Comparison of Pulmonary Artery and Pulmonary Venous Wedge Pressure in Congenital Heart Disease

Br Heart J: first published as 10.1136/hrt.35.4.386 on 1 April 1973. Downloaded from

British Heart_Journal, I973, 35, 386-39I. Comparison of pulmonary artery and pulmonary venous wedge pressure in congenital heart disease

Richard E. Hawker' and John M. Celermajer From The Adolph Basser Institute of Cardiology, Royal Alexandra Hospitalfor Children, Sydney, Australia

A statistical comparison ofpulmonary artery andpulmonary vein wedge pressures has been made by the correlation coefficient method, in patients with congenital heart disease. The correlation was poor (r = o066) in patients with a pulmonary artery mean pressure in the normal range, but good (r= o.86) in patients with mild to moderate pulmonary hypertension. An analysis of patients in whom catheterization of the pulmonary artery may not always be possible, showed a good correlation in transposition of the great arteries (r= o084) and ventricular septal defect, with banding of the mainpulmonary artery (r=o093), but not as good in tetralogy ofFallot andpulmonary stenosis. Despite the poor statistical correlation in the overallgroup, it wasfound that patients with pulmonary vein wedge mean pressure below the upper limits of normal had a pulmonary artery mean pressure below this level in 8I of87 cases and minimally raised pulmonary artery mean pressure in the remaining 6. Likewise, a raised pulmonary artery pressure was found in 33 out of 34 cases with high pulmonary vein wedge mean pressure. Therefore it is concluded that (i) if the pulmonary artery is not entered, a pulmonary vein wedge pressure within the normal range makes it very unlikely that the pulmonary artery pressure is raised; (2) if the pul- http://heart.bmj.com/ monary vein wedge pressure is raised, the pulmonary artery pressure is almost certainly raised. However, this rise cannot be quantitatedfrom the pulmonary vein wedge trace; and (3) the pulmonary vein wedge mean pressure is a useful indication ofpulmonary artery mean pressure in selected cases of congenital heart disease.

The pulmonary vein wedge pressure is sometimes the systolic or diastolic pressure was selected for used as an indirect measurement of pulmonary statistical analysis as it was thought that the correla- artery pressure in cases of congenital heart disease tion of mean pressures would be more meaningful when catheterization of the pulmonary artery is because it might be less affected by the damping on September 29, 2021 by guest. Protected copyright. unsuccessful. A recent publication (Rao and Siss- effect on the pressure wave transmitted through the man, I97I) examined the relation of pulmonary pulmonary vascular bed. Moreover, the mean pul- venous wedge to pulmonary arterial pressures. How- monary arterial pressure is used in the calculations ever, a statistical correlation of pulmonary vein of pulmonary vascular resistance and resistance wedge and pulmonary artery pressures in congenital ratios, and so in the assessment of pulmonary vascu- heart disease has not to our knowledge been pub- lar disease. lished, though it is known that the correlation is poor inthe presence ofpulmonary hypertension (Zimmer- Subjects and methods man, I966). It is the purpose ofthis paper to present The reports of cardiac catheterizations performed be- a statistical correlation between mean pulmonary tween February I968 and September 1970 in the Car- vein wedge and mean pulmonary artery pressures ob- diac Department of the Royal Alexandra Hospital for tained at routine and emergency cardiac catheteriza- Children were reviewed. In I87 of these, a pulmonary tion of infants and children with congenital heart vein wedge pressure was recorded and in 135 both pul- disease, and to assess the usefulness of pulmonary monary artery and pulmonary vein wedge pressures records. The mean rather than were available. The pulmonary vein wedge pressure vein wedge pressure recording was considered adequate only if there was a Received 7 July 1972. recognizable arterial type wave form present and if 1 Present address: The Johns Hopkins Hospital, Baltimore, there was the expected wave form change on with- Maryland, U.S.A. drawal of the catheter from the wedge position to the Br Heart J: first published as 10.1136/hrt.35.4.386 on 1 April 1973. Downloaded from

