Boulton, A.J.

MA NU

(2010). Impact of chronic kidney disease on survival after amputation in individuals with diabetes. Diabetes Care 33, 2365-2369.

Lavie, C.J., De Schutter, A., and Milani, R.V. (2015). Healthy obese versus unhealthy

ED

lean: the obesity paradox. Nat Rev Endocrinol 11. 55-62.

PT

Leng, G.C., Fowkes, F.G., Lee, A.J., Dunbar, J., Housley, E., and Ruckley, C.V.

CE

(1996). Use of ankle brachial pressure index to predict cardiovascular events and

AC

death: a cohort study. BMJ 313, 1440-1444. Li, H.Y., Jiang, Y.D., Chang, C.H., Chung, C.H., Lin, B.J., and Chuang, L.M. (2012). Mortality trends in patients with diabetes in Taiwan: a nationwide survey in 2000-2009. J Formos Med Assoc 111, 645-650. Morbach, S., Furchert, H., Groblinghoff, U., Hoffmeier, H., Kersten, K., Klauke, G.T., Klemp, U., Roden, T., Icks, A., Haastert, B., Rumenapf, G., Abbas, Z.G., Bharara, M., and Armstrong, D.G. (2012). Long-term prognosis of diabetic foot patients and their limbs: amputation and death over the course of a decade. Diabetes Care 35,

21

ACCEPTED MANUSCRIPT 2021-2027.

T

Moulik, P.K., Mtonga, R., and Gill, G.V. (2003). Amputation and mortality in

RI P

new-onset diabetic foot ulcers stratified by etiology. Diabetes Care 26, 491-494

SC

Naschitz, J.E., and Lenger, R. (2008). Why traumatic leg amputees are at increased risk for cardiovascular diseases. QJM 101, 251-259.

MA NU

Norgren, L., Hiatt, W.R., Dormandy, J.A., Nehler, M.R., Harris, K.A., Fowkes, F.G., and Rutherford, R.B. (2007). Inter-society consensus for the management of peripheral arterial disease. Int Angiol 26, 81-157.

ED

Oreopoulos, A., Ezekowitz, J.A., McAlister, F.A., Kalantar-Zadeh, K., Fonarow, G.C.,

PT

Norris, C.M., Johnson, J.A., and Padwal, R.S. (2010). Association between direct

CE

measures of body composition and prognostic factors in chronic heart failure. Mayo

AC

Clin Proc 85, 609-617.

Pauley, T., Devlin, M., and Heslin, K. (2006). Falls sustained during inpatient rehabilitation after lower limb amputation: prevalence and predictors. Am J Phys Med Rehabil 85, 521-532. Rooke, T.W., Hirsch, A.T., Misra, S., Sidawy, A.N., Beckman, J.A., Findeiss, L.K., Golzarian, J., Gornik, H.L., Halperin, J.L., Jaff, M.R., Moneta, G.L., Olin, J.W., Stanley, J.C., White, C.J., White, J.V., and Zierler, R.E. (2011). ACCF/AHA Focused update of the guideline for the management of patients with peripheral artery disease

22

ACCEPTED MANUSCRIPT (updating the 2005 guideline): a report of the American College of Cardiology

T

Foundation/American Heart Association Task Force on practice guidelines.

RI P

Circulation 124, 2020-2045.

SC

Schenkeveld, L., Magro, M., Oemrawsingh, R.M., Lenzen, M., de Jaegere, P., van Geuns, R.J., Serruys, P.W., and van Domburg, R.T. (2012). The influence of optimal

MA NU

medical treatment on the 'obesity paradox', body mass index and long-term mortality in patients treated with percutaneous coronary intervention: a prospective cohort study. BMJ Open 2, e000535.

ED

Siu, A.L., Penrod, J.D., Boockvar, K.S., Koval, K., Strauss, E., and Morrison, R.S.

PT

(2006). Early ambulation after hip fracture: effects on function and mortality. Arch

CE

Intern Med 166, 766-771.

