The initial query from 1/19/2017 to 9/25/2018 identified 1857 patient studies. Once exclusion criteria were applied a total of 503 unique patient studies were included in our analysis. (Figure 3) Of the 503 patients, 279 were female (55.5%) and 224 were male (44.5%), with an average age of 59 (58.5) years (Range 18 – 91 years-old, SD 15.8).
In total there were 260 extra-spinal degenerative findings, 9 patients with femoro-acetabular impingement (FAI), 3 patients with extra-pulmonary thoracic findings, 14 patients with pulmonary findings, 3 patients with vascular findings, 2 patients with abdominal findings, 12 patients with bony abnormalities, and 1 patient with a finding we categorized as “Other”, for a total of 304 IESF. In total there were 280 Minor findings, 18 Moderate, and 6 Major findings (Table 1). Secondary comparison of the IESF to the C-RADS  classification system is presented in Table 2. A summary of each of the IESF by system is reported below.
260 patients (149 female, 53.8%, average age 64-years-old) had reported degenerative changes in extra-spinal joints including knees, hips, CMC joints, ankles, glenohumeral joints, and acromioclavicular joints. All degenerative change was categorized as Minor (see Table 1) and only one area of degenerative change was counted per patient even if more than one joint had evidence of degenerative change on the EOS study. Radiologists did not recommend follow up for any of the findings of degenerative change.
9 patients (3 female, 30%, average age 31.5-years-old) had a dictated finding of femoro-acetabular impingement, such as abnormal femoral neck morphology, femoral neck-head offset, or CAM type lesions of the femoral neck. All FAI findings were categorized as Minor. Any patient over 50-years-old with dictation of FAI was not included. Radiologists did not recommend follow up for any of the FAI findings.
Thoracic, Extra-Pulmonary Abnormalities
3 patients (1 female, 33%, average age 66-years-old) had a dictated finding of thoracic, extra-pulmonary abnormalities including paratracheal fullness, widened mediastinum, and supra-hilar density. All findings were classified as Moderate. All dictations recommended follow up imaging.
14 patients (8 female, 57%, average age 64-years-old) had a dictated finding of pulmonary abnormality including granuloma (6), pulmonary nodule (4), multiple nodules v. apical thickening (1), pleural plaque (1), pleural effusion (1), and pulmonary nodule v. artifact (1). All granulomas, pleural effusion, and pleural plaque were classified as Minor. 3 pulmonary nodules were 9, 9, and 10 mm, respectively, and, classified as Major, and the other one nodule was 5 mm and classified as Moderate. The abnormality “apical thickening v. pulmonary nodule” was 4-5 mm and also classified as Moderate. Finally, the nodule v. artifact was classified as Moderate. All dictations for Moderate and Major findings recommended follow up imaging.
3 patients (3 female, 100%, average age 65-years-old) had a dictated finding of vascular abnormality. All 3 vascular abnormalities were described as enlarged or prominent cardiac silhouettes. All 3 were listed as Moderate. 2 of the 3 dictations recommend follow up imaging. Of note, vascular calcifications were not included in our study.
2 patients (1 female, 50%, average age 62-years-old) had a finding of non-vascular calcification located within the abdomen. Both were classified as Moderate and both dictations recommended follow up.
12 patients (7 female, 58.3%, average age 61-years-old) had a dictated finding of bony abnormality including bone infarct (3), lytic/lucent long bone lesion (3), sclerotic long bone lesion (2), rib abnormality (3) or loosening of hardware (1). All lytic lesions were classified as Major, all bone infarcts categorized as Minor, and all other findings classified as Moderate. Dictations recommended follow up in 2 of 3 Minor findings, all Moderate and 2 of 3 Major findings.
1 patient (1 female, 100%, age 50-years-old) had a dictated finding of possible discontinuity in a ventriculoperitoneal (VP) shunt. The dictation recommended clinical correlation. The finding was classified as Moderate because of significant implications if the shunt was indeed in discontinuity.
We reviewed the electronic medical record (EMR) of patients with 24 IESF (i.e. classified as Moderate or Major) (Table 3). We did not review the EMR for any Minor IESF because, by definition, these findings are not clinically significant and unlikely to have adverse health effects for the patient. Therefore, we did not review the EMR for the following abnormalities: degenerative change, FAI, pleural plaque, pleural effusion, granuloma, or bone infarct. The clinical significance of these findings in our study is unknown.
One of the 3 thoracic, extra-pulmonary findings was followed up with additional imaging. The paratracheal fullness, concerning for mediastinal lymphadenopathy, on further chest x-ray and CT was determined to be without corresponding finding and classified as “Insignificant”.
All 3 Major pulmonary nodules were classified as Insignificant after further workup. 2 were referred for CT studies and determined to not require additional intervention. The other Major nodule was known to the patient and the imaging was sent to their primary care provider at an outside facility. Of the 3 Moderate findings, 1 was determined to be a granuloma (Insignificant) by an outside provider using an older CT study. The finding of nodule v. artifact was never followed up, per chart review. The patient with 4-5 mm nodules (Figure 5 A) was referred to a pulmonologist who diagnosed possible Interstitial Lung Disease (ILD) and continued to see the patient for 2 years of regular follow up and imaging (Figure 5 B shows follow up CT chest of nodules). This finding was deemed “Significant”.
Of the Vascular IESF, 3 enlarged cardiac silhouettes were classified as Moderate, and none of the 3 were further investigated (No follow up), per the EMR.
Of the Abdominal IESF, 1 finding was not further followed up. The other finding was thought to be a renal abnormality and additional studies including MR renal and ultrasonography resulted in diagnosis of a simple, exophytic renal cyst (Insignificant).
Of the Bony IESF, all 3 lytic bone lesions occurred in patients with history of metastatic cancer (Significant) although the findings were not followed up with further imaging, presumably because the patients’ clinicians were already aware of the presence of metastatic disease. Figure 6 is a magnified EOS image of one of the 3 patients with lytic lucencies. One of the 2 sclerotic lesions was not followed up while the other was evaluated by orthopedic providers (including an orthopedic oncologist) and determined to be a Non-ossifying fibroma (Insignificant). As for the 3 rib abnormalities, 1 received no further workup while 2 did. One rib lesion was followed up with CT chest at outside facility and determined to be consistent with old trauma (Insignificant) and the other was followed up with CT chest at our facility and no specific clinical action was taken (Insignificant). The finding of hardware loosening was in the femoral component of a total hip replacement and was not followed up.
The term “Other” was used to classify a dictation with concern for discontinuity of a VP shunt. The patient presented to neurosurgery clinic, for a different reason, after their EOS study where the shunt was evaluated by providers without mention of abnormality (Insignificant).
Comparison of IESF between the RCR and C-RADS classification systems yields mostly concordant categorizations. Although, 3 pulmonary nodules classified as Major according to our modified Fleischner and RCR criteria were classified as “E3” findings because the nodules were less than 1 cm.
The overall rate of IESF for our study (including all Minor, Moderate and Major findings) was 60% (60.4). The rate of clinically significant IESF was 0.8% and consisted of 1 pulmonary nodule and 3 lytic bone lesions. 24 IESF (4.8%) were classified as either Major or Moderate and were recommended by radiologist dictations for further workup. 14 of these 24 IESF (58.3%) were further investigated, per review of the EMR, with either additional imaging, discussion amongst providers, or referral to a subspecialist. There was no evidence in the medical record that the remaining 10 findings were further evaluated at our institution or at an outside institution. The rate of false positives (i.e. Insignificant column in Table 3) was 42%.