(Circulation. 1995;92:223-227.)
© 1995 American Heart Association, Inc.
Articles |
From the Department of Surgery, Columbia University College of Physicians and Surgeons, New York, NY.
Correspondence to Mehmet C. Oz, MD, Department of Surgery, Columbia-Presbyterian Hospital, Milstein 7-435, 177 Fort Washington Ave, New York, NY 10032.
| Abstract |
|---|
|
|
|---|
Methods and Results In 15 donor hearts preserved in University of Wisconsin solution, heart weight (HW) was obtained immediately after explantation and after transport before implantation. Left ventricular mass (LVM) was calculated separately in 18 patients with the use of epicardial two-dimensional echocardiograms obtained both before explantation from the donor and after transplantation and weaning from cardiopulmonary bypass. While changes in LVM could be due to preservation or reperfusion injury, changes in HW can only be due to edema occurring during transport. HW averaged 339±24 g (mean±SE) before and 340±24 g after transport (P=NS); however, LVM increased 14 g, from 164±8 to 178±11 g (P<.05, paired t test). LVM increased in 10 of 18 patients (56%). No correlation was demonstrated between duration of ischemia (mean, 172±13 minutes) and changes in HW or LVM. Two patients died as a result of primary graft failure. In the first, HW increased 54 g, 2 SD above the mean. In the second, LVM increased 66 g, 2 SD above the mean, but HW changed minimally.
Conclusions While current preservation methods result in minimal change in HW during transport, reperfusion injury frequently increases LVM. LVM determination by two-dimensional echocardiography may prove valuable in detecting allograft injury.
Key Words: echocardiography edema transplantation reperfusion
| Introduction |
|---|
|
|
|---|
Structural changes after reperfusion of ischemic myocardium include disruption of the microtubules,19 formation of contraction bands in the contractile proteins, and appearance of calcific granules within the mitochondria. Cellular edema also occurs, associated with the disruption of sarcoplasmic and mitochondrial membranes20 and with impaired left ventricular (LV) diastolic function.21 Other changes include intracellular calcium flux and loading and formation of oxygen free radicals.22 23 24
Neutrophil chemotaxis and activation also are implicated in reperfusion injury.25 Ischemia degrades myocyte ATP, with accumulation of AMP, adenosine, inosine, hypoxanthine, and xanthine. Washout of these metabolites during reperfusion prevents restoration of ATP, depresses myocardial function, and contributes to synthesis of oxygen free radicals,26 27 proteases, and lipases, which further damage the myocardium.
Little information is available regarding the extent of reperfusion injury and myocardial edema after human allograft transplantation. Accordingly, this study was designed to discriminate between injury caused by myocardial preservation, manifested as an increase in heart weight (HW), and injury resulting from reperfusion, detectable as changes in LV mass (LVM) as measured by two-dimensional echocardiography (2-DE).
| Methods |
|---|
|
|
|---|
After completion of 2-DE imaging, the left inferior pulmonary vein and the inferior vena cava were partially divided to decompress and vent both ventricles. The aorta was cross-clamped, and 1 L of cold (4°C) University of Wisconsin (UW) solution was perfused through the aortic root. Pressure applied to the cardioplegia bag was kept under 150 mm Hg, resulting in an estimated root pressure of less than 60 mm Hg. LV distension was carefully avoided. Cold (4°C) normal saline was poured into the pericardium for topical hypothermia. Maintaining sterility, the arrested heart was excised, trimmed, drained dry, and weighed (scale XL-1800, Denver Instrument Co). The heart was transported in a multiple-bag UW solution bath surrounded by ice, but direct contact between the ice and the heart was avoided.
After transport to Presbyterian Hospital, the heart was unpacked, drained, and reweighed using the same scale. After graft implantation, institution of inotropic support, and separation from cardiopulmonary bypass, the LV SAC was again imaged by 2-DE and recorded.
Ischemic time of the allograft, defined as the time between aortic cross-clamping for donor heart explanation and initiation of reperfusion after transplantation, was recorded.
