Journal of Cardiovascular Translational Research

, Volume 4, Issue 1, pp 99–105

Relationship between Retrograde Coronary Blood Flow and the Extent of No-Reflow and Infarct Size in a Porcine Ischemia–Reperfusion Model

Authors

  • Stavros Stavrakis
    • Department of Clinical TherapeuticsUniversity of Athens School of Medicine “Alexandra” Hospital
  • John Terrovitis
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
  • Elias Tsolakis
    • Department of Clinical TherapeuticsUniversity of Athens School of Medicine “Alexandra” Hospital
  • Stavros Drakos
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
  • Argirios Dalianis
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
  • Michael Bonios
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
  • Dimitrios Koudoumas
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
  • Konstantinos Malliaras
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
    • 3rd Department of CardiologyUniversity of Athens School of Medicine “Laiko” Hospital
Article

DOI: 10.1007/s12265-010-9240-4

Cite this article as:
Stavrakis, S., Terrovitis, J., Tsolakis, E. et al. J. of Cardiovasc. Trans. Res. (2011) 4: 99. doi:10.1007/s12265-010-9240-4
  • 76 Views

Abstract

Recanalization of an infarct-related artery does not predictably reflect tissue reperfusion. We examined the relationship between coronary blood flow (CBF) pattern during reperfusion and infarcted (IA) and no-reflow (NR) area in a porcine ischemia–reperfusion model. The mid-left anterior descending artery of 18 pigs was occluded for 1 h and reperfused for 2 h. CBF during reperfusion was measured with a transit-time ultrasound flowmeter, while systemic arterial and left atrial pressures were monitored. IA and NR were measured with triphenyl tetrazolium chloride and thioflavin staining, respectively. In 13 pigs, early systolic retrograde CBF developed within the first 30 min and persisted throughout reperfusion. No retrograde CBF was observed in five pigs. Mean retrograde CBF at 2 h of reperfusion predicted a larger IA (r = 0.71; p = 0.001). Time-to-development of retrograde CBF was inversely related to IA (r = −0.55; p = 0.019) and NR (r = −0.62; p = 0.006). A larger IA (OR 1.12, 95% CI 1.01–1.24, p = 0.037) and NR (OR 1.09, 95% CI 1.01–1.18, p = 0.037) predicted the presence of retrograde CBF. Retrograde CBF during recanalization of the infarct-related artery predicts IA and NR and might be used as an index of successful reperfusion at the tissue level.

Keywords

Myocardial ischemiaMyocardial infarctionCoronary reperfusionCoronary blood flowNo-reflow phenomenon

Introduction

A timely reperfusion of the myocardium is essential when patients present with an acute myocardial infarction (MI) and ST-segment elevation on the surface electrocardiogram [1]. However, the restoration of epicardial coronary blood flow (CBF) by thrombolysis or by primary coronary intervention does not predictably indicate that the previously ischemic myocardium has been completely reperfused [2]. The “no-reflow” (NR) phenomenon is the incomplete reperfusion of ischemic myocardium in presence of coronary epicardial flow [3] and has been associated with microvascular injury [2, 3]. In experimental models of ischemia–reperfusion, the NR phenomenon develops mostly within the first and, to a lesser degree, the second hour of reperfusion [4, 5] and persists for at least 4 weeks [6]. Confirming experimental observations, several clinical studies, using imaging techniques to visualize myocardial reperfusion, such as contrast echocardiography or magnetic resonance imaging (MRI), have shown areas of low perfusion in the myocardial territories supplied by the infarct-related artery [79]. Importantly, the extent of NR has been correlated with long-term clinical outcomes [712].

