Relationship and prognostic value of coronary artery calcification by electron beam computed tomography to stress-induced ischemia by single photon emission computed tomography

Gautam Ramakrishna, Todd D. Miller, Jerome F. Breen, Philip A. Araoz, David O. Hodge, Raymond J. Gibbons

Research output: Contribution to journalArticle

52 Scopus citations


Background: Stress single photon emission computed tomography (SPECT) is commonly performed in patients with abnormal electron beam computed tomography (EBCT) to define risk stratification, but the published prognostic data for patients undergoing both SPECT and EBCT are limited. The objective of the study was to examine the association and prognostic value between EBCT, coronary artery calcium score (CACS), and stress SPECT imaging. Methods: We identified 835 patients (age 54.8 ± 10.0 years, 77% male) who underwent EBCT and stress SPECT within a 3-month period. Coronary artery calcium score was categorized as normal (0), minimal (1-10), mild (11-100), moderate (101-400), and severe (>400). Single photon emission computed tomography summed stress score (SSS) was categorized as normal, low risk, intermediate risk, and high risk per Cedar Sinai criteria. Average follow-up was 4.8 ± 3.2 years. End points were all-cause death, death/myocardial infarction (MI), and death/MI/late revascularization. Results: The correlation of CACS to SSS was weak but statistically significant (r = +0.19, P < .001). The percentage of high-risk SSS increased with higher CACS scores; 4% of patients with normal EBCT and 18% with severe CACS had high-risk SSS. Coronary artery calcium score (χ2 = 11.4, P < .001), diabetes mellitus (χ2 = 4.6, P = .031), and chest pain class (χ2 = 8.7, P = .003) were independently associated with high-risk SPECT. The SSS (χ2 = 6.9, P = .009) and CACS (χ2 = 7.8, P = .005) were independently associated with mortality, as well as with both secondary end points of death/MI and death/MI/late revascularization. Only CACS predicted mortality in the 408 asymptomatic patients (χ2 = 5.2, P = .02), but these patients had an annual mortality of only 0.4% over the next 5 years. Conclusions: In selected patients undergoing both EBCT and SPECT, CACS is weakly correlated with SPECT SSS, likely reflecting the different information provided by EBCT and SPECT. Coronary artery calcium score is independently associated with high-risk SPECT after adjustment for clinical variables. Coronary artery calcium score and SSS are complementary for the prediction of mortality in symptomatic patients. Only CACS predicted mortality in the asymptomatic patients, but they had a low annual mortality.

Original languageEnglish (US)
Pages (from-to)807-814
Number of pages8
JournalAmerican heart journal
Issue number5
StatePublished - May 1 2007


ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine

Cite this