We investigated the effect and mechanism of action of potassium in flushing solution on lung graft function. Five groups of rabbits with different potassium concentration solutions were evaluated : Group 1 (n = 7), 0mEq/L ; Group 2 (n= 6), 5mEq/L ; Group 3 (n= 6), 10mEq/L ; Group 4 (n=6), 15mEq/L and Group 5 (n=10), 20mEq/L. The grafts were topically flushed
in vivo with the respective solutions and harvested. After 18 hours preservation at 8°C, graft function assessment was performed. We used the system in our department which utilizes autologous blood. The graft was placed in a closed circuit for 30minutes. We defined this period as the first reperfusion. The circuit was then intercepted, and the grafts were perfused with preserved autologous blood for 120 seconds. We defined this period as the second reperfusion.
During flushing, the mean pulmonary artery pressure (PAP) was significantly higher (p < 0.05-0.01) in Group 5 than in the other groups. During the 1st reperfusion, 2 cases in Group 2 were excluded due to extreme values on Smirnov test. Two cases in Group 4 and 4 cases in Group 5 were excluded due to lung edema. During the 1st reperfusion, a significant indirect correlation was found between average PAP (avPAP) in each group (Y) and potassium concentration (X) (Y = 24.2 +0.85X ; r = 0.912, p <0.01). During the 2nd reperfusion, a significant correlation was found between mean PAP (Y) in each group and potassium concentration (X) at 60 sec and 120 sec.
(60sec : Y=17.674+2.322X ; r=0.917, p<0.01)
(120sec : Y=21.112+2.738X ; r=0.919, p<0.01)
An indirect correlation between mean pO
2 (Y) and potassium concentration (X) was observed at both 60 sec and 120 sec.
(60sec : Y=162.552-3.615X ; r=0.964, p<0.01)
(120sec : Y=147.888-3.708X ; r=0.913, p<0.01)
Based on F-test for avPAP in the 1st reperfusion and PAP at two points during the 2nd reperfusion, the 5 groups could be divided into two groups : A) Group 1-3 and B) Group 4-5. On comparison of Group A with Group B, a significant difference (p < 0.05-0.01) was found in avPAP in the 1st reperfusion and pO
2 and PAP during the 2nd reperfusion. The above results suggest that potassium ion may injure the graft and thereby affect its function.
We presumed that the above findings were related to the action of potassium ion channels, therefor added Ba
2+ into the solution for Group 1 (Group 1 + Ba
2+ (n=6)) and Group 5 (Group 5+ Ba
2+ (n=6)). During flushing, the PAP (mean± SD) in Group 1+ Ba
2+ (12.5±1.00) was significantly higher (p <0.05-0.01) than in Group 1 (11.2 ±0.86). However, there was no significant difference between the PAP in Group 5 + Ba
2+ and Group 5 without Ba
2+. Although there was no significant difference in the results between Group 1+ Ba
2+ and Group 1 during graft function assessment, the results of average PAP during the 1st reperfusion and pO
2 and PAP during the 2nd reperfusion of Group 5+ Ba
2v were significantly better than those of Group 5.
We conclude that the potassium concentration in flush solution should be within 10mEq/L, or it may be desirable to have a flush solution without potassium. This special characteristic of lung flush solutions is likely to be related to the potassium ion channel.
View full abstract