Phasic oscillations in CBF can be recorded during coronary angiography, using an intracoronary Doppler guidewire [13]. This technique, applied in patients presenting with acute MI, has shown characteristic changes in CBF after recanalization of the infarct-related artery, including a decrease in antegrade flow during systole and the development of abnormal retrograde flow in early systole. These observations have been correlated with microvascular dysfunction and the presence of NR phenomenon [1416]. Furthermore, the CBF pattern during reperfusion is more sensitive and specific than angiography, electrocardiography, or biochemical assays in predicting the presence of NR phenomenon [17]. Although the relationship between CBF pattern during myocardial reperfusion and microvascular injury has been previously reported [18], the magnitude of the association between CBF pattern and the extent of tissue necrosis and the time course of CBF pattern changes during reperfusion have not been well characterized. In the present study, we aimed at determining quantitatively the relationship between CBF pattern during reperfusion and the extent of NR and infarct size, as well as the time course of the CBF pattern changes during reperfusion, in a porcine model of ischemia–reperfusion. We used an invasive experimental, clinically relevant, animal model, in order to obtain quantitative data that cannot be easily or accurately acquired in clinical investigations.

Μethods

The 18 pigs used in this study received humane care, in compliance with the guidelines for care and use of laboratory animals of the US National Institutes of Health. Animals weighing between 25 and 35 kg were premedicated with ketamine hydrochloride 15 mg/kg intramuscularly (IM), midazolam 0.5 mg/kg IM, and atropine 0.5 mg (if weighing <30 kg) or 1.0 mg (if >30 kg) IM; anesthetized with thiopental sodium 9 mg/kg intravenously (IV) and fentanyl citrate 500 mg IV; and ventilated with a Soxitronic™ (Soxil, S.P.A.; Segrate, Italy) volume ventilator. Thiopental sodium, 3.0 mg/kg/h, and fentanyl citrate, 0.03 mg/hg/h, were continuously infused intravenously throughout the experiments, and 2.0 mg of pancuronium bromide was administered IV every 15 min. After placement of catheters in the (a) right atrium from the right internal jugular vein, (b) right carotid artery, and (c) left jugular vein, the chest was opened via a midline sternotomy and the heart was suspended in a pericardial cradle. A fourth catheter was placed in the left atrium directly through the left atrial appendage. The left anterior descending artery (LAD) was dissected free after the bifurcation of the second diagonal and instrumented with a loose ligature and a transit-time ultrasound flowmeter probe (Μedi-Stim Ιnc, Oslo, Norway), for the continuous monitoring of CBF. After completion of the preparation, the fluid losses were replaced, and 15 min was allowed for hemodynamic stabilization of the animals, before the recording of baseline heart rate, aortic and left atrial pressures, and mean and peak CBF. The LAD was occluded for 1 h, followed by 2 h of reperfusion. Hemodynamic parameters and CBF were continuously monitored and recorded every 15 min throughout the experiments. After 2 h of reperfusion, thioflavin was injected into the left atrium via the left atrial catheter. The LAD was re-occluded, and 1.5 mg/ml of 1% gentian violet was injected into the left atrium over 30 s. The heart was electrically fibrillated within the next 5 s and immediately excised. The left ventricle, including the septum, was separated from the remainder of the heart and cut into 1-cm thick sections, perpendicular to the apex–base axis. The left ventricular area at risk was identified by the absence of gentian violet staining. The NR area was identified by the absence of vital thioflavine stain under a fluorescent lamp [3]. The borders of the area at risk and NR were traced with Sinica dye on the heart slices. Tracings of both sides of each ventricular section, outlined by the gentian violet and thioflavin dye, were drawn on transparent plastic sheets, and the areas at risk and NR were measured by planimetry. The ventricular sections were placed into a 1% triphenyltetrazolium chloride solution to identify the infarcted area (IA). Tracings of both sides of each ventricular section, outlined by the triphenyltetrazolium chloride reaction, were drawn on transparent plastic slides, and the IA was measured by planimetry. All ventricular sections were weighed. The myocardium at risk was expressed as a percentage of the whole left ventricle, whereas the IA and NR were expressed as a percentage of the myocardium at risk. Coronary flow waveforms were analyzed off-line. CBF values were calculated as the average of 20 consecutive cardiac cycles. Retrograde CBF was defined as any flow below the zero baseline of the flow waveform. Peak retrograde CBF was defined as the largest absolute retrograde flow value. Mean retrograde coronary blood flow was defined as the area under the retrograde CBF curve.

