Monoclonal Gammopathy of Undetermined Significance and Risk of Cardiovascular Disease in US Veterans

Background: Monoclonal gammopathy of undetermined significance (MGUS) has been associated with worse overall survival. The cause of this association is unclear but may result from increased risk of cardiovascular disease (CVD) events. Population-based studies have demonstrated associations between MGUS and CVD, but these were in largely Non-Hispanic White (NHW) populations. As such, we aim to examine the association between MGUS and CVD risk in the US Veterans Health Administration (VHA) which has the largest population of non-Hispanic Black patients (NHB). Additionally, this database can capture patient outcomes with an extensive electronic medical record system. The identification of a novel risk factor for CVD will improve our ability to prevent CVD events further.

Methods: This was a population-based retrospective cohort study of US Veterans using linked data from the national Veterans Health Administration and Medicare. Veterans free from CVD events were included in the analytic cohort and were followed from age 50 until they either develop CVD, they expire, or the date 12/31/2022 is reached. Veterans who died or were lost to follow up within 3 years after age 50 were excluded. The exposure is presumed MGUS diagnosis before CVD event or ever. The MGUS diagnosis was captured by a natural language processing (NLP) algorithm (previously developed by our group) based on the presence of a monoclonal protein less than 3 gm/dL and bone marrow plasma cells less than 10%. The outcome of interest was CVD event after age 50 which was identified with International Classification of Diseases (ICD)-9 and -10 codes atherosclerotic cardiovascular diseases and congestive heart failure. A multivariable logistic regression adjusting for sex, race/ethnicity, region of residence, income level, marital status, type 2 diabetes before CVD diagnosis or ever, and BMI at age 50 was performed to estimate odds ratios (ORs) and 95% confidence intervals (CIs).

Results: The analytic cohort contained 523,604 Veterans: 503,098 without MGUS, among whom 25.8% developed CVD and 20,506 with MGUS, among whom, 40% developed CVD (p<0.0001). The CVD events included heart failure with 18.6% (n=3,804) in the MGUS group and 8.7% (n=43,955) in the group without MGUS (p<0.0001). The mean age of MGUS (n=20,506) was 69.2 years and 66.8 in patients who had a CVD event (n=137,918). Compared to individuals without MGUS, individuals with MGUS had a higher proportion of being male (95.5 versus 86.5%, p<0.0001), non-Hispanic Black (29.7 versus 22.0%, p<0.0001), non-Hispanic White (55.2 versus 50.0%, p<0.0001), divorced/separated/widowed (36.2 versus 19.6%, p<0.0001), single/never married (10.6 versus 6.0%, p<0.0001), diabetes mellitus (38.4 versus 22.0%, p<0.001). The odds ratio of having a CVD in the setting of MGUS was 1.23 (95% CI: 1.19-1.26).

Conclusions: MGUS was associated with an increased risk of diagnosis of cardiovascular disease for individuals aged 50 years or older. Clinicians should consider evaluation of patients with MGUS for CVD risk factors. Further studies are required to evaluate whether addressing modifiable CVD risk factors is associated with decreased risk of developing CVD in MGUS patients.