Preliminary observations on haemodynamics during physiological ...

European Journal of Cardio-thoracic Surgery 20 (2001) 262±269

www.elsevier.com/locate/ejcts

Preliminary observations on haemodynamics during physiological stress conditions following `double-ori®ce' mitral valve repair q Valentino Borghetti a, Marco Campana b, Carla Scotti b, Giovanni Parrinello c, Roberto Lorusso a,* a

b

Cardiac Surgery Division, Civic Hospital, Brescia, Italy `Salvatore Maugeri' Foundation IRCCS, Division of Cardiology, Gussago, Italy c Section of Medical Statistics and Biometrics, Brescia, Italy

Received 17 October 2000; received in revised form 18 April 2001; accepted 30 April 2001

Abstract Objective: The `double-ori®ce' (DO) technique has been proposed to simplify mitral valve repair (MVRep) in particular settings of mitral insuf®ciency. However, the haemodynamic effects of such a redesigned valve are poorly documented, particularly during stress conditions. Thus, we sought to evaluate the haemodynamic changes during exercise conditions after the DO procedure. Methods: Twenty-seven selected patients were enrolled for this study. Mean age was 60 ^ 14 years (range 31±80 years). All patients had had severe mitral regurgitation and normal LV function preoperatively, and had undergone DO as isolated procedure for MVRep. Annular remodelling was associated in 24 cases (Carpentier classic ring in 13 patients and autologous pericardium in 11 patients, respectively). Postoperatively, haemodynamic data were recorded at baseline and during supine exercise test at submaximal workloads by means of transthoracic echocardiography. A logistic regression analysis was applied to evaluate the association between the observed haemodynamic changes and surgical technique. Results: Mean follow-up was 47 ^ 20 months. Signi®cant residual mitral insuf®ciency (grade three over four) was found in ®ve patients at baseline assessment, and in six patients at peak exercise. Transmitral pressure gradient showed a signi®cant P , 0:001 increase in both peak and mean values at peak exercise (from 7 ^ 4 to 17 ^ 10 and from 3 ^ 2 to 8 ^ 6 mmHg, respectively). Pulmonary hypertension was observed in six patients (moderate in all cases) at rest and in 13 patients (moderate in seven cases and severe in ®ve cases) at peak exercise. We did not ®nd any correlation between the haemodynamic data and surgical factors. Conclusions: This study indicates that though effective mitral valve competence is achieved in the majority of operated patients, DO repair may induce impaired diastolic mitral dynamism in some cases, particularly during exercise conditions. Further investigations are required to thoroughly elucidate the overall mechanics of a DO valve, especially at strenuous cardiocirculatory states. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Mitral valve repair; Double-ori®ce mitral valve; Mitral incompetence; Stress-test echocardiography

1. Introduction Reconstruction of regurgitant mitral valve because of anterior lea¯et defects or complex anatomical lesions undoubtedly represents a surgical challenge. Several techniques have been proposed to correct the underlying pathological ®ndings and ensure mitral valve repair (MVRep), often being demanding, time-consuming and in some instances, not easily reproducible [1±5]. The application of the `double-ori®ce' (DO) technique has been introduced by Al®eri since 1992 at our institution mainly to simplify the correction of the failing anterior lea¯et [6,7]. The creaq Presented at the 14th Annual Meeting of the European Association for Cardio-thoracic Surgery, Frankfurt, Germany, 7±11 October, 2000. * Corresponding author. Tel.: 139-030-3995636; fax: 139-0303995004. E-mail address: [email protected] (R. Lorusso).

tion of an àcquired' DO valve, by suturing the prolapsing or injured portion of the anterior lea¯et to the opposing edge of the posterior lea¯et, has been shown to provide an easy, effective and reproducible technique. The procedure has been initially applied in routine cases, but, subsequently, showed short- and mid-term successful outcome also in dif®cult pathological settings [8±13]. Debate, however, persists on the potential drawbacks of this procedure, namely, the occurrence of mitral stenosis after the surgical correction, the dynamic behaviour of such a redesigned valve and lastly the long-term ef®cacy and durability of this procedure. Our study speci®cally addressed the evaluation of the haemodynamic effects of a DO repair in a selected group of patients to provide useful insights on such a novel valve con®guration that inevitably implies in¯uence on valve function and related ¯uid dynamics. Furthermore, we

1010-7940/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S10 10-7940(01)0077 4-6

V. Borghetti et al. / European Journal of Cardio-thoracic Surgery 20 (2001) 262±269

assessed the effects of the DO procedure during stressful conditions in order to better elucidate haemodynamic and functional characteristics in this peculiar setting, which could have some impact on the surgical strategy.

