Background The erythrocyte sedimentation rate (ESR) is an inexpensive laboratory test for assessing the inflammatory response but is influenced by a wide range of factors, many of them unrelated to inflammation.
Objectives To assess whether ESR levels correlate with the level of disease activity at each visit and whether a change in ESR could be useful in predicting future changes in disease activity.
Methods 34000 visits in a prospective SLE cohort were analyzed for correlation between ESR and level of disease activity. A regression model of 12,000 pairs of visits separated by 80-100 days, with change in ESR as the independent and change in disease activity as the dependent variable was used to assess whether a change in ESR correlates with a concurrent change in disease activity. 6141 visit triplets (three consecutive visits where adjacent visits are separated by 80-100 days) were used for analysis of whether a change in ESR between the 1st and 2nd visit as the independent variable can predict a change in disease activity between the 2nd and 3rd visit as the dependent variable. Multiple linear regression models were used to adjust for age, race, gender, body mass index, complement levels, hematocrit, dsDNA, prednisone use, hydroxychloroquine use, and immunosuppressant use. Follow-up visits while patients had cancer, bacterial infection, pregnancy or were in renal failure were excluded.
Results After adjusting for confounding factors, ESR correlated with SLEDAI, physician global assessment (PGA), fatigue, renal, joint, rash, serositis and hematological visual analogue scale as well as with proteinuria (p<0.0001), but not with neurological or pulmonary VAS or hematuria. A change in ESR between two visits was highly correlated with a concurrent change in PGA, renal, fatigue and joint visual analogue scale (p<0.0001) but not with changes in other organ activity indices. This correlation remained strong after adjustment for changes in body mass index, complement levels, hematocrit, dsDNA levels, and prednisone, hydroxychloroquine or immunosuppressant doses. There was no statistically significant correlation between change in ESR between the 1st and 2nd visit and change in disease activity in between the 2nd and 3rd visit.
Conclusions ESR is one of the oldest markers of inflammation, but still retains its utility as a marker of disease activity and is strongly associated with disease activity in SLE measured by SLEDAI, PGA and with organ specific activity including serositis, rash, joint, renal and hematological visual analogue scale. A change in ESR between two visits was highly correlated with a change in PGA, renal, fatigue and joint VAS but not with changes in other organ activity indices. Changes in ESR between two visits do not predict changes in disease activity in a future visit. Until more specific biomarkers are validated, serial ESR may join the ranks of useful markers for assessing changes in renal disease activity.
Disclosure of Interest None Declared