Angiojet rheolytic thrombectomy in massive pulmonary embolism: locally efficacious but systemically deleterious?

fluoroscopy time, which is under control of the operator, as a primary outcome measure. Although not ideal, it allowed us to make a meaningful estimate of the number of patients who would need to be included in the study. It is possible that we might have overcome the variability by completing a much larger study, but we wanted to focus on those factors that might be easily modifiable by the operator. We did not find a difference in the kerma-area product in our study groups despite a difference in fluoroscopy time. Dr. Miller speculates that we might have collimated to half the field in those cases in which unilateral embolizations were performed, and this would explain the failure to find a difference. We did not, in fact, consciously do that. We routinely collimated the field during the embolization to the uterus and the very immediate adnexa, and we did not vary it consciously when performing a unilateral or bilateral embolization. Although we likely increased collimation somewhat during the unilateral procedures, we did not measure the difference and doubt it was as much as 50%. This is because the vessels on one side can supply leiomyomas across the midline, and it is also possible to have arteriovenous shunting on either side near the end of the procedure. Arteriovenous shunting is one sign of the appropriate endpoint of embolization, and we believe it is important to monitor it. We believe that it is more likely that the variability of body habitus, uterine size, and uterine location were greater than the estimated difference in the measured fluoroscopy time and imaging. This was shown in the earlier study by Bratby et al (1) we referenced in our study. They used an anthropomorphic phantom and simulated unilateral and bilateral procedures based on values from their study. They did find a difference in dose-area product and cumulative dose when body size factors were held constant (1). We did not repeat that confirmation, as we did not have sufficiently detailed data to duplicate a typical procedure. For example, we did not record the time of anteroposterior fluoroscopy, fluoroscopy time with tube angulation, the degree of tube angulation, or the average collimation dimensions. We still believe that, all other things being equal, less fluoroscopy and imaging use are likely to reduce patient exposure, and this reduction cannot be completely mitigated by more aggressive collimation in uterine artery embolization procedures. The need to provide adequate monitoring of the procedure prevents a perfect trade-off of fluoroscopy time and field collimation. Having said that, limiting the tissue area exposed is an important component of radiationsafe practice, and we thank Dr. Miller for raising the issue.