| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2006;29:517-524
© 2006 Elsevier Science NL
a Department of Cardiothoracic Surgery, University Hospital Erlangen, University of Erlangen-Nuremberg, Krankenhausstraße 12, 91054 Erlangen, Germany
b Department of Anesthesia, University Hospital Erlangen, University of Erlangen-Nuremberg, Krankenhausstraße 12, 91054 Erlangen, Germany
Received 25 September 2005; received in revised form 23 December 2005; accepted 29 December 2005.
* Corresponding author. Address: Klinik für Herzchirurgie, University Hospital Erlangen, Krankenhausstrasse 12, 91054 Erlangen, Germany. Tel.: +49 9131 8533319; fax: +49 9131 8532768. (Email: olaf.roerick{at}herz.imed.uni-erlangen.de).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Key Words: Low-flow perfusion • Deep hypothermic circulatory arrest • Aortic arch • Collateral blood flow
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
2.2 Surgical procedure
The piglets were operated via a median sternotomy using cardiopulmonary bypass. Heparin was administered intravenously with a dosage of 400 U/kg after opening the pericardium. The activated clotting time (ACT) was measured and, if necessary, additional heparin was administered in piglets that had an ACT below 500 s after the first bolus. The right common carotid artery was dissected and an ultrasound flow probe (T 206 Transonic Systems Inc.) was fixed to the vessel wall. The cardiopulmonary bypass was started after cannulation of the ascending aorta (12 F Fem-Flex II, Edwards Lifesciences, Unterschleissheim, Germany) and the right atrium (24 F, Medtronic dlp®, Minneapolis, USA). For providing cardiopulmonary bypass, a roller pump (Stöckert, Munich) and a standard pediatric tubing system (Sorin, Milan, Italy) were used. The priming consisted of 500 ml HAES 6%, heparin 100 U/kg, 300 ml donor porcine blood, 20 ml sodium bicarbonate 8.4% and 30 mg/kg prednisolone hemisuccinate [11] (SoluDecortin H, Merck, Darmstadt, Germany). In order to achieve a hemoglobin level of above 4.6 g/dl, the priming was hemofiltrated and in some cases adding of more donor blood was necessary. The flow of the cardiopulmonary bypass was kept at 100 ml/(kg min). Circulating blood was permanently hemofiltrated (Haemoconcentrator DHF O2, Dideco, Mirandola, Italien), hemoglobin was kept above 5.0 g/dl, and base excess was adjusted with additional bicarbonate administration if exceeding –5 mmol/l.
2.3 Deep hypothermic circulatory arrest
The animals that received the operation under the conditions of deep hypothermic circulatory arrest were cooled down to a temperature right below 20 °C, which is at the upper limit of DHCA, as DHCA is commonly used with temperatures between 15 and 20 °C in the clinical setting. The heart was arrested with Bretschneider's cardioplegic solution (4 °C, 30 ml/kg bodyweight), administered via the aortic root. Afterwards, circulation was arrested for 60 min. Reperfusion at 20 °C was carried out with the same perfusate temperature for 5 min before continuous rewarming to 37 °C rectal temperature. For rewarming, a maximal temperature gradient of 10 °C between perfusate and patient temperature was respected. If the heart did not start beating spontaneously at a temperature of 35 °C, it was defibrillated. As soon as contractions were observed during rewarming, dobutamine at a concentration of 5 µg/(kg min) was given. All animals were treated using this scheme, so comparability concerning the neurologic outcome exists. Circulation was stabilized by volume infusion when needed or by adjusting dobutamine levels to achieve adequate myocardial function. As soon as the rectal temperature achieved 37.0 °C, rewarming was finished. Afterwards, a modified ultrafiltration according to Jonas and Elliott [3], using our own modification [12], was carried out.
2.4 Continuous antegrade low-flow perfusion via the innominate artery
The cooling by means of CPB was stopped as soon as a temperature of right below 25 °C was achieved. The aortic cannula was advanced into the innominate artery, snared around with a tourniquet, and the aorta was cross-clamped. Bretschneider's cardioplegic solution was infused as mentioned above. A pump flow of 30 ml/(min kg) was maintained for 60 min, and blood pressure was measured in the right subclavian artery. The level was adjusted pharmacologically below 50 mmHg in order to preserve cerebral auto-regulation. Continuous antegrade low-flow perfusion via the innominate artery was applied without ultrafiltration. After 60 min, the aortic cannula was repositioned into the ascending aorta. Reperfusion was carried out in the same way as in the deep hypothermic circulatory arrest group with the only difference of starting at a temperature of 25 °C.
