Erythrocyte-containing versus crystalloid cardioplegia in the rat: effects on myocardial capillaries

Background The purpose of this study was to investigate the effects of crystalloid and erythrocyte-containing cardioplegia on capillary morphology of the isolated erythrocyte-perfused rat heart. Methods Hearts from adult Sprague-Dawley rats were perfused throughout with resuspended sheep erythrocytes and subjected to the following protocols (n = 6, all groups): (1) 15 minutes nonworking and 30 minutes working heart mode (control; group 1); (2) as for group 1, with 30 minutes erythrocyte-containing (BL) or crystalloid (CR) cardioplegic arrest without reperfusion (groups 2BL and 2CR); (3) as for group 2, with 30 minutes nonworking reperfusion (groups 3BL and 3CR); and (4) as for group 3, with 30 minutes working heart mode (groups 4BL and 4CR). After each protocol troponin I from coronary effluent was measured. Corrosion casts were then made of the coronary microvasculature. Cast density was calculated as cast volume per left ventricular dry weight. Casts also underwent scanning electron microscopy. Analysis was by analysis of variance. Values are mean ± standard deviation. Results Prearrest working heart coronary flow averaged 15.1 ± 4.7 mL/min without any differences among groups. Coronary flow in group 4 working hearts was the same before and after either cardioplegia. Cardiac outputs were similarly consistent in all groups. Cast density in group 1 (control) was 9.60 ± 1.17 × 10−2 mm3/mg. It was unaltered by erythrocyte-containing cardioplegia, but after crystalloid cardioplegia (group 2CR), it was 6.52 ± 0.93 × 10−2 mm3/mg (p = 0.0001 versus group 1 and p = 0.0007 versus group 2BL). With 30 minutes of nonworking reperfusion (group 3CR, there was slight improvement in cast density at 7.60 ± 0.90 × 10−2 mm3/mg (p = 0.0072 versus group 1; p = 0.0242 versus group 3BL). No further improvement was seen in group 4CR. Electron micrographs showed circumferential angularities or narrowings in crystalloid-perfused, arrested hearts, consistent with ischemic damage. Troponin I rose significantly after reperfusion in all groups, but it was higher in crystalloid-perfused, arrested hearts: 0.054 ± 0.013 μg/L versus 0.024 ± 0.017 μg/L (p = 0.0273). Conclusions Erythrocyte-containing cardioplegia maintained capillary density and morphology. Crystalloid cardioplegia produced capillary loss, visible abnormalities, and higher troponin I release. These hearts may be more vulnerable to myocardial damage during reperfusion than hearts perfused with erythrocyte-containing cardioplegic solution.