Eur J Cardio-thorac

Surg (1992) 6: 579-585

surgery

0 Springer-Verlag 1992

Induced hypothermia in the management of refractory low cardiac output states following cardiac surgery in infants and children N. E. Moat, R. K. Lamb, J. C. Edwards, J. Manners, B. R. Keeton, and J. L. Monro Wessex Cardiothoracic

Centre, Southampton

General Hospital,

Tremona Road, Southampton,

SO9 4XY, UK

Abstract. Post-operative low cardiac output states remain a major cause of mortality following cardiac surgery in infants and children. Since 1979 we have used moderate induced whole-body hypothermia in the management of low-output states refractory to conventional modes of therapy. This is based not only upon the relationship between body temperature and oxygen consumption, but also on experimental work showing a beneficial effect of cooling upon myocardial contractility, particularly when there is pre-existing impairment of ventricular function. Between July 1986 and June 1990,20 children with refractory low-output states were cooled by means of a thermostatically controlled water blanket to a rectal temperature of 32-33 OC. The median age was 12 months (1 week-11 years) with a median weight of 6 kg (3.5-33 kg). Ten children survived to leave hospital while a further two made a haemodynamic recovery. There was a marked reduction in heart rate (P < 0.001). The mean arterial pressure rose (P = 0.037) while there was a fall in mean atria1 pressure (P < 0.001). There was a significant improvement in the urine output (P = 0.002). A fall in the platelet count (P < 0.001) was not accompanied by any change in the white cell count (P = 0.15). Although it is impossible to say whether cooling influenced the outcome in any of these children, it was usually effective in stabilising their clinical condition. The technique is simple and has a sound theoretical basis. We believe it is a useful addition to the therapeutic interventions available for the management of refractory low-output states in infants and children. [Eur J Cardio-thorac Surg (1992) 6:579-5851 Key words: Hypothermia

- Infant

- Post-operative

care

Postoperative low cardiac output states still occur after cardiac surgery in infants and children despite optimal myocardial preservation. Conventional management [IO] involves the exclusion of a residual surgical problem, optimising ventricular pre-load and after-load, correction of hypoxia, anaemia and acidosis and inotropic agents to enhance myocardial contractility. Despite these measures, the cardiac output may still be inadequate with a continuing deleterious imbalance between tissue oxygen supply and demand. Since our initial report in 1980 [16], we have used induced moderate (32-33 “C) whole-body hypothermia in the post-operative management of children with refractory low-output states. At these temperatures, tissue oxygen consumption and, hence, demand are much reduced [8]. In addition, there are certain theoretical advantages of this degree of hypothermia on myocardial function [l, 4, 9, 15, 18, 191.

Presented at the 5th Annual Meeting of the European Association for Cardio-thoracic Surgery, London, UK, September 23-251991 Correspondence to: Mr. James L. Monro, FRCS

Material and Methods Between July 1986 and June 1990,603 children underwent corrective or palliative surgery, utilizing cardiopulmonary bypass at the Wessex Regional Cardiothoracic Centre. The overall mortality was 7.6%. There were 261 infants (mortality 13%) and 342 children over 1 year (mortality = 3.5%). The data were compiled by retrospective analysis of the case notes and intensive therapy unit (ITU) charts. Prior to July 1986 the records were inadequate for detailed data retrieval. Only those children formally cooled to 32-33°C were included in the study. Induced whole body hypothermia was used in 20 infants and children. During this period of time a number of children were also cooled from moderate or severe hyperthermia to 35-37°C to alleviate the deleterious effects of such high core temperatures, but they are not considered in this paper. The 20 patients were aged between 1 week and 11 years (median 12 months) and weighed 3.5-33 kg (median 6 kg). They encompassed a wide range of congenital heart disease (Tables l-4). It is clearly important to attempt to define inclusion criteria for the institution of cooling. Given that this was a retrospective study, there was no strict protocol of enrohnent into this treatment mode. They were judged to have a low cardiac output based upon hypotension, poor peripheral perfusion, a diverging core to toe temperature differential and oliguria in the presence of normal, or more usually raised, atrial pressures. The inadequacy of their cardiac output is

580 Table 1. Induced hypothermia

in infants

Diagnosis and/or operation

Indication

1) 9; 9.4 kg

Arterial correction multiple VSDs

Low cardiac output

Died: persistent output

low cardiac

2)

RVOT reconstruction intact septum

Low cardiac output

Died: persistent output

low cardiac

Low cardiac output +complete AV block

Died: persistent output

low cardiac

Intractable

Survived

4; 5.5 kg

of Taussig - Bing heart with for pulmonary

3) 4; 4.3 kg

Repair of complete atrioventricular

4)

11; 8 kg

Total correction

5)

3; 3.5 kg

Closure of ventricular

6)

5; 5.7 kg

Correction

7)

