Objectives. This study was designed to evaluate the indications for performing multidetector computed tomographic coronary angiography (MDCTA) after catheter-based coronary angiography. Background. Appropriateness criteria for MDCTA apply exclusively to patient evaluation prior to catheterbased angiography. Methods. All MDCTA performed after catheterbased angiography at a tertiary referral center were reviewed. Results. Fourteen of a total of 2,000 MDCTAs (0.7%) fulfilled the criteria: 14 were performed after catheter-based angiography. The indications were: 1) inability to selectively cannulate a native vessel or graft; 2) severe pressure damping limiting safe angiography; 3) ostial disease; 4) course of anomalous vessels; 5) relationship of bypass grafts to the sternum; 6) graft morphology; 7) chronic total occlusions. In all cases, the MDCTA enabled definitive treatment. Conclusions. In a variety of scenarios, MDCT coronary angiography may provide essential information as an adjunctive tool after catheter-based angiography. Indications for MDCTA should be expanded to accommodate these clinical necessities. J INVASIVE CARDIOL 2008;20:1–6 Key Words: computed tomographic angiography; coronary angiography; coronary artery disease

Figure 2.

In a 56-year-old male with angina, the anomalous right coronary artery could not be selectively cannulated. MDCTA revealed the vessel arising from the left coronary sinus, passing anteriorly between the aorta and pulmonary artery, as shown in the maximum- intensity pixel globe (A). The curved MPR displays total occlusion of the mid portion and ostial stenosis (B). Cross section of the straightened MPR demonstrates an MLA of 4.0 mm2. In combination with a severe mid left anterior descending stenosis, bypass surgery was recommended. MDCTA = multidetector computed tomographic angiography; MLA = minimum luminal area; MPR = multiplanar reconstruction.

       Multidetector computed tomographic coronary angiography (MDCTA) has rapidly gained in popularity and applicability.¹ Most recently, the concept of MDCTA-guided percutaneous intervention has been introduced, with MDCTA evaluation of minimum luminal area dictating the need for intravascular ultrasound (IVUS), irrespective of the catheter-based appearance.² However, all published studies have dealt exclusively with its applicability to evaluation prior to catheter-based coronary angiography. The role of MDCTA as an adjunctive tool after catheter-based coronary angiography has not been addressed.

Methods
       All MDCTA performed after catheter-based angiography at a tertiary referral center were reviewed.
       MDCTA protocol. The MDCTA were acquired on the Philips Brilliance-64 scanner (Philips Medical Systems, Cleveland, Ohio) using the 64 x 0.625 mm detector configuration, 120–140 kVp, 600–1000 mAs, 0.2 pitch and standard or sharp filters (Philips CC and CD filters). Nonionic contrast (ioversol 350 mg/ml at 5–6 ml/second for a total of 75–90 ml) was used, followed by 50 ml of saline at the same rate using a double-head injector (Optivantage DH, Mallinkrodt, Cincinnati, Ohio). The estimated effective radiation dose was 13 mSv for men and 18 mSv for women. Metoprolol 50–100 mg p.o. and/or 5 mg intravenous (IV) x 4 was administered to reduce the heart rate (HR) to < 60 beats per minute. The cardiac phase best demonstrating each artery (usually 75% of the RR interval) is analyzed using a dedicated CT workstation (Philips CT Extended Brilliance Workspace, Philips Medical Systems, Cleveland, Ohio) and a cardiac adaptive multisegmentreconstruction algorithm. The axial and multiplanar reformatted (MPR) images of all arteries were analyzed for the presence of measurable obstructive coronary artery disease. The curved MPR were rotated to the angle displaying the narrowest diameter, and percent diameter stenosis was quantitatively measured using a normal proximal or distal reference point, avoiding areas of positive or negative remodeling. Minimum luminal area was determined from cross-sectional analysis of the straightened MPR, with a combination of computer-aided and operator-modified analysis of Hounsfield unit gradients (Figures 2, 3, 5 and 7). The study was approved by the Institutional Review Board of Lenox Hill Hospital.

Results

Table 1.

Figure 1.

Selective injections of the right internal mammary graft to the diagonal could not be obtained. Subsequent MDCTA revealed a patent right internal mammary to the diagonal in the sagittal ( A) and coronal ( B) views. Axial view ( C) demonstrated the graft’s location immediately behind the sternum, posing a risk of transection during repeat bypass surgery. MDCTA = multidetector computed tomographic angiography.