Pulmonary artery and pulmonary venous wedge pressures 387 pulmonary vein in cases where a withdrawal trace was Comparisons were also made of the pulmonary artery recorded. Cases known to have peripheral pulmonary and pulmonary vein wedge mean pressure in (i) trans- artery stenosis were excluded since it could not be deter- position of the great arteries, (2) pulmonary stenosis and mined if pulmonary artery pressures had been recorded tetralogy of Fallot, and (3) ventricular septal defect, beyond the obstruction. The pressure traces of all cases with banding of the main pulmonary artery, because in in which there was a greater than 20 per cent discrep- these conditions the pulmonary artery may be difficult ancy between pulmonary artery and pulmonary vein to catheterize. Eleven additional patients with these mal- wedge pressures were reviewed and I3 cases were rejec- formations were added to the number drawn from the ted on the basis of incomplete information or technically original series to increase the significance of the results. inadequate pulmonary vein wedge traces. Statistical comparison ofpairedobservations was made by The ages of the patients ranged from i8 hours to 13 using the regression equation and correlation coefficient years. Neonates were curarized, intubated, and ven- method. tilated during the procedure, while infants and older Finally, cases in which the pulmonary artery was not children were studied under local anaesthesia with pre- entered were examined to see if any significance could medication of omnopon and scopolamine except for be placed on the recorded pulmonary vein wedge those weighing less than 6 kg who were given morphine. pressures. A 5 or 6 French gauge endhole catheter' was used to record both pulmonary artery and pulmonary vein wedge Results pressure. The catheters were connected via Statham Group A: normal and subnormal mean pul- 23 De pressure transducers to a Sanborn 8-channel monary artery pressures (Fig. i) photographic recorder. The reference level for zero pressure was at the midchest position. There was a poor correlation in this group (correla- Comparison was made between the pulmonary artery tion coefficient r = o-66, standard error of the estim- and pulmonary vein wedge mean pressures except in ate (SE est.) = 2.49, slope = o-63). cases with banding of the main pulmonary artery in However, in only one case was the pulmonary vein which right or left pulmonary artery mean pressure and wedge mean pressure greater than 20 mmHg. The the pulmonary vein wedge mean pressure from the greatest discrepancy between mean pulmonary corresponding side was used. The paired results were artery and pulmonary vein wedge was 8 mmHg. The considered as follows. slope of the line being less than i indicates the over- Group A: 82 cases with mean pulmonary artery pressure all tendency for the pulmonary artery pressure to below and in the normal range of Kjellberg et al. (I955) be higher than the pulmonary vein wedge pressure. (i.e. 8 to I9 mmHg). http://heart.bmj.com/ Group B: I9 cases with mean pulmonary artery pressure Group B: mean pressure 20 to 39 mmHg (Fig. 2) between 20 and 39 mmHg. The = Group C: 21 cases with mean pulmonary artery pressure correlation was good in this group (r o-86, above 39 mmHg. SE est. = 2.53). The range of pulmonary vein wedge 1 U.S.C.I. Lehman.

35 on September 29, 2021 by guest. Protected copyright.

24 0 r=O- E c- E 25

E 0 0 0 . 0 p 0 2 i 62 !D . , 20 0 0 *0 3 *0 0 .# f *0. 15 0 . 0 00 0

>~ 8 & 0 0 a lo , 4,

0%W 0 4 8 12 16 20 24 28 32 0 5 10 15 20 25 30 35 40 Mean PA pressure (mmHg) Meon PA pressure (mmHg)

FIG. I Statistical comparison of mean pulmonary FIG. 2 Statistical comparison of mean pulmonary artery and pulmonary vein wedge pressures: Group A artery and pulmonary vein wedge pressures: Group B (mean pulmonary artery pressure less than 20 mmHg). (mean pulmonary artery pressure 20-39 mmHg). Br Heart J: first published as 10.1136/hrt.35.4.386 on 1 April 1973. Downloaded from