6; 4.9 kg

8) 9)

septal defect

of tetralogy of Fallot

JET

Low cardiac output + JET

Died: low output + intractable JET

Low cardiac output + JET

Survived

Correction of complete AV septal defect with severe AV valve regurgitation

Low cardiac output

Survived

0.5; 3.5 kg

Tetralogy of Fallot: acute revision of blocked mod. BT shunt

Low cardiac output + JET

Survived

0.25; 3.7 kg

Correction

Low cardiac output

CVS recovery but brain dead (died)

Table 2. Preoperative

4% (months)

1) 2) 3) 4) 5) 6) 7) 8) 9)

atresia with

for cooling

Outcome

Age (months); weight (kg)

9 4 4

1 3 5 6 0.5 0.25

septal defect

of DORV with subpulmonary

stenosis

of type II truncus arteriosus

characteristics

of infants in Table 1

Mean arterial pressure (mmHg)

Mean atria1 pressure (mmHg)

Heart rate (per min)

Urine output (ml/kg/h)

Core:periphery temperature gradient (“C)

40 50 46 100 50 45 56 36 30

14 17 11 13 18 9 18 16 15

140 190 130 240+ + 220 + 190 250 + 160 150

0 1.5 0 1.6 0.6 0.5 0.2 1.5 0.3

6 5.2 4 9.5 7 8.5 0.5 1 7

+ =junctional ectopic tachycardia + + = junctional tachycardia alone

and

low

cardiac

output;

reflected in the large core to peripheral temperature gradient of the group pre-cooling: 6.5k3.6”C (mean + SD) (range 0.5-13.1). The pre-operative haemodynamic characteristics of the patients are shown in Tables 2 and 4. There were four children whose low-output states were exacerbated by the presence of a junctional ectopic tachycardia (JET) and one in whom cooling was instituted for uncontrollable JET alone. All of the children were initially managed conventionally. If the rhythm was not sinus then atrioventricular (AV) sequential pacing was instituted by means of atria1 and ventricular epicardial wires that were routinely placed at operation. Echocardiography was used in an attempt to exclude a persisting mechanical problem, and two of these children were cooled after having returned to theatre for revision of some aspect of their surgery. Three more were re-explored because of a suspicion of tamponade, although none was found, and in one of these the sternum was splinted open. The diagnosis of a refractory low output state was made if there was a continuing clinical deterioration despite appropriate pharmacological support. This included the use of catecholamines. nitrates and

other vasodilators and, later in the study, phosphodiesterase inhibitors. The majority of patients had, by this time, a peritoneal dialysis catheter in place, having often already been on a frusemide infusion. All were continued on their per&operative broad spectrum intravenous antibiotics, these being changed every 8-10 days. Hypothermia was induced by surface cooling with a thennostatically controlled water blanket. The rectal temperature (indwelling thermistor) was reduced to 32-33 “C usually over a period of 3060 min. Cooling was instituted a median 15 h (range 2-240) postreturn to the ITU. The duration of cooling varied between 24 and 190 h (median 68 h). Every child was sedated with an opiate infusion and paralyzed. All but four children required total parenteral nutrition. Haematological, biochemical and clotting studies were evaluated at least once every 24 h, with regular screens for sepsis. The decision to rewarm a patient was based upon clinical parameters aided by sequential echocardioagraphic assessment. It was extremely unusual to be able to consider rewarming within 48 h of cooling. The rate of rewarming was variable and more gradual than the rate of cooling. It was often staged, taking the temperature up to 34-35°C then re-evaluating the patient at the higher core temperature. The data were compiled by comparing a 4-h period preceding cooling the child and a second 4-h period, commencing 6 h after the onset of cooling. This 6-h delay was taken so that there could be effective cooling with establishment of a steady state at the new temperature. The means and standard deviations of haemodynamic variables were calculated for each of these periods. Any change in inotropic or vasodilator dose was recorded and the data were excluded if there was any increase in inotrope or if more than 10% of the estimated circulating volume was infused during this period of time. Atria1 pressures were recorded from the atrium with the higher pre-cool level irrespective of whether this was right, left, systemic or pulmonary venous atrium. Urine output data was deemed invalid if more diuretics were given in the post-cooling period. Haematological variables were compared between the last pre-cooling count and the counts after 48 h of cooling. Survival was defined as the child leaving hospital, with haemodynamic recovery being the ability to rewarm the child and discontinue inotropic support. After confirming that each data set conformed to a normal distribution, they were compared using Student’s t-test for paired data.