Statistical Analysis

The results are expressed as mean±standard error of the mean. Repeated measures analysis of variance was used to examine the changes in mean and peak CBF during reperfusion and in hemodynamic measurements, with a Bonferroni post-test to compare pairs of means at different time points. A linear regression analysis was used to examine the relationship between IA and mean and peak retrograde CBF at the end of reperfusion, and between IA and NR and time-to-development of retrograde CBF. Logistic regression analysis was used to assess predictors of retrograde CBF. A p value <0.05 was considered statistically significant.

Results

Retrograde Coronary Blood Flow

Retrograde CBF was observed in early systole within the first 15 min in seven pigs and between 15 and 30 min in six pigs and persisted throughout the period of reperfusion (Fig. 1a). No retrograde CBF was observed in five pigs (Fig. 1b). In the former group, the absolute mean and peak retrograde CBF increased progressively over time, peaked at 90 min of reperfusion, and decreased during the last 30 min of reperfusion (Fig. 2). Similar results were obtained when mean and peak retrograde CBF were expressed as a percentage of the respective mean and peak anterograde CBF at the same time point (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs12265-010-9240-4/MediaObjects/12265_2010_9240_Fig1_HTML.gif
Fig. 1

Typical coronary flow waveforms from an animal with early systolic flow reversal during reperfusion (a) and another without flow reversal (b). Shaded areas represent the mean retrograde coronary blood flow

https://static-content.springer.com/image/art%3A10.1007%2Fs12265-010-9240-4/MediaObjects/12265_2010_9240_Fig2_HTML.gif
Fig. 2

Time course of absolute mean and peak retrograde coronary blood flow (CBF) during reperfusion

Hemodynamic Measurements

The mean hemodynamic measurements made at baseline, during the first hour of ischemia and during the 2 h of reperfusion are shown in Fig. 3. The baseline mean arterial pressure (MAP) was 100.9 ± 3.4 mmHg and remained stable during coronary occlusion and reperfusion (p = 0.59). Heart rate was 101.7 ± 4.3 bpm at baseline and changed significantly over time (p < 0.0001), increasing during reperfusion by 31–36 bpm (p < 0.05). The baseline mean left atrial pressure was 9.0 ± 0.9 mmHg and changed significantly over time (p = 0.004), increasing during coronary occlusion (p < 0.05), but remaining stable thereafter (p > 0.05) and during reperfusion.
https://static-content.springer.com/image/art%3A10.1007%2Fs12265-010-9240-4/MediaObjects/12265_2010_9240_Fig3_HTML.gif
Fig. 3

Mean arterial blood pressure (BP) (a), heart rate (b), and left atrial pressure (c) at baseline and during ischemia and reperfusion

The baseline mean and peak CBF were 39.0 ± 3.9 and 72.2 ± 6.7 ml/min, respectively. During reperfusion, after an initial period of reactive hyperemia following the release of the occlusion, mean and peak CBF decreased gradually over time (p < 0.0001; Fig. 4).
https://static-content.springer.com/image/art%3A10.1007%2Fs12265-010-9240-4/MediaObjects/12265_2010_9240_Fig4_HTML.gif
Fig. 4

