THE VENTILATORY RESPONSE TO HYPOXIA DURING EXERCISE IN CYANOTIC CONGENITAL HEART DISEASE

Transcription

1 Clinical Science and Molecular Medicine (1973) 45,99-5. THE VENTILATORY RESPONSE TO HYPOXIA DURING EXERCISE IN CYANOTIC CONGENITAL HEART DISEASE M. R. H. TAYLOR Department of Paediatrics, Institute of Diseases of the Chest, Brompton Hospital, London (Received 5 February 1973) SUMMARY 1. In contrast to the diminished ventilatory response to hypoxia which has been found at rest in cyanotic congenital heart disease, hyperventilation was noted on exercise in children who were cyanosed. 2. Sixteen children had low arterial oxygen saturations on exercise and thirteen of these hyperventilated by an amount similar to that reported in normal adults breathing hypoxic gas mixtures. 3. The three children who had little ventilatory response in relation to the increase of hypoxia during exercise all had a triad of long-standing cyanosis starting early in life, high haemoglobin concentration and low arterial oxygen saturation at rest in air. Key words : ventilation, exercise, hypoxia, congenital heart disease. Normal adults hyperventilate in response to hypoxia (Asmussen, 1967) but Sorensen & Severinghaus (1968) and Edelman, Lahiri, Braudo, Cherniack & Fishman (1970) have shown a diminution of the ventilatory response to hypoxia at rest in cyanotic congenital heart disease. However, in this laboratory, children with cyanotic congential heart disease were found to hyperventilate during exercise. Hyperventilation was most marked in those children who became deeply cyanosed. This paper records the ventilatory changes during exercise in Mteen children with varying degrees of cyanosis due to Fallot s tetralogy, and in two children with ventricular septa1 defects and right to left shunts. MATERIALS AND METHODS Details of the seventeen patients studied are given in Table 1. Parental consent was obtained for the study in each case. None of the children were receiving drug therapy at the time of the study. Exercise studies were performed in the upright position on an electrically braked Correspondence: Dr Mervyn R. H. Taylor, Department of Paediatrics, Trinity College, Dublin. 99

2 0 M. R. H. Taylor bicycle ergometer as described elsewhere (Godfrey, Davis, Wozniak & Barnes 1971a). Steadystate measurements were made at rest, one-third and two-thirds of the maximum work load that the child had completed in a previous assessment of his maximum working capacity. In a few children with severe exercise intolerance steady-state measurements were made at rest and one-half the maximum work load reached. The inspired gas was air throughout each study. Expired gas was flushed through a Tissot spirometer and analysed continuously for 0, and COz. When heart rate, minute ventilation TABLE 1. Details of the patients studied. = Fallot s tetralogy; VSD = ventricular septa1 defect; TCF = total correction of Fallot s tetralogy. Patient Age Age of onset Sex Height Diagnosis and operations no. (years) of cyanosis (cm) 1 I lo a 13b l+ years 8 months years At birth 2 years 11 years 5 days At birth At birth 1 week 1 week 5 years months M 122 M 9 F 112 M 1 F 1 F 1 M 136 F 124 M 1 M 1 F 4 F 1 M 135 M 1 M 145 F 1 F 1 M 125. Blalock shunt, aged 4 years. Pulmonary infundibular resection, aged 4 years Eisenmenger VSD Eisenrnenger VSD TCF, aged 4 years TCF, aged 8 months TCF, aged 8 months TCF, aged 9 years TCF, aged 8 years TCF, aged 6 years TCF, aged 24 years and expired gas concentrations were all steady a collection of expired gas was made in the spirometer over at least 1 min and analysed immediately. While the expired gas was being collected, a sample of blood was taken from an ear lobe for estimation of ph, Po, and Pco,. The blood was collected, analysed and the ph and PCO, corrected to arterial values as described by Godfrey, Wozniak, Courtenay-Evans & Samuels (1971b). The mean oxygen difference between ear and arterial samples using this method is 2.1 mmhg (SD 2.5) at rest and 0.9 mmhg on exercise (SD 3.1). Arterial oxygen saturation was calculated as described by Severinghaus (1966). Blood oxygen content was calculated from the oxygen saturation and haemoglobin concentration. A ventilatory index was used to relate the children s results to normal values. This was calculated as follows: observed minute ventilation = x 0 expected minute ventilation The expected minute ventilation was calculated from the child s oxygen consumption, height and sex using the data from normal children reported by Godfrey et al. (1971a).

