Abstract
1655
Objectives Pulmonary ventilation/perfusion scintigraphy (V/Q) in pulmonary embolism (PE) diagnosis has diminished, because of multiple factors, in the last decade while Computed Tomography Pulmonary Angiography (CTPA) use has increased. These trends are supported by limited published data. Well known concerns regarding CTPA radiation exposure have been published particularly in the PE diagnosis of women. This study contributes insurance claim documented medical record, historic experience in PE diagnostic imaging to understand any clinically relevant differences, around the choice of a PE imaging modality, in a large patient cohort.
Methods Data on patients with no recent V/Q or CTPA scans were extracted using EMRClaims+® (eMAX Health, White Plains, NY) containing insurance claims linked to electronic medical records between 2005 and 2015 for over 2 million patients across 13 hospitals in the Midwestern United States. The sample included all adult patients with either scan who met the inclusion and exclusion criteria. Demographics were recorded at baseline, and the Charlson Comorbidity Index (CCI) was used to identify comorbidities during the 12 month pre-index period. Baseline differences were assessed using Chi-square and T-test analysis.
Results A total of 79,442 patients (2,396 V/Q and 77,046 CTPA) were included in the dataset. There was a significantly greater proportion of women in V/Q group compared to CTPA (63% vs. 56%, p<0.0001). The ratio of patients with renal disease in the V/Q group was 6 times patients in the CTPA group (31% vs. 5%, p<0.0001). The percentage of patients with diabetes in the V/Q group was twice the percentage in the CTPA group (31% vs. 15%, p<0.0001). The CCI score was higher in the V/Q group (2.82 vs. 1.26, p<0.0001), while a greater proportion of CTPA studies was performed in the 18-45 year population vs V/Q. Some of these findings were expected, while other study data showed non-alignment with published guidelines. There was a greater proportion of younger women being imaged using CTPA.
Conclusions This study found clinically relevant and statistically significant differences in the distribution of populations depending on the imaging modality selected for diagnosing PE. This study adds important information, over a recent 10 year interval in ~80,000 patients, about who are getting V/Q scans. The data also reveals discordance between actual imaging modality selection and practice guidelines. Given the lower radiation exposure, V/Q imaging should be considered first in some patients—especially young women as suggested by a number of published guidelines. This real world data may better inform needed educational efforts and further development of practice guidelines. Additional analyses are indicated.