The TARIF study included 40 adult patients with quiescent IBD and chronic fatigue consecutively from the outpatient clinic at Aarhus University Hospital, Denmark. Eligible patients had had a diagnosis of IBD for more than 12 months and had disease in remission. Fatigue severity were assessed using the Inflammatory Bowel Disease-Fatigue Questionnaire (IBD-F) section I. Patients with a fatigue score > 12 and fatigue duration > 6 months were included (Czuber-Dochan et al. 2014; Bager et al. 2018). Patients with anemia, iron deficiency, folate acid deficiency, vitamin-B12 deficiency, or vitamin-D deficiency were excluded. We excluded pregnant women and patients with co-morbidity that could explain a high level of fatigue (e.g. cancer, chronic kidney disease, chronic heart disease, diabetes). Patients were allocated 1:1 to either; (1) high-dose oral thiamine for 4 weeks, followed by 4 weeks of washout, followed by 4 weeks of oral placebo; or (2) oral placebo for 4 weeks, followed by 4 weeks of washout, followed by 4 weeks of high-dose oral thiamine.

A comparison group was included parallel to inclusion in the intervention study. We included 20 controls with quiescent IBD and no fatigue (fatigue score ≤ 12 on the IBD-F scale), matched on gender, age, and IBD disease type (Fig. 1, Table 1). Blood samples were drawn, and the patients answered questionnaires regarding fatigue at each study visit (baseline, week 4, week 8, and week 12). The controls had only drawn blood samples and answered fatigue questionnaires at baseline.

TARIF trial profile. EOT, End of trial; * Controls had quiescent IBD and a fatigue score ≤ 12 on the IBD-F scale; ** High dose (600–1800 mg day) for 4 weeks

We assessed fatigue severity using the IBD-F questionnaire section I. It yields a score between 0 and 20 where higher score indicates more fatigue. Patients with a fatigue score > 12 were classified as being fatigued (Czuber-Dochan et al. 2014) as 12 was equivalent to the 95th percentile for fatigue reported by a background population (Bager et al. 2018). A clinical important improvement of fatigue was defined as ≥ 3 points decrease in the IBD-F score.

The study was conducted according to the principles of the Helsinki declaration, and the protocol and informed consent forms were approved by the Ethics Committee in Central Denmark Region (j.no. 64207) and the Danish Medical Agency (EudraCT j.no. 2018-002324-17). Written informed consent was obtained from all participants.

Blood samples were collected at the time points described above. All plasma samples were stored at − 80 °C and analyzed in one batch after study completion. Plasma were analyses for relevant B-vitamins and metabolites (listed in Table 1) by Bevital AS, Norway (http://bevital.no) (Midttun et al. 2009).

Baseline data from fatigued patients and non-fatigued patients were compared. Changes in fatigue score before and after actual high dose thiamine treatment were calculated. Data from the group of patients with ≥ 3 points decrease in fatigue were compared to the group with < 3 points decrease.

Data were analyzed by nonparametric statistics, using the Mann–Whitney ranked sum test. Data are presented as the median and interquartile range (IQR). P-values < 0.05 were considered statistically significant. Data analysis was conducted in Stata (version 18.0, StataCorp, College Station, Texas).