2421. Effect of Metal Speculum During Hysterosalpingography on Patient Radiation Dose
Authors * Denotes Presenting Author
  1. Nancy Ann Little *; University of Wisconsin Hospitals and Clinics
  2. Sean Rose; University of Wisconsin School of Medicine and Public Health
  3. John Vetter; University of Wisconsin School of Medicine and Public Health
  4. Jessica Robbins; University of Wisconsin Hospitals and Clinics
  5. Meghan Lubner; University of Wisconsin Hospitals and Clinics
Hysterosalpingography (HSG) technique varies in our department regarding the speculum; some leave the speculum in place while others remove the speculum during contrast injection and image capture. This raised concern whether the presence of this metallic object would affect the machine’s automatic exposure control (AEC) and increase patient radiation dose. An argument to leave the speculum in place is to minimize the risk of displacing the catheter during speculum removal and an argument to remove the speculum is to minimize radiation dose. The purpose of this study is to determine whether radiation dose of HSG is increased when a speculum is present.

Materials and Methods:
In this retrospective cohort study, we collected demographic, technique, and radiation dose data for all HSGs performed by abdominal radiologists between 8/1/2018 and 8/26/2020. HSGs were performed on a Siemens Luminos Agile Max fluoroscopy machine. Dose data was collected for the first AP single shot of each HSG. Of the 77 unique patient encounters, 12 were excluded due to missing data (n=8), variation in technique (n=3), or high body mass index (BMI) outlier (n=1). The remaining 65 encounters represented HSGs performed by one of 8 attending radiologists in which at least one AP single-shot image was obtained (33 with metal speculum, 32 without). Linear regression analysis was applied to compare patient radiation dose, as measured by reference air kerma, between images with and without a speculum in frame. Subsequent F-test assessed whether presence of the speculum impacted reference air kerma. We additionally investigated the response of our fluoroscopy system’s AEC to a metal speculum using an acrylic and aluminum phantom simulating a 20cm patient thickness.

After adjusting for source to detector distance (SID), magnification mode, and BMI, the average reference air kerma was 0.89mGy [0.81, 0.98] for images without and 0.84mGy [0.77, 0.92] for images with a speculum (95% confidence intervals in brackets); there is no significant difference in reference air kerma (p=0.44). The fluoroscopy system’s AEC uses two laterally located regions of interest (ROIs) in our HSG protocol. In our phantom study, we obtained a reference air kerma of 0.38mGy with no speculum in the X-ray beam. With the speculum placed at locations simulating clinical practice, reference air kermas ranged from 0.38mGy-0.41mGy (0-8% increase). Attempting to fully block one of the AEC ROIs with a speculum increased the reference air kerma from 0.38mGy to 0.57mGy (50% increase); however, this position would not be used clinically.

In our study population, there was no increase in entrance radiation dose with metal speculum left in place, and phantom testing revealed at most an 8% increase for clinically feasible speculum locations. It is critical to emphasize this result is due to specific protocols that do not increase machine output with centrally-located radiopacity and physicist colleagues should be consulted to ensure appropriate local examination protocols.