INTRODUCTION

The International Diabetes Federation Diabetes Atlas 7th Edition estimated that approximately 415 million were living with diabetic mellitus (DM) worldwide; the reported prevalence was 8.8%, and 5 million patients died due to complications of DM annually.1 In the past 30 years, the prevalence of DM in China had increased significantly. In 2015, the number of diabetic patients in China ranked first in the world with a prevalence rate of 10.6%; in addition, approximately 1.3 million Chinese persons died of DM.1 Diabetic foot ulceration (DFU) was one of the most serious complications of diabetic patients, but a lack of awareness of DFU and its clinical relevance are thought to contribute to underreporting of this important diabetic complication.

The all-cause and cardiovascular disease (CVD) mortality among persons living with DM is approximately twice in persons with diabetes.2 Research also indicates that mortality rates are even higher among diabetic persons who develop DFU.3,4 Therefore, the aim of this study was to analyze all-cause and cardiovascular-related death rates in Chinese persons with DM alone and in those with DM and DFU.

METHODS

A single-center, retrospective case-control study was conducted in the Endocrinology Department of the Sun Yat-sen Memorial Hospital in Guangzhou, China. We reviewed medical records of patients receiving care from January 1, 2004, to December 31, 2010. Subjects were divided into 2 groups based on a diagnosis of DM (both groups) with or without foot ulcerations. Inclusion criteria were inpatients 18 years and older with a confirmed diagnosis of DM. Diagnostic criteria used to identify DM were based on 1999 World Health Organization (WHO) diagnostic criteria.5 The case group (DM and DFU) had a diagnosis of Wagner grade 1-5 diabetic foot ulcer during the medical review period.6 Exclusion criteria included those lost to follow-up during the data collection period and patients who had undergone amputation at time of study enrollment. Study procedures were reviewed and approved by the Medical Research Ethics Committee of the Sun Yat-sen Memorial Hospital (approval #SYSEC-KY-KS-2019-071).

Controls

Based on literature review and clinical experience, we matched controls to patients with DM and DFU as closely as possible. Specifically, patients were selected who matched based on the following criteria: age within 5 years of cases (+/-5 years), sex, similar duration of DM within 5 years, and similar duration of DM (+/- 5 years). We also selected controls with similar glycemic control level (mean glycosylated hemoglobin, HbA_1c, values were determined based on all available results from the laboratory test for use as the baseline level of diabetes control). These control levels were divided into 3 categories based on the mean HbA_1c (<7%, 7%-9%, >9%). The purpose of this 1-to-2 matching was to control for confounding factors that may influence death rates other than the variable of interest, presence of a diabetic foot ulcer.

Study Procedures

Data were recorded on a form designed for use in this study; records were reviewed between January 1, 2004, and December 31, 2010. Baseline demographic data collected were sex and age. In addition to the presence of DM with or without concurrent DFU, clinical data included history of cigarette smoking, alcohol intake as a dichotomous variable, body mass index, waist to hip ratio, HbA_1c, and relevant comorbid conditions such as dyslipidemia, hypertension, coronary artery disease, and secondary DM complication nephropathy or retinopathy. Outcome data collected were survival outcome, date, and cause of death. Baseline data were collected from medical records; follow-up data were collected through the electronic medical record (EMR) or via telephone. Mortality diagnosis, the study endpoint, was coded according to the ICD-10 (International Classification of Diseases, Tenth Revision) and based on the direct cause of death identified on the EMR. For purposes of data analysis, cause of death was discussed dichotomously as death related to CVD versus all other causes.

Data Analysis

All statistical analyses were performed using the SPSS version 20.0 software (SPSS, Chicago, Illinois). The Kaplan-Meier analysis was used to analyze survival times. Survival curve in groups was compared by log-rank analysis. Univariate and multivariate Cox proportional hazard model analyses were used to analyze factors influencing all-cause and cardiovascular-related deaths.

