Coronary artery calcification detected in lung cancer screening predicts cardiovascular death

Figures & data

Table I. Summary of published studies regarding non-gated CAC score outcome in asymptomatic persons included in the meta-analysis.

Reference	Patients, n	Mean age, years	Male gender (%)	Mean follow-up, months	CT slice thickness, mm	CAC measurement	CAC reference	Adjusted HR (95%CI)	Event
NELSON (16)	7.557	59.5	83	22	3.1	AS CAC > 1000	AS CAC = 0	11 (2.4–51)^a	All-cause mortality
								7.8 (2.4–25)^a	CAD events*
PROVIDI (15)	10.410	62	58	18	3.0–10	Visual CAC ^ε	Visual CAC = 0^ε	5.0 (3.2–7.7)^b	CAD events*
I-ELCAP (28)	8.782	64.8	49	72	1.25–5	Visual CAC > 4^ε	Visual CAC = 0^ε	2.1 (1.4–3.1)^c	Cardiac death
MILD (29)	1.159	57.5	68	36	5	mCAC > 400^†	mCAC ≤ 400^†	3.7 (1.1–13)^d	All-cause mortality
								2.9 (1.1–7.3)^d	CAD events**
DLCST	1.945	57.4	55	85	3	AS CAC > 400	AS CAC = 0	2.1 (1.0–4.8)^a	All-cause mortality
								3.8 (1.0–15)^a	CAD mortality

AS, Agatston score; VS, Volume score; CAC, Coronary artery calcium; CAD, Coronary artery disease; HR, Hazard ratio; mCAC, modified CAC; NELSON, NEderlands-Leuvens Longkanker Screenings Onderzoek; PROVIDI, PROgnostic Value of Unrequested Information in Diagnostic Imaging; I-ELCAP, International Early Lung Cancer Action Program; MILD, Multicentric Italian Lung Detection; DLCST, Danish Lung Cancer Screening Trial.

*defined as “combined fatal MIs and non-fatal CAD admissions.”

**defined as “MI, unstable angina pectoris, PCI, or CABG.”

***unadjusted HR.

^εVisual score from 0 to 3 of each of the 4 main coronary arteries (left main, left anterior descending artery, circumflex, and right) according to CAC deposition (0: absent; 1: mild; 2:moderate; and 3:severe). Accordingly, a maximum visual CAC score of 12 could be reached.

†modified CAC: Agatston score measured with 5-mm CT slice thickness.

^aadjusted for age, gender, diabetes, hypertension, hypercholesterolemia, and smoking status.

^badjusted for age, gender, indication for CT, image quality, and type of medical center where CT was performed.

^cadjusted for age, gender, diabetes, and pack-years of smoking.

^dadjusted for age, gender, BMI, diabetes, hypertension, smoking status, and duration of smoking.

^eadjusted for age, smoking status, smoking history, and cardiovascular history.

Table II. Summary of clinical characteristics in the DLCST.

	Overall (N = 1945)	Men (N = 1075)	Women (N = 870)	P value*
Age, years	57 (53–61)	58 (54–61)	57 (53–60)	< 0.0001
BMI, kg/m^2	25 (23–27)	25 (23–28)	24 (22–27)	< 0.0001
Medical treatment of
Hypertension, N (%)	270 (14%)	138 (13%)	132 (15%)	0.14
Hypercholesterolemia, N (%)	138 (7%)	87 (8%)	51 (6%)	0.06
Diabetes, N (%)	34 (2%)	26 (2%)	8 (1%)	< 0.05
Previous stroke, N (%)	32 (2%)	23 (2%)	9 (1%)	0.06
Current smoking, N (%)	1468 (76%)	813 (76%)	655 (75%)	0.86
Pack-years, N (years)	34 (27–42)	36 (29–45)	31 (26–39)	< 0.0001
Lung function
No COPD, N (%)	1109 (57%)	614 (57%)	495 (57%)
Mild COPD, N (%)	549 (28%)	310 (29%)	239 (28%)	0.60
Moderate–severe COPD, N (%)	286 (15%)	151 (14%)	135 (16%)
CAC score 0, N (%)	1035 (53%)	431 (40%)	604 (69%)	< 0.0001
CAC score > 400 (%)	132 (7%)	105 (10%)	27 (3%)	< 0.0001
Median Agatston CAC score	0 (0–50)	9 (0–107)	0 (0–5)	< 0.0001
Follow-up (years)	7.1 (6.8–7.5)	7.1 (6.8–7.5)	7.2 (6.8–7.5)	0.17

BMI, Body Mass Index; COPD, Chronic Obstructive Pulmonary Disease; CAC, Coronary Artery Calcium.