388 Hawker and Celermajer pressure was I3 tO 35 mmHg, and the greatest dis- 60. crepancy between pulmonary artery and pulmonary vein wedge was 7 mmHg. o =E 50. C Group C: mean pulmonary artery pressures 1- greater than 39 mmHg (Fig. 3) a 40- I range This group showed a very poor correlation (r = 0 I34, SE est. = 7 78). However, in all cases the I50*0 * mean CL mean pulmonary vein wedge pressure was 20 mmHg c or greater. The slope of the regression line is nega- E0 20 tive, indicating that in this range the mean pulmonary vein wedge pressure actually becomes less the 10 higher the mean pulmonary artery pressure rises. x No mean pulmonary vein wedge pressure higher than 50 mmHg was recorded. 0 III <8 8-19 20-29 30-39 40-49 >50 Meon PA pressure (mmHg) 60 FIG. 4 Statistical comparison of mean pulmonary - 50- artery and pulmonary vein wedge pressures: the dis- CE crepancy between mean pulmonary artery and pul- 40- monary vein wedge pressures increases with increasing pulmonary artery pressure. ° 30- r.n 0-314 3: 20- >. 60 cxa I0 io. c, 50 http://heart.bmj.com/ I E 0 10 20 30 40 50 60 70 80 40 Mean PA pressure (mmHg) 6) 30Q FIG. 3 Statistical comparison of mean pulmonary 3. artery and pulmonary vein wedge pressures: Group C a> 20' (mean pulmonary artery pressure greater than 39 ac 0 mmHg). I 10 on September 29, 2021 by guest. Protected copyright.

0 The tendency for the discrepancy between mean 0 10 20 30 40 50 60 70 80 pulmonary artery and mean pulmonary vein wedge Mean PA pressure (mmHg) pressures to become greater with increasing pulmonary artery pressure is shown in Fig. 4. FIG. 5 Statistical comparison of mean pulmonary artery and pulmonary vein wedge pressures: trans- Transposition of the great arteries - 27 cases position of the great arteries. (Fig. 5) A good correlation was obtained but there was a considerable scatter at the higher pulmonary artery Ventricular septal defect with banded pul- pressures (r = o-84, SE est. = 6-42). monary artery - 6 cases (Fig. 7) A good correlation was obtained for this small Tetralogy of Fallot and pulmonary valve group (r=0o93, slope= i-i6, SE est. = 179). The stenosis - 29 cases (Fig. 6) slope of the regression line is greater than I, indi- The correlation was poorer for this group (r = 0-7, cating the tendency for the pulmonary vein wedge SE est. = 2-66, slope = o-6o8). However, the greatest pressure to be higher than the pulmonary artery discrepancy was only 7 mmHg. pressures. Br Heart J: first published as 10.1136/hrt.35.4.386 on 1 April 1973. Downloaded from

