Long-term haemodynamic evaluation of aortic

Transcription

1 British Heart journal, 1978, 40, Long-term haemodynamic evaluation of aortic pericardial xenograft ANAND P. TANDON, DONALD R. SMITH, WILLIAM WHITAKER, AND MARIAN I. IONESCU From the Departments of Cardiology and Cardiothoracic Surgery, The General Infirmary, Leeds SUMMARY Haemodynamic studies were performed in 30 patients at a mean interval of 43-4 (range 27 to 59) months after aortic valve replacements-withpericardial xenografts. Five valve sizes- 19, 21, 23, 25, and 27 mm annulus diameter-were used. Of these 30 patients, 16 had preoperative haemodynamic investigations. Comparison of the pre- and postoperative data showed a marginal but significant increase in cardiac output (P < 0.05). There was a significant reduction in the pulmonary wedge and left ventricular end-diastolic pressures at rest and on exercise (P < 0-05 and P < 0-01). At the postoperative study the mean peak systolic gradient was 8-3 mmhg at rest and 12-3 mmhg on exercise in patients with the smallest xenograft inserted (19 mm) and decreased with each corresponding increase in graft size, so that across the 27 mm graft it was hardly measurable. The calculated xenograft surface area ranged from 1 1 to 2 1 cm2 at rest and 1-3 to 2-4cm2 during exercise. Aortic root angiography showed competent valves in all 30 patients, but a grade 1 to 2 perivalvular leak was shown in 8 patients. This long-term study has indicated significant circulatory improvement and very small transvalvular gradients in patients with aortic pericardial xenografts. Clinical background Clinical results of heart valve replacement with patients, 17 had early diastolic murmurs. Periglutaraldehyde stabilised pericardial xenografts valvular leaks were confirmed in 14 patients either have already shown a very good survival record, at angiography (12 patients) or at reoperation (2 low thrombogenicity without anticoagulants, and patients). Clinically 96 per cent of patients are in maintenance of structural integrity of the grafts grade 1 (NYHA) at the latest evaluation. over a period in excess of 54 years (Ionescu et al., 1977). Haemodynamic confirmation of the func- Subjects and methods tional performance of the valve was necessary. This report presents the results of a late haemodynamic Haemodynamic investigations were performed in 30 assessment of 30 patients who had aortic valve patients at a mean interval of 43-4 (range 27 to 59) replacement with the Ionescu-Shiley pericardial months, postoperatively. The criteria for selection xenograft.' were an interval of at least 2 years from valve replacement and the informed consent of the patient. An effort was made to include patients with early diastolic murmurs in order to investigate them Since March 1971, 176 patients have had single angiographically. In our belief, these 30 patients aortic valve replacements with pericardial xeno- were in no other way different from the remaining grafts. The hospital and late mortalities were uninvestigated patients with aortic pericardial 6-2 and 5'1 per cent, respectively. Actuarial xenografts. The clinical details of the 30 patients analysis of the expected survival rate and individual studied are outlined in Table 1. Additional mitral event-free curves for reoperation, bacterial annuloplasty and implantation of permanent epiendocarditis, and thromboembolism have been cardial pacing systems were undertaken in 2 patients carried out by the method of Anderson et al. (1974) each at the time of the original operation. Of the 30 and the results are shown in Fig. 1. Of the total 176 patients reported herein, 16 had preoperative cardiac catheterisations. The number of patients with 'Shiley Laboratories, Irvine, California. different xenograft sizes is shown in Table 2. Received for publication 18 May All patients were admitted to hospital 24 hours

