Home Adult site Impact of lumbar spinal stenosis on the incidence of metabolic syndrome in community-dwelling adults in the Aizu Cohort Study (LOHAS)

Impact of lumbar spinal stenosis on the incidence of metabolic syndrome in community-dwelling adults in the Aizu Cohort Study (LOHAS)


Study population

This prospective cohort study was based on Locomotive Syndrome and Health Outcomes in the Aizu Cohort Study (LOHAS), with a follow-up duration of up to 6 years. The LOHAS was designed in 2008 to study the risk of cardiovascular disease, quality of life, medical costs and mortality attributable to musculoskeletal dysfunction. The study recruited community-dwelling individuals from the cities of Minami-Aizu and Tadami in Fukushima Prefecture, Japan, who were enrolled in a national health insurance plan and who participated in annual health checkups ( specific health visits targeting people aged 13.14. The LOHAS baseline survey included 2725 participants. The inclusion criteria for this study were to have participated in LOHAS and to be aged

Basic survey

Special health checks and questionnaires were provided during the baseline survey. All assessments were conducted between April and July 2008. Each participant underwent all examinations on the same day. Special health checkups included an interview about medical history and lifestyle habits such as smoking and alcoholism; measurement of weight, height and blood pressure; physical examination; and blood biochemical tests for serum triglycerides, serum total cholesterol, serum low-density lipoprotein cholesterol, hemoglobin A1c, and other parameters, without prior instruction to fast. The baseline self-reported questionnaire for participants included information on the following characteristics: marital status, occupational status, dietary habits, physical activity, health-related quality of life, and locomotor dysfunction.

Metabolic Syndrome Definitions

Metabolic syndrome has been defined as the presence of at least three of the following risk factors15, identified during special or standard annual health checks: (1) enlarged waist circumference (≥ 80 cm in women or ≥ 90 cm in men); (2) increased triglyceride levels (≥ 150 mg/dL, or taking medication prescribed for high triglycerides); (3) low high-density lipoprotein cholesterol (

Main explanatory variable

The main explanatory variable was lumbar spinal stenosis, which was identified using a self-reported questionnaire, developed as an aid tool in the diagnosis of lumbar spinal stenosis.16. This questionnaire consisted of 10 items (Q1 to Q10). According to the clinical prediction rule, those with a total score ≥ 4 points in Q1–Q4 or a score ≥ 1 in Q1–Q4 and ≥ 2 in Q5–Q10 were identified as positive for the presence of lumbar spinal stenosis. The sensitivity and specificity of positive endpoints in a validation dataset were 84% and 78%, respectively.

Other Variables

During the special health check or health check, height and weight were measured and the body mass index (BMI) was calculated. We recorded data on age, sex (male or female), smoking status (current or never/former) and smoking habits (daily, sometimes, rarely or never). Data collected by questionnaires included information on employment (yes or no) and the number of metabolic syndrome components at baseline (0, 1 or 2). We also recorded the mental health domain score in the SF-36 health survey (a higher score indicates better mental health)17 and level of physical activity as assessed by the Japanese version of the International Physical Activity Questionnaire (IPAQ)18. The IPAQ is one of the most widely used physical activity assessment tools in the world, and questionnaire responses were categorized as low, moderate, or high activity, depending on the scoring program.19.

statistical analyzes

Basic demographic data has been summarized. Continuous variables are presented as mean and SD, and categorical variables as frequency and percentage. The you-test and chi-square test analyzed continuous and categorical variables, respectively. The follow-up was managed as a special or standard annual check-up. The endpoint of the present study was the development of metabolic syndrome or the last health check.

Univariate and multivariate Cox proportional hazard regression models were used to calculate the hazard ratio (RR) and 95% confidence interval (CI) for the incidence of metabolic syndrome during 6-year follow-up in association with baseline lumbar stenosis. After analysis not available, we performed a multivariate analysis including age, sex, smoking status, alcohol consumption, number of metabolic syndrome components at baseline, and mental health as confounders. In the sub-analysis, the incidence of each of the five components of metabolic syndrome and obesity (BMI ≥ 25 kg/m2) was analyzed in participants by a Cox multivariate proportional hazards regression model. Outcome variables were the incidence of each of the five components of obesity at annual special health checkups or health checkup results. The explanatory variables were the same components at inclusion, lumbar stenosis and possible confounding factors (age, sex, smoking status, alcohol consumption, number of components of the metabolic syndrome at inclusion and mental health). A subgroup analysis stratifying participants by age (

Selection bias and loss of information due to missing data in primary explanatory variables and covariates were managed by multiple imputations under the missing-at-random assumption as a sensitivity analysis20. The variables used in the regression or logistic regression analyzes were used to generate 20 sets of imputed data. Rubin’s rules were applied to combine estimates and standard errors21.

We also compared full case analyzes with multiple imputation results. All analyzes were performed using STATA, version 17.0 (StataCorp, College Station, TX, USA). All tests were bilateral.