Predictors of inferior vena cava filter retrieval in a population-based Canadian cohort.

BACKGROUND

The objective of this study was to determine the predictors of inferior vena cava (IVC) filter retrieval in a contemporary North American cohort of patients who received an IVC filter.

METHODS

A retrospective population-based cohort study was conducted using Ontario administrative health data. Physician service fee codes were used to identify all adults with an IVC filter placement from January 1, 2010, to December 31, 2019. The cumulative incidence of filter retrieval over time was calculated, accounting for death as a competing risk. Multivariable sub-distribution hazard regression models were constructed to quantify the association between covariates and the likelihood of filter retrieval.

RESULTS

A total of 5617 people received an IVC filter during the study period. Median follow-up was 1.8 years (interquartile range, 0.2-5.4 years). The probability of filter retrieval plateaued under 40% with most retrievals (96%; n = 2049 of 2135) occurring within 1 year of placement. Filter placement in a teaching hospital (hazard ratio, 1.85; 95% confidence interval, 1.60-2.02), and placement after 2016 were associated with a greater likelihood of filter retrieval. Older age and greater comorbidity were associated with a lower likelihood of filter retrieval.

CONCLUSIONS

In this population-based study of IVC filter retrieval, less than 40% of filters were retrieved, mostly within 1 year of insertion. Better coordination and standardization of services responsible for follow-up of patients with IVC filters are needed.