2 Haemodynamics of aortic pericardial xenograft 603 0/0 Aortic replacement *9±l12q0... (4 960± 530/o (2) '''''''''''' 93-0±5-30/ (2) (930+±5.30/% Fig. 1 Actuarial analysis of late deaths (5.1%) (1) 901 ± 7.50/o results after aortic pericardial xenograft valve replacement. The 6 mths. mesenteric vein thrombosis data are expressed as percentage 9 myocardial infarction of expected survival rate and 9 S.B.E. heart failure free from: individual event-free curves. The 10 S.B.E. heart failure -X- t. embolism n s r 17 dissection descending aorta S.B.E. numbers in parentheses along the myocardial infarction - re-op horizontal axis denote the number 32 re-op R.V. laceration of patients at the beginning of 45 sudden death each year. Ṭ 60 brain tumour survival (165) (138) (116) (80) (49) (25) No.of patients Years Table 1 Preoperative clinical details of 30 patients assessment, a chest radiograph, a 12-lead electrocardiogram, and a phonocardiogram were obtained No. of patients in all patients. Left ventricular voltage was measured Male 20 by the sum of SV1 and RV5 from the electrocardio- Female 10 grams performed preoperatively and at the time of Age (years) Mean 49.7 haemodynamic study. Right, retrograde left, and (±24) transseptal catheterisations were performed in the Range Preoperative clinical diagnosis post-absorptive state without any previous sedation. Stenosis 15 Pulmonary and systemic pressures were transduced Regurgitation 10 by strain-gauge manometers (SEM 486), integrated Mixed valvular disease 5 New York Heart Association status electronically, and recorded on a multichannel ultra- Grade 1 1 violet light recorder (SEM 3012) with the zero Grade 3 13 level set 5 cm below the sternal angle. Cardiac output was measured by the direct Fick method. Haemodynamic data were obtained during a 4- before the haemodynamic study. On admission, a minute period of rest and between the 4th and 6th detailed clinical history, estimation of functional minute of a 6-minute period of supine leg exercise capability, physical examination, haematological on a bicycle ergometer at a predetermined maximal Table 2 Postoperative haemodynamic data as related to xenograft size (mean values + SEM) Xenograft In vitro No. of AVF PWP LVEDP PSG ESG XSA size surface patients (ml/s) (mmhg) (mmhg) (mmhg) (mmhg) (cm2) (annulus area R E R E R E R E R E R E diameter (cm2) in mm) ± ±4-1 ±1-8 ± ±2-7 ±3-2 ±6-4 ±0 07 ± ±119 ±16-5 ±1-4 ±2-9 ±2-0 ±3-2 ±16 ±2-3 ±1-2 ±6-2 ±01 ± ±13-8 ±13-2 ±05 ±04 ±05 ±1 9 ±1-6 ±2-6 ±1-6 ±1-6 ±0 1 ± ±19-3 ±22-2 ±11 ± ± ±2-5 ±1-2 ±19 ±0-08 ± * All patients ±11 ±13-7 ±0 5 ±11 ±0 7 ±1-6 ±10 ±1-6 ±0 9 ±1-6 ±0 07 ±0 1 *Had additional mitral annuloplasty at the time of aortic valve replacement. Abbreviations: R, rest; E, exercise; AVF, aortic valve flow; PWP, pulmonary wedge pressure; LVEDP, left ventricular end-diastolic pressure; PSG, peak systolic gradient; ESG, ejection systolic gradient; XSA, calculated xenograft surface area.

3 604 A. P. Tandon, D. R. Smith, W. Whitaker, and M. I. Ionesc Table 3 Pre- and postoperative haemodynamic data (mean values ± SEM and statistical significance) 02 uptake Cardiac index Stroke index PWP LVEDP (ml/min per mi) (I/min per m2) (m,/m2) (mmhg) (mmhg) R E R E R E R E R E Preoperative (16 patients) ±12-5 ± ±03 ±3-4 ±3-7 ±1-7 ±5-2 ±2-8 ±5-8 Postoperative (30 patients) ±2-7 ±7 9 ±0 1 ±0-2 ±1-6 ±2-3 ±0 5 ±11 ±0 7 ±1-6 P value <001 <001 <005 <005 <001 <001 <005 <001 <0.01 <0.01 SEM, standard error of the mean; PWP, pulmonary wedge pressure; LVEDP, left ventricular end-diastolic pressure; PSG, peak systolic gradient R, rest; E, exercise. load. Aortic root angiography was performed in all 2 5 patients at the end of the study. Pulmonary and systemic vascular resistances E xerci se were calculated using standard formulae. The E 20 ejection systolic gradient across the aortic peri-, \ cardial xenograft was measured by planimetric integration of at least five simultaneously recorded 0 phasic left ventricular and aortic root tracings. The a Rest pericardial xenograft surface area was calculated Rs according to the hydraulic formula of Gorlin and Gorlin (1951) using the ejection systolic gradient. I Results The details of the haemodynamic data are given in Tables 2 and 3 and Fig. 2 and 3. 0 r=0982 (rest) r=0-887 (exercise) Valve sizes I Ejection systolic gradient (mm Hg) Steady-state flow Fig. 3 Linear regression curve of ejection systolic C temperature 21 gradient related to calculated xenograft surface area. Valve Valve 32 size orifice area (sq mm) ELECTROCARDIOGRAPHIC AND I CAG RADIOLOGICAL CHANGES E A highly significant reduction, from a preoperative E mean value of 53'2 (±1-3) per cent to 47-7 (±0 9) c per cent was noted inthe cardiothoracicratio at the * postoperative study (P < 0 001) The mean left ventricular voltage before operalb - tion 12 investigation (P < 0O001). 