2. Methods From June 1992 to October 1999, 75 consecutive patients underwent DO repair at our institution because of mitral insuf®ciency secondary to anterior mitral lea¯et defect or complex valvular lesions. Twenty-seven patients were selected for the echocardiographic assessment at rest and during exercise conditions and represented the study group. Inclusion criteria for the study enrolment consisted of: (1) preoperative presence of central valve prolapse because of anterior mitral lea¯et, of both lea¯ets or of posterior lea¯et with calci®ed posterior mitral annulus; (2) preserved left ventricular function; (3) preserved respiratory function; (4) absence of postoperative angina syndrome or ischaemic mitral valve regurgitation; (5) absence of ventricular arrhythmia; (6) absence of functional impairment or orthopaedic disease which could adversely affect exercise capacity. Preoperative patient pro®le is displayed in Table 1. Preoperative mitral regurgitation had been severe (1111) in all patients. The mechanisms of regurgitation had been anterior lea¯et prolapse in 11 patients (40%), prolapse of both lea¯ets in eight patients (30%) and posterior lea¯et prolapse in eight patients (30%). The etiology had been degenerative Table 1 Preoperative patient pro®le a Number of patients

27

Gender Age (years)

20 M, 7 F 60 ^ 14 (range 31±80)

NYHA class I±II III±IV

15 (55%) 12 (45%)

Cardiac rhythm SR AF PM

21 (78%) 5 (18%) 1 (4%)

Etiology Degenerative Bacterial endocarditis Rheumatic Mechanism of regurgitation ALP PLP BLP a

BD 12 (44.5%) FED 2 (7.5%) 7 (26%) 6 (22%) 11 (40%) 8 (30%) 8 (30)

NYHA, New York Heart Association; SR, sinus rhythm; AF, atrial ®brillation; PM, pacemaker; BD, Barlow disease; FED, ®bro-elastic de®ciency; ALP, anterior lea¯et prolapse; PLP, posterior lea¯et prolapse; BLP, bilea¯et prolapse.

263

in 14 patients (52%), infective in seven patients (26%) and rheumatic in six patients (22%). All patients had undergone conventional DO repair as originally described [7]. Annuloplasty had been carried out in 24 patients by the application of a rigid classic Carpentier ring in 13 patients (48%) and glutaraldeheyde pretreated autologous pericardial ring in 11 patients (41%), respectively. No annular remodelling had been used in three patients (11%). In three patients, commissural edge-to-edge technique had also been performed [7], whereas one patient had received conventional posterior quadrangular resection. Associated procedures had been aortic valve replacement (two patients) and CABG (two patients). After MVRep, the left ventricle had been forcefully ®lled in with saline water to test the valve competence. Intraoperative transoesophageal echocardiography had not been performed routinely, but all the patients had been submitted to postoperative echocardiographic assessment before hospital discharge. During the perioperative course, four patients had had transient pulmonary hypertension and one patient had had temporary right ventricular dysfunction. Two patients had shown residual mild mitral valve regurgitation. Bacterial endocarditis had occurred in one patient (medically treated with no residual functional or anatomical impairment) and one patient had undergone permanent pacemaker implantation because of the onset of third degree atrio/ventricular block. All patients had been on anticoagulant therapy for the ®rst 3 months. 2.1. Echocardiographic study protocol All patients underwent bicycle ergometre stress test in the supine position, with a 25 W incremental step every 3 min, until 70% of maximal age-predicted heart rate was reached. Real-time phased array echo-Doppler recordings were obtained at rest and during maximal exercise with a GE Vingmed CFM 750 CV (2.75 or 3.25 MHz transducer). The examinations were carried out by two experienced investigators. Digital images were stored on a magneto-optical disk for subsequent analysis. LV diameters and volumes (area±length method) were determined from two-dimensional parasternal long-axis view and from apical four-chamber view, respectively. Calculations of LV fractional shortening and LV ejection fraction were then derived from diameters and volumes, respectively. Continuous-wave Doppler was used to obtain transmitral pressure gradient (TPG); both mean and peak gradients were considered. When regurgitant jet was present, maximal velocity of tricuspid Doppler tracing was measured to estimate systolic pulmonary artery pressure. A ®xed value of 10 mmHg has been added as estimated right atrial pressure, after determination of inspiratory collapse of the inferior vena cava [14]. Measurements of all echo-Doppler examinations were performed in a single day, by only one investigator. Mean value of three measurements was considered for each variable.

