Prevalence、Incidence and Survival Analysis of Somatic andutonomic Diabetic Neuropathy in Community-Based Studies
Date Issued
2009
Date
2009
Author(s)
Hsu, Wei-Chih
Abstract
Background：ecause the prevalence of diabetes is increasing, the diabetes-related complications, including macrovascular and microvascular complications, lead to significant medical and economic burdens to patients, their families and whole society. Therefore, early detection and identification of complications among diabetic patients in community and intervention for modification of risk factors are important. Regarding to microvascular complications, the epidemiological studies of neuropathy have been rarely addressed in the literature. urpose：his thesis aimed to conduct a series of community- based epidemiological studies for diabetic neuropathies, including somatic sensorimotor and autonomic neuropathy. These studies encompassed prevalence rate, incidence rate, important correlates affecting the occurrence of neuropathies and the sequel to mortality, survival analysis and computer simulation of diabetic neuropathies in the community setting. aterials and Methods：atients enrolled in this thesis were from two community-based integrated screening programs in Keelung and Matsu. The Keelung cohort included a large numbers of participants, so a two-stage design was performed for identifying subjects with diabetic neuropathy. The first step in the two-stage study design used the Neurological Symptom Score (NSS) questionnaire to identify positive cases. These positive cases were further confirmed by nerve conduction tests in the second stage. A validation study was conducted for detecting the sensitivity, specificity of questionnaire. Another 587 diabetics were selected for main study. Subjects who screened positive in the first stage were referred to nerve conduction test for confirmation. Potential risk factors were assessed, including fasting plasma glucose, HbA1c, body mass index, retinopathy, age, sex, diabetic duration, total cholesterol, triglyceride, hemoglobin and life styles.he second cohort, Matsu, is a small community. Therefore, all potential subjects with somatic sensorimotor and autonomic neuropathy were investigated. The somatic neuropathy was diagnosed by nerve conduction tests and the autonomic neuropathy was confirmed by 5-min resting electrocardiograph for heart rate variability.his thesis consists of ：（1）investigating prevalence of somatic and autonomic neuropathy at Matsu cohort; besides, Bayesian analysis, using validation data and publications data as prior, was employed to estimate the prevalence rate; （2）incidence rate of diabetic neuropathy was estimated from those who were found to be free of diabetic neuropathy at screening programs in 2001. Information about the time and diagnosis of peripheral nerve disorders at outpatient clinics after screen activity was obtained till the end of 2004 from National Health Insurance;（3）correlates of somatic and autonomic neuropathy were investigated;（4）survival analysis was performed on 708 diabetic patients and 326 diabetic patients who accepted nerve conduction study in 2001. Those patients were linked to National Health Insurance till the end of 2006, information of date and cause of death can be gathered for deceased. Kaplan-Meier test was done by the presence of somatic neuropathy or not. Important correlates and life styles for predicting all-cause and diabetes-related mortality were also studied;（5）Monte-Carlo simulation was used to predict the disease burden. The disease course of the 1000 patients in 10 years was randomly assigned. esults：1）A number of 143 persons was found to have high fasting plasma glucose (>110 mg/dl) or with past history of type 2 diabetes in Matsu cohort. For 133 subjects who accepted NCS, 12 subjects (12/133=9.0%) were categorized into definite somatic neuropathy, 27 subjects 27/133=20.3%) were probable somatic neuropathy and 94 subjects (94/133=70.7%) were classified as no somatic neuropathy. Among 118 subjects who completed validated heart rate variability test, results of SDNN consist of : 17（17/118=14.4%）diabetics was categorized into low SDNN level, 64 （54.2%） patients were middle SDNN and 37（31.4%）patients were high SDNN level. （2） Among of the 326 diabetic patients who accepted nerve conduction study, 218 patients were classified as no somatic neuropathy in 2001. After linking to outpatient clinics dataset of National Health Insurance, 28 subjects were diagnosed as having peripheral nerve disorders till the end of 2004. Three of these 218 patients died, and 160 patients remained asymptomatic for peripheral neuropathy. Besides, a number of 27 had diagnosis of peripheral nerve lesion already before the screening date. The incidence rate of diabetic neuropathy was 4.8 per hundred person years. After adjusted for the 27 already existed cases before screening among 218 cases, the incidence rate was 5.5 per hundred person years. Because prevalence rate of diabetic neuropathy was 28.46~36.30%, the duration of developing symptomatic diabetic neuropathy was 7.2-10.3 years.