AC

Suckow, B.D., Goodney, P.P., Cambria, R.A., Bertges, D.J., Eldrup-Jorgensen, J., Indes, J.E., Schanzer, A., Stone, D.H., Kraiss, L.W., and Cronenwett, J.L. (2012). Predicting functional status following amputation after lower extremity bypass. Ann Vasc Surg 26, 67-78. Sun, J.H., Tsai, J.S., Huang, C.H., Lin, C.H., Yang, H.M., Chan, Y.S., Hsieh, S.H., Hsu, B.R., and Huang, Y.Y. (2012). Risk factors for lower extremity amputation in diabetic foot disease categorized by Wagner classification. Diabetes Res Clin Pract 95, 358-363.

23

ACCEPTED MANUSCRIPT Tamakoshi, A., Yatsuya, H., Lin, Y., Tamakoshi, K., Kondo, T., Suzuki, S., Yagyu, K.,

T

and Kikuchi, S. (2010). BMI and all-cause mortality among Japanese older adults:

RI P

findings from the Japan collaborative cohort study. Obesity (Silver Spring) 18,

SC

362-369.

Tsai, C.Y., Chu, S.Y., Wen, Y.W., Hsu, L.A., Chen, C.C., Peng, S.H., Huang, C.H.,

MA NU

Sun, J.H., and Huang, Y.Y. (2013). The value of Doppler waveform analysis in predicting major lower extremity amputation among dialysis patients treated for diabetic foot ulcers. Diabetes Res Clin Pract 100, 181-188.

ED

Tseng, C.H., Chong, C.K., Tseng, C.P., Cheng, J.C., Wong, M.K., and Tai, T.Y. (2008).

PT

Mortality, causes of death and associated risk factors in a cohort of diabetic patients

CE

after lower-extremity amputation: a 6.5-year follow-up study in Taiwan.

AC

Atherosclerosis 197, 111-117. Wannamethee, S.G., Shaper, A.G., Lennon, L., and Whincup, P.H. (2007). Decreased muscle mass and increased central adiposity are independently related to mortality in older men. Am J Clin Nutr 86, 1339-1346.

24

ACCEPTED MANUSCRIPT Figure legends

T

Figure 1: Adjusted survival curve via a Cox proportional hazard model separated by

SC

RI P

limb preservation status.

Compared to the limb-preserved group, the adjusted mortality hazard ratios were 0.92

MA NU

(95% CI 0.74－1.16, P = 0.49) for minor LEA and 1.34 (95% CI 1.07－1.68, P = 0.01) for major LEA.

Covariates included continuous variables of age, diabetes duration, body mass index,

ED

and HbA1c; and categorical variables of hypertension, major adverse cardiac events,

CE

PT

peripheral artery disease, and renal function.

AC

Figure 2: Median survival time of the patients with DFUs according to limb preservation status and BMI category.

There was an increasing trend of median survival time with BMI category in all patients and the limb-preserved and minor LEA groups, but not in the major LEA group. Underweight includes subjects with a BMI < 18.5, normal weight BMI: 18.5－ 23.9, overweight BMI: 24－26.9, and obesity BMI ≥ 27.

25

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 1

26

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

Fig. 2

27

ACCEPTED MANUSCRIPT Table 1. Demographics and clinical characteristics of the patients with diabetic foot ulcers

Mean (standard deviation) or No. (%) Characteristic Limb-preserved

Minor LEA

(n=1346)

(n=858)

(n=249)

729

426

130

6.14 (5.63－6.65)

7.09 (6.42－7.76)

6.96 (5.67－8.25)

3.25 (2.23－4.27)

57%

61%

60%

39%

Age (years)

63.84

Male gender

747

(±12.51) (56%)

Diabetes duration (years)

11.48

(±8.08)

Body mass index

24.53

(±4.24)

256

(19%)

812

(60%)

Major adverse cardiac event

323

(24%)

Peripheral artery disease

440

(33%)

71

(16%)

Hypertension a

Revascularization

c

a

(±4.44)

(59%)

134

(±11.27)