BackgroundThe different antirheumatic drugs reduce inflammation in RA patients, causing alterations in cholesterol levels (mainly decreasing LDL and increasing HDL cholesterol levels), but HDL and LDL structure and function predict cardiovascular disease better than LDL and HDL cholesterol levels.ObjectivesThe investigation of the qualitative changes of LDL and HDL lipoprotein subclasses in RA patients who are good responders, depending on the anti-inflammatory treatment (sDMARDs vs anti-TNFa+sDMARDs) and their in between associations.Methods85 patients ((89%) 76females) with established RA (mean disease duration ≥5 yrs), mean age 57 yrs (SD: 12yrs), without known cardiovascular disease, D/M and thyroid disorders, on sDMARD (MTX, LEF, SSZ, low dose prednisolone or combination) and naïve to biologic treatment, were divided into two groups: the 1st one of 43 patients who had DAS28>3.2 and were given in addition a TNFα inhibitor (23 patients had golimumab and 20 patients had certolizumab-pegol) for at least 54weeks (mean treatment duration 18 months) with good clinical response according to Eular and the 2nd group of 42 patients (disease control group) with DAS28≤3.2,who continued on sDMARDs and followed closely the same time period so as to ensure, as well, good clinical response without biologic therapy. Plasma electrophoresis was performed in two time points, before and after anti-TNFα administration and on the same time for the control group, with non-denaturing polyacrylamide gel (ND-PAGE) for size-based separation of lipoprotein subclasses, a standard laboratory technique that indentify various HDL subspecies separable on the basis of average diameter/size into 6 distinct subclasses (HDL-1, HDL2a, HDL2b, HDL3a, HDL3b, HDL3c), as well as LDL lipoprotein into 2 subfractions (LDL-B, LDL-A.)ResultsThe RA patients on DMARDs had a percent reduction of LDL-B subfraction by 11.9% (p=0.014), but a significant increase of HDL-3c subclass by 2.38% (p=0.049). In a multivariate model of stepwise logistic regression the after (treatment)-LDL-B subfraction in these patients was found to have significant positive association only with pro-LDL-B subfraction (odds ratio: 10.95, %CI (1.59–62.79)). In the RA patients who took TNFα inhibitors was observed a prominent percent decrease of the more lipoprotein subclasses: LDL-B by 2.33% (p=0.005), LDL-A by 2.33% (p=0.0001), HDL3a by 11.63% (p=0.072) and HDLc by 4.65% (p=0.035). Accordingly, in stepwise logistic regression the after-(treatment) LDL-B subfraction had a significant positive association only with pro-LDL-B subfraction (odds ratio: 12.91,95% CI (2.2–75.83)), as well as, after-LDL-A with pro-LDL-A subfraction (odds ratio:36.16, 95% CI (5.05–258.89)), whereas the reduced after (treatment)-HDL3c subclass was significantly positively associated with pro-HDL3a subclass (odds ratio:8.15, 95% CI (1.04–63.65)) and after-HDLa subclass (odds ratio:22.09, 95% CI (3.56–137.18)).ConclusionsThe discrepancy of qualitative modification of lipoprotein subclasses (LDL and HDL) after treatme...