Pulmonary artery and pulmonary venous wedge pressures 389

TABLE I Mean pulmonary artery pressure in cases 24 - / 70 with pulmonary vein wedge mean pressure. r.19r tmmHg I 20 x Pulmonary artery No. of cases 16 x pressure (mmHg) > 12 I9 8i 20-21 5 22-23 I >23 0 )4 x Follot Total 87 PS 0 0 4 8 12 lb 20 24 28 TABLE 2 Mean pulmonary artery pressure in cases Mean PA pressure (mmHg) with pulmonary vein wedge pressure. > i9 mmHg FIG. 6 Statistical comparison of mean pulmonary artery and pulmonary vein wedge pressures: tetralogy Pulmonary artery No. of cases of Fallot and pulmonary stenosis. pressure Normal I 24 - .a093 Raised 33 Total 34 20- I E E 16 Pulmonary vein pressure when pulmonary artery not entered e 12 Table 3 shows a group of I4 patients with raised http://heart.bmj.com/ 8' mean pulmonary vein wedge in whom the pulmon- a.> ary artery was not entered. In the 9 cases with transposition of the great arteries, the pressure was 1 4- recorded before balloon septostomy. The case with pulmonary valve fusion aged i week is, therefore, O L the only case out of I4 in which raised pulmonary 0 4 8 12 16 20 24 28 artery pressure would not be anticipated. Mean PA pressure (mmHg) TABLE 3 Cases with pulmonary vein wedge mean on September 29, 2021 by guest. Protected copyright. FIG. 7 Statistical mean pulmonary comparison of pressure. >1T9 mmHg in which pulmonary artery artery and pulmonary vein wedge pressures: ventricular septal defect with banding of main pulmonary not entered artery. Diagnosis No. of cases from Prediction ofpulmonary artery pressure 'Transposition of great arteries 9 pulmonary vein wedge pressure Neonates Pulmonary atresia with intact inter- The ability to predict from the pulmonary vein ventricular septum, persistent ductus arteriosus I wedge pressure record whether the pulmonary Truncus arteriosus I artery pressure is raised or normal is shown in Tables i and 2. It is seen that 8i out of 87 patients rComplicated transposition of J a raised vein mean pres- Older children great arteries (i year) 2 without pulmonary wedge of Fallot with mean Tetralogy right sure did not have raised pulmonary artery L Blalock's anastomosis* I pressure and no patient in this group had a pulmonary artery mean pressure greater than 23 mmHg. Total I4 On the other hand, raised mean pulmonary vein wedge pressure was associated with a raised pul- * Mild concentric hypertrophy of pulmonary arterioles found monary artery pressure in 33 out of 34 cases. at necropsy. Br Heart J: first published as 10.1136/hrt.35.4.386 on 1 April 1973. Downloaded from

39o Hawker and Celermajer

Table 4 shows that at least 25 out of the 29 cases of patients without pulmonary hypertension but with normal pulmonary vein wedge might have been did not record their results. Most recently, Rao and expected to have a normal pulmonary artery Sissman (I97i) have compared pulmonary venous pressure. wedge to pulmonary arterial pressures in 6o children. Though their figures were not subjected to statistical analysis the results and conclusions were TABLE 4 Cases with pulmonary vein wedge mean almost identical with those of the present study. pressure. < i9 mmHg in which pulmonary artery not Despite these conflicting reports we had expected entered to find better statistical correlation than we did, at least in the normal range of pressures. The simplest Diagnosis No. of explanation for the discrepancy between the pul- cases monary artery and pulmonary vein wedge pressure in the normal range is the lack of contemporaneous Transposition of great arteries + ventricular septal defect + pulmonary stenosis* 8 readings, for the minute-to-minute changes in Tetralogy of Fallot 7 haemodynamics under the conditions of catheter- Transposition of great arteries + persistent ductus ization were probably considerable. Reflex pulmon- arteriosus (neonates) 3 ary arterial constriction in response to pulmonary Pulmonary valve stenosis 4 Pulmonary atresia 3 vein distension has been well established (Hy- Tricuspid atresia 3 man and Sanchez, I964; Lloyd and Schneider, Transposition of great arteries + ventricular septal i969), but it seems unlikely that wedging of a defect + persistent ductus arteriosus (i year) I 5 or 6 French gauge catheter would distend a pul- Total 29 monary vein sufficiently to trigger these reflexes and produce an artefactually low pulmonary vein wedge pressure. Gensini et al. (i955) had * Includes cases with either single ventricle or ventricular noted that the septal defect, and either pulmonary stenosis or atresia. pulmonary vein wedge pressure exceeded the pulmonary artery pressure in their case of pulmonary stenosis and concluded that the catheter had signifi- cantly decreased the area ofthe narrowed pulmonary Discussion valve. Therefore, the inclusion of these cases and http://heart.bmj.com/ The value of the pulmonary vein wedge pressure those with tetralogy of Fallot and pulmonary steno- recording is related to its ability to predict the pul- sis which showed this phenomenon may also have monary artery pressure at cardiac catheterization. impaired the correlation in Group i. In I952, Weissel, Salzmann, and Vetter had ob- Gensini considered that increasing pulmonary served the arterial form of the pulmonary vein vascular resistance did not invalidate the pulmonary wedge pressure wave, but could not relate its pres- artery - pulmonary vein wedge relations, but ad- sure or contour to that in the pulmonary arteries. mitted that he had no case with very high resistance.