       Of a total of 2,000 MDCTA studies, 14 (0.7%) were performed after catheter-based angiography. The indications for the procedures are summarized in Table 1, and examples are presented below.
       Failure of image acquisition. Inability to selectively cannulate a native vessel or graft. Anomalous coronary arteries, right internal mammary grafts and native coronary arteries in the setting of extreme aortic dilatation may be difficult to cannulate. Frequently, angiographic assessment is limited. A right internal mammary graft could not be located in a patient with chest pain; the graft was found to be patent on MDCTA (Figure 1), and medical therapy was pursued. An anomalous right coronary artery (RCA) coursing anteriorly between the aorta and pulmonary artery from the left coronary sinus could not be selectively injected in a patient with severe left anterior descending artery (LAD) disease. Demonstration of a proximal total occlusion in the RCA by MDCTA resulted in a recommendation for bypass surgery rather than stenting of the LAD (Figure 2). 

Figure 4.

A 55-year-old male with typical angina underwent selective angiography revealing severe distal left main stenosis (A, arrow). Severe damping accompanied right coronary cannulation, and subselective injection was inadequate (B). MDCTA curved MPR confirmed the left main disease (C, arrow) and yielded detailed images of the RCA demonstrating subtotal occlusion of the midvessel (D, E, arrow) and severe disease in the PL branch (E, arrowhead), necessitating separate grafts for the PDA and PL branches. MDCTA = multidetector computed tomographic angiography MPR = multiplanar reconstruction; PL = posterolateral; PDA = posterior descending artery; RCA = right coronary artery.

Figure 3.

A 61-year-old male presented with atypical chest pain and a nuclear stress test with significant ST depression but scintigraphically normal images. Selective angiography revealed critical RCA disease ( A); cannulation of the left main coronary yielded severe pressure damping and n o n selective injection was inadequate ( B). MDCTA curved MPR confirmed the RCA disease ( C) and demonstrated critical left main stenosis as well as significant LAD disease (D, arrows). Crosssection of the straightened MPR confirmed the critical nature of the left main stenosis (MLA 2.1 mm2) and the proximal (MLA 2.6 mm2 and mid (MLA 2.1 mm2) LAD lesions ( D). Bypass surgery was performed, with separate grafts for the LAD and LCx mandated by the LAD disease. LAD = left anterior descending; LCx = left circumflex; MDCTA = multidetector computed tomographic angiography; MLA = minimum luminal area; MPR = multiplanar reconstruction; RCA = right coronary artery.

       Severe pressure damping. Pressure decrease on coronary cannulation may represent critical atherosclerosis, coronary spasm or congenital abnormalities; injection of intracoronary nitroglycerin to resolve spasm may be dangerous if a critical stenosis is present, particularly in the left main coronary artery. Severe damping was noted on left main cannulation in a patient with a critical mid- RCA stenosis. Rather than pursue further attempts at left coronary visualization, MDCTA was performed and revealed critical left main, as well as proximal and mid-LAD, stenosis (Figure 3). The patient underwent bypass surgery on the basis of the MDCTA, with the LAD lesions prompting separate grafts for the LAD and circumflex arteries.
       Severe damping of the RCA was noted in a patient with critical left main disease. To avoid the risk of further attempts at selective injection with possible complications in this criticalsetting, MDCTA was performed. There was no significant ostial RCA disease but there was a subtotal occlusion of the mid RCA (Figure 4). In addition, a distal lesion in the posterolateral branch was noted, necessitating separate grafts for the posterolateral and posterior descending branches.
       Requirement for additional information. Even after successful catheter-based visualization of all vessels and grafts, further clarification may be required.
       Ostial disease versus spasm. Persistent ostial narrowing after intracoronary nitroglycerin injection may represent refractory spasm or true disease, or a combination thereof. In a 45-year-old female with left main stenosis post nitroglycerin administration, MDCTA was performed and revealed a normal vessel (Figure 5).

Figure 6.

A 45-year-old male with critical aortic stenosis and chest pain underwent selective angiography which revealed an anomalous left coronary arising from the right coronary sinus, but the relationship to the aorta and pulmonary artery could not be definitively established. MDCTA maximum intensity pixel globe demonstrated a benign posterior course between the Ao and LA. Calcification is noted in the aortic valve. Ao = aorta; LA = left atrium; LM = left main; PA = pulmonary artery.

Figure 5.

In a 42-yearold female with atypical c hes t p ain s elec tive angiogra phy re vealed le ft main stenosis unchanged after intracoronary nitroglycerin a ccompanied by mild pressure damping ( A ). MDCTA curved MPR revealed a normal left ma in ( B ) , and c ross section of the straightened MPR demonstrate d no rmal minimum luminal areas and the absence of plaque ( C ). MDCTA = multidetect or c omputed tomog rap hic angio grap hy; MPR = mult iplanar reconstruction.