Table III. Clinical characteristics of participants in the DLCST, who were alive or died from cardiovascular and non-cardiovascular causes.

	Alive (N = 1873)	Cardiovascular death (N = 19)	Non-cardiovascular death (N = 53)
Male sex, N (%)	1027 (55%)	17 (89%)	31 (58%)
Age, years	57 (53–61)	57 (53–61)	60 (56–63)
BMI, kg/m^2	25 (23–27)	25 (23–27)	26 (23–29)
Medical treatment of
Hypertension, N (%)	252 (13%)	5 (26%)	13 (25%)
Hypercholesterolemia, N (%)	130 (7%)	3 (16%)	5 (9%)
Diabetes, N (%)	31 (2%)	2 (11%)	1 (2%)
Previous stroke, N (%)	29 (2%)	1 (5%)	2 (4%)
Current smoking, N (%)	1410 (75%)	15 (79%)	43 (19%)
Pack-years (years)	34 (27–42)	42 (31–68)	37 (29–45)
Lung function
No COPD, N (%)	1066 (57%)	16 (84%)	27 (49%)
Mild COPD, N (%)	532 (28%)	1 (5%)	16 (30%)
Moderate–severe COPD, N (%)	274 (15%)	2 (11%)	10 (21%)
CAC score = 0, N (%)	1011 (54%)	5 (26%)	19 (36%)
CAC score > 400, N (%)	122 (7%)	5 (26%)	5 (9%)
Median Agatston CAC score	0 (0–48)	68 (0–501)	16 (0–68)

BMI, Body Mass Index; COPD, Chronic Obstructive Pulmonary Disease; CAC, Coronary Artery Calcium.

Figure 1. Kaplan–Meier survival curves in the DLCST for all-cause mortality according to CAC category (log-rank < 0.001).

Table IV. Mortality rate in addition to unadjusted and adjusted Cox regression HRs according to CAC score category in the DLCST.

	CAC score = 0 (N = 1035)	CAC score = 1–400 (N = 778)	CAC score > 400 (N = 132)	P value
All-cause mortality rate, N (%)	24 (2%)	32 (4%)	10 (8%)	< 0.001*
Cardiovascular mortality rate, N (%)	5 (0%)	9 (1%)	5 (4%)	< 0.01*
Unadjusted Cox HR—all-cause (95%CI)	1	2.1 (1.3–3.6)	3.4 (1.6–7.2)	< 0.01^αβ
Adjusted Cox HR—all-cause (95%CI)^§	1	1.7 (1.0–2.9)	2.1 (1.0–4.8)	0.06 ^αβ
Unadjusted Cox HR—cardiovascular (95%CI)	1	2.4 (0.8–7.6)	8.3 (2.4–29)	0.11^α < 0.001^β
Adjusted Cox HR—cardiovascular (95%CI)^§	1	1.6 (0.5–4.9)	3.8 (1.0–15)	0.43^α
				< 0.05^β

CAC, coronary artery calcium; HR, hazard ratio; 95%CI, 95% confidence interval.

*Chi-Square test.

§Cox regression model adjusted for gender, age, smoking, hypertension, hypercholesterolemia, and diabetes.

*p-value for CAC = 1–400; *p-value for CAC > 400; *p-value for CAC = 1–400 and CAC > 400.

Figure 2. Meta-analysis of published studies including data from the DLCST. Please notice that the definitions of low and high CAC were different in different studies, refer to Table I.

Jacobs PC, Gondrie MJ, van der Graaf Y, de Koning HJ, Isgum I, van Ginneken B, et al. Coronary artery calcium can predict all-cause mortality and cardiovascular events on low-dose CT screening for lung cancer. AJR Am J Roentgenol. 2012;198:505–11.

 PubMed Web of Science ®Google Scholar

Jacobs PC, Gondrie MJ, Mali WP, Oen AL, Prokop M, Grobbee DE, et al. Unrequested information from routine diagnostic chest CT predicts future cardiovascular events. Eur Radiol. 2011;21:1577–85.

 PubMed Web of Science ®Google Scholar

Shemesh J, Henschke CI, Shaham D, Yip R, Farooqi AO, Cham MD, et al. Ordinal scoring of coronary artery calcifications on low-dose CT scans of the chest is predictive of death from cardiovascular disease. Radiology. 2010;257:541–8.

 PubMed Web of Science ®Google Scholar

Sverzellati N, Cademartiri F, Bravi F, Martini C, Gira FA, Maffei E, et al. Relationship and prognostic value of modified coronary artery calcium score, FEV1, and emphysema in lung cancer screening population: the MILD trial. Radiology. 2012;262:460–7.

 PubMed Web of Science ®Google Scholar