264


Table 2 Echocardiographic monitoring data at rest a Parameter

Preoperative

Postoperative

P

LAD LVEDD LVESD EF

54 ^ 10 mm 66 ^ 19 mm 41 ^ 10 mm 61 ^ 9%

37 ^ 11 mm 53 ^ 5 mm 36 ^ 7 mm 60 ^ 7%

, 0.001 , 0.01 ns ns

a LAD, left atrial diameter; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; EF, ejection fraction.

2.2. Statistical analysis All values are expressed as mean ^ standard deviation. We used the Wilcoxon rank sum test, given the skewed distribution of peak TPG (PTPG), to analyze possible differences of this variable in relation to the type of ring. A logistic regression was also performed to evaluate the prediction role of PTPG in relation to the categorised variable pulmonary hypertension (1 present, 0 absent). Student's t-test for unpaired variables was used for the analysis of volumes, diameters and contractile performance of the left ventricle pre- and postoperatively. A P value less than 0.05 was considered as signi®cant.

3. Results Mean follow-up was 47 ^ 20 months (range 4±81 months). Clinical examination at the time of echocardiographic study showed 23 patients in NYHA class I±II and four patients in NYHA class III. Ten patients were on diuretics and ACE-inhibitors. Cardiac rhythm at the time of

investigation was sinus rhythm in 24 patients, atrial ®brillation in two patients and one patient was pacemaker-dependent because of the persistent complete atrio/ventricular block. Table 2 shows the haemodynamic data recorded at rest preoperatively and at the time of investigation, as measured by transthoracic echocardiography. Signi®cant mitral valve regurgitation (111/1111) was shown in ®ve patients (18%). Pulmonary hypertension (moderate in all cases) was detected in six patients (22%). Left ventricular contractility was normal in all patients. Supine exercise signi®cantly increased heart rate and systemic blood pressure. Table 3 shows functional and haemodynamic data obtained during exercise. All patients but one showed enhancement of left ventricular contractility during exercise. In one patient left ventricular ejection fraction fell from 54 to 47% at peak exercise. One patient showed mitral systolic anterior movement with left ventricular out¯ow tract obstruction with a signi®cant intraventricular gradient (55 mmHg). No substantial modi®cations were elicited by stress conditions in terms of mitral valve insuf®ciency. Conversely, TPG showed a signi®cant P , 0:001 increase in both peak and mean values at peak exercise (from 7 ^ 4 to 17 ^ 10 and from 3 ^ 2 to 8 ^ 6 mmHg, respectively) (Fig. 1). Furthermore, 12 patients showed PTPG values which doubled at peak exercise if compared with resting data (Fig. 2). Similarly, the number of patients who developed pulmonary hypertension during exercise rose from six to 13 cases P , 0:001, which con®rmed the trend towards a restrictive behaviour of a DO valve during strenuous cardiocirculatory conditions (Fig. 3). Both peak systolic values of pulmonary hypertension and PTPG were matched with the presence of signi®cant mitral valve regurgitation, in order to evaluate the in¯uence of such

Table 3 Degree of mitral valve regurgitation and haemodynamics during exercise a Rest