RESULTS

The sample comprised 264 patient records: 88 cases comprised patients with DM and DFU and 176 patients comprised the controls (DM without DFU matched based on parameters described in the "Methods" section). The average follow-up time among cases was 4.64 (SD = 2.78) years versus an average of 6.98 (SD = 2.50) years in the control group. Patients tended to be older; the mean age was 64.59 (SD = 9.72) years. Groups were evenly divided based on sex: 48.9% (n = 129) were males and 51.1% (n = 135) were females. Univariate analysis of groups showed no significant differences based on sex, age, duration of DM, and glycemic control levels (Table 1). Cases (those with DM and DFU) were more likely to have a history of cigarette smoking and intake of alcohol. Conversely, controls (those with DM only) were more likely to have comorbid conditions anemia, hypoproteinemia, dyslipidemia, hypertension, along with diabetic nephropathy and retinopathy. Cases had a higher prevalence of previous hypertension, whereas nonsignificant differences were observed with regard to coronary heart disease and stroke between the 2 groups (Table 1).

TABLE 1. Baseline Clinical Characteristics in Diabetes Mellitus Versus Diabetic Foot Ulceration

Half of the cases (n = 44/88) had a diabetic foot ulcer for more than 1 month; the mean duration of living with a diabetic foot ulcer was 1.60 (SD = 2.42) months. More than half (n = 51; 58.0%) had a single diabetic foot ulcer; 65 cases (73.8%) had a diabetic foot ulcer. Wagner grade ranged from 1 to 3. Almost three-fourths (n = 65) were treated for wound infection, 68.2% (n = 60) had diabetic neuropathy, and 64.8% (n = 57) had lower extremity arterial disease. In addition, 14 patients (16.0%) underwent amputation during data collection.

The Figure shows estimated survival rates (Kaplan-Meier curves) for all-cause and cardiovascular-related survival in the case and control groups. Table 2 lists causes of deaths for cases and controls. Cardiovascular disease was the leading cause of death in the sample. The proportion of CVD-specific deaths was similar in the case and control groups (25.6% in the case group vs 30.0% in the control group). In addition, 25.6% of deaths in the case group were caused by sepsis originating from the foot.

Figure. Kaplan-Meier survival curves for cases (DFU + DM) versus controls (DM alone). (A) Estimated survival rates for all-cause deaths. (B) Estimated survival curves for cardiovascular-related deaths. DFU indicates diabetic foot ulceration; DM, diabetes mellitus.

TABLE 2. Cause-Specific Death in Diabetes Mellitus Versus Diabetic Foot Ulceration

Cox proportional hazard regression analyses were completed to analyze the influence of DFU on all-cause and cardiovascular-related deaths. Univariate analysis indicated age, education level, work status, anemia, hypertension, stroke, diabetic nephropathy and retinopathy, peripheral neuropathies, lower extremity arterial disease, and DFU were predictors of all-cause death in diabetic patients. Analysis also revealed that age, work status, duration of DM, anemia, hypertension, stroke, lower extremity arterial disease, and DFU were predictors of CVD-related deaths. In contrast, multivariate analysis found that DFU, age, diabetic nephropathy, and hypertension were predictors of morbidity (Table 3). Next to age, the presence of DFU had the highest hazard ratio (HR) for all-cause and cardiovascular-related related deaths.

TABLE 3. Cox Multivariate Analysis of All-Cause and Cardiovascular Disease Death

The crude all-cause mortality rate in the case (DFU + DM) and control (DM alone) groups was 48.9% and 22.7%, respectively. The adjusted HR for the association between DFU and all-cause mortality was 3.126 (95% CI, 1.998-4.891; P = .000). The crude CVD all-cause mortality rate in the DFU and DM groups was 12.5% and 6.8%, respectively. The adjusted HR for the association between DFU and CVD mortality was 3.277 (95% CI, 1.392-7.715; P = .007).

