Background We previously observed an association between systemic lupus erythematosus (SLE) disease activity and heart rate variability (HRV) with a preliminary observation of consistency between these two measures when disease activity changed between two timepoints (1).
Objectives To prospectively test the hypothesis that HRV reflect longitudinal changes in clinical status and patient reported outcomes.
Methods The current project evaluated HRV measurements using a 5 min ECG in SLE patients who completed a minimum of 2 visits in an ongoing clinical trial. HRV parameters were calculated in the time domain (RMSSD and pNN50) and the frequency domain [high frequency (HF) as well as low frequency to high frequency (LF/HF) ratio]. A mixed effects linear model, with generalized estimating equations to account for clustering of visits for each patient, was used to compare changes in HRV between paired visits and to examine linear associations between HRV parameters and clinical scores. All models were adjusted for baseline values of each HRV parameter.
Results Forty nine patients (age 44.9±11.7, 46 female), followed in a total of 413 paired visits (median time between visits 1 month), were included. Global BILAG score was inversely associated with RMSSD (regression coefficient β=-1.39±0.01; p<0.0001). BILAG, SLEDAI and PGA were directly associated with the LF/HF ratio (β=0.96±0.02; p<0.0001, 0.42±0.10; p<0.0001 and 0.83±0.09; p<0.0001, respectively). Changes in BILAG were inversely associated with changes in RMSSD and pNN50 (β=-7.0±1.9; p=0.003 and -1.6±0.04; p<0.0001, respectively). BILAG changes were also directly associated with changes in the LF/HF ratio (β=0.78±0.05; p<0.0001). Categorical improvement, defined as ≥1 letter grade improvement in BILAG and no new BILAG A or B scores, occurred in 77 (19%) visit pairs (group 1) and either no improvement or worsening in 335 (81%) group 2. RMSSD and HF increased in group 1 compared to group 2 (group difference=-33.3±10.1; p=0.001 and -30.9±4.1; p<0.0001, respectively), and the LF/HF ratio decreased in group 1 compared to group 2 (group difference=3.1±0.8; p=0.002).
Changes in Physical Component Summary (PCS) of SF36v2 were inversely related to changes in SLEDAI and PGA (β=-0.39±0.14; p=0.006 and -0.19±0.02; p<0.0001, respectively). Changes in Mental Component Summary (MCS) were inversely related to changes in BILAG, SLEDAI and PGA (β=-0.23±0.07; p=0.0001, -0.31±0.10; p=0.002 and -0.08±0.03; p=0.008, respectively). PCS was related to HF (β=0.67±0.28, p=0.01) whereas MCS was inversely related to the LF/HF ratio (β=-0.11±0.03, p=0.0001). Changes in PCS were related to changes in pNN50 (β=0.21±0.05, p=0.0001) and changes in MCS were related to changes in HF (β=1.57±0.18; p<0.0001).
Conclusions Changes in HRV reflect changes in clinical status and patient reported outcomes in patients with SLE. These data suggest that HRV may be a simple non-invasive tool used to gage or predict clinical improvement in SLE. Further studies are warranted.
Thanou A, Stavrakis S, Dyer JW, Munroe ME, James JA, Merrill JT. Impact of heart rate variability, a marker for cardiac health, on lupus disease activity. Arthritis Res Ther. 2016 Sep 2;18:197.
Disclosure of Interest None declared