       Course of anomalous vessels. A 40-year-old patient with aortic stenosis requiring valve replacement underwent angiography demonstrating anomalous origin of the left coronary artery from the right coronary sinus, but the anterior/ posterior relationship to the aor ta was unclear. MDCTA revealed a benign posterior course (Figure 6), and concomit ant bypass surgery was not performed.
       Relationship of bypass grafts to the sternum. In pat ient s undergoing repeat bypass surgery, the proximity of pr ee xis t ing internal mammary and vein grafts to the sternum and the risk of surgical transection cannot be easily determined by conventional angiography. MDCTA readily determined this relationship in a patient who required a second surgical procedure (Figure 1).
       Graft morphology. A pseudoaneurysm followed by severe stenosis was suspected in a vein graft to the marginal branch. Percutaneous intervention was considered, but MDCTA revealed straightforward aneurysmal dilatation ratherthan a pseudoaneurysm, and there was no significant stenosis (Figure 7). Medical therapy was pursued in this patient.
       Chronic total occlusions. The use of MDCTA to predict successful intervention for chronic total occlusions (CTOs) has been previously described.³ More importantly, MDCTA can be utilized to directly guide the procedure in the catheterization laboratory. In particular, attempted opening of flush occlusions may result in fruitless attempts to locate the entrance to the CTO without the guidance provided by MDCTA mapping.

Figure 8.

Figure 7.

A 64-year-old male post bypass surgery presented with chest pain. On selective angiography ( A), a pseudoaneurysm followed by severe stenosis was suspected in a vein graft to the marginal branch. Percutaneous intervention was considered, but MDCTA curved MPR ( B) revealed straightforward aneurysmal dilatation rather than a pseudoaneurysm. Quantitative measurements of the presumed stenosis yielded 55% and 61% respectively, using proximal and distal reference points and avoiding the aneurysmal segment. Cross-sectional analysis of the straightened MPR ( C) using mean rather than minimum diameters revealed the stenosis measurements to be 33% and 45%, respectively, with more than adequate MLA. Medical therapy was pursued.

       Angiography demonstrated flush occlusion of the LAD in a symptomatic 65-year-old female with anterior ischemia. MDCTA mapping (TrueView, Philips Medical Systems) was imported to the catheterization laboratory monitor and electronically linked to the C-arm; the C-arm was rotated, with accompanying automatic rotation of the TrueView map to the angle predetermined by the MDCTA to best demonstrate the origin and course of the CTO without overlapping branches. The guidewire was introduced to the precise origin of the flush occlusion, followed by successful recanalization (Figure 8).

Discussion
       In a high-volume tertiary referral interventional center, MDCTA has proved to be an extremely valuable adjunctive tool after coronary angiographyto resolve issues not satisfactorily addressed by the invasive procedure. While representing only 0.7% of the total MDCTA volume, in each case it has been essential to appropriate decision making in cases in which there was no alternative method to obtain the necessary information. The additional radiation dose of the MDCTA and its hazards are an unfortunate but necessary concomitant of the decision-making imperative. High-quality MDCTA interpreted by experienced operators is essential. The indications are summarized in Table 1.
       Failure of image acquisition. While inability to cannulate may be an infrequent occurrence in the hands of experienced angiographers, its occurrence is inevitable, leaving no alternative but MDCTA, the accuracy of which is unaffected by anatomical variations. Severe damping in the setting of critical disease is operator-independent, and persistent attempts at selective injection may be catastrophic. In each case, MDCTA provided a clear diagnosis, allowing definitive treatment to be implemented with confidence. The clarity of distal vessel delineation enables the appropriate choice of surgical targets (Figure 4). The ability to calculate minimum luminal areas confirms the critical ostial disease suggested by the severe pressure damping (Figure 2).
       Requirement for additional information. Although coronary spasm is usually resolved by intracoronary nitroglycerin, persistent ostial narrowing can only be evaluated by nontraumatic visualization by MDCTA (Figure 5). Twodimensional catheter-based angiography may not be able to provide the three-dimensional data intrinsic to MDCTA that may be necessary for tracking the relationship of anomalous arteries to the aorta and pulmonary artery (Figure 6) or of an internal mammary graft to the sternum (Figure 1). Graft dilatation and tortuosity may mimic a pseudoaneurysm, and comparison of normal to dilated segments may give the false impression of significant stenosis. MDCTA reconstruction allows for straightening of the graft and for area measurements that confirm simple dilatation and the absence of significant narrowing (Figure 7).
       The increasing interest in opening CTOs can be greatly facilitated by the consistent visualization by MDCTA of the diseased wall of the occluded segment and of the relationship of this segment to adjacent branch vessels. With the importation of the coronary tree into the catheterization laboratory, the operator is empowered to rotate the C-arm to the precise angle for optimizing entry into the CTO. Flush occlusions pose a particularly difficult problem that can be solved by utilization of TrueView, as illustrated in Figure 8 in the first reported case in the world’s literature of MDCTA-guided CTO intervention in a flush occluded vessel.

Conclusions
       In a variety of scenarios, MDCTA coronary angiography provides essential information as an adjunctive tool after catheter-based angiography. Indications for MDCT should be expanded to accommodate these clinical necessities.