Data Analysis
Two-dimensional echocardiographic SAC was
used to determine LVM, ejection fraction (EF), and
end-diastolic volume (EDV) as previously described by
an observer blinded to true values for HW.30 Videotaped
2-DE images were analyzed with a computerized light pen (Varian
3000 phased-array ultrasonograph). End-diastolic
image boundaries were hand drawn in images frozen at the maximum
presystolic area. The epicardial area (Aepi)
was drawn at the myocardium-pericardium interface, and
the endocardial area (EDA) was drawn at the
myocardium-blood interface. The numerical
difference between the epicardial and endocardial areas was used to
calculate Am, the area of the myocardial ring.
Values for Am and EDA were converted to
angiographically validated values for LVM and LVEDV by the following
relations: (1) LVM=6.6 Am+21 and (2) LVEDV=3.7
EDA+40.
EF required additional planimetry of the endocardial area at end systole (ESA), defined as the smallest endocardial area. EF was then calculated as a shortening fraction31 : EF=100x(EDA-ESA)/EDA.
Results for each measure of function were calculated for three representative beats and averaged. A two-tailed paired t test was used to compare changes in LVM, LVEDV, EF, and HW. In addition, regression analysis between ischemic time and these measures was performed.
| Results |
|---|
|
|
|---|
|
Comparison of HW (339±24 versus 340±24 g), LVEDV (101±4 versus 95±3 mL), and EF (51±3% versus 53±3%) by paired t test showed no significant differences over time. In contrast, there was a significant 14±6 g increase (164±8 versus 178±11 g, P<.05) in LVM before explant versus after transplant. Ten of 18 cardiac allografts (56%) exhibited an increase in LVM after transplantation. No correlation was demonstrated between ischemic time and changes in LVM or HW. Similarly, no correlation was found between LVM and HW.
Two of the 22 recipients died as a result of primary graft failure. The
first received a cardiac allograft that sustained a HW increase of 54
g, 2 SD (SD=22 g) above the mean HW change. This was the largest weight
gain in this series. No technical problems or depression of LV function
was noted during harvesting. No posttransplant 2-DE images were
obtained, because the patient could not be separated from
cardiopulmonary bypass. The second patient received a
cardiac allograft that sustained a 66-g increase in LVM but only a 9-g
increase in HW. The increase in LVM was 2 SD (SD=25 g) above the mean
LVM change. This was the largest LVM increase in this study. Figs
1
and 2
illustrate 2-DE images in this patient
before explantation in the donor and after transplantation. The
thickness of the LV myocardial ring increases after reperfusion.
|
|
| Discussion |
|---|
|
|
|---|
No significant change in HW was observed despite a mean ischemic time of almost 3 hours. This supports the view that UW does not promote edema formation when used as a storage medium. UW is specifically structured to prevent edema through the use of impermeants and other factors; the clinical data reported here support its efficacy as a storage medium. Increased mass during reperfusion could indicate inadequate myocardial protection unmasked by reperfusion or could indicate subsequent injury during weaning from bypass.
Despite the fact that UW was usually efficacious in preventing weight gain, one heart in this series exhibited a large weight gain of 54 g and was associated with recipient death as a result of an acute power failure. This large weight gain is disturbing in retrospect, particularly in light of the small changes observed in most other patients. However, this event occurred early in this study, and its clinical relevance is unclear. With no apparent difficulties during the organ harvest and in the absence of other reasonable therapeutic options (the recipient had already undergone cardiectomy), implantation of the allograft appeared to be a reasonable course. What should be done in the future if a similar weight gain is observed, and whether weight gained during transport will prove clinically relevant, remains a problem. All residual fluid must be drained from the heart before weighing.