DISCUSSION

This case-control study showed that patients with DM and DFU had higher risk for all-cause and cardiovascular-related deaths than a matched group of patients with DM alone. Research indicates that development of a diabetic foot ulcer is associated with an elevated risk for a lower extremity amputation rate and a high 5-year mortality rate.7 A review of literature from 1980 to 2013 found that the 5-year mortality rate of patients with DFU was about 40.0%.8 Other studies have reported 5-year cumulative all-cause mortality in DFU patients ranging from 13.7% to 45.8%.9-12 These findings are similar to ours that showed a 5-year all-cause mortality rate among patients with DM and DFU of 35.2%.

Study findings also indicated that DFU was associated with an increased risk of all-cause mortality when compared to patients with DM alone. Specifically, we found that the risk of all-cause mortality in patients DM and DFU was 3.126 times higher than in patients with DM alone. This finding is consistent with that of Junrungsee and colleagues, who reported an HR of 3.51 (95% CI, 1.03-11.96).13 Reported HR for all-cause mortality in patients with DFU ranged from 1.470 to 3.980.9,13-15 Brownrigg and associates14 completed a systematic literature review and meta-analysis and reported a crude death rate (per 1000 people per year) of persons with DFU versus those with DM and no foot ulcers as 99.9 versus 41.6 (RR = 1.89; 95% CI, 1.60-2.23). Another meta-analysis from China reported similar findings.16

Cardiovascular disease was the leading cause of death in our sample. We further found that the CVD-related mortality rate of the subjects was 8.7%, of which the all-cause mortality rate was 12.5% in the DFU group and 6.8% in the DM group. This result is consistent with findings reported by Pinto and colleagues10 and Brownrigg and coinvestigators,14 who reported a 5-year CVD-related mortality rate of DFU patients as 12.7% and 5.1% versus 7.3% and 1.1%, respectively. This finding also supports the 2016 WHO Global Diabetes Report that advocated for integration of DM and CVD prevention and management.17

The top 3 causes of death in the DFU group were CVD, foot infection, and kidney failure, respectively. Research clearly indicates that DM is associated with an increased risk for CVD- and cardiovascular-related death.18-29 These findings are consistent with studies conducted in China.19,24 For example, one study found that between 1991 and 2005, CVD was the leading cause of death in Chinese persons living with type 2 DM, accounting for 39.0% of all causes of death.19

Similar to the guidelines of the 2016 WHO Global Diabetes Report, research indicates that more can be done to manage DM and related CVD in China.30-32 For example, Young and colleagues33 showed that aggressive cardiovascular risk management improved survival of persons with DFU, resulting in a reduction in mortality from 48% to 27%, but more evidence was needed. Nevertheless, additional research is needed to determine the optimal management of DM and CVD in patient with and without DFU.

LIMITATIONS

Several limitations must be considered when interpreting study findings. Data were collected from records at a single facility, thus limiting generalizability of results. In addition, data were collected retrospectively, resulting in the need to match cases and controls as compared to a prospective cohort study that would have allowed a longitudinal evaluation of the effects of DFU on all-cause and CVD-related mortality in patients with DM.

CONCLUSION

A case-control study of 176 patients with DM found that presence of a diabetic foot ulcer was associated with an elevated all-cause and CVD risks. Our findings are consistent with prior research indicating an increased risk for CVD and deaths among persons living with DM. However, the mechanisms of how DFU increases the CVD death rate remained to be explored and we recommend additional research aimed at clarifying the association between DFU and CVD-related morbidity and mortality in particular. We also recommend foot and nail care clinicians continue their effort to prevent and effectively treat DFU in this vulnerable population.

ACKNOWLEDGMENTS

The authors acknowledge the following individuals: Yang Chuan and Xie Wen contributed to the conception of the study and critically reviewed the study proposal; Mai Lifang and Mao Xiaoqun assisted with the official help; and Qiu Liangzhi, Li Yongjie, and Zhu Lisi contributed significantly to the data collection and data analysis. The authors thank the Sun Yat-sen Memorial Hospital, Sun Yat-sen University, and extend the most sincere thanks to all.

REFERENCES