Clinical relevance of raised inflammatory markers in primary care: a prospective cohort study using CPRD

Problem

Inflammatory markers including C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and plasma viscosity (PV) are commonly used in primary care. Most previous studies describe laboratory findings for a single disease, and most are based in secondary care. In contrast GPs are often faced with a test result and try to work out which disease (if any) is causing the abnormality.

Approach

A prospective cohort study of 160,000 patients from Clinical Practice Research Datalink (CPRD) who had inflammatory marker blood tests in 2014. The primary outcome was incidence of infections, autoimmune conditions and cancers in those with raised versus normal inflammatory markers. A comparison cohort of 40,000 age, sex and practice matched patients without inflammatory marker testing in 2014 was identified, to compare incidence of disease in tested versus untested patients.

Findings

Of the tested cohort 73% had a CRP test, 59% had an ESR test, and 10% had PV test; 34% of these had at least one raised inflammatory marker. Patients with pre-existing cancer (n=4197), pre-existing autoimmune conditions (n=11,476), and recent infections (n=10,127) were excluded. A total of 1,351 out of 44,721 patients with raised inflammatory markers developed cancer in 1 year after testing (2.9%; 95% CI 2.8-3.1) compared to 1,192 out of 93,003 with normal inflammatory markers (1.3%; CI 1.2-1.4) and 307 out of 37,591 with no inflammatory marker (0.8%; CI 0.7-0.9). For autoimmune conditions 5.1% (CI 4.9 to 5.3) of the raised inflammatory marker, 1.6% (CI 1.5-1.7) of the normal inflammatory marker and 0.5% (CI 0.46-0.60) of the no inflammatory marker patients were diagnosed with a new autoimmune condition in 1 year after testing. For infections 5.9% (CI 5.6-6.1) of the raised inflammatory marker, 2.9% (CI 2.8-3.0) of the normal inflammatory marker and 2.0% (CI 1.8-2.2) of the no inflammatory marker patients were diagnosed with a new infection in 1 month after testing. The overall sensitivity of inflammatory markers for detecting cancer, infection or autoimmune condition was 47.7% for CRP 45.8% for ESR and 47.2% for PV. Specificities were 75.4% for CRP, 74.3% for ESR and 73.4% for PV.

Consequences

This is the first study to explore multiple disease outcomes following inflammatory marker testing in primary care. Results will be clinically useful for GPs interpreting inflammatory marker test results. Interpretation of tests will be contingent on the clinical reason for testing. Raised inflammatory markers are a risk marker for cancer, and may warrant further investigation for underlying malignancy. Presentation of disease risk by gender, age, and level of raised inflammatory marker will help identify those at highest risk. The overall sensitivity of inflammatory marker tests is low, suggesting they are not clinically useful to ‘rule-out’ serious disease.