Peak

P

HR SBP EDV ESV EF

78 ^ 13 137 ^ 14 87 ^ 24 35 ^ 11 60 ^ 7

126 ^ 14 184 ^ 25 83 ^ 18 29 ^ 8 67 ^ 7

, 0.001 , 0.01 ns ns , 0.01

TPG MTPG PTPG

3.5 ^ 2.5 7.5 ^ 4.7

8.0 ^ 6.1 17.7 ^ 10.6

, 0.001 , 0.001

Mitral valve regurgitation Absent or trivial (1) Moderate (1 1 ) Severe (111/1111)

10 patients (37%) 12 patients (44%) 5 patients (18.5%)

10 patients (37%) 11 patients (40%) 6 patients (22%)

ns ns ns

Systolic pulmonary hypertension Moderate Severe

6 patients (22%) 0 patients

7 patients (26%) 6 patients (22%)

, 0.001

a

HR, heart rate; SBP, systolic blood pressure; EDV, end-diastolic volume; ESV, end-systolic volume; EF, ejection fraction; TPG, transmitral pressure gradient; MTPG, mean transmitral pressure gradient; PTPG, peak transmitral pressure gradient.


Fig. 1. Diagram represents the increase of TPG in both mean and peak values during exercise.

a variable on haemodynamics and no correlation was found. No statistical relation was also found between haemodynamic changes and the timing of postoperative follow-up. We did not evaluate the in¯uence of cardiac rhythm on haemodynamics since only three patients were in atrial ®brillation. A statistically signi®cant correlation P , 0:05 was found between PTPG reached at peak exercise and development of systolic pulmonary hypertension (Fig. 4). Furthermore, we compared PTPG with the type of the ring implanted, either prosthetic (13 patients) or biologic (11 patients). A more restrictive pattern was observed in

265

Fig. 3. Graphic shows the number of cases of pulmonary hypertension at basal conditions and at peak exercise, respectively. An increase in absolute peak pulmonary pressure values was also seen.

the presence of a rigid ring, although correlation did not reach statistical signi®cance. Logistic regression analysis showed a signi®cant correlation between PTPG and probability to develop systolic pulmonary hypertension during exercise. The associated reconstructive techniques to achieve mitral competence (commissural closure) apparently did not affect haemodynamic response to exercise after DO repair. Indeed, no pulmonary hypertension was detected in

Fig. 2. Mitral and tricuspid Doppler signals at rest and during exercise conditions are shown. A marked increase of both transmitral ¯ow velocities (upper panel) and onset of tricuspid regurgitant jet (lower panel) in a case of exercise-induced pulmonary hypertension are shown.

266


Fig. 4. Notched boxes are drawn. Patients are subdivided into two groups on the basis of presence (yes) or absence (no) of pulmonary hypertension. The ®gure shows that patients who developed pulmonary hypertension had higher PTPG at peak exercise. If the notches on two boxes do not overlap, this indicates a difference in a location roughly at 5% signi®cance level. Correlation showed a statistical signi®cance between two groups of patients.

two cases either at rest or at peak exercise, whereas the third patient developed moderate pulmonary hypertension only during stressful conditions. 4. Discussion DO repair was proposed to enhance and simplify MVRep in dif®cult pathological settings. The apposition of the prolapsing or injured portion of the mitral lea¯et to the opposite lea¯et edge leading to the creation of acquired DO valve appeared to provide ease and durable ef®cacy in mitral reconstruction results [13]. Soon thereafter, concern was expressed by some investigators regarding the possibility of peculiar shortcomings, namely mitral stenosis and reduced mechanical properties of such a redesigned valve. Maisano et al. [15] reported on the computational model of blood ¯ow through a DO valve and showed that blood velocity was exclusively related to the total valve area irrespective of the conformation and equality of the two ori®ce areas. Accordingly, echo-Doppler measurements were shown to provide reliable estimation of transvalvular pressure gradient by simply analyzing blood ¯ow velocity in one ori®ce. Nonetheless, a major limitation of this study was linked to the evaluation of a virtual and therefore inanimate model of valvular physiology. The effects of exercise on the normal mitral valve physiology have been clearly reported by Rassi et al. [16] who showed that exercise state induces augmentation of transmitral blood ¯ow by an increase in mean diastolic cross-sectional area. These ®ndings suggest that the maintenance or the least limitation of natural diastolic motion of

