Table 7

Summary of Studies Examining the Value of Myocardial Perfusion Imaging for Preoperative Assessment of Cardiac Risk

Referencen% of Patients With IschemiaEvents % (MI/Death)Perioperative EventsComments
Ischemia: % Positive Predictive ValueNormal: % Negative Predictive Value
Vascular Surgery
Boucher et al., 1985 (192)4833 (16)6 (3)19 (3/16)100 (32/32)First study to define risk of thallium redistribution
Cutler and Leppo, 1987 (193)11647 (54)10 (11)20 (11/54)100 (60/60)Only aortic surgery
Fletcher and Kershaw, 1988 (194)6722 (67)4 (3)20 (3/15)100 (56/56)
Sachs et al., 1988 (195)4631 (14)4 (2)14(2/14)100 (24/24)
Eagle et al., 1989 (44)20041 (82)8 (15)16 (13/82)98 (61/62)Defined clinical risk
McEnroe et al., 1990 (196)9536 (34)7 (7)9 (3/34)96 (44/46)Fixed defects predict events
Younis et al., 1990 (197)11136 (40)7 (8)15 (6/40)100 (51/51)Included long-term follow-up
Mangano et al., 1991 (198)6037 (22)5 (3)5 (1/22)95 (19/20)Managing physicians blinded to scan result
Strawn and Guernsey, 1991 (199)68N/A6 (4)N/A100 (21/21)
Watters et al., 1991 (200)2658 (15)12 (3)20 (3/15)100 (11/11)Included echocardiographic (TEE) studies
Hendel et al., 1992 (201)32751 (167)9 (28)14 (23/167)99 (97/98)Included long-term follow-up
Lette et al., 1992 (202)35545 (161)8 (30)17 (28/161)99 (160/162)Used quantitative scan index
Madsen et al., 1992 (203)6569 (45)8 (5)11 (5/45)100 (20/20)
Brown and Rowen, 1993 (204)23133 (77)5 (12)13 (10/77)99 (120/121)Prognostic utility enhanced by combined scan and clinical factors
Kresowik et al., 1993 (205)17039 (67)3 (5)4 (3/67)98 (64/65)
Baron et al., 1994 (206)45735 (160)5 (22)4 (7/160)96 (195/203) NFMI onlyDid not analyze for cardiac deaths; no independent value of scan
Bry et al., 1994 (207)23746 (110)7 (17)11 (12/110)100 (97/97)Cost-effectiveness data included
Koutelou et al., 1995 (208)10644 (47)3 (3)6 (3/47)100 (49/49)Used adenosine/SPECT thallium imaging
Marshall et al., 1995 (209)11747 (55)10 (12)16 (9/55)97 (33/34)Used adenosine thallium and sestamibi; size of ischemic defect enhanced prognostic utility
van Damme et al., 1997 (210)14234 (48)2 (3)N/AN/AUsed dobutamine SPECT sestamibi and echocardiographic imaging; echocardiographic and nuclear scan prognostic utility was equivalent
Huang et al., 1998 (211)10636 (39)5 (5)13 (5/39)100 (24/24)Dipyridamole thallium SPECT
Cohen et al., 2003 (212)15331 (48)4 (6)4 (2/48)100 (21/21)Dipyridamole SPECT sestamibi imaging; perioperative and long-term follow-up. Perfusion defect in the LAD territory was best predictor of long-term death/MI.
Harafuji et al., 2005 (213)30230 (92)1.3 (4)2 (2/92)100 (210/210)SPECT thallium with adenosine stress in 239 patients; dipyridamole in 63. Summed stress score greater than or equal to 14 (20-segment model) best multivariate predictor of events.
Nonvascular Surgery
Camp et al., 1990 (214)4023 (9)15 (6)67 (6/9)100 (23/23)Diabetes mellitus, renal transplant
Iqbal et al., 1991 (215)3141 (11)11 (3)27 (3/11)100 (20/20)Exercise 86%, diabetes mellitus, pancreas transplant
Coley et al., 1992 (216)10036 (36)4 (4)8 (3/36)98 (63/64)Define clinical risk factors in patients with known or suspected CAD
Shaw et al., 1992 (217)6047 (28)10 (6)21 (6/28)100 (19/19)Used adenosine
Takase et al., 1993 (218)5328 (15)11 (6)27 (4/15)100 (32/32)Patients with documented or suspected CAD; included rest echocardiogram
Younis et al., 1994 (219)16131 (50)9 (15)18 (9/50)98 (87/89)Intermediate- to high-risk CAD
Stratmann et al., 1996 (220)22929 (67)4 (10)6 (4/67)99 (1/92)Used dipyridamole sestamibi and noted fixed defect had more prognostic utility than transient defect
Zoghbi et al., 2003 (221)878 (7)2 (2)14 (1/7)97 (1/79)Liver transplant cohort
Patel et al., 2003 (222)17431 (54)7 (12)15 (8/54)97 (116/120)Renal transplant cohort

All studies except those by Coley et al. (216) and Shaw et al. (217) acquired patient information prospectively. Only in reports by Mangano et al. (198) and Baron et al. (206) were attending physicians blinded to scan results. Patients with fixed defects were omitted from calculations of positive and negative predictive value.

CAD indicates coronary artery disease; LAD, left anterior descending coronary artery; MI, myocardial infarction; n, number of patients who underwent surgery; N/A, not available; NFMI, nonfatal myocardial infarction; SPECT, single-photon emission computed tomography; and TEE, transesophageal echocardiography.

  • Positive and negative predictive values are predictive value (number of patients/total number of patients whose test results indicated either ischemia or a normal scan).

  • Studies utilizing pharmacological and/or exercise thallium testing.