Impact of individual microvascular disease on the risks of macrovascular complications in type 2 diabetes: a nationwide population-based cohort study

Data source

The Taiwanese government implemented the National Health Insurance (NHI) in 1995 to provide a mandatory comprehensive medical care plan. About 99% of 23 million citizens were enrolled in the NHI program by 2000 [11]. Patient information obtained during medical visits, including age, gender, residential area, diagnosis, procedures, and hospitalizations, is recorded in the National Health Insurance Research Database (NHIRD). Disease diagnosis of the NHIRD is according to the International Classification of Diseases, Ninth/Tenth Revision, Clinical Modification (ICD-9/10-CM). NHIRD has a link with the National Death Registry to substantiate the cause of death. This study identified patients from the NHIRD. All information from patients and care providers was de-identified and encrypted before release to protect individual privacy. This study was approved by the Research Ethics Committee of China Medical University and Hospital (CMUH104-REC2-115-CR4), and the Research Ethics Committee approved a request to waive informed consent.

Study population and procedures

We identified patients with newly diagnosed type 2 diabetes from January 1, 2009, to December 31, 2018, and followed them up to December 31, 2019. The diagnosis of T2D was based on the ICD codings (Additional file 1: Table S1), with at least 3 outpatient visits within 1 year or one hospitalization [12]. Exclusion criteria were as follows: (1) missing information on age or sex; (2) age below 18 or above 80 years; (3) previous diagnosis of type 2 diabetes before January 1, 2009; (4) diagnosis of type 1 diabetes; (5) diagnosis of chronic kidney disease, dialysis, diabetic retinopathy, or diabetic neuropathy before the diagnosis of T2D; (6) diagnosis of coronary artery disease (CAD), stroke, atrial fibrillation, or heart failure before the index date; (7) death or follow-up for less than 180 days after the index date.

Patients with T2D diagnosis within one year were identified and categorized as (1) without microvascular disease, (2) diabetic kidney disease, including proteinuria, diabetic and hypertensive nephropathy, chronic kidney disease, renal failure, dialysis and renal replacement [13, 14], (3) diabetic retinopathy, including background and proliferative diabetic retinopathy, macular degeneration, retinal edema and detachment, vitreous hemorrhage, and vision loss [15], (4) diabetic neuropathy, including mononeuropathy and polyneuropathy, cranial nerve and peripheral nerve palsy, autonomic neuropathy, neuralgia, and Charcot’s arthropathy [16, 17], (5) diabetic kidney disease and retinopathy, (6) diabetic kidney disease and neuropathy, (7) diabetic retinopathy and neuropathy, (8) diabetic kidney disease, retinopathy, and neuropathy (Additional file 1: Table S2). We defined the 366th day after the diagnosis of type 2 diabetes as the index date of this study and assessed the outcomes during the follow-up period.

Demographics, comorbidities, and medications

The study included the following variables: age, gender, obesity, smoking, alcohol-related disorders, hypertension, dyslipidemia, peripheral arterial disease (PAD), chronic obstructive pulmonary disease (COPD), liver cirrhosis, connective tissue diseases, cancers, psychosis, depression, dementia, Charlson Comorbidity Index (CCI) score [18], antidiabetic agents, antihypertensive drugs, statin, and aspirin.

The outcomes of interest

The outcomes of interest were the development of CAD, stroke, heart failure, and cardiovascular death. Stroke, CAD, and heart failure were diagnosed by at least two outpatient visits within 1 year or one hospitalization during the study period [19, 20]. The adjudication of cardiovascular death was via the link with the National Death Registry. Each patient was tracked to the occurrence of outcomes, discontinuation of the NHI program, or the study observation end date of December 31, 2019, depending on which occurred first.

Statistical analyses

Multivariable-adjusted Cox proportional-hazard models were used to measure the risk of outcomes between patients with and without microvascular disease. Propensity score matching was used to construct matched pairs of patients with diabetic kidney disease and retinopathy, diabetic kidney disease and neuropathy, and diabetic retinopathy and neuropathy. We adopted the non-parsimonious multivariable logistic regression to appraise the propensity score for every patient, and 42 clinically relevant variables (Additional file 1: Tables S4–S6) were determined as independent variables. We adopted the nearest-neighbor algorithm to identify matched pairs and considered a standardized mean difference (SMD) < 0.1 to be a negligible difference between the matched pair of patients. We have performed subgroup analyses to see the influence of one, two or three microvascular diseases on composite major adverse cardiovascular events (CAD, stroke, heart failure and cardiovascular death) in the subgroups of patients with and without PAD, with and without insulin, aspirin and statin use.

The analyzed results were shown as hazard ratios (HRs) and 95% confidence intervals (CIs) for the compared groups. The two-tailed P value of less than 0.05 was considered statistically significant. SAS v9.4 (SAS Institute, Inc., Cary, NC, USA) was used for analysis.

留言 (0)

沒有登入
gif