the mitral valve is an important component of MVRep to ensure unrestrictive transvalvular ¯ow rates, especially during high ¯ow cardiocirculatory conditions. From a purely surgical standpoint, the procedure implies the testing, at the end of the repair, by which the dimensions of the obtained ori®ces are calculated. So, the induction of haemodynamically signi®cant mitral stenosis is extremely unlikely under a resting condition [11±13]. However, though acceptable global valve area is usually maintained after DO repair, the functional consequences of two ori®ces were yet to be de®ned during exercise. These premises prompted us to study a selected group of patients previously submitted to DO repair. Our ®ndings demonstrated that a DO valve may predispose to restrictive transmitral blood ¯ow and hence to pulmonary hypertension, which is more evident during exercise conditions. Echocardiographic estimation of LV function at rest and during exercise in our study group showed preserved functional state which excludes the potential correlation between increased transvalvular gradient and impaired diastolic ventricular ®lling or systolic performance. Moreover, the extent of mitral insuf®ciency did not in¯uence and was not statistically related to exercise-induced pulmonary hypertension, since patients with no mitral incompetence at rest and at high ¯ow conditions developed pulmonary hypertension. On the other hand, haemodynamically signi®cant mitral regurgitation, which was assessed at resting conditions in a few patients, did not change during the exercise test despite increased pulmonary resistance, making the hypothesis of increased pulmonary hypertension due to augmented valve incompetence inconsistent. Furthermore, the appearance or deterioration of pulmonary hypertension was apparently related to the type of annuloplasty performed, despite the absence of linear correlation. We speculate that limitation of mitral annular excursion and valve opening reserve due to the central suture might partially explain the restricted transmitral blood ¯ow. Umana et al. [17] speci®cally addressed annular mechanics in a sheep model of DO repair in the setting of acute ischaemic mitral insuf®ciency, and showed preserved annular motion and ventricular function as compared to conventional MVRep. This experimental model, however, did not include annuloplasty procedures. Our study suggests that rigid ®xation of the mitral annulus in DO repair may further predispose to reduced mitral valve diastolic motion since rigid prosthetic ring induced a more marked increase in transvalvular gradient during exercise test as compared to ¯exible biological annular support. This ®nding is in accordance with previous investigations performed at our institution and by others on the effects of different approaches in terms of annular remodelling in MVRep [18±20]. As suggested by Umana et al. [11], annular remodelling during DO repair should be achieved with minimal intervention and, in our opinion, with ¯exible and limited (C-type) devices to partially


preserve the diastolic mechanical contribution of the native annulus and to counterbalance the centripetal effect of the central suture. Recurrent mitral insuf®ciency may be encountered in some patients at follow-up, but the in¯uence of such redesigned valves on ¯uid dynamics, which may affect ef®cacy and durability of MVRep is still unknown. A thorough analysis of our complete patient population is under way to speci®cally elucidate the precise mechanism of recurrent mitral regurgitation and to disclose potential determinants of unfavourable outcome after the DO repair. The DO repair had a favourable impact on several anatomical and functional cardiac parameters in our study population. Remarkable reduction in preoperative LA and LV dimensions were clearly documented at follow-up. Furthermore, the majority of patients showed competent valve, with no modi®cation even under stressful haemodynamic conditions. Therefore, we can con®rm that DO repair may be a valuable technique to face complex mitral lesions which could otherwise lead to valve replacement or to extremely complex reparative techniques with uncertain outcome. 4.1. Study limitation Our study has been carried out on a limited patient population; therefore, the statistical analysis could have been adversely in¯uenced by inappropriate study power. Patients were not randomised according to the application or to the type of annuloplasty procedure. No homogeneity in terms of etiology of the disease and patient age was present, making the impact of these inherent differences in mitral valve behaviour possible. All selected patients underwent exercise test, also in the presence of haemodynamically signi®cant mitral incompetence, which could have affected some data interpretation. Nonetheless, these patients represented only a minority and the behaviour of such valve regurgitation in the presence of a DO valve was considered anyhow important to allow appropriate decision making for hypothetical indication of reoperation. In conclusion, our study showed the haemodynamic effects of a redesigned DO valve for MVRep recorded at rest and during exercise conditions. A tendency towards restrictive patterns of transmitral blood ¯ow was shown, particularly under strenuous cardiocirculatory conditions. Our data also suggest that the postoperative functional mitral stenosis does not relate to the application of the DO valve per seÂ as the only causal factor; rather, it is the result of a combination of several factors, particularly related to annular remodelling. These insights may be meaningful to design a more ef®cient application of a DO repair (unrestrictive annuloplasty) to correct complex mechanism of mitral insuf®ciency. Further studies are required to better elucidate the modi®cations of the actual DO mitral dynamism and ¯uid dynamics following DO

267

repair, particularly without the in¯uence of prosthetic annular support.