（3）Correlates associated with somatic sensorimotor and autonomic neuropathies in pre-diabetic and diabetic subjects are different. In multivariate analysis, systolic blood pressure (OR=1.07 ; 95% CI=1.00-1.14) and fasting blood glucose (OR=1.07; 95% CI=1.03-1.11) accounted for somatic sensorimotor neuropathies where as no significant factors were found in autonomic neuropathy group. A total of 93 subjects died among 708 diabetic patients after 5-year follow up. Among these 708 diabetic patients, 326 patients who accepted nerve conduction study for screening somatic neuropathy, a total of 44 patients died. The statistically significant correlates for all-cause mortality in 93 deceased were: age（HR=1.06），male sex（HR=0.38），BMI（HR=0.90），BUN （HR=0.92），creatinine（HR=6.92），prior cardio-and cerebrovascular diseases（HR=2.25）in multivariate analysis. For diabetes-related death, the statistically significant correlates were: age（HR=1.06），creatinine level（HR=7.97），prior cardio- and cerebrovascular diseases （HR=2.99）. Targeting the 326 subjects with nerve conduction tests, diabetic neuropathy is the strongest predictor for all cause mortality and the second strongest predictor for diabetes-related mortality, next to prior cardio-/cerebrovascular disease in univariate analysis. In selected model, presence of diabetic neuropathy （HR=4.38）, fasting glucose（HR=1.01） and creatinine level （HR=13.23）and serum cholesterol（HR=0.99） remained statistically significant for all-cause mortality. For disease-related mortality, presence of neuropathy（HR=5.69）, fasting glucose（HR=1.01）, hemoglobin（HR=0.70）, serum BUN （HR=0.84）and creatinine（HR=26.99） levels are statistically significant. （4）Kaplan-Meier curves during the 5 years of follow up showed that diabetic neuropathy was a statistically significant factor for all-cause （p<0.001）and diabetes-related mortality（p<0.001）by log-rank test. （5）By simulated cohort, 59.7% of diabetic patients will progress to somatic neuropathy after 10-year follow up. For patients with somatic neuropathy, 14.7% of diabetic patient will die of diabetes-related causes and 9.2% of these will die of non-diabetes related causes. For patients without somatic neuropathy, 3% of these will die of diabetes-related causes and also 3% will die of non-diabetes related causes. onclusion:he present thesis provided an insight into estimating the prevalence rates, incidence rate, survival analysis, important correlates and medical burdens of diabetic neuropathy, including both types of somatic sensorimotor and autonomic neuropathies, by using a population-based screening program. We finds that the prevalence rates of somatic neuropathy in type 2 diabetes were approximate 30% in Keeling and Matsu community and autonomic neuropathy even doubled this figure. The incidence rate of developing diabetic neuropathy was 4.8-5.5 per hundred-person years. Correlates of both types of neuropathy and for predicting mortality were different. Somatic neuropathy is an independent risk factor for all-cause and diabetes-related mortality. After 10-years’ time in simulation study, prevalence rate of somatic neuropathy will be doubled among diabetic patients. The high prevalence rate will cause great medical and economic burdens on patient, family and whole society. Prevention of diabetic complications, and diabetic neuropathy in particular, were dependent on early detection, monitor and control blood sugar strictly and modification of life style in these vulnerable subjects. Screening programs for diabetic neuropathy, including cost-effective and cost-benefit analysis, should be carried out in the future for identifying the benefits in public health.
Subjects
Diabetic complication
autonomic neuropathy
peripheral neuropathy
epidemiology
SDGs
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-98-D93842010-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):8003013a79f0d2a6f2e4131913790263