| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 1999;15:515-518
© 1999 Elsevier Science NL
`Wessex Cardiothoracic Centre', Southampton, UK
Received 21 September 1998; received in revised form 26 January 1999; accepted 2 February 1999.
Corresponding author. Department of Cardiothoracic Surgery, Southampton General Hospital, Tremona Road, Southampton, SO16 6YD UK. Tel.: +44-170-379-6241; Fax: +44-170-379-6614.
 |
Abstract |
|---|
 Top Abstract Introduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
Key Words: Cooling • Hypothermia • Cardiac failure • Paediatrics • Cardiac surgery
|
Introduction |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
Induced hypothermia is a well established technique used primarily during operations involving cardiopulmonary bypass to reduce tissue oxygen demand and provide a degree of cellular protection against ischaemia [1] [2] [3] [4] [5] [6]. We initially reported our postoperative experience with induced hypothermia in 1980 [7] and subsequently in 20 patients between 1986 and 1990, in 1992 [5]. We now report our cumulative experience with the use of induced hypothermia for the treatment of intractable heart failure unresponsive to conventional therapy in 57 patients following surgery for congenital heart disease.
|
Methods |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
Data was taken retrospectively from ITU charts. Only patients formally cooled to 32–33°C for a persistent low cardiac output state in the postoperative period were included in the review. Patients cooled for a SVT (four) and those who returned to theatre for further corrective surgery (eight) were excluded.
A low cardiac output state was diagnosed by continuing non-responsive hypotension, poor peripheral perfusion, a widening core-peripheral temperature, oliguria in the presence of normal (or raised) right atrial pressure and a developing base deficit, despite optimization of pharmacological support, biochemical and haematological parameters.
Inotropic support was instituted in all patients and included the use of infusions of adrenaline, dobutamine, noradrenaline, dopamine and nitrates as thought necessary. Metabolic abnormalities such as acidosis, electrolyte imbalance and anaemia were also corrected as far as possible.
All patients were intubated and mechanically ventilated and their blood gases optimized, the FiO2 being kept to a minimum level to ensure an adequate PaO2. Positive end-expiratory pressure (PEEP) and inverse ratio ventilation were used if appropriate. Patients were sedated with infusions of midazolam (10–100 µg/kg per h) and morphine (10–100 µg/kg per h) and paralyzed with an infusion of vecuronium (20–100 µg/kg per h).
Patients in a rhythm other than sinus were paced using sequential atrioventricular pacing through epicardial pacing wires.
Regular post-operative echocardiography was performed to exclude any potentially correctable mechanical defect and to exclude the presence of cardiac tamponade. Oliguria (<0.5 ml/kg per h) unresponsive to both adequate atrial pressure and diuretics was treated by peritoneal dialysis.
If the patient continued to deteriorate in the presence of maximal intensive care support, moderate hypothermia (32–33°C) was induced using a thermostatically controlled water-filled cooling blanket placed under the child. This usually took about 1 h to achieve but could be speeded up by placing a wet sheet over the patient and fanning. It is important not to overshoot, cooling should not go below 33°C. Core temperature was measured using a rectal thermistor and peripheral temperature using a skin thermistor placed on the big toe.
Haemodynamic parameters were measured for a 4-h period before cooling and compared with another 4-h period 6 h after the start of cooling. This 6-h period was chosen to allow adequate cooling of all tissues and the attainment of a physiological steady state. Parameters measured were heart rate, right atrial pressure, left atrial pressure (in patients with a left atrial catheter placed during surgery), systemic arterial blood pressure, urine output and core-peripheral temperature difference. The blood gases and acid-base status were also measured. Platelet count and white blood cell count were measured pre-cooling and at 48 h post cooling (or prior to rewarming if this occurred earlier) and were compared with the pre-cooling values.
Operative mortality was defined as death within 30 days or in hospital.
|
Statistical analysis |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
|
Results |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
|
|
A significant increase in urine output from 0.78±0.9 to 3.07±2.1 ml/kg per h occurred following cooling (P<0.001).
Cooling resulted in a significant drop in the platelet count from 202±97 to 70±63 (x109/l), (P<0.001). Thrombocytopaenia was treated with platelet transfusions although no fixed limit was used below which transfusion was instigated. There was no evidence of disseminated intravascular coagulation in any child. The fall in platelet count was not associated with increased bleeding. There was no significant change in white blood cell count post-cooling.
There was only a significant difference in the increase in urine output between survivors and non-survivors (P<0.01).
|
Discussion |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
The protective effects of hypothermia are thought to be due to a fall in metabolic rate associated with a slowing of temperature-dependent enzymatic reactions [1]. Oxygen consumption falls by around 8% for each degree drop in temperature and at a core temperature of 33°C, oxygen demand falls to approximately 70% of basal rate [1] [8]. The cytoprotection seen at moderate hypothermia (33–35°C) is however disproportionate to the degree of hypothermia. This additional protection is thought to result from other mechanisms including reduced anaerobic metabolism whose products are themselves cytotoxic, suppression of oxygen free radicals [9], preservation of a high-energy phosphate store [10] and a specific action on lipid peroxidation [11].
In addition to the possible beneficial metabolic effects, there is also evidence that moderate hypothermia may have a direct beneficial effect on myocardial function, particularly when there is pre-existing ventricular dysfunction [4] [12] [13] [14]. There may also be some added beneficial direct effect on the failing myocardium, as a result of the decrease in heart rate caused by the direct effect of cold on the sino-atrial node [13] [15].