2.5 Blood gas management during cardiopulmonary bypass
As it was an intention to compare alpha-stat with pH-stat principles in blood gas management, each protocol was applied to half of the piglets of the DHCA and the LF groups. The latest modifications of the gas management according to Jonas and co-workers [13] were used. If necessary, the piglets got additional carbon dioxide insufflated via the oxygenator.
2.6 Measurements
For measuring the flow at one femoral artery and one internal carotid artery, ultrasound flow probes as mentioned above were used. The sole measurement of subdiaphragmal blood flow was the right femoral artery. Laboratory values were measured by the clinical laboratory of the University of Erlangen-Nuremberg. Somatosensory-evoked potentials were acquired as explained above.
2.7 Statistical analysis
Statistics were performed using statistical software (EXCEL, Microsoft Corporation, Redmond, USA, and SPSS for Windows, SPSS Inc., Chicago, USA). The data are expressed as mean standard deviation of the mean values. Tests on significance have been carried out by the Student's t-test or the non-parametric U-test, according to the task. A statistically significant difference was considered to exist at a probability value of less than 0.05 (p
< 0.05), a highly significant difference at a probability value of less than 0.01 (p
< 0.01).
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
|
The average body size of the animals in the circulatory arrest group was 71.5 ± 1 cm with a weight of 12.4 ± 0.3 kg compared to 68.6 ± 0.7 cm and 11.0 ± 0.2 kg in the low-flow group (p < 0.05). The temperature course during the operation is presented in Fig. 1 . In the DHCA group – during cardiac arrest – a temperature of right below 20 °C was achieved and maintained. For the low-flow group the temperature was right below 25 °C.
|
3.2 Measurement of blood flow
Blood flow through the carotid artery was never significantly different between the groups except during the circulatory arrest period that resulted in total loss of any flow. However, in the low-flow perfusion group a flow of 120.5 ± 9.0 ml/min was measured in the carotid artery during the low-flow period. This means that with initiation of antegrade brain perfusion the flow was doubled as compared to the flow measured before cross clamping the aorta. Blood gas management had no influence on the carotid flow.
Blood flow through the femoral artery was nearly the same in both groups at any time during the operations. Even during low-flow perfusion via the innominate artery only negligible flows of 1–3 ml/min in the femoral arteries were detected.
3.3 Blood pressure and hemodynamic measurements
Mean arterial blood pressure was kept in physiologic ranges and showed only slight differences between the groups. During DHCA, there was certainly no measurable pressure, whereas the arterial pressure in the right subclavian artery was regulated pharmacologically to about 50 mmHg or lower during low flow. Hemodynamic measurements showed stable conditions throughout. Fig. 1 represents mean arterial pressures in the femoral arteries.
3.4 Somatosensory-evoked potentials
The appearance of the cortical SSEP-answer was visually analyzed by a blinded specialist. Both low-flow perfusion group and the circulatory arrest group lost their somatosensory-evoked potentials during cardiopulmonary bypass during the cooling period. In none of the animals operated under the condition of DHCA any reoccurrence of SSEPs was present, whereas 10 animals of the low-flow group re-established their cortical potentials through stimulation of the nervus medianus within first 5 min of reperfusion, representing a nearly full functional recovery. The remaining two animals of the low-flow group that showed no reoccurrence of SSEPs belonged to the alpha-stat group (p
< 0.05).
3.5 Laboratory values
3.5.1 Base excess
As shown in Fig. 2
, the base excesses of the two groups showed only little and insignificant differences until the end of reperfusion and rewarming. At the end of CPB (t
6) and after modified ultrafiltration (t
7) negative base excess values were significantly higher in the low-flow animals compared to DHCA animals, corresponding to the lactate values. The differentiation between various sites of measurement has been carried out for lactate only.
|
|
|
|
For lactate concentrations in the inferior vena cava, low-flow-perfused animals that received pH-stat management showed nearly the same values as circulatory arrest animals that received alpha-stat management. Circulatory arrest piglets that received pH-stat management had the lowest lactate concentrations in the inferior vena cava, whereas low-flow animals with alpha-stat management had the highest values during the whole experiment.
3.7 Gamma-glutamyl transpeptidase (gamma-GT)
The gamma-glutamyl transpeptidase, a highly sensitive indicator of liver cell damage, revealed no significant differences between low flow and circulatory arrest animals and never rose significantly during the test.
