Survival of Patients Following First Diagnosis of Diabetic Foot Complications: A Nationwide 15-Year Longitudinal Analysis
Chia-Hung Lin 1,2,3, David G. Armstrong 4, Pi-Hua Liu1,5, Cheng-Wei Lin1, Chung-Huei Huang1 and Yu-Yao Huang1,3,6*
Background and Aims: The long-term survival in people with type 2 diabetes following first diagnosis of diabetic foot complications (FDDFC) is unclear. The object is to evaluate the mortality rate in subjects with type 2 diabetes following FDDFC and the impacts of the major cardiovascular comorbidities.
Methods: Nationwide data were analyzed for subjects with T2D and DFC between 2003 and 2017 according to ICD-9 coding. DFC was defined with the codes of ulcers, infections, or severe peripheral artery disease that required intervention (PAD) to mimic the real world diagnosis. Criteria of FDDFC were preceded by a period without any DFC for at least 5 years. Major cardiovascular comorbidities: established PAD and cardiovascular diseases (CVD: including coronary heart disease (CHD), stroke, or heart failure) before the index date as well as lower-extremity amputations (LEA) at the index episode were analyzed.
Results: Among 300,115 subjects with DFC, a total of 103,396 patients had FDDFC. The mean 5-year survival rate of these subjects was 81.05%. Using subjects without associated major cardiovascular comorbidity as baseline, the adjusted hazard ratios (aHR) were1.43 (95% confidence interval 1.38–1.49) in group PAD-/CVD+, followed by 1.70 (1.59–1.80) in PAD+/CVD- and 1.98 (1.89–2.08) in PAD+/CVD+. The aHR was further increased in patients with PAD who additionally had heart failure (3.77, 3.50–4.05), stroke (2.06, 1.95–2.18), or CHD (1.89, 1.79–2.00). Subjects with PAD rather than other CVD were associated with LEA at FDDFC. Patients with major LEA (above the ankle) at FDDFC episode had lower 5-year survival rate (65.01%) followed by those with minor LEA (72.24%) and without LEA (81.61%).
Conclusions: Cardiovascular comorbidity as well as LEA status at the event of FDDFCs were both associated with patient survival outcomes. Earlier identification of this large population could lead to higher survival rates.