8)a //2. / was 50 3 (±2 3) mm and showed a significant reduction to 34-3 (±2-1) mm at the postoperative 8-29 CARDIAC OUTPUT AND OXYGEN UPTAKE 31 The cardiac index showed a marginal but significant 4//./ - / 33 increase (Table 3) at the postoperative study both at 2 rest and during exercise (P < 0 05). Cardiac output 0, response to exercise was normal. The oxygen uptake b lb showed significant reduction both at rest and on Flow (I/min) exercise (P<0-01). Fig. 2InvitrflowadgradentacrssperiardiaThe stroke volume at rest increased from the Fig. 2 vitro and gradient across pericardial mea proeaievleo 92m 2 to4- mm2 xenografts measured in a continuous flow rig. The mean preoperative value of 29s2ml/ma to 40i 4 mi/ curves represent mean values obtained from measurements at the postoperative study. A similar increase was made with at least two valves of each size. noted during exercise. The increase in stroke

4 Haemodynamics of aortic pericardial xenogreaft 605 'SG Vascular resistance (dyne s cm-5m-') culated surface area increased during exercise with 'mmhg) Pulmonary Systemic the increased flow. Further, the calculated value? E R E R E was lowest with the smallest xenograft implanted and rose with increasing xenograft size (r=0o996 ±100 - ±3497 ±23-7 ±38106 ±159-2 at rest and r=0-951 during exercise). As shown in t19-1 tl Table 2 the calculated areas in vivo are smaller than < Not significant the in vitro surface area but a significant correlation was noted between the two values (r=0995 at rest and r=0934 on exercise). volume both at rest and during exercise was Regression analysis showed a linear relation significant (P < 001). between the calculated xenograft surface area and the ejection systolic gradient as shown in Fig. 3. PULMONARY WEDGE AND LEFT VENTRICULAR END-DIASTOLIC PRESSURES AORTIC VALVE FLOW Mean pulmonary wedge and left ventricular end- The mean flows through the pericardial xenograft diastolic pressures were normal at rest at the post- at rest and during exercise were 225 ± 11 and operative study and the reduction from preoperative 322 ± 13X7 ml/s, respectively, at the postoperative values was significant (P < 005 and P < 001). The study. The flows across various sizes of pericardial corresponding mean values during exercise, though xenograft are given in Table 2. A significant corabnormally high, showed a significant reduction relation was noted between aortic valve flow and the when compared with the preoperative data annulus diameter of the xenograft implanted (P <0.01). (r=0861 at rest and r=0'998 during exercise). VASCULAR RESISTANCE AORTOGRAPHY No significant change was noted in either pulmonary Aortic root angiography showed that all the perior systemic vascular resistance at the postoperative cardial xenografts studied were competent. Of the study (Table 2). 30 patients, 8 were known to have early diastolic murmurs from the time of valve replacement. TRANSVALVULAR GRADIENT These showed grade 1 or 2 regurgitation of contrast The hydraulic characteristics of the pericardial material into the left ventricle (Brandt et al., 1969), xenograft were studied in vitro. Pressure gradient the site of the regurgitation being clearly shown to was measured across the xenografts in a continuous be perivalvular on multiple plane angiograms. The flow rig and the results are shown in Fig. 2. At a haemodynamic data were not significantly different flow of 12 1/min, the gradient across the 17 mm between the 8 patients with perivalvular leak and the valve was only 12 mmhg. remaining 22. At the postoperative investigation both peak and ejection systolic gradients were measured in all Discussion patients. The mean values are shown in Tables 2 and 3. This long-term haemodynamic investigation has The mean peak systolic gradient was 8-3 mmhg shown significant circulatory improvement after at rest and 12-3 mmhg on exercise in patients with aortic valve replacement and maintenance of the xenograft size 19 and it decreased with the cor- function of the pericardial xenograft in the aortic responding increase in the graft size, so that with position up to 59 months postoperatively. the 27 mm graft it was hardly measurable. A highly The cardiac output showed a marginal but significant correlation was noted between the peak statistically significant increase and its response to systolic gradient and the annulus diameter of the exercise was normal. Oxygen utilisation improved xenograft (r=0-954 at rest and r=0-937 during after valve replacement. Similar findings have been exercise). A similar correlation was found between reported by Olin (1970), Bjork et al. (1971, 1974), mean ejection systolic gradient and the xenograft Bristow and Kremkau (1975), Ionescu et al. (1974), size (r=0 991). and Dubiel and Cullhed (1975). Pulmonary wedge and left ventricular end- CALCULATED XENOGRAFT SURFACE AREA diastolic pressures showed significant reduction The calculated xenograft surface area varied from both at rest and during exercise. The abnormal rise 1-0 cm2 to 2-6 cm2, the mean values being 1-6 cm2 of these pressures during exercise appears to be at rest and 2-0 cm2 during exercise. Thus the cal- related to the size of the xenograft inserted. How-