Wilson, Hoseth, and Dempsey (i955) recognized Connolly and Wood (i955), however, recognized a on September 29, 2021 by guest. Protected copyright. that the wedging phenomena depended on the lack worsening of the correlation of mean pressure and of anastomosis between pulmonary arterioles and wave form with increasing pulmonary vascular re- the large veins, the absence of valves in the veins, sistance and felt that the structural changes in the and the rich capillary network in the lungs. By pulmonary arteries, which tend to decrease flow, occluding flow, the catheter becomes an extension interfere with the correlation. The good correlation of a column of blood which transmits pressure from in Group B in this series suggests a satisfactory trans- the opposite side of the lung bed. Wilson produced mission of pressure waves at this level of pulmonary an excellent correlation of simultaneously recorded artery pressure. The statistical correlation in Group pulmonary artery and pulmonary vein wedge pres- B is better than in Group A, simply because of the sure in patients undergoing thoracotomy for lung higher numerical value ofthe pressure samples with- disease or for mitral valvotomy (Wilson et al., i955) out the discrepancy between pulmonary artery and and in open chest normal dogs (Wilson et al., 1953), pulmonary vein wedge pressures being greater (see while Connolly and Wood (i955) could not obtain Fig 4). The results in Group C certainly support good correlation at catheterization in 9 patients with Connolly's findings. It is possible that there is a atrial septal defect and partial anomalous pulmonary critical arteriolar diameter below which pressure venous return. However, Gensini, Balchum, and wave transmission becomes distorted. This distor- Blount (I955) obtained good correlation in 5 similar tion probably results from changes in the sites of patients, and Kjellberg et al. (i955) stated that his reflection of the primary pressure wave. Unlike the group obtained a good correlation in a large series situation in the systemic circulation where reflection Br Heart J: first published as 10.1136/hrt.35.4.386 on 1 April 1973. Downloaded from