581 Table 3. Induced hypothermia Age (months); weight (kg)

in children over 1 year of age

Diagnosis and/or operation

Indication

1) 2; 8.2 kg

Repair of complete AVSD + transanmtlar RVOTO

2) 1.3; 4.5 kg

Closure of VSD + transannular

3) 3; 12 kg

Fontan

4) 6; 20 kg

Modified

5) 2.2; IO kg

Univentricular heart (RV morphology): Fontan operation

6) 1.5; 11 kg

patch for

patch for RVOTO

for cooling

Outcome

Low cardiac output (RV failure)

Died: persistent

low CO

Low cardiac output

Survived

for tricuspid atresia

Low cardiac output

Survived: spastic quadraparesis

operation

Low cardiac output (LV dysfunction)

Died: perioperative

Low cardiac output + serous effusions

Survived

Reconstruction of complex RVOTO and closure of secundum ASD

Low cardiac output + JET

Survived

7) 3.5; 8 kg

Modified

Low cardiac output

Died: persistent low CO 2 months postop

8) 3; 10 kg

Correction of truncus arteriosus right pulmonary artery

Low cardiac output + intermittent JET

Haemodynamic recovery: died of respiratory failure

9) 1.3; 6 kg

Complex tricuspid atresia: Fontan then revised to a shunt

Low cardiac output

Survived

10) 11; 33 kg

Correction of pulmonary PA and MAPCAs

Low cardiac output (7-h CPB)

Survived

11) 5; 20kg

Modified phology)

Low cardiac output

Died: persistent output

Table 4. Preoperative

Age (months)

1)2 2) 1.3 3) 3 4) 6 5) 2.2 6) 1.5 7) 3.5 8) 3 9) 1.3 10) 11 11) 5

operation Fontan

Fontan

Fontan

characteristics

for tricuspid atresia modified

for tricuspid atresia and VSD with disconnected

atresia + absent central

for univentricular

heart (LV mor-

Mean atria1 pressure (mmHg)

Heart rate (per min)

Urine output (ml/kg/h)

70 43 38 50 47 53 65 37 73 35

14 14 14 16 20 18 14 20 15 22

140 165 130 150 150 190+ 165+ 165 160 90

1.5 0 0 0 0.4 0.9 0.9 0 1.2 0

+ =junctional ectopic tachycardia and - = inefficient or uninterpretable data

low

Core:periphery temperature gradient (“C) 4.5 7.5 4 6 9.3 9.5 13.5 _ 8.5 11

cardiac

low cardiac

Overall outcome

of children in Table 3

Mean arterial pressure (mmHg)

MI

The outcome for each individual patient is demonstrated in Tables 1 and 3. Ten children survived to leave hospital. Of the survivors two have a significant neurological deficit. Both of these children had sustained a cardiopulmonary arrest in the post-operative period. A further two made a haemodynamic recovery. Post-mortem examination confirmed the clinical findings that these children died of a massive neurological insult (following a prolonged arrest prior to the onset of cooling) and respiratory failure, respectively.

Cardiovascular output;

Results

The results of the group are considered as a whole. An analysis has been made, excluding those patients whose myocardial insufficiency was contributed to by an arrythmia. This had no effect on the overall pattern of change seen nor on any of the statistical analyses. Therefore, for descriptive purposes and for graphical representation they are included in the whole and they will be commented on separately in the discussion.

effects

The typical pattern of change on cooling a patient is illustrated in Fig. 1, depicting the ITU chart of one of the children. There was an invariable fall in the heart rate after cooling (Fig. 2: P< 0.001). Of the five patients with a JET prior to cooling, all had their tachycardia brought under control with four reverting to sinus rhythm. One died with persistent low output and an intractable JET on rewarming, while another developed complete AV dissociation during cooling which reverted to sinus rhythm on rewarming. Three other children developed AV dissociation after cooling, necessitating AV sequential pacing, before reverting to sinus rhythm on rewarming. There was a significant increase in mean arterial pressure on cooling the patients (Fig. 3: P=O.O37). This was accompanied by a significant fall in atria1 pressure (Fig. 5:

582 18 +a, 16 $ 14 7 12

i

10 P 8 ,; 9

140; 120; o^ 1004 80 s. 2 60 f T

COOL

i

I

I

12

I

I

I

3

4

5

I

I

I

6 7 8 time (hours1

I

1

1

1

9

10

11

12

Fig. 1. Reproduction of the haemodynamic changes seen on the ITU chart of one of the children in this series, indicating the typical changes that occur

pre-cool

port~cool

Fig. 3. The effect of cooling on the mean systemic arterial and the mean atria1 pressures. N=17; P=O.O37 for l . N=17; P~O.001 for A

200

125

r ; AZ 100

r

I

pre-cool

post-cool

Fig. 2. The effect of cooling on the heart rate. N= 17; mean f SEM; P

Induced hypothermia in the management of refractory low cardiac output states following cardiac surgery in infants and children.

Post-operative low cardiac output states remain a major cause of mortality following cardiac surgery in infants and children. Since 1979 we have used ...
850KB Sizes 0 Downloads 0 Views