Abstract
Introduction
Individuals with multiple myeloma (MM) are at a greater risk of thrombotic complications than those with other cancers. Cancer-related venous thromboembolism (VTE) including pulmonary embolism and deep vein thrombosis is associated with increased morbidity and mortality. Both incidence of MM and VTE differ by race/ethnicity. Blacks have a higher incidence of MM compared to Whites, whereas Asian/Pacific Islanders (API) and Hispanics have a lower incidence. In the general population, the risk of VTE and prevalence of its associated risk factors are higher in Blacks. The hypercoagulable state induced by MM and its treatment could further modify any pre-existing VTE risk in different racial/ethnic groups. Our objective was to describe the incidence of VTE following diagnosis and treatment of MM by race/ethnicity.
Methods
We conducted a retrospective cohort study using data from the Surveillance, Epidemiology, and End Results (SEER) - Medicare linked database. Individuals 66 years or older with myeloma as their first primary malignancy diagnosed between 2001 - 2011 were identified. These individuals were required to have age related eligibility for Medicare, continuous enrollment in the 12 months period prior and post myeloma diagnosis, and Medicare as their primary payer. Using SEER registries and administrative claims, we collected data on race/ethnicity, cancer characteristics and treatment, and chronic comorbidities. VTE events were defined as having an inpatient hospitalization or 2+ outpatient visits with ICD-9 diagnosis codes 415.1x, 451.xx, and 453.xx. Long-term risk of VTE was assessed by determining the annualized incidence of VTE at 12, 24, and 36 months following MM diagnosis. In each 12-month period, MM patients were followed from diagnosis until the earliest of the following: a VTE event, death, or end of 12-month period since diagnosis. Additionally, we calculated VTE incidence rates in the 12-month period following stem cell transplantation for those that received it. Overall and race-specific crude incidence rates with 95% confidence intervals (CIs) were calculated in each period using person-time contributed per observation period. Poisson regression was used to compare age- and gender-adjusted incidence rate ratios (IRRs) and 95% CIs comparing different racial/ethnic groups with White patients.
Results
In a final analytic cohort of 9,480 patients, most were White (73%) and fewer patients were Black (15%), Hispanic (6%) or API (4%). The median age at diagnosis was 77 years (interquartile range [IQR]: 71-82). Half the cohort (51%) were women, 28% had diabetes and 21% had a history of heart failure at diagnosis. Overall, the median time to first VTE event post-MM diagnosis was 140 days (IQR: 46-409). More patients that developed VTE were female, Black, younger at MM diagnosis, and treated with stem cell transplantation. The overall incidence rate of VTE during the first 12 months following the diagnosis of MM was 170 per 1,000 person-years (95% CI 160-180). Adjusting for age and gender, Blacks had a higher incidence of VTE compared to Whites (IRR 1.24; 95% CI 1.07-1.44) and APIs had a lower incidence (IRR 0.55; 95% CI 0.37-0.81). In the second post-diagnosis year, there were 6,257 individuals remaining that contributed 5,443 person-years. Overall, the incidence rate was 82 per 1,000 person-years (95% CI 75-90). Incidence was greater in Blacks vs. Whites (adjusted IRR 1.25; 95% CI 0.98-1.59). Consistent with findings from the first analysis period, APIs continued to have lower incidence vs. Whites (adjusted IRR 0.44; 95% CI 0.22-0.89). In the third post-diagnosis year, the overall incidence rate was 80 per 1,000 person-years (95% CI 72-89) and any racial disparities observed in age and gender adjusted IRRs were no longer significant. For patients in the first year following stem cell transplantation, annualized incidence rates of VTE were elevated (overall incidence rate of 148, 95% CI 117-188).
Conclusion
In this large population-based cohort of older MM patients, we observed racial disparities in the incidence of VTE within the first 24 months of diagnosis. A risk-adapted method of VTE prophylaxis should be considered. Understanding risk factors and groups vulnerable to development of thromboembolic events can help guide clinical decision making regarding VTE prophylaxis.
Disclosures
No relevant conflicts of interest to declare.
Epidemiologic trends for isolated tibia shaft fracture admissions in The Netherlands between 1991 and 2012