References [1] Smedira NG, Selman R, Cosgrove DM, McCarthy PM, Lytle BW, Taylor CP, Apperson-Hansen C, Stewart RW, Loop FD. Repair of anterior lea¯et prolapse: chordal transfer is superior to chordal shortening. J Thorac Cardiovasc Surg 1996;112:287±292. [2] Salati M, Moriggia S, Scrofani R, Santoli C. Chordal transposition for anterior mitral prolapse: early and long-term results. Eur J Cardiothorac Surg 1997;11:268±273. [3] Frater RWM, Vetter HO, Zussa C, Dahm M. Chordal replacement in mitral valve repair. Circulation 1990;82(Suppl IV):IV-125±IV-130. [4] Di Mattia DG, Lemma M, Scrofani M, Mangini A, FundaroÁ P. Treatment of anterior mitral lea¯et prolapse: chordal suture plication and free-edge remodeling in 34 patients. G Ital Cardiol 1998;28:630±635. [5] FundaroÁ P, DiMattia D, Salati M, Santoli C. Free edge suture plication and remodeling: a technique for anterior mitral lea¯et prolapse repair. Ann Thorac Surg 1997;63:1186±1188. [6] Fucci C, La Canna G, Berra P, Pardini A, Sandrelli L, Al®eri O. Reconstructive surgery of mitral valve for bacterial endocarditis. G Ital Cardiol 1995;25:335±340. [7] Fucci C, Sandrelli L, Pardini A, Torracca L, Ferrari M, Al®eri O. Improved results with mitral valve repair using new surgical techniques. Eur J Cardio-thorac Surg 1995;9:621±627. [8] Maisano F, Torracca L, Oppizzi M, Stefano PL, D'Addario G, La Canna G, Zogno M, Al®eri O. The edge±edge technique: a simpli®ed method to correct mitral insuf®ciency. Eur J Cardio-thorac Surg 1998;13:240±246. [9] Lorusso R, Fucci C, Pentiricci S, Coletti G, La Canna G, Zogno M. ªDouble-ori®ceº technique to repair extensive mitral valve excision following acute endocarditis. J Card Surg 1998;13:24±26. [10] Mc Carthy JF, McCarthy PM, Starling RC, Smedira NG, Scalia GM, Wong J, Kasirajan V, Goormastic M, Young JB. Partial left ventriculectomy and mitral valve repair for end-stage congestive heart failure. Eur J Cardio-thorac Surg 1998;13:337±343. [11] Umana JP, Salehizadeh B, DeRose Jr JJ, Nahar T, Lotvin A, Homma S, Oz MC. ªBow-tieº mitral valve repair: an adjuvant technique for ischemic mitral regurgitation. Ann Thorac Surg 1998;66:1640±1646. [12] Totaro P, Tulumello E, Fellini P, Rambaldini M, La Canna G, Coletti G, Zogno M, Lorusso R. Mitral valve repair for isolated prolapse of the anterior lea¯et: an 11-year follow-up. Eur J Cardio-thorac Surg 1999;15:119±126. [13] Maisano F, Schreuder JJ, Oppizzi M, Fiorani B, Fino C, Al®eri O. The double-ori®ce technique as a standardised approach to treat mitral regurgitation due to severe myxomatous disease: surgical technique. Eur J Cardio-thorac Surg 2000;17:201±205. [14] Simonson J, Schiller NB. Sonospirometry: a new method for noninvasive estimation of mean right atrial pressure based on two-dimensional echographic measurements of the inferior vena cava during measured inspiration. J Am Coll Cardiol 1998;11:557±564. [15] Maisano F, Redaelli A, Pennati G, Fumero R, Torracca L, Al®eri O. The hemodynamic effects of double-ori®ce valve repair for mitral regurgitation: a 3D computational model. Eur J Cardio-thorac Surg 1999;15:419±425. [16] Rassi Jr A, Crawford MH, Richards KL, Miller JF. Differing mechanisms of exercise ¯ow augmentation at the mitral and aortic valves. Circulation 1988;77:543±551. [17] Umana JP, DeRose J, Choudhri A, Madigan J, Spanier T, Nahar T, Mihalatos D, Homma S, LoõÂvin A, Dickstein M, Sun B, Oz M. ªBowtieº mitral valve repair successfully addresses subvalvular dysfunction in ischemic mitral regurgitation. Surg Forum 1997;48:279±280. [18] David TE, Komeda M, Pollick C, Burns RJ. Mitral valve annulo-

