Date: Monday, October 22, 2018
Session Title: Measures and Measurement of Healthcare Quality Poster II
Session Type: ACR Poster Session B
Session Time: 9:00AM-11:00AM
Background/Purpose: In clinical practice, the two most commonly used markers of systemic inflammation are the serum CRP level and ESR. The economic costs of these tests are substantial, with Medicare allowable charges in Los Angeles in 2018 for CRP level and ESR being $38.04 and $62.76, respectively. An algorithm that reduces the need for these tests in the care of patients with systemic rheumatic diseases could lead to considerable cost savings without sacrificing the quality of patient care.
Methods: The electronic medical records of two independent cohorts (discovery and validation) of patients with systemic rheumatic diseases seen between May 2015 and June 2017 in the rheumatology clinics at a single academic medical center were retrospectively reviewed. Correlations and receiver operator characteristic (ROC) curves between serum CRP level and ESR vs serum globulin gap (the difference between levels of total protein and albumin) and albumin-to-globulin (A:G) ratio were determined.
Results: The discovery (263 subjects, 446 entries) and validation (438 subjects, 1959 entries) cohorts were predominantly female (89.0% and 82.9%) and Hispanic (93.9% and 87.9%), with median ages of 47.4 and 51.6 years at the time of sample collection and the majority (52.7% and 67.1%) of entries coming from RA or SLE patients. In each of these independent cohorts, the globulin gap and A:G ratio correlated significantly (p < 0.001) with CRP level and ESR, with the respective correlation coefficients being greater for ESR (discovery: 0.472 and -0.672; validation: 0.509 and -0.596) than for CRP level (discovery: 0.308 and -0.374; validation: 0.225 and -0.310). ROC curve analyses demonstrated better respective abilities of globulin gap and A:G ratio to discriminate between normal and elevated ESR (discovery area-under-curve [AUC]: 0.726 and 0.823; validation AUC: 0.738 and 0.771) than between normal and elevated CRP level (discovery AUC: 0.681 and 0.726; validation AUC: 0.602 and 0.656). These relationships were independent of sex (female vs male) or ethnicity (Hispanic vs non-Hispanic), and similar results were obtained when only RA or SLE was considered. Elevated globulin gap (≥4.0 g/dl) and low A:G ratio (<0.8) had respective positive predictive values (PPVs) of only ≥0.554 and ≥0.788 for elevated CRP level, whereas the respective PPVs for elevated ESR were ≥0.962 and ≥0.960. When only the first entry for a given patient was considered, the correlations between ESR and globulin gap (r = 0.531) and A:G ratio (r = -0.648) and the corresponding AUCs (0.742 and 0.795) were similar to those observed for all entries. Moreover, the abilities of elevated globulin gap and low A:G ratio to predict an elevated ESR remained very high, with PPVs of 0.976 and 0.966, respectively. Among patients with high globulin gap, the change in globulin gap over time faithfully reflected changes in ESR.
Conclusion: In patients with systemic rheumatic disease, elevated globulin gap (readily calculable from the routinely-obtained comprehensive metabolic panel) is a highly reliable marker of elevated ESR. Ordering an ESR (or CRP) test in such patients may frequently be unnecessary, resulting in an estimated savings of $7.56 million per million patients with systemic rheumatic disease.
To cite this abstract in AMA style:Stohl W, Kenol B, Kelly A, Ananth Correa A, Panush R. Elevated Serum Globulin Gap As a Reliable and Cost-Savings Marker of Inflammation in Patients with Systemic Rheumatic Diseases [abstract]. Arthritis Rheumatol. 2018; 70 (suppl 10). https://acrabstracts.org/abstract/elevated-serum-globulin-gap-as-a-reliable-and-cost-savings-marker-of-inflammation-in-patients-with-systemic-rheumatic-diseases/. Accessed September 27, 2021.
« Back to 2018 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/elevated-serum-globulin-gap-as-a-reliable-and-cost-savings-marker-of-inflammation-in-patients-with-systemic-rheumatic-diseases/