Current issues of ACP Journal Club are published in Annals of Internal Medicine


Angioscopy was more sensitive than angiography for intracoronary thrombus

ACP J Club. 1995 Mar-April;122:45. doi:10.7326/ACPJC-1995-122-2-045

Source Citation

den Heijer P, Foley DP, Escaned J, et al. Angioscopic versus angiographic detection of intimal dissection and intracoronary thrombus. J Am Coll Cardiol. Sep 1994



To compare coronary angioscopy with angiography for the detection of intimal dissection and intracoronary thrombus.


Blinded comparison of coronary angioscopy with angiography.


Catheterization laboratories of the University Hospital, Groningen, and the Thoraxcenter, Rotterdam, the Netherlands.


52 patients (mean age, 61 y) who had both coronary angiography and coronary angioscopy during the same cardiac catheterization session. Angioscopy was done before (n = 15) or after (n = 37) coronary balloon angioplasty.

Description of Test and Diagnostic Standard

Angioscopy was done with the Image-Cath system (Baxter, Interventional Cardiology Division). Angioscopic thrombus was defined as a red, white, or mixed red and white, intraluminal, superficial, or protruding mass, adherent to the vessel surface but clearly a separate structure. Angioscopic dissection was defined as visible cracks or fissures on the lumen surface or mobile protruding structures that were contiguous with the vessel wall and of homogeneous appearance with the vessel wall. Standard coronary angiography was done. Angiographic intimal dissection was defined as intraluminal filling defects, extravasation of contrast material, or linear lumen density staining. Angiographic intracoronary thrombus was defined as the presence of an intraluminal central filling defect or lucency surrounded by contrast material, seen in multiple projections and with no calcifications within the defect.

Main Outcome Measures

Sensitivity, specificity, and positive and negative predictive values.

Main Results

Angiographic and angioscopic findings were in agreement in 21 patients (40%) when thrombus was absent and in 6 patients (12%) when thrombus was present. 25 angioscopically observed thrombi (48%) were undetected with angiography. When angioscopy was used as the diagnostic standard, the sensitivity of angiography for detecting thrombus was 19%, the specificity was 100%, the positive predictive value was 100%, and the negative predictive value was 46%. Angiographic and angioscopic findings were in agreement in 10 patients (19%) when dissection was absent and in 13 patients (25%) when dissection was present. 2 angiographic dissections were undetected by angioscopy, and 27 angioscopic dissections were undetected by angiography.


Coronary angiography was less sensitive than angioscopy for detecting intracoronary thrombus. Angioscopy and angiography were complementary diagnostic tools for detecting intimal dissections.

Source of funding: In part, Commission of the European Communities.

For article reprint: Dr. P. den Heijer, Department of Cardiology, Igna Fius Hospital, 4800 RK Breda, the Netherlands. FAX 31-76-5217250.


The study by den Heijer and colleagues is of interest to both interventional cardiologists and internists. The availability of low-profile, flexible, and steerable angioscopic systems has allowed the routine percutaneous application of this technology. Interpretable images can now be obtained in most patients having interventional procedures, although the technique still has important limitations. Problems remain with aiming the angioscope, and a clear imaging field cannot always be obtained. The widespread availability of second-generation interventional technologies, including stents, atherectomy catheters, and local drug delivery systems, give the interventionist a choice of therapies. Precise knowledge of coronary morphology may have an important effect on the decision of which technique to use to treat a given lesion.

The study by den Heijer and colleagues suggests that, in a few patients, the angioscopic information was useful to the clinician. These results, however, cannot be extrapolated to all patients receiving interventions. Future research is required on more patients.

What is the likely clinical utility of coronary angioscopy? As of 1995, it should be used selectively when the interventionist needs to know if a suspicious lesion represents coronary thrombus. This would be important if stent placement or intracoronary thrombolysis is being considered. I do not believe that angioscopy is likely to provide additional clinical value for coronary dissection above that provided by angiography. The angioscope appears to identify surface flaps more readily than does angiography, but the more severe and dangerous deep intramural dissections are likely to be more apparent by radiographic contrast imaging. Only further clinical research, however, can definitively answer this question.

Will coronary angioscopy become a tool for every patient having coronary intervention? This is unlikely, although the more options the interventionist has, the more important precise diagnostic information will be. I expect that over the next decade, as the number of pharmacologic and device options increase, so will the application and quality of coronary angioscopy.

Frank Litvack, MD
Cedars-Sinai Medical Center Los Angeles, California