Studies have shown that coronary artery calcium (CAC) incidentally identified on a noncontrast chest computed tomography (NCCT) performed for noncardiac indications has diagnostic and prognostic value. The frequency by which radiologists report incidental CAC and its impact on patient management are unknown. This study included 204 consecutive patients (63 ± 17 years, 59% men) without a history of coronary artery disease referred for an NCCT for noncardiac indications. The presence of CAC was determined by an expert cardiologist and compared with the radiology report. For each patient, the medical record was reviewed for changes in medications. Physicians caring for these patients were surveyed regarding their awareness and the clinical importance of incidental CAC after their patients had been discharged from the hospital. There were 108 of 201 patients (53%) with a CAC score >0 as determined by an expert reader. The interpreting radiologist reported the presence of CAC in 74 of 108 patients (69%). Of the 74 patients, there was an increase in stain and aspirin prescription of 4% and 5%, respectively. Of the 132 physicians surveyed, 54% of physicians surveyed believed that CAC on an NCCT scan was analogous to the presence of coronary artery disease, 23% were aware that incidental CAC was reported, and only 4% said they would make medical management decisions based on the finding of incidental CAC. In conclusion, incidental CAC is under-reported by the interpreting radiologists and suggests an integral role for a cardiovascular imaging specialist. When incidental CAC is reported, physicians are not cognizant of the meaning and importance of this finding. This lack of knowledge is reflected in the negligible impact reported incidental CAC has on clinical management decisions.
Coronary artery calcium (CAC) identified on a noncontrast chest computed tomography (NCCT) study is considered diagnostic for the presence of coronary artery disease (CAD). The prognostic value of CAC detected by a dedicated ECG-gated CAC scan has been demonstrated in different patient populations. Recent study has shown that CAC incidentally detected on nongated NCCT scans also has prognostic value. With the decades-long increase in the use of chest CT and the concerns regarding cumulative lifetime radiation dose to patients, it is important that physicians use all the data that are available to them from tests their patients undergo. This is underscored when considering CAD, which is highly prevalent in the United States. Previous studies have shown that the finding of incidental CAC seen on NCCT is not uniformly reported in the clinical setting. When reported, the impact incidental CAC has on patient management is unclear. To answer this question, we prospectively enrolled patients who underwent a nongated NCCT for noncardiac indications and the physicians who ordered the examinations with the following aims: (1) to ascertain the frequency in which incidental CAC on NCCT is reported by radiologists, (2) to elucidate what the finding of incidental CAC means to the physicians that ordered the tests, and (3) to determine if the finding of incidental CAC affects medical management of their patients.
Methods
We reviewed 304 consecutive NCCT scans of patients admitted to the Internal Medicine service at our Institution from May 2013 to August 2013. Exclusion criteria included a history of CAD (n = 45), an NCCT ordered for a cardiac indication (n = 4), and patient expiration during the hospitalization (n = 4). After excluding these patients, the study cohort included 204 patients (63 ± 17 years, 59% men). This study was approved by the Institutional Review Board of the Mount Sinai St. Luke’s and Roosevelt Hospital.
All NCCT scans were performed using commercially available CT scanners (Aquillon 4-, 16-, 64-detector; Toshiba America Medical Systems, Tustin, California, or Somatoform Sensation 64-detector; Siemens Medical Solutions USA, Malvern, Pennsylvania). Images were nongated and nonenhanced with a slice thicknesses ranging from 1.0 to 3.0 mm. Each CT scan was initially interpreted in the clinical setting by a radiologist not involved with the study. After the patients were discharged from the hospital, all NCCT scans were blindly reviewed by an expert cardiologist for the presence of CAC using commercially available software (Vitrea, Vital Images; Toshiba Medical Systems, Minnetonka, Minnesota). The presence of CAC abnormality based on these research readings were compared with the clinical reports to assess the comparative frequency of CAC abnormality noted by the interpreting radiologists.
For each patient included in the study, the medical chart was reviewed for baseline clinical demographics and patient medication regimen at admission and change in medication regimen and any cardiac testing performed during hospitalization.
To link the presence of incidental CAC and the subsequent management decisions, we identified the 132 attending physicians and house staff who were directly involved with the medical management decisions for the 74 patients with incidental CAC reported. All these physicians completed a survey concerning their attitudes to the utility of incidental CAC. Of those surveyed, 40 (30%) were attending physicians, 28 (21%) were postgraduate year 3, 20 (15%) were postgraduate year 2, and 44 (34%) were postgraduate year 1 or physician assistants. The survey consisted of questions to assess: (1) physicians attitudes toward equating CAC abnormality with CAD, (2) awareness of the incidental CAC that had been reported on CAC scanning, and (3) what management decisions should follow from the reporting of incidental CAC abnormality. The survey is included in the Supplementary Material . All surveys were conducted after their patient had been discharged from the hospital, thus ensuring the survey would not interfere with the medical management patients underwent. Each provider surveyed was required to provide a 1-time written informed consent and research authorization as required by the institutional review board.
Continuous data were analyzed using Student t test and Paired t test where applicable. Categorical variables were analyzed using the chi-square test. One-way analysis of variance with a post hoc Bonferroni test was used to compare means of continuous variables among multiple groups. Continuous data are presented as mean ± SD. Categorical data are presented as absolute numbers or percentages. For all analyses, a 2-tailed p value ≤0.05 was considered significant. All analyses were performed using SPSS for Windows (version 16; SPSS Inc., Chicago, Illinois).
