A wearable perspiration biosensor for continuous biomarker monitoring in inflammatory bowel disease: Feasibility, concordance, and exploratory temporal dynamics
In brief
Wearable sweat sensor had 95% to 97% negative predictive value for inflammation
In 52 people with inflammatory bowel disease, sweat-based tests had high negative predictive values against standard blood and stool inflammation tests, but positive results were less reliable. Sweat C-reactive protein tracked blood levels moderately, while calprotectin tracked stool levels only in ulcerative colitis, not Crohn's disease. The early findings support an adjunct, not a replacement, and need confirmation against endoscopy.
- Journal
- Clinical and translational medicine (Q1)
- Published
- 1 November 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 4, Very Low (CEBM 4)
- Authors
- Annapoorna Ramasubramanya, Kai-Chun Lin, Sarah Shahub, Alexandra R McDermott, Zachary D Fine, Robert P Hirten, et al.
- PMID
- 43034281
- DOI
- 10.1002/ctm2.70896
Why clinicians should know about it
- Picked for Biochemistry (medical) (paper of the day, 20 November 2026): Evaluates wearable biomarker agreement and analytical performance against standard tests
Abstract
BACKGROUND: Serum C-reactive protein (CRP) and faecal calprotectin, used to monitor inflammatory bowel disease (IBD) between endoscopies, require intermittent blood or stool sampling. We assessed whether a skin-worn perspiration sensor could measure CRP and calprotectin continuously; TNFα and IL-6 were exploratory analytes. METHODS: At three US centres, perspiration biomarkers were recorded continuously for 3-30 days in 52 IBD patients (27 ulcerative colitis, 24 Crohn's disease, 1 indeterminate colitis) and 19-32 days in 9 controls, with serum CRP and faecal calprotectin sampled at baseline and study end. Agreement was assessed using correlation and Bland-Altman analysis; ROC-derived cutoffs matched serum CRP 5 mg/L and faecal 250 µg/g. Circadian rhythmicity and ultradian (4-12 h) power were compared among biochemically active IBD (serum CRP > 5 mg/L; n = 15), inactive (n = 37), and controls. RESULTS: 47 of 52 IBD participants (90.4%) completed the protocol. Across 99 paired observations, perspiration CRP correlated with serum CRP (r = .52-.58); perspiration and faecal calprotectin correlated in ulcerative colitis (r = .59-.62) but not in Crohn's disease. A perspiration CRP cutoff of 1551 pg/mL (serum 5 mg/L) gave 85.7% sensitivity and 89.4% specificity; a calprotectin cutoff of 385 (µg/g)·min (faecal 250 µg/g) gave 92.3% and 67.9%. Negative predictive values were high (95%-97%), positive modest (57%). Circadian MESOR and amplitude of both markers were higher in active than inactive IBD. Ultradian power differed by group for calprotectin (Kruskal-Wallis p = .0005) and CRP (p = .0037)-highest in active disease, lowest in controls-but the active-inactive difference was non-significant after false-discovery-rate correction. CONCLUSIONS: Perspiration CRP agreed moderately with serum CRP, and perspiration calprotectin with faecal calprotectin in ulcerative colitis only. Their high negative predictive values support the study as a non-invasive adjunct to, not a substitute for, standard-of-care testing. All findings are preliminary, and the temporal-dynamics results are hypothesis-generating, needing confirmation in a larger, prospectively designed, endoscopy-anchored cohort.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.