Safety assessments were completed at every visit, and there were no missing cases. The results of the 2008 pilot study were blinded until interim analysis in 2009, revealing a non-significant trend towards a reduction in Indinavir sulfate fatigue in the patients on the active drug. The hospital pharmacy could produce Methimazole placebo in syringes identical to the active drug in 2010, with a durability that allowed the patients to receive a 14 days supply of the assigned treatment at week 0 and week 2. The patients were trained to self administer the injection of either active drug or placebo, and administered the first injection under supervision at week 0. The patients registered the time of every injection, and the registration form and the empty syringes were collected at the study visit at week 2 and week 4. Safety assessments were completed at every visit. The primary outcome measure was a group-wise comparison of fatigue scores at week 4, adjusted for baseline values. Secondary outcome measures were change in fatigue scores within each treatment group during the study, and safety and tolerability of anakinra in pSS. The proportion of patients achieving a 50% reduction in fatigue from baseline to week 4 in the active treatment group, as compared to the placebo group, was analysed as a posthoc outcome. We regarded this as a robust measure as a 20% reduction in fatigue VAS score had been used in another recent study on rituximab-treated pSS patients. This randomised, double-blind, placebo-controlled trial of IL-1 blockade did not show a significant reduction in fatigue after treatment with IL-1 inhibition based on the primary endpoint analysis, while the post-hoc analysis indicate a possible positive effect. There are several possible explanations for this finding: One explanation is that IL-1 does not substantially influence fatigue in pSS, and that other factors leading to fatigue are more important in this setting such as depression, sleep-disorders and autonomic dysfunction. As our intention was to investigate somatic factors associated with fatigue, patients with moderate to serious depression were excluded, and only one of the patients used prescription drugs against insomnia. Autonomic dysfunction was not evaluated systematically, but none of the patients spontaneously reported these types of symptoms. Therefore, we believe these factors are less important in this study. Another explanation is that the small number of subjects led to low statistical power and therefore we could not confirm an actual effect of IL-1 inhibition on fatigue. In the post-hoc analysis, half of the patients in the active drug group did report a 50% reduction in fatigue VAS score, compared to one patient in the placebo group. Although not supported by the primary outcome measure, this may indicate that IL-1 inhibition influences fatigue in patients with pSS. This observation is in line with our primary hypothesis that a blockade of the IL-1 receptor should lead to reduction of fatigue.