Remote sampling of biomarkers of inflammation with linked patient generated health data in ... rheumatic & musculoskeletal diseases, 2022, Druce et al

Full title: Remote sampling of biomarkers of inflammation with linked patient generated health data in patients with rheumatic and musculoskeletal diseases: an Ecological Momentary Assessment feasibility study

Abstract

Background
People with rheumatic diseases experience troublesome fluctuations in fatigue. Debated causes include pain, mood and inflammation. To determine the relationships between these potential causes, serial assessments are required but are methodologically challenging. This mobile health (mHealth) study explored the viability of using a smartphone app to collect patient-reported symptoms with contemporaneous Dried Blood Spot Sampling (DBSS) for inflammation.

Methods
Over 30 days, thirty-eight participants (12 RA, 13 OA, and 13 FM) used uMotif, a smartphone app, to report fatigue, pain and mood, on 5-point ordinal scales, twice daily. Daily DBSS, from which C-reactive Protein (CRP) values were extracted, were completed on days 1–7, 14 and 30. Participant engagement was determined based on frequency of data entry and ability to calculate within- and between-day symptom changes. DBSS feasibility and engagement was determined based on the proportion of samples returned and usable for extraction, and the number of days between which between-day changes in CRP which could be calculated (days 1–7).

Results
Fatigue was reported at least once on 1085/1140 days (95.2%). Approximately 65% of within- and between-day fatigue changes could be calculated. Rates were similar for pain and mood. A total of 287/342 (83.9%) DBSS, were returned, and all samples were viable for CRP extraction. Fatigue, pain and mood varied considerably, but clinically meaningful (≥ 5 mg/L) CRP changes were uncommon.

Conclusions
Embedding DBSS in mHealth studies will enable researchers to obtain serial symptom assessments with matched biological samples. This provides exciting opportunities to address hitherto unanswerable questions, such as elucidating the mechanisms of fatigue fluctuations.

Open access, https://bmcmusculoskeletdisord.biomedcentral.com/articles/10.1186/s12891-022-05723-w

 

"Second, we selected CRP as our measure of inflammation here because a) it is a measure typically used in clinical assessments and research studies and b) analysis of CRP is cost-effective in a feasibility study such as ours. While we have shown that it is possible to extract CRP values using DBSS, we also showed that there was little variance in CRP despite high variance in fatigue. We note that this, in conjunction with our self-selection recruitment process, may suggest that (particularly for RA participants) we have recruited only those who are healthy and who have well controlled disease. However, it may also suggest that alternative fatigue-specific inflammatory markers (e.g. TNF-α, IL-1, IL-6 and IFN-γ [18, 32,33,34,35,36]) may better account for variation in fatigue, and we have not ascertained how viable DBSS is for their extraction. However, there is no plausible reason why this method of sample collection could not be used to extract other potential markers in a larger cohort in the future."

Reference 33 is to Cytokine signature associated with disease severity in chronic fatigue syndrome patients, 2017, Montoya et al, https://www.pnas.org/doi/full/10.1073/pnas.1710519114

 

The debate will continue because in BPSland no one trusts results that disprove a beloved belief. Because debates are for entertainment. Experts follow the evidence and the evidence for "mood", whatever is meant here, has always been "this dude says so and this other dude agrees somewhat".

 

Three of the authors work for a company with interests in e-data.
This looks like a very poor quality study and it is disappointing to see Versus Arthritis involved yet again. It isn't even clear to me what the point of this was supposed to be.

The idea that if CRP shows nothing it is worth trying cytokines is laughable. CRP is far easier to measure in RA and none of them will be abnormal in the other groups.

 

Apologies for humour here, but I pictured "remote sampling" as a lab tech using a VERY long needle to gather a blood sample.

 

My main reason for posting was that I thought that the study design was of interest, in that it provides a seemingly successful model in how something similar might be done in pwME. The question as to whether we can measure something that might be of value using this model I'll leave to others to answer. Lactate perhaps?