To the Editor:
In the July 1, 1997, issue of Circulation, Piek and coworkers1 provided very interesting intracoronary data on the pharmacological responsiveness of the human coronary collateral circulation. In 38 patients with 1-vessel coronary artery disease and visible (n=24) or recruitable (n=14) collaterals, a collateral flow index, indices for collateral resistance, and index for the resistance in the vascular bed downstream of the collateral receiving artery were investigated regarding their response to intracoronary adenosine or nitroglycerin injected into the collateral donor artery. The authors found that spontaneously visible and recruitable collaterals substantially differed in their response to intracoronary adenosine or nitroglycerin in that the former were able to decrease their resistance and thus increase the flow through it, whereas the latter were unable to do so. Although the direct study of collateral hemodynamics in humans is very important because almost no data exist in this area of research, the mentioned results of the present study have to be taken cautiously for some methodological reasons.
1. The authors’ assertion is incorrect that they obtained collateral flow by measuring diastolic occlusive time velocity integral (dVi, cm). In fact they determined just an index for collateral flow that did not account for the influence of changes in the caliber of coronary collaterals, which may develop during pharmacological stress.
2. More importantly, an analysis of combined intracoronary Doppler- and pressure-derived collateral indices in our laboratory2 has indicated that the calculation of a collateral index using just diastolic integrated velocity signals does not yield accurate results when compared with pressure-derived, intracoronary, ECG-validated values (n=21 patients with bidirectional flow velocity signals). In our opinion, flow velocity signals have to be added over the entire cardiac cycle irrespective of their direction in order to provide correct results of collateral flow indices.2 Intuitively, this makes sense, since coronary epicardial flow indicated by the velocity signal will finally reach the myocardial bed irrespective of the direction it has taken at the time of its detection upstream. Systolic velocity signals originating from increased epicardial backward or forward flow due to enhanced myocardial wall stress can be recognized easily by their high pitch, high peak velocity, and high acoustic intensity onset and their short duration. These signals should not be accounted for in the computation of collateral flow indices. However, all other flow velocity signals during systole should be used in the calculation of Doppler indices of collateral flow.
3. As a minor point, it has to emphasized that Equation 3 in the appendix of the article is given incorrectly. It should probably read Pw=dVi∗R4 (and not Pw+dVi∗R4).
- Copyright © 1998 by American Heart Association
Piek JJ, van Liebergen AM, Koch KT, de Winter RJ, Peters RJG, David GK. Pharmacological modulation of the human collateral vascular resistance in acute and chronic coronary occlusion assessed by intracoronary blood flow velocity analysis in an angioplasty model. Circulation. 1997;96:106–115.
Seiler C, Fleisch M, Meier B. Intracoronary occlusive pressure compared with velocity measurements for the assessments of collaterals. Eur Heart J. 1997;18(suppl):240. Abstract.
Only a limited number of studies have been conducted on the hemodynamics of the coronary collateral circulation in humans. Consequently, the data reported in our articleR1 require confirmation by other investigators. The critical comments by Seiler are well appreciated and may stimulate further research on this topic of interest. The comments raised are addressed in numerical order.
1. The currently available technical equipment in interventional cardiology hampers the assessment of collateral vascular resistance in an angioplasty model by measurement of collateral volume flow. This requires a simultaneous assessment of both collateral blood flow velocity and the diameter of the arterial segment. It is correct that collateral flow was estimated in the study as an index using the diastolic blood flow velocity integral. The selective administration of vasodilators (adenosine or nitroglycerin) in the donor coronary artery may induce vasodilation of the recipient coronary artery.R2 This indicates that the observed changes in collateral blood flow velocity after the administration of vasodilators in patients with spontaneously visible collateral vessels are rather an underestimation of the true alterations in volume flow and hence of the collateral vascular resistance. On the other hand, our conclusion that recruitable collateral vessels, in general, do not respond to vasodilators may be incorrect due to this phenomenon. These methodological drawbacks were addressed in the limitations section of the article.
2. The use of the diastolic blood flow velocity integral for the calculation of the collateral vascular resistance was based on the consideration that systolic blood flow velocity signals can be generated by myocardial contraction even in the absence of recruitable collateral vessels.
The interpretation of these systolic signals is rather undefined, taking into account the limited number of studies performed on this subject.R3 R4 The diastolic blood flow velocity integral as an index of collateral flow is arbitrary.
Table 2 of the original article depicts the alterations in the coronary blood flow velocity after the administration of vasodilators using the diastolic blood flow velocity integral as well as the total blood flow velocity integral. Bidirectional signals were added for the calculation of this total blood flow velocity integral. That table shows that the blood flow velocity alterations are not different using the diastolic or total blood flow velocity integrals.
For completeness, we have calculated the changes in collateral vascular resistance based on the total blood flow velocity integral rather than the diastolic velocity integral. Again, the following Table⇑ shows that this collateral flow index does not yield different conclusions, with the exception of the effect of nitroglycerin on the peripheral vascular resistance index of the recipient coronary artery, which shows a trend toward significance.
3. The authors apologize for the incorrect Equation 3 (Pw+dVi∗R4) in the appendix that should read as: Pw=dVi∗R4.
Piek JJ, van Liebergen RAM, Koch KT, de Winter RJ, Peters RJG, David GK. Pharmacological modulation of the human collateral vascular resistance in acute and chronic coronary occlusion assessed by intracoronary blood flow velocity analysis in an angioplasty model. Circulation. 1997;96:106–115.