The minimal weight gain reported in this series should be contrasted with weight gained commonly observed with clinical cardioplegia solutions, particularly Plegisol.32 Cardioplegia-induced edema has been associated with impaired diastolic filling in laboratory studies.32
The increase in LVM based on 2-DE was highly significant and almost certainly reflects edema caused by cardiac injury during reperfusion and weaning from bypass. This could be a delayed manifestation of ischemic damage during transport or an effect of hemodilution and inotropic support. In addition to parenteral crystalloid administered during cardiac allografting, many of these patients suffer from preoperative fluid retention. Thus, a graft from a normal donor may be reperfused with blood that is hyposmolar and hypooncotic relative to its native environment. Studies of myocardial water content by Gross33 suggest that this phenomenon alone could cause substantial increases in LVM during reperfusion. Prior studies also suggest that high levels of inotropic stimulation can cause subendocardial injury, which also contributes to edema formation.
The technique used for the arrest, harvest, and preservation of the hearts was unchanged during this study, as was the immunesuppression protocol. There was no evidence of hyperacute rejection in any of the observed patients; including the two patients who died.
Measurement of changes in LVM by 2-DE during cardiac surgery in animals and after transplantation have been reported previously,30 32 34 35 36 37 but statistically significant changes of the magnitude reported here have not been observed previously. The use of single-section algorithms for measurement of LVM is inferior to multiple-section algorithms when accurate sections can be obtained. However, we find the reproducibility of long-axis LV sections to be inferior to short-axis sections because of difficulties in localizing the apex. Admittedly, single-section measurements of LVM are volume dependent. Recently, a mathematical method for correcting single-section LVM calculations for changes in LVEDV has been described (Cabreriza et al, unpublished data, with permission). That correction has not been applied to the present data because the change in LVEDV in the present study is not statistically significant.
Recent studies involving the use of neutrophil adhesion molecule blockingagents38 or reperfusion with leukocyte-depleted blood39 offer possible approaches to the prevention of reperfusion injury and myocardial edema. Byrne et al40 demonstrated that antibodies to CD18 or ICAM-1 reduce posttransplant reperfusion injury in rabbits. In the absence of treatment, myeloperoxidase activity, myocardial water content, and coronary vascular resistance were associated with increased myocardial inflammation, edema, low reflow, and a leftward shift of the end-diastolic pressure-volume relation.
Sun et al41 demonstrated in transplanted hearts in rats that free radical scavengers (superoxide dismutase and catalase) with glucose-insulin-potassium improved LV end-diastolic pressure, rate of rise of LV systolic pressure, myocardial blood flow, coronary resistance, and myocardial ATP. The effectiveness of free radical scavengers, confirmed in other studies,42 43 44 45 may result from adenosine-mediated coronary vasodilation. Other approaches to reperfusion injury include prevention of adenosine transport and ATP depletion,46 47 reduction of intracellular calcium flux with calcium antagonists48 or 2,3-butanedione monoxime,49 and the addition of cGMP to reduce ventricular relaxation abnormalities.50
Summary
While changes in LVM by 2-DE may be due to effects of
preservation
or reperfusion injury, changes in HW can only be due to
preservation-related edema. The major benefit of the present
study is that it provides clinical evidence that myocardial edema
formation occurs predominantly during reperfusion. In addition, this
study demonstrates that current preservation techniques with UW are
generally adequate in preventing myocardial edema. While reperfusion
injury and myocardial edema may occur after human allograft
transplantation, the extent of initial LV dysfunction is variable.
The present study reveals that observed increases in LVM after
separation from cardiopulmonary bypass are statistically
significant but do not adversely effect clinical outcome in the
majority of cases studied. LVM determination by 2-DE and measurement of
HW before and after transport could be valuable in detecting allograft
injury clinically and provide a useful tool with which to evaluate
future techniques of myocardial preservation and stratagies for
reducing reperfusion injury.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. M. R. Amirhamzeh, D. A. Dean, C.-X. Jia, S. E. Cabreriza, J. P. Starr, M. J. Sardo, N. Chalik, M. L. Dickstein, and H. M. Spotnitz Iatrogenic Myocardial Edema: Increased Diastolic Compliance and Time Course of Resolution In Vivo Ann. Thorac. Surg., September 1, 1996; 62(3): 737 - 743. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |