OBJECTIVES

To evaluate incidence of complications associated with central venous access devices (CVADs) in PwHA.

METHODS

This retrospective cohort study was conducted using claims data from MarketScan Commercial Research Database from 07.01.2005–03.31.2019. The study cohort comprised PwHA and included CVAD cases (≥1 CVAD insertion claim), and controls with no CVAD insertion claim through the study period. Patients were required to have continuous enrollment for 6 months pre- and at least 3 months post-index date. Index date was defined as the first date of port insertion for CVAD cases and first hemophilia A (HA) diagnosis for controls during the study period. HA was identified using a previously validated claims-based algorithm. CVAD use and complications (all-cause infections, thrombosis and hematoma) were identified using ICD-9-CM/ICD-10-CM diagnosis/procedure and CPT codes. Patients were followed until first outcome, plan switch or over a 2-year post-index period. Incidence and rates of complications among CVAD cases and controls were evaluated using Cox proportional-hazards models (adjusted for age, region, comorbidity score, and insurance type).

RESULTS

The study cohort comprised 862 PwHA; 61 (7%) had evidence of CVAD use. CVAD cases were significantly younger than controls (mean age ± SD: 4.7±5.3 years vs. 25.9±17.5 years, p<0.001) and had a significantly higher Elixhauser comorbidity score (mean ± SD: 0.9±0.6 vs. 0.5±0.8, p<0.001). In the post-index period, a significantly higher proportion of CVAD cases (vs. controls) had all-cause infections (44.3% vs. 26.7%, p<0.001) and thrombosis (13.1% vs. 1.1%, p<0.001). No CVAD cases had evidence of hematoma; it occurred in 1.5% of controls. Cox models revealed that CVAD cases had 2.3 times (95%CI: 1.5–3.6) and 9.2 times (95%CI: 2.4–35.6) higher rate of all-cause infections and thrombosis, respectively.

CONCLUSIONS

CVAD use in HA is associated with higher rates of complications, underscoring the need for novel non-intravenous treatments which remove the need for CVADs.