Mean (a) and peak (b) coronary blood flow (CBF) during reperfusion

Morphometric Analyses

The mean percentage of left ventricular myocardium at risk was 26.8 ± 1.5%. The mean IA and NR, expressed as a percentage of the left ventricular myocardium at risk, were 68.9 ± 3.5% and 53.4 ± 4.5%, respectively. As expected, a significant correlation was found between NR and IA (r = 0.64; p = 0.004). In order to demonstrate the functional relationship of retrograde CBF with IA and investigate if a large retrograde CBF in absolute values is associated with larger IA, we plotted IA as a function of the magnitude of retrograde CBF at the end of reperfusion (Fig. 5a). The absolute values of mean and peak retrograde CBF at the end of reperfusion were significantly correlated with IA (r = 0.71; p = 0.001 and r = 0.69; p = 0.001, respectively). In order to exclude the confounding effect of MAP during reperfusion on the relationship between retrograde CBF and IA, we examined this relationship after categorizing the animals into two groups, according to whether their MAP during reperfusion was above or below the median value for all animals. Interestingly, the lines of regression were almost identical for the two groups, suggesting that MAP was neither a confounder, nor an effect modifier of the relationship between retrograde CBF and IA. Similar results were obtained for the relationship of peak retrograde CBF and IA. Since the NR phenomenon is a dynamic process, which increases over time, we investigated the relationship between the time-to-development of retrograde CBF and NR or IA. The time-to-development of retrograde CBF was inversely correlated with IA (r = −0.59; p = 0.010) and NR (r = −0.61; p = 0.008).
https://static-content.springer.com/image/art%3A10.1007%2Fs12265-010-9240-4/MediaObjects/12265_2010_9240_Fig5_HTML.gif
Fig. 5

Linear regression between infarct size (IA-dependent variable) as a percentage of the area of myocardium at risk (AR) and the mean retrograde coronary blood flow (CBF-independent variable). The correlation was statistically significant (r = 0.81; p < 0.0001)

Determinants of Retrograde CBF

Univariate analyses were performed for IA, NR, mean and peak CBF, mean arterial pressure, left atrial pressure, and heart rate at the end of reperfusion (Table 1). The presence of retrograde CBF was significantly correlated to a larger IA (odds ratio 1.12, 95% confidence interval 1.01–1.24, p = 0.037) and NR (odds ratio 1.09, 95% confidence interval 1.01–1.18, p = 0.037). The small number of observations precluded the conduction of a reliable multivariate analysis and therefore the results of the univariate analysis should be interpreted with caution.
Table 1

Determinants of retrograde coronary blood flow

Parameter

Odds ratio

95% Confidence interval

P value

Infarct size

1.12

1.01–1.21

0.037

No-reflow area

1.09

1.01–1.18

0.037

Mean CBF

0.97

0.92–1.03

0.30

Peak CBF

1.00

0.97–1.03

0.90

Mean arterial pressure

1.21

0.98–1.48

0.08

Left atrial pressure

1.21

0.88–1.66

0.25

Heart rate

1.00

0.97–1.04

0.82

CBF coronary blood flow

Discussion

In the present study, we have shown that CBF reversal in early systole is common during reperfusion in acute myocardial infarction. We also report for the first time, to the best of our knowledge, a strong relationship between the presence of this flow reversal, its timing and magnitude, with infarct size, microvascular injury, and no-reflow area. Although previous studies have described a relationship between retrograde coronary blood flow and NR, quantification of this direct relationship between retrograde coronary blood flow and the extent of myocardial tissue necrosis has not been previously reported. In addition, the association between the time course of the CBF pattern during reperfusion and the extent of myocardial necrosis has not been previously described. The present study therefore confirms in a carefully controlled experimental model the previous studies and provides quantitative data about the magnitude and time course of retrograde CBF and infarct size that cannot be derived from clinical observations.

Retrograde CBF and Microvascular Injury

Our observations are concordant with previous clinical studies, which found a correlation between the appearance of retrograde CBF immediately after primary angioplasty for acute MI and the amount of NR ascertained by myocardial contrast echocardiography [14, 15]. The changes in CBF pattern were attributed to the increased impedance of the injured coronary microvasculature [14, 19]. Under normal conditions, CBF in the large epicardial vessels is antegrade throughout the cardiac cycle. However, in the endomyocardial arteries, such as the septal branches of the LAD, CBF is reversed during early systole [20, 21]. These differences are explained by the higher capacitance of the large epicardial vessels [20, 21]. When the microvasculature is diffusely injured, as often happens during ischemia and after reperfusion, the microvascular impedance increases, while the intramyocardial blood volume decreases. In this situation, the increased myocardial stress during systole, combined with the increased microvascular impedance, results in less effective propulsion of the intramyocardial blood into the venous circulation. The intramyocardial blood is thus pushed back to the epicardial coronary artery to produce early systolic retrograde CBF [14, 19].