Results
The baseline characteristics of the patient population are listed in Table 1 . Figure 1 shows the frequency of CAC noted on NCCT scan by the reporting radiologists versus the expert cardiac readers. Of the 74 patients who had CAC abnormality noted by the radiologists, the expert reader concurred in 72 (97%). However, 36 patients (18%) had CAC abnormality identified by the expert readers that was not reported by the radiologist on the patients’ clinical reports.
| Variable | n = 204 |
|---|---|
| Age (years) | 63 ± 17 |
| Male | 121 (59%) |
| Hypertension | 115 (56%) |
| Diabetes mellitus | 53 (26%) |
| Hyperlipidemia | 35 (17%) |
| Smoker | 80 (40%) |
| Medications | |
| Statin | 46 (23%) |
| Aspirin | 43 (21%) |
| ACE-I/ARB | 64 (32%) |
| Beta blockers | 53 (26%) |
| Calcium channel blockers | 56 (28%) |
| Antiplatelet | 3 (2%) |
| Insulin | 26 (13%) |
| Oral hypoglycemics | 30 (15%) |
| Indications for CT scanning: | |
| Respiratory | 136 (67%) |
| Aortic | 22 (11%) |
| Chest Pain | 4 (2%) |
| Esophageal | 3 (2%) |
| Other | 39 (18%) |
Table 2 lists the change in medications between baseline and discharge according to whether CAC abnormality was identified on the clinical report. No significant difference in change in any medication category was noted in the patients with reported CAC versus not reported with CAC. The initiation of new statin use during hospitalization both for those with and without reported CAC was quite small (3% vs 7%, p = 0.2) as was that for initiation of new aspirin use (5% vs 10%, p = 0.4).
| Variable | Patients with no CAC reported (n = 130) | Patients with CAC reported (n = 74) | ||||
|---|---|---|---|---|---|---|
| Baseline | At discharge | p Value | Baseline | At discharge | p Value | |
| Statin | 23 (18%) | 26 (20%) | 0.6 | 23 (31%) | 26 (35%) | 0.7 |
| Aspirin | 24 (19%) | 30 (23%) | 0.4 | 19 (26%) | 23 (31%) | 0.6 |
| ACE-I/ARB | 35 (27%) | 35 (27%) | 1.0 | 29 (39%) | 31 (42%) | 0.9 |
| Beta antagonists | 30 (23%) | 33 (26%) | 0.7 | 23 (31%) | 21 (28%) | 0.9 |
| Calcium channel antagonists | 32 (25%) | 37 (29%) | 0.5 | 24 (32%) | 24 (32%) | 1.0 |
| Antiplatelet | 1 (0.8%) | 1 (0.8%) | 1.0 | 2 (3%) | 3 (4%) | 1.0 |
| Insulin | 12 (9%) | 14 (11%) | 0.7 | 14 (19%) | 15 (20%) | 1.0 |
| Oral hypoglycemic | 16 (12%) | 17 (13%) | 0.9 | 14 (19%) | 16 (22%) | 0.8 |
Of the 74 patients who had CAC, we surveyed 132 physicians who were involved in their care ( Table 3 ). Overall, 54% of physicians thought that CAC on NCCT should be treated as CAD. The frequencies were comparable among house staff and attending physicians. Only 23% of physicians were aware that CAC had been reported on NCCT. Of the 15 attending physicians who were aware that CAC was reported on NCCT, only 3 thought that the finding of CAC should be a cause for initiating medical therapy. Whereas, of the 25 attending physicians who reported being unaware of the reported CAC, all of them reported that incidental CAC should not be a cause for initiation of medical therapy. A comparable number of house staff reported likewise.
| Variable | All (n=132) | PGY1/PA (n=44) | PGY2 (n=20) | PGY3 (n=28) | Attending (n=40) | p Value |
|---|---|---|---|---|---|---|
| Do you think CAC reported on a NCCT should be treated as CAD? | 71 (54%) | 26 (59%) | 11 (55%) | 14 (50%) | 20 (50%) | 0.8 |
| Were you aware of the CAC reported in the NCCT of your patient? | 30 (23%) | 6 (14%) | 5 (25%) | 4 (14%) | 15 (38%) | 0.04 |
| Did you make any medication changes based on the CAC reported on your pt? | 6 (4%) | 0 (0%) | 1 (5%) | 2 (7%) | 3 (8%) | 0.3 |
| What do you think should be done when incidental CAC is reported? | ||||||
| Further evaluate with a stress test regardless of symptoms | 23 (18%) | 6 (14%) | 5 (25%) | 5 (18%) | 7 (18%) | 0.7 |
| Evaluate with stress test only if symptomatic | 52 (39%) | 22 (50%) | 8 (40%) | 9 (32%) | 13 (33%) | 0.3 |
| Initiate medical management for CAD | 48 (36%) | 12 (27%) | 10 (50%) | 11 (39%) | 15 (38%) | 0.3 |
| Outpatient Cardiology consult | 15 (11%) | 5 (11%) | 0 (0%) | 3 (11%) | 7 (18%) | 0.3 |
| Inpatient cardiology consult | 44 (33%) | 15 (34%) | 6 (30%) | 12 (43%) | 11 (28%) | 0.6 |
| Do nothing | 11 (8%) | 3 (7%) | 1 (5%) | 3 (11%) | 4 (10%) | 0.8 |
| Did you refer the patient for cardiologist? | 0.4 | |||||
| Inpatient cardiology consult | 9 (7%) | 2 (5%) | 0 (0%) | 3 (11%) | 4 (10%) | |
| Outpatient cardiology consult | 16 (12%) | 4 (9%) | 5 (26%) | 3 (11%) | 4 (10%) | |
| Informal curbside consult | 2 (2%) | 1 (2%) | 1 (5%) | 0 (0%) | 0 (0%) | |
| No, I did not refer the patient | 103 (79%) | 36 (84%) | 13 (69%) | 22 (78%) | 32 (80%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