3.8 Aspartate amino-transferase (AST)
The aspartate amino-transferase is contained in muscle cells and liver cells and is being released in case of their damage. It was clearly evident that in case of low-flow perfusion AST had higher values than under the condition of circulatory arrest (see Fig. 6
).
|
3.10 Alpha-amylase
As an indicator for pancreatic tissue damage, alpha-amylase was measured. The values before the actual beginning of the operation were higher than the normal human range, which is typical of the species. The animals operated under the conditions of deep hypothermic circulatory arrest had slightly higher values already at the beginning of the operation. The difference was insignificant right after sternotomy, but significantly higher values were detected in case of the circulatory arrest animals as compared to the low-flow animals already after 20 min of cardiopulmonary bypass. The elevation persisted and became even highly significant after the modified ultrafiltration was performed.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Though a blood flow of about 10 ml/(kg min) via the innominate artery cannula is generally considered to produce a sufficient neuroprotective cerebral blood flow at 25 °C, we used the threefold value in order to provide enough flow for perfusion of abdominal organs and of the lower limbs via natively present collaterals [14]. However, the flow measured at the femoral arteries did not exceed 1–3 ml/min and seems to be far away from a sufficient collateral perfusion.
The main reason for using femoral artery flows as the only site for measurement for subdiaphragmatic blood flow was to simplify our experimental protocol. During our preliminary experiments, flow measurements at the level of several abdominal arterial vessels (arteria hepatica communis, arteria mesenterica superior) after median laparotomy were tested in five animals. We observed similar flow values in all abdominal vessels as in the right femoral artery, which convinced us that femoral artery flow measurements are sufficient enough to estimate subdiaphragmal arterial flows.
Nevertheless, we can assume that in the case of the abdominal organs there is a protective effect of low-flow perfusion via the innominate artery for abdominal organs through native collaterals, because our data prove that the animals operated under conditions of low flow have no increased values for gamma-GT, alkaline phosphatase, creatinine (not listed above), ALT, and alpha amylase as compared to animals operated under conditions of deep hypothermic circulatory arrest, although their temperature was adjusted to 25 °C during low-flow perfusion as compared to the more protective temperature of 20 °C in the DHCA group. AST was the only measured laboratory parameter representing the damage of liver tissue and muscle cells that was clearly elevated in the low-flow animals after the low-flow period. We attribute this pattern of enzyme elevation to relative ischemia in the muscles of the lower limbs because it was not accompanied by an appropriate elevation of ALT and gamma-GT as it would be in the case of significant liver cell damage. This observation is supported by the elevated lactate levels after reperfusion in all low-flow animals. Clear evidence for the subdiaphragmal origin of elevated serum-lactate is the fact that at the beginning of reperfusion (t 4), the amounts of lactate in the circulatory arrest group and in the low-flow group as measured in the jugular bulb are similarly high, but in the low-flow animals, lactate is clearly elevated in the samples from the inferior vena cava right after the removal of the aortic clamp. Thirty minutes after reperfusion (t 5), both values approached the jugular bulbs once again, probably due to mixing of the blood. This means that the ‘trickle-flow’ of 1–3 ml/min in the femoral arteries is too low to avoid ischemic muscle cell damage in the lower limbs, resulting in a metabolic acidosis and lactate elevation that was higher in the low-flow group than in the circulatory arrest group due to the less protective temperature of 25 °C in the low-flow group as compared to 20 °C in the circulatory arrest group. These temperatures were chosen consistent with the most common clinical settings. Measuring creatine kinase, especially its skeletal muscle isoenzyme CK-MB, would have been helpful for tracing the origin of lactate elevations even more precisely, but creatine kinase and its isoforms are difficult to quantify in piglets due to the fact that these animals have multiple other isoforms of this enzyme than the human species.
There is a circumstance that makes it difficult to transfer the data received from this study to the clinical setting of aortic arch operations in young children: the most common operations concerning the aortic arch in newborns aim at the reconstruction of stenoses or hypoplasias and for corrections of arch interruptions. As these malformations usually imply the formation of collaterals, these children might have a much higher potential of receiving sufficient collateral flow when being perfused with a continuous low-flow protocol through the innominate artery. To explore the potential of those collaterals in pathological conditions, it would be interesting to carry out the same experiment after previous surgically induced stenoses of the proximal or distal aortic arch in order to stimulate the formation of collaterals.