As with any retrospective series spanning a decade that documents the effect of a treatment, there is wide scatter in the pre- and post treatment variables of patients that are subjected to that treatment. Statistically proving that any treatment is solely responsible for a change in outcome requires large numbers. Thankfully the scenario of a low cardiac output state refractory to all conventional treatment is rare and therefore sufficient numbers are not available, given the data scatter to achieve statistical significance in variables between survivors and non-survivors. Similarly it is impossible to scientifically demonstrate the different benefit between survivors and non-survivors. A proposed explanation of how we believe hypothermia increases cardiac output is however incumbent upon us.
The rise in mean arterial pressure is due to peripheral vasoconstriction causing an increase in peripheral vascular resistance. Any increase in afterload does not seem adversely to affect cardiac function because there is a reduction in atrial pressure. The decrease in heart rate following cooling is partly due to a direct effect of cold on the sinoatrial node. Usually both the blood pressure and urine output increase before the heart rate decreases, suggesting that mechanisms other than control of heart rate may be important. We believe this observation is due to the direct benefit of cooling on myocardial function. The improved urine output reflects quantitatively the additional benefit produced by cooling in those patients in whom cardiac function has been enhanced.
Moderate hypothermia has several potentially harmful effects. The oxygen dissociation curve is shifted to the left [2], resulting in an increased affinity of haemoglobin for oxygen at a given PaO2. This reduced oxygen delivery to tissues is partly offset by an increase in free dissolved oxygen by as much as 10% at 33°C [1] [2]. Hypothermia increases muscle tone and at the degree of hypothermia used in these patients is usually sufficient to trigger shivering. This increase in muscle activity can increase oxygen consumption by several fold [16] but is prevented by the use of muscle relaxants. Despite known detrimental effects of cooling, this technique however appears safe. The only adverse change we noted was a falling platelet count which was treated by platelet transfusions and was not associated with increased mortality or morbidity from bleeding. Furthermore, cooling has been widely used in the treatment of persistent supraventricular tachycardia with no unwanted side effects.
There are certain points which we have learned following our experience. It is important not to overshoot on cooling, we avoid going below 32°C. Once cool, complete paralysis is essential, the increased oxygen consumption produced by shivering is detrimental. It is sometimes tempting to rewarm early, when the haemodynamics appear more favourable, we now usually avoid rewarming for at least 48 h.
We acknowledge that proving that cooling affected the outcome in these cases is impossible, for the reasons stated, however, subjecting patients to a cheap non-interventional mode of treatment which appears to be of significant benefit can be justified considering the alternatives. The message is simple, when the outlook is appalling, think of cooling in our experience, it has reduced mortality by over 50%.
It is our belief that the use of moderate hypothermia has played a significant part in reducing the mortality in this extremely ill group of patients.
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
Mr Dalrymple-Hay: We've used ECMO when we failed to get the patient off cardiopulmonary bypass and taken them to ITU on ECMO. However, in the scenario where we have a persistent low cardiac output state that we can't correct in ITU, we wouldn't use ECMO, we'd induce hypothermia.
Dr G. Stellin (Padua, Italy): As Dr. Jim Monro gave us the idea of the surface cooling to treat severe low output syndrome in the postoperative period about 8 years ago, we have treated over 40 patients with such a technique and we are sharing similar clinical results. However, it is still not clear to us why surface cooling improves cardiac output, renal function reversing arterial blood gases acidosis. We are undertaking experimental studies in animals at our University in order to scientifically validate these data. I wonder if you have any answer to all my questions?
Mr Dalrymple-Hay: I don't have any experimental animal data. I can however offer an opinion, because we've clearly all thought about it. Aside from the benefit of reducing the metabolic rate with hypothermia due to changes in temperature-dependent enzymatic reactions, I think there is now quite a lot of evidence showing that you get a disproportionate benefit from the reduction in temperature. This is thought to result from other mechanisms, because you get reduced anaerobic metabolism, preservation of high energy phosphate stones, less free radical production. In addition, I think from the myocardial point of view, there is quite good evidence in dogs that you see improvements in myocardial function with cooling if they had abnormal myocardial function precooling. Furthermore, I think the cooling directly slows the heart rate via a direct effect on the sinoatrial node. Therefore, the disproportionate balance of improvement in myocardial function, while raising the peripheral resistance by cooling, allows us to get away with the cooling to 32°C.
Dr M. Wojtalik (Poznan, Poland): Did you use any vasodilators to control rising systemic vascular resistance? That those patients with better urine output survived, don't you think it means that those with better cardiac output survive and those with worse do not?
Mr Dalrymple-Hay: We don't have to use any vasodilators for the rising arterial pressure. I'm sure you're correct that the increase in urine output is a manifestation of better residual myocardial function with cooling.
Dr C. Knott-Craig (Oklahoma City, OK, USA): I noticed that your mean blood pressure prior to cooling was 50 and your mean age of the patients was 8 months, which is probably not a really bad blood pressure if one can equate blood pressure with cardiac output. Does this presuppose that your patients are on a number of inotropes before you decide to cool them? If so, what do you do with your inotropic support once you introduce cooling?
Mr Dalrymple-Hay: Each case clearly is treated on its own merit. Yes, they're on a lot of inotropes when they're cooled. We've persisted with maximum medical therapy before we cool. Once they are cooled, we would manipulate the inotropes as we felt the haemodynamic parameters permitted through the cooling process and once they were cooled.
|
References |
|---|
|
TopAbstractIntroduction
Methods
Statistical analysis
Results
Discussion
Appendix A. Conference...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