3 Ventilation in hypoxic exercise 1 Fall in arteriol oxygen saturation from resting value FIG. 1. The relationship between ventilatory index and the fall in arterial oxygen saturation from the resting value (y = 4.081x+96.96: patients 2, 3 and 7 omitted). 95% confidence limits are given on either side of the regression line. The results from Asmussen s (1967) data are marked (A) _ C g I60 I 80 I I 1 I I I I I.I I I I I I I *I Arteriol Po2 (mmhg) FIG. 2. The relationship between ventilatory index and arterial Po2. The values from Asmussen s (1967) data are marked (A). The trend line was drawn by eye.

4 2 M. R. H. Taylor RESULTS At rest there was no significant difference between the mean minute ventilation of these children (8.2fSEM 1.0 litres/min) and that of normal children studied in this laboratory (7.8 f SEM 0.69 litres/min). Fig. 1 shows the relationship between the ventilatory index on exercise and the change in arterial oxygen saturation from the resting value. Fig. 2 shows the relationship between the ventilatory index on exercise and arterial Po,. There are no values available for the ventilatory response of normal children to hypoxia during exercise for comparison with this data. However, the results calculated from the normal adult data of Asmussen (1967) (assuming normal resting arterial oxygen saturation and a 2% venous admixture on exercise) agree with the present results (Fig. 1). The ventilatory index on exercise was significantly related to the arterial ph, but statistical assessment of the multiple regression analysis of ventilatory index, change in arterial oxygen TABLE 2. Haemoglobin concentration (Hb), oxygen consumption ( Vo,), ventilatory index, arterial Pco, (Pa,co,), arterial Poz(Pa,oz), % arterial oxygen saturation (Sa,o,) and arterial ph (ph) before and during exercise Patient Work Hb Vo, Ventilatory Pa,coz Pa,ol Sa,o, ph no. (W) (g/loo ml) (litres/min) index (mmhg) (mmhg) (%)

5 Ventilation in hypoxic exercise TABLE 2 (continued) 3 Patient Work Hb Vo2 Ventilatory Pa,co2 Pa,02 Sa,02 ph no. (W) (g/0 ml) (litreslmin) index (mmhg) (mmhg) (%) a 13b TABLE 3. Statistical analysis of the data. ASa,02 = change in arterial oxygen saturation from rest, Ca,02 = oxygen content of arterial blood (in vo1./0 ml). Other abbreviations are as in Table 2. In the upper section of the table, patients 2, 3 and 7 have been omitted; in the lower section all patients are included. Dependent variable (y) Independent variable (x) Correlation coefficient P value Pa,co2 Pa,02 Sa,02 ASa,02 Ca,ol PH > 03 < 0001 < < < < Sa, < 0001 ASa, < 0001 Ca, < 0.001