Cortical somatosensory-evoked potentials clearly recovered in 10 of the 12 animals operated under the conditions of low-flow perfusion via the innominate artery, but not in any of the 14 animals operated under the conditions of deep hypothermic circulatory arrest. This proves that antegrade low-flow perfusion to the brain is more neuroprotective as compared to deep hypothermic circulatory arrest. This result appears even more striking when bearing in mind that the animals operated under the conditions of low-flow perfusion were adjusted to a less protective rectal temperature of 25 °C during the low-flow period, whereas the deep hypothermic circulatory arrest group was adjusted to 20 °C, indicating that temperature alone is not the only protecting factor in the low-flow group: the flow itself can even compensate for the higher tissue temperature, which confirms the results of previous works on this topic.
As far as the SSEPs are concerned, not even the pH-stat principle for adjustment of blood gases, currently known to be the best neuroprotective method in DHCA [15], could help any of the circulatory arrest animals to retrieve cortical SSEPs. On the contrary, pH-stat principles turned out to be advantageous in those operations performed under low-flow perfusion, resulting in full functional cerebral recovery for all animals compared to a lower recovery rate in the alpha-stat group (100% vs 80% SSEP recovery; p < 0.05). This is consistent with the animal experiments of Priestley et al. [15] confirming the advantageous neurological outcome in piglets comparing the pH-stat method versus an otherwise identical CBP-alpha-stat protocol. Surprisingly, management of the acid–base balance did not have any significant influence on the flow in the carotid artery despite the vasoactive potential of CO2. The sole neuroprotective effect of the pH-stat method, therefore, seems to be improved oxygen delivery to the perfused brain tissue supported by a general acidosis, which results in overall less anaerobic metabolism. Lactate values were therefore significantly lower during reperfusion using the pH-stat management as compared to alpha-stat management. As already explained, we attribute this to a much better pre-oxygenation during the cooling period under these rather ‘unphysiologic’ conditions.
During the entire experiment it was observed that despite a blood flow of 300–450 ml/min was applied to the innominate artery, only 120.5 ± 9.0 ml/min was measured in the right internal carotid artery. We assume that most of the difference is a runoff to the subclavian artery, which is obviously not reaching the femoral vessels via native collaterals.
|
5. Conclusion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
However, according to our study, there is some evidence that low-flow perfusion at 25 °C leads to some tissue damage in the musculature of the lower limbs. Despite this, its superior potential in preserving the functional integrity of the spinal cord and brain justifies the expanding application of this perfusion technique in congential aortic arch surgery. Great care must be taken if higher temperatures than 25 °C are applied to the patients, because borderline perfusion in the presence of mild hypothermia (like 28–30 °C) may not be protective enough in the absence of bigger collateral vessels.
The combination of low-flow perfusion and pH-stat principles resulted in the best neurological outcome as measured with SSEPs and resulted in the lowest lactate elevation indicating the least tissue damage.
To draw a clinical consequence from our experiments, we would like to place a word of caution that some children may not be well protected by low-flow perfusion via the innominate artery, although their neurological outcome seems to be better. Especially, children already presenting with a severe acidosis before surgery seem to be at increased risk, as they probably have insufficient collaterals not warranting an adequate perfusion of subdiaphragmal organs and lower limbs.
|
Appendix A |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Dr V. Tsang (London, United Kingdom): Did you look at the renal function?
Mr Roerick : We looked at the renal function only in measuring the creatinine, which was the same in both groups. So the question is maybe whether the creatinine can really represent renal function in that short time. But the levels were the same in both groups during the whole operation.
Dr Tsang : And what was the reason for randomizing the two groups to either pH-stat or alpha-stat? Were you introducing another variable?
Mr Roerick : Actually, I didn’t want to present this now, as there was not much time, but I can tell it to you now. We saw in the pH-stat groups, on low flow with pH-stat and circulatory arrest with pH-stat, definitely better results as compared to alpha-stat. For example, as far as lactate is considered, this was, in the low flow group, highly significantly better in the pH-stat groups.
Dr P. Vouhe (Paris, France): What are your clinical implications following this study?
Mr Roerick : Following this study, we want to give a word of caution as far as especially patients are concerned who have, for example, elevated lactate values or decreased pH before operation. So if you have any hint that there must be a lower perfusion of subdiaphragmatic organs, which would give us the idea, for example, to try cannulation of the descending aorta as well. This would be a good idea in these patients.
|
Footnotes |
|---|
Presented at the joint 19th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 13th Annual Meeting of the European Society of Thoracic Surgeons, Barcelona, Spain, September 25–28, 2005. |
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