In previous experimental studies, the NR phenomenon developed within the first and to a lesser degree within the second hour of coronary reperfusion [4, 5]. Furthermore, the NR phenomenon is a dynamic process, which evolves during reperfusion, nearly tripling in size between 2 min and 3.5 h [22]. In the present study, retrograde CBF was observed within the first 30 min of reperfusion and increased significantly in absolute value during reperfusion. The similar time course of the NR phenomenon and retrograde CBF suggests a common pathophysiology. An equally important observation made in our study was a larger infarct size and greater extent of NR associated with an earlier appearance of retrograde CBF during reperfusion. These results indicate a functional relationship between the severity of retrograde CBF and the extent of NR phenomenon and infarct size. The earlier appearance of retrograde CBF might indicate more extensive microvascular injury, resulting in a larger NR area and infarct size.

Retrograde CBF and Infarct Size

Previous studies have suggested that changes in the CBF pattern during reperfusion of the previously ischemic myocardium are indicative of the extent of myocardial injury. In an experimental model of ischemia–reperfusion, the presence of early systolic retrograde CBF was related to the pathologic characteristics of the reperfused myocardium, such as coagulation necrosis, marked cellular vacuolar degeneration, and diffusely distributed red blood cells in the interstitial tissue [23]. The present study showed, for the first time, that the presence of retrograde CBF is related, not only to more extensive microvascular injury, but also to a larger myocardial infarct size. Furthermore, there was a linear relationship between the severity of retrograde CBF at 2 h of reperfusion and the infarct size, which further strengthens their association. We did not perform ligation at different sites of the coronary tree (more distal LAD or different coronary artery branches) in order to investigate if the association between retrograde CBF and infarct size is modified by the size of the area at risk. Although it is reasonable to assume that a larger area at risk, by leading to a larger infarct, would increase the absolute values of retrograde flow, we do not have direct data to support this.

We have also shown that infarct size predicts the presence of retrograde CBF. This observation is consistent with a previous study by Reffelman et al. [4], who showed that infarct size is the major determinant of reflow at a given time point of reperfusion (and hence no-reflow, which in turn translates into retrograde CBF).

Clinical Implications

The results of our study indicate that the presence of early retrograde systolic CBF during myocardial reperfusion is associated with severe myocardial and microvascular injury and that its observation might be a clinical index of unsuccessful tissue reperfusion. Several lines of evidence suggest that platelet and thrombus embolization, associated with primary coronary intervention, constitute a major cause of no-reflow in humans, and this can impact no-reflow independently of the size of the area at risk [2, 24]. In this setting, prompt recognition of unsuccessful tissue reperfusion becomes of paramount importance. CBF can be measured with a Doppler guidewire, a technique applicable immediately after the completion of primary coronary intervention for acute MI [13], without markedly lengthening the procedure. Therefore, it might provide important prognostic information more expeditiously than myocardial contrast echocardiography or MRI, diagnostic techniques that are not always available, require the injection of contrast agents during the acute phase of MI and, in the case of MRI, can only be performed when the patient is clinically stable. The use of CBF pattern is limited by its application to the left coronary artery only, since the right coronary artery flow pattern follows that of the aorta, peaking during systole [25, 26]. However, it is for patients with large infarcts in the territories perfused by the LAD that this additional prognostic information is likely to be important, since it may contribute to the identification of patients needing an aggressive approach, including interventions to reduce the NR phenomenon [27, 28].

Conclusions

The presence of early systolic retrograde CBF during recanalization of an infarct-related epicardial artery indicates the presence of a large area of NR and myocardial infarct size. This observation could be used immediately after primary coronary intervention to risk-stratify patients presenting with an acute MI.

Copyright information

© Springer Science+Business Media, LLC 2010