5 606 A. P. Tandon, D. R. Smith, W. Whitaker, and M. I. Ionescu Table 4 Postoperative haemodynamic data in patients with aortic valve substitute according to annulus diameter Authors Valve substitute Peak systolic gradient Calculated surface area t Mean t Mean Bjork (1970) Bjork-Shiley valve (39) (20) (10) (17) (17) Bjork et al. (1971) Bjork-Shiley valve Sigwart et al. (1976) Lillehei-Kaster valve - 30* 32-7* 21-7* * (48 5)* (48 5)* Cohn et al. (1976) Hancock porcine xenograft Morris et al. (1976) Hancock porcine xenograft (37) Hannah and Reis (1976) Hancock porcine xenograft (70) Present series Ionescu-Shiley pericardial xenograft (12-3) (13-3) (10) (5 7) (1) (9 6) The figures in parentheses are systolic gradients during exercise. *Mean ejection systolic gradient. Annulus diameter (mm) of the valve substitute. ever, the number of patients in each subgroup is too surgical techniques for annulus enlargement. small for this trend to be subjected to statistical Consequently the use of the pericardial valve analysis. should be advantageous in patients with a small Published data regarding systolic gradient and aortic annulus. calculated valve surface area with other currently A separate series of sequential haemodynamic used aortic valve substitutes are given in Table 4. studies, performed at mean intervals of 9-9 and Pericardial xenografts, when compared with these 42 2 months after aortic valve replacement, showed valves, offer the least resistance to forward flow. that the postoperative circulatory improvement In the present investigation the systolic gradient was maintained and the transvalvular gradient did was inversely related to the annulus diameter of not change with the passage of time (Tandon et al., the xenograft inserted. As expected, augmentation 1977). of the calculated xenograft surface area was noted This study has shown significant haemodynamic with increasing annulus diameter of the valve. improvement as a result of aortic valve replacement Several authors (Cohn et al., 1976; Hannah and with the Ionescu-Shiley pericardial xenograft and Reis, 1976; Jones, 1976; Morris et al., 1976) have has substantiated the results of the clinical assesscommented on the high systolic gradients associated ment. The transvalvular gradients are insignificant with the smaller sizes of Hancock porcine xeno- and predictable. Repeat circulatory studies are grafts, and there is general agreement that for contemplated at longer intervals in order to assess clinical use valves larger than 23 mm annulus continually the durability and functional perfordiameter should be preferred for aortic valve re- mance of this valve. placement. High gradients have also been reported with the small sizes of Lillehei-Kaster prostheses. References Nicoloff (1976) found a mean peak systolic gradient of 46 mmhg (range 28 to 70) in 10 patients with Anderson, R. P., Bonchek, L. I., Grunkemeier, G. L., No. 14 Lillehei-Kaster valve (21 mm annulus presentation Lambert, L. E., of and surgical Starr, results A. (1974). by actuarial The analysis methods. and diameter). Sigwart et al. (1976) propose that Journal of Surgical Research, 16, Lillehei-Kaster valves of at least 18 mm intemal Bjork, V. 0. (1970). A new central-flow tilting disc valve diameter (25 mm annulus diameter) should be used prosthesis: one year's clinical experience with 103 patients. in the aortic position in adults in order to minimise Bjork, Journal V. of O., Thoracic Henze, and A., Cardiovascular and Holmgren, Surgery, A. (1974). 60, Five prosthesis-induced stenosis. years' experience with the Bjork-Shiley tilting disc valve Because of the high gradients produced by such in isolated aortic valvular disease. Journal of Thoracic and valves, elaborate surgical manoeuvres have been Cardiovascular Surgery, 68, advocated and used to enlarge the aortic annulus in Bjork, (1971). V. Clinical O., Holmgren, and haemodynamic A., Olin, C., results and Ovenfors, of aorticc. valve 0. order to accommodate a larger valve (Najafi et al., replacement with Bjork-Shiley tilting disc valve prosthesis. 