268


plasty: the effect of the type on left ventricular function. Ann Thorac Surg 1989;47:524±528. [19] Okada Y, Shomura T, Yamaura Y, Yoshikawa J. Comparison of the Carpentier and Duran prosthetic rings used in mitral valve reconstruction. Ann Thorac Surg 1995;59:658±663. [20] Borghetti V, Campana M, Scotti C, Domenighini D, Totaro P, Coletti G, Pagani M, Lorusso R. Biological versus prosthetic rig in mitral valve repair: enhancement of mitral annulus dynamics and left-ventricular function with pericardial annuloplasty at long term. Eur J Cardio-thorac Surg 2000;17:431±439.

Appendix A. Conference discussion Dr D. Adams (Boston, MA, USA): Your two papers beg one question. I am very concerned about your failure rate, as more than 20% of patients have severe residual regurgitation. I do not believe the double-ori®ce repair technique should replace the remodeling techniques of Professor Carpentier but rather be an adjunct in selected cases. Do you have a comparative group in your hospital that has undergone repair using Carpentier's techniques? Dr Borghetti: Undoubtedly this is just our preliminary report. This study has been conducted on just half of the patients who underwent doubleori®ce repair in our center. So the high percentage of failure with this

Appendix B. Editorial Comment `There are no sects in geometry' Voltaire. In this issue of the Journal, Dr Borghetti and associates raise a question that has been in the mind of most surgeons faced with this innovative technique. They report the postoperative echocardiographic ®ndings in a group of 27 patients followed for a mean of 47 months after edge-toedge suture of the mitral lea¯ets (Àl®eri's stitch'). An annuloplasty had been added in 88% of the patients. At peak exercise, an important transmitral gradient became apparent with pulmonary hypertension in 13 cases. Signi®cant regurgitation was detected in six cases (22%). One case of signi®cant systolic anterior movement (SAM) was observed. These rather poor results must be carefully analyzed and should not immediately disqualify a technique that because of its simplicity, has probably saved many mitral valves. A criticism of these results might be the case selection. Out of the 75 patients with the procedure, only 27 nonrandomized cases were studied. Although the inclusion criteria are described, whether a natural tendency towards studying patients with a poor clinical outcome is possible. Also, the small number of patients minimizes the value of the study. However, early reports of a new technique tend to be limited in the number of patients. This argument plays both ways since the early publications by Al®eri's group did not report many cases either [21]. It can also be argued that although the sizes of the annuloplasty rigid rings used by the authors were large, they might further the diastolic restriction imposed by the lea¯et suture. This `double-ori®ce' technique has recently become quite popular, primarily because of its technical simplicity. As

