Serum Lipid Profiles Are Associated With Disability and MRI Outcomes in Multiple Sclerosis

Bianca Weinstock-Guttman; Robert Zivadinov; Naeem Mahfooz; Ellen Carl; Allison Drake; Jaclyn Schneider; Barbara Teter; Sara Hussein; Bijal Mehta; Marc Weiskopf; Jacqueline Durfee; Niels Bergsland; Murali Ramanathan

Disclosures

J Neuroinflammation. 2011;8(127) 

In This Article

Methods

Study Population

Ethics Statement The study was approved by the University at Buffalo Human Subjects Institutional Review Board. The Institutional Review Board approval waived the requirement for informed consent.

Study Design Single-center, retrospective, longitudinal study.

Study Population The study population included consecutive patients, followed at the Baird MS Center, State University of New York, Buffalo, NY, with clinically definite MS patients according to the McDonald criteria[23] with available baseline EDSS assessment within ± 6 months of lipid profile testing and a follow-up EDSS assessment ≥ 6 months from the baseline clinical visit. Patients with CIS and neuromyelitis optica were not included.

The collected data included demographic and clinical information, statin use history, height and weight and fasting lipid profile laboratory values: HDL, LDL, triglycerides, total cholesterol and cholesterol to HDL ratio.

The exclusion criteria consisted of: any relapse with corticosteroid treatment at the time or within one month preceding study entry or MRI examination, pre-existing medical conditions known to be associated with brain pathology (e.g., neurodegenerative disorders, cerebrovascular disease, positive history of alcohol abuse, etc.), and insufficient quality of the MRI scan for quantitative analysis.[24]

MRI Analysis

Quantitative MRI analysis obtained within ± 3 months from the baseline clinical visit (yielding EDSS and fasting cholesterol levels) was available for 210 of 492 patients at baseline. MRI image analysis was conducted at the Buffalo Neuroimaging Analysis Center using approaches previously described.[25,26] MRI analysts were blinded to lipid profile and clinical status. The standardized acquisition and analysis methods for obtaining contrast-enhancing lesion volume (CE-LV), CE lesion number (CEL number), T2-LV, T1-LV and brain parenchymal fraction (BPF) are detailed in Additional File 1.

Data Analysis

SPSS (SPSS Inc., Chicago, IL, version 15.0) statistical program was used for all statistical analyses.

One-way ANOVA followed by post-hoc independent sample t-tests were used to test for differences in means of continuous demographic variables such as age, age of onset, and disease duration. The 2 test was used for analysis of count variables for categorical data and the Fisher exact test was used where appropriate.

The MS Severity Scale (MSSS) was calculated from the EDSS and disease duration values using software downloaded from http://www-gene.cimr.cam.ac.uk/MSgenetics/GAMES/MSSS/Readme.html. The global reference data set provided with the software was used for calculations.

The difference between EDSS at follow-up and EDSS at baseline was analyzed as the dependent variable in regression analysis with gender, disease duration at baseline EDSS, EDSS at baseline, time difference between follow-up and baseline EDSS assessments, statin use and a lipid profile variable of interest (either HDL, LDL, triglycerides, total cholesterol or cholesterol to HDL ratio) as predictor variables. The difference between MSSS at follow-up and MSSS at baseline was analyzed in the same manner as the EDSS; however, the MSSS at baseline was included as a predictor in place of EDSS at baseline and the disease duration was not included as a predictor variable. Similar regression analyses were also conducted in the subset of patients who were not on statins to assess the contributions of lipid profile variables in the absence of statin treatment.

Baseline EDSS was dichotomized into two groups based on EDSS < 4.0 and ≥ 4.0. The baseline EDSS groups were analyzed using logistic regression with sex as a factor and disease duration and lipid profile variable of interest.

The CE-LV, T2-LV and T1-LV data were normalized by cube-root transformation to reduce skew. The cube-root-transformed T2-LV and T1-LV values were analyzed as dependent variables using multiple linear regression. The presence/absence of CE lesions (CEL) was analyzed with logistic regression and the CEL number was analyzed with Poisson loglinear regression and the transformed CE-LV values were analyzed with Tweedie regression.[27] All regression MRI analyses included sex, disease duration at time of MRI, statin use, and a lipid profile variable of interest (either HDL, LDL, triglycerides, total cholesterol or cholesterol to HDL ratio) as predictor variables. Regression analyses were also conducted in the subset of patients who were not on statins to assess the contributions of lipid profile variables in the absence of statin treatment.

To correct for the multiple testing involved, a conservative Type I error level of 0.01 was used to assess significance; a trend was assumed if the Type I error level ≤ 0.10.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....