1969; Nicks et al., 1970; Konno et al., 1975; Blank Scandinavian Journal of Thoracic and Cardiovascular et al., 1976). The present study has established that Surgery, 5, the gradient across pericardial xenografts, even those Blank, R. H., Pupello, D. F., Bessone, L. N., Harrison, E. E., ofsls.iatand ngin Sbar, S. (1976). Method of managing the small aortic of small size, is insignificant and therefore the annulus during valve replacement. Annals of Thoracic insertion of this valve does not require additional Surgery, 22,

6 Haemodynamics of aortic pericardial xenograft 607 Brandt, P. W. T., Roche, A. H. G., Barratt-Boyes, B. G., and porcine xenograft aortic valve (abstract). American Journal Lowe, J. B. (1969). Radiology of homograft aortic valves. of Cardiology, 37, 157. Thorax, 24, Najafi, H., Ostermiller, W. E., and Javid, H. (1969). Narrow Bristow, J. D., and Kremkau, E. L. (1975). Hemodynamic aortic root complicating aortic valve replacement. Archives changes after valve replacement with Starr-Edwards of Surgery, 99, prostheses. American Journal of Cardiology, 35, Nicks, R., Cartmill, T., and Bernstein, L. (1970). Hypoplasia Cohn, L. H., Sanders, J. H., and Collin, J. J. (1976). Aortic of the aortic root. Thorax, 25, valve replacement with the Hancock porcine zenograft. Nicoloff, D. M. (1976). In discussion of paper by: Starek, Annals of Thoracic Surgery, 22, P. J. K., McLaurin, L. P., Wilcox, B. R., and Murray, G. F. Dubiel, W. T., and Cullhed, I. (1975). Aortic valve replace- Clinical evaluation of the Lillehei-Kaster pivoting-disc ment with frame supported autologous fascia lata grafts. valve. Annals of Thoracic Surgery, 22, III. Haemodynamic and angiographic findings. Scandina- Olin, C. (1970). Evaluation of the Kay-Shiley disc valve vian Journal of Thoracic and Cardiovascular Surgery, 9, prosthesis in the aortic position. Scandinavian Journal of Thoracic and Cardiovascular Surgery. Suppl. 7. Gorlin, R., and Gorlin, N. G. (1951). Hydraulic formula for Sigwart, U., Schmidt, H., Gleichmann, U., and Borst, H. G. calculation of the area of the stenotic mitral valve, other (1976). In vivo evaluation of the Lillehei-Kaster heart valve cardiac valves, and control circulatory shunts. American prosthesis. Annals of Thoracic Surgery, 22, Heart Journal, 41, Tandon, A. P., Smith, D. R., Mary, D. A. S., and Ionescu, Hannah, H., III., and Reis, R. L. (1976). Current status M. I. (1977). Sequential hemodynamic studies in patients of porcine heterograft prostheses: a 5-year appraisal. having aortic valve replacement with the Ionescu-Shiley Circulation, 54, Suppl. III, pericardial xenograft. Annals of Thoracic Surgery, 24, Ionescu, M. I., Pakrashi, B. C., Mary, D. A. S., Bartek, I. T., and Wooler, G. H. (1974). Long term evaluation of tissue valves. Journal of Thoracic and Cardiovascular Surgery, 68, Addendum Ionescu, M. I., Tandon, A. P., Mary, D. A. S., and Abid, A. (1977). Heart valve replacement with the Ionescu-Shiley Since the preparation of this manuscript, in Novempericardial xenograft. Journal of Thoracic and Cardio- ber 1976, the number of patients who received aortic vascular Surgery, 73, Jones, E. L. (1976). In discussion of paper by: Blank, R. H., pericardial xenograft valves has icreased to 216 and Pupello, D. F., Bessone, L. N., Harrison, E. E., and Sbar, S. 6 more patients had haemodynamic investigations. Methods of managing the small aortic annulus during The clinical results of the entire series, at 7 years valve replacement. Annals of Thoracic Surgery, follow-up, are essentially similar to those described Konno, S., Imai, Y., Iida, Y., Nakajima, M., and Tatsuno, K. (1975). A new method for prosthetic valve replacement i this paper in congenital aortic stenosis associated with hypoplasia of the aortic valve ring. Journal of Thoracic and Cardiovascular Requests for reprints to M. I. Ionescu, Esq., Surgery, 70, FACS,Dpr eto Morris, D. C., Wickliffe, C. W., King, S. B., Douglas, adohrccsrey J. S., F.AGS, Department of Cardiothoracic Surgery, and Jones, E. L. (1976). Hemodynamic evaluation of the The General Infirmary, Leeds LS1 3EX. Br Heart J: first published as /hrt on 1 June Downloaded from on 2 September 2018 by guest. Protected by copyright.