Pulmonary artery and pulmonary venous wedge pressures 391

occurs from a single site at arteriolar level References (O'Rourke, I967), reflection occurs at several differ- Braun, K., and Stern, S. (I967). Functional significance of the ent sites in the normal pulmonary circulation pulmonary venous system. Americanjournal of Cardiology, (Braun and Stem, I967; Wiener et al., I966). 20, 56. Carr, I., and Wells, B. (I966). Coaxial flow-guided catheteriza- The concept of a critical pulmonary artery pressure tion of the pulmonary artery in transposition of the great beyond which pressure transmission is poor is sup- arteries. Lancet, 2, 3i8. ported by one of our patients with ventricular septal Celermajer, J. M., Venables, A. W., and Bowdler, J. D. (I970). defect and pulmonary hypertension. While breath- Catheterization of the pulmonary artery in transposition of the great arteries. A simple method. Circulation, 41, ing room air the mean pulmonary artery pressure I053. was 6o mmHg and the pulmonary vein wedge pres- Connolly, D. C., and Wood, E. H. (I955). The pulmonary sure 42 mmHg. On breathing approximately 60 per vein wedge pressure in man. Circulation Research, 3, 7. cent oxygen the mean pulmonary artery pressure Gensini, G., Balchum, 0. J., and Blount, S. G., Jr. (I955). The transmission of the pulmonary artery pressure across fell to 48 mmHg while the pulmonary vein wedge the capillary bed of the lungs. American Heart journal, 49, pressure remained unchanged. 507. In the present study systolic and diastolic levels in Hyman, A. L., and Sanchez, G. (i964). Alterations in pul- the pulmonary venous wedge pressure wave were monary vasculature and urine produced by acute distension of a single pulmonary vein in intact dogs (abstract). measured but not subjected to detailed analysis Circulation, 30, Suppl. III, 98. for reasons mentioned above. In general, however, Kjellberg, S. R., Mannheimer, E., Rudhe, U., and Jonsson, B. the results for the systolic and diastolic pressures (1955). Diagnosis of Congenital Heart Disease. Year Book paralleled those for mean pressure as they did also Publisher, Chicago. Lloyd, T. C., and Schneider, A. J. L. (i969). Relation of pul- in the study by Rao and Sissman (I97I). monary arterial pressure to pressure in the pulmonary Despite the statistically poor correlation between venous system. Journal of Applied Physiology, 27, 489. pulmonary artery and pulmonary vein wedge pres- O'Rourke, M. F. (i967). Pressure and flow waves in systemic sure demonstrated here, two useful facts emerge arteries and the anatomical design of the arterial system. journal of Applied Physiology, 23, I39. from this study. Rahimtoola, S. H., Ongley, P. A., and Swan, H. J. C. (i966). Percutaneous suprasternal puncture (Radner technique) of I) A raised pulmonary vein wedge pressure is a the pulmonary artery in transposition of the great vessels. reliable indication of a raised pulmonary arterial Circulation, 33, 242. pressure (Table 2) and this may be vital information Rao, P. S., and Sissman, N. J. (i97i). The relationship of pul- http://heart.bmj.com/ in where monary venous wedge to pulmonary arterial pressures. patients pulmonary artery catheterization Circulation, 44, 565. has proved impossible. Weissel, W., Salzmann, F., and Vetter, H. (I952). Pulmonary 2) The demonstration of a normal pulmonary vein capillary arterial pressure pulse in man. British Heart wedge pressure is a reliable indication that the pul- Journal, 14, 47. monary Wiener, F., Morkin, E., Skalak, R., and Fishman, A. P. (1966). artery pressure is normal or, at most, Wave propagation in the pulmonary circulation. Circula- mildly raised. tion Research, I9, 834. Wilson, R. H., Hoseth, W., and Dempsey, M. (I955). The Nevertheless, the pulmonary vein wedge pressure inter-relations of the pulmonary arterial and venous wedge on September 29, 2021 by guest. Protected copyright. should not be relied upon if entry to the pulmonary pressures. Circulation Research, 3, 3. Wilson, R. H., McKenna, W. T., Johnson, F. E., Jensen, N. artery is possible by special techniques, e.g. via a K., Mazzitello, W. F., and Dempsey, M. E. (I953). Sig- systemic to pulmonary artery shunt or by use of the nificance of the pulmonary arterial wedge pressure. Jour- floppy wire (Celermajer, Venables, and Bowdler, nal of Laboratory and Clinical Medicine, 42, 408. I970) or other techniques (Carr and Wells, I966; Zimmerman, H. A. (I966). Intravascular Catheterization, 2nd Rahimtoola, Ongley, and Swan, I966). ed. Charles C. Thomas, Springfield, Illinois. Thus, provided the limitation of the measurement is clearly recognized, the pulmonary vein wedge Requests for reprints to Dr. John M. Celermajer, pressure recording can prove a valuable aid to the Adolph Basser Institute of Cardiology, Royal Alexandra haemodynamic assessment of some cases of con- Hospital for Children, Pyrmont Bridge Road, Camper- genital heart disease. down 2050, Sydney, New South Wales, Australia.