6 4 M. R. H. Taylor saturation and ph showed that the contribution made by ph was not significant when the ventilatory index and change in arterial oxygen saturation were known. Increase in arterial Pco2 (Pa,co2) and fall in ph both stimulate respiration and in theory could have been responsible for the raised ventilatory index on exercise. The highest Pa,co2 during exercise was 35 mmhg and Pa,co2 on exercise was not significantly related to the ventilatory index (Table 3). Therefore the increase in minute ventilation was not due to an increase in Pa,co2. Three children (patients 2, 3 and 7) had a lower ventilatory response than the others. Their results fell outside the 95% confidence limits estimated from the results of the other children. Direct measurement of ventilatory capacity in these children showed that they were all capable of a ventilation rate which would have brought their results within the 95% confidence limits. In these children Pa,co2 during exercise was not lower than that of the other children (Table 2). As all the children with a poor response during exercise had high haemoglobin values it was possible that their arterial oxygen content might be higher than was suggested by their arterial oxygen saturation, and the ventilatory response of all the children might be more closely related to oxygen content than saturation. However, statistical analysis showed that the ventilatory response was more closely related to both oxygen saturation and change in oxygen saturation than to oxygen content (Table 3). DISCUSSION The three patients (2,3 and 7), who had a poor ventilatory response in relation to the increase of hypoxia on exercise, all had haemoglobin concentration above 16 g/loo ml, arterial oxygen saturation (Sa,02) of less than 88% at rest in air, and a history of cyanosis of at least 43 years duration starting at or before years of age. Of the children who did not have a low ventilatory response on exercise, two patients (13 and ) had a similar history of cyanosis and a low saturation at rest, but a haemoglobin concentration below 16 g/loo ml, and one Opatient 11) had a low oxygen saturation at rest and a haemoglobin concentration of 18 g/0 ml, but a late onset of cyanosis. It would appear that a raised haemoglobin concentration, arterial desaturation in air at rest and long-standing cyanosis starting early in life were associated with insensitivity to hypoxia during exercise, though it is not possible to say which is cause and which is effect from the data available at present. Sorensen 8c Severinghaus (1968) using a steady-state technique have shown blunting of the ventilatory response to hypoxia at rest in adults with surgically corrected Fallot s tetralogy. Edelman et al. (1970) using transient hypoxia showed a diminished response in cyanotic congenital heart disease. In the present study it would have been interesting to compare the ventilatory response to hypoxia at rest with that on exercise but it was not felt that the induction of hypoxia at rest by breathing hypoxic mixtures was justifiable in these children. The ventilatory response on exercise was not diminished as compared with that reported by Asmussen (1967) in normal adults. It is possible that the ventilatory response to hypoxia is merely blocked at rest and not damaged as has been suggested (British Medical Journal, 1970). The results of the studies of totally corrected Fallot s tetralogy by Sorensen & Severinghaus (1968) and Edelman et al. (1970) are not in agreement. Sorensen & Severinghaus (1968) found a diminished ventilatory response to hypoxia at rest while Edelman et al. (1970) found a normal

7 Ventilation in hypoxic exercise 5 or near-normal response. This difference could be due to the different techniques used (steadystate and transient hypoxia) or to the fact that the three post-operative cases studied by Edelman el al. (1970) had normal arterial oxygen saturations and haemoglobin concentrations, while four of the five patients studied by Sorensen & Severinghaus (1968) were still hypoxic and had high haematocrits after total surgical correction. From the present study it appears that, except in patients with a triad of long-standing cyanosis from early life, a raised haemoglobin concentration and a low arterial oxygen saturation at rest in air, the ventilatory response during exercise in cyanotic congenital heart disease is not diminished as other workers have shown it to be at rest. ACKNOWLEDGMENTS I would like to thank the consultants of the Brompton and Guy s Hospital for allowing me to study patients under their care, my colleagues in the Institute of Diseases of the Chest for advice and assistance and Mr Andrew Nunn for statistical advice. This work was carried out during the tenure of the Eden Fellowship in Paediatrics. REFERENCES ASMUSSEN, E. (1967) Exercise and the regulation of ventilation. Circulation Research, 2CL21, Suppl., BRITISH MEDICAL JOURNAL (1970) Control of breathing. British Medical Journal, iii, EDELMAN, N.H., LAHIRI, S., BRAUDO, N.S., CHERNIACK, N.S. & FISHMAN, A.P. (1970) Blunted ventilatory response to hypoxia in congenital heart disease. New England Journal of Medicine, 282, GODFREY, S., DAVIS, C.T.M., WOZNIAK, E.R. & BARNES, C.A. (1971a) Cardio-respiratory response to exercise in normal children. Clinical Science,, GODFREY, S., WOZNIAK, E.R., COURTENAY-EVANS, R.J. & SAMUELS, C.S. (1971b) Ear lobe blood samples for blood gas analysis at rest and during exercise. British Journal of Diseases of the Chest, 65, SEVERINGHAUS, J.W. (1966) Blood gas calculator. Journal of Applied Physiology, 21, SORENSEN, S.C. & SEVERINGHAUS, J.W. (1968) Respiratory insensitivity to hypoxia persisting after correction of tetralogy of Fallot. Journal of Applied Ph.vsiology, 25,