particular technique is maybe due to the dif®culty in the selection of patients and interpretation of the mechanism of regurgitation. In fact, last year we presented a similar study carried out in a population of patients who had undergone conventional posterior quadrangular resection and no or trivial mitral valve regurgitation had been found. Dr S. Collins (Umea, Sweden): As you know, we have other variables which in¯uence the mitral transgradient and pulmonary hypertension. Have you considered such variables, such as left ventricular function and so on? Dr Borghetti: Yes. We have taken into account other variables such as the degree of mitral regurgitation or cardiac rhythm, which could in¯uence the transmitral pressure gradient, but no correlation was found. Moreover, we tried to correlate the presence of left ventricular dysfunction with the development of pulmonary hypertension, which would be in¯uenced by that, but again, no correlation was found. Dr O. Al®eri (Milan, Italy): How could you include in your study a population of patients who had severe mitral insuf®ciency and pulmonary hypertension at baseline evaluation? You have simply demonstrated that a patient who did not have a good repair does not tolerate an effort. Dr Borghetti: The patients had been chosen with strict inclusion criteria and we excluded patients with preoperative pulmonary hypertension. Moreover, I have mentioned before that other variables, able to in¯uence our ®ndings during exercise, have been taken into consideration and we failed to show any correlations between them.

with all new surgical techniques, their originators, because of their dedication to the subject, careful technique and experience that eliminates earlier mistakes, tend to report better results than their followers. These early good results encourage the optimistic view that the new approach can solve most problems. Usually, time and hard reality modulate the initial enthusiasm until a balance is achieved and the more limited and precise indications become progressively established. Therefore, it is bene®cial to learn of the problems that an òutsider' has encountered when applying this new technique. Contrary to many emerging techniques, this one is not limited by its technical complexity. Confusion might have arisen about the need for adding an annuloplasty to the double-ori®ce technique. Although initially Dr Al®eri limited the surgery to only his stitch, more recent reports show a progressive increase in the use of a contention annuloplasty. In his recent publication (corresponding to his presentation at the last Annual Meeting of the Society [21] 91% (75/82) of the patients had an annuloplasty while at the presentation 69% (47/68) had an annuloplasty. The actual reasons that led to this increase should be known. Also, it was understood that Dr Al®eri considered this technique as a partial suture closure at any point of the mitral coaptation line which included the commissures [22] and therefore not necessarily a `double-ori®ce' technique. Appropriately, the authors of the present report include three such cases. Two main questions about this technique must be addressed today: its durability and its precise indications. A recent study from the Stanford group [23], in an experimental model of ischemic regurgitation, has shown an abnormal cyclic increase in tension (particularly during diastole) at the level of the Al®eri stitch. The clinical impli-


cations of these ®ndings are unknown. Only time will elucidate this question. Another important point is whether this technique can and should be applied to all cases of mitral regurgitation. Again, while the initial reports included rheumatic cases, present information tends to concentrate on the excellent results obtained in myxomatous disease. A possibly clearer indication is prolapse in the presence of a calci®ed annulus where lea¯et resection is hazardous although more complex techniques such as chordal replacement are available. It is probably unwise to expect good results in rheumatic valves where ®brotic lea¯ets do not allow for their partial closure without impunity. Twenty-two percent of the patients reported by Borghetti and associates had a rheumatic etiology. An interesting area where this technique is receiving attention is in cases of `functional' regurgitation where our ignorance of its mechanism limits the bene®t of a simple reducing annuloplasty. The simplicity of the method, possibility of being performed through an open ventriculotomy and extra time needed for an annuloplasty has encouraged some groups to use it systematically in ischemic and dilated cardiomyopathies.

269

Dr Borghetti and collaborators should be congratulated for raising questions that encourage fruitful discussion. In this ®eld close to geometry, dispassionate reporting of positive and negative results will determine its value and decide whether to incorporate it to the mitral repair armamentarium. So far, it should be observed carefully. Carlos M.G. Duran International Heart Institute of Montana, 554 West Broadway, Missoula, MT 59802-4008, USA [1] Maisano F, Schreuder JJ, Oppizzi M, Fiorani B, Fino C, Al®eri O. The double-ori®ce technique as a standardized approach to treat mitral regurgitation due to severe myxomatous disease: surgical technique. Eur J Cardio-thorac Surg 2000;17:201±205. [2] Al®eri O. Invited commentary. Umana et al. `Bow-tie' mitral valve repair: an adjuvant technique for ischemic mitral regurgitation. Ann Thorac Surg 1998;66:1645±1646. [3] Nielsen SL, Timek TA, Liang DH, Daughters DG, Ingels NB, Miller DC. What is the tension on Al®eri's stitch used for repair of ischemic mitral regurgitation? (abstract). Circulation 2000;102(II): 461.