Catheterization and Cardiovascular Diagnosis 2 3 172-1 76 (1991)
Fluoroscopic Guidance in Transseptal Catheterization for Percutaneous Mitral Balloon Valvotomy P. Rocha, MD, J. Berland, MD, M. Rigaud, MD, F. Fernandez, MD, J.P. Bourdarias, MD, FACC, and B. Letac, MD, FACC A percutaneous mitral balloon valvotomy (PMBV) was attempted on 190 patients with fluoroscopic guidance of atrial septal puncture for transseptal catheterization; in 3 cases, the procedure could not be performed. The left atrium was always reached on the first attempt, when the relationship of the Brockenbrough needle to the aortic catheter was previously observed in 3 fluoroscopic views: anteroposterior, 45" right anterior oblique, and lateral. The atrial septal puncture site was located immediately below the aortic valve level, probably in the fossa ovalis, for the first 80 patients, and at mid distance between the aortic valve level and the diaphragm for the last 110. Hemodynamic data were similar in both groups. Fluoroscopic guidance for atrial septal puncture seemed capital for patients with scoliosis or in whom a vascular distortion (e.g., advanced pregnancy, right inferior vena cava absence) prevented a perfect parallelism between the needle curve and the needle outer index.
Key words: fluoroscopic incidences, accidents, transseptal catheterization, percutaneous mitral balloon valvotomy
mitral stenosis and an echographic mitral valve area < 1.5 cm2. There were 142 women and 48 men, whose Transseptal left heart catheterization was introduced mean age was 44 2 17 years. Thirty patients had had a almost simultaneously by Cope [I] and by Ross and prior surgical mitral commissurotomy ; 70 patients had Braunwald  in 1959. Frequent complications associatrial fibrillation; the others had normal sinus rhythm. ated with this technique [3-51 and the increasing use of Mild mitral regurgitation was present in 65 patients. pulmonary wedge pressure to assess left atrial pressure Most of patients were in the NYHA functional class 111. led to a decline in the use of this procedure in the mid to Mitral calcification was present in 71 patients. Patients late 1960s. Percutaneous valvotomies brought a revival with recent peripheral embolism, severe mitral regurgiof the transseptal approach to mitral valvotomy  in tation, or evidence of left atrial thrombus on echocardio1984, and later to aortic valvotomy , when retrograde graphic study were not considered for the procedure. catheterization was not possible . Several recent inPatients were separated into two groups, according to novations in equipment and refinement of the transseptal the level at which the atrial septum was punctured, as technique have rendered it a safer procedure . will be explained in ' 'Transseptal catheterization. " In O'Keefe et al.  recommended that the needle position group I, composed of the first 80 patients, mitral cathebe observed in the anteroposterior and the lateral fluoroter-balloons with a straight tip were used for the first 38; scopic views before septal puncture. Croft and Lipscomb a pigtail tip was used for the other 42, as well as for all [ 101 recommended the right anterior oblique (40" to 50") group I1 patients. For 8 patients in group I, the mitral view to locate left atrial limits. In this report, fluoroscopic guidance of the transseptal catheterization technique employed in 190 consecutive percutaneous mitral valvotomies is described and com- From the Physiology and Cardiology departments, Ambroise Par6 mented upon. Hospital, Boulogne, France (P.R., M.R., F.F., J.P.B.) and the CardiINTRODUCTION
Percutaneous mitral balloon valvotomy (PMBV) was attempted on 190 consecutive patients with symptomatic 0 1991 Wiley-Liss, Inc.
ology Department, Charles Nicolle Hospital, Rouen, France (J.B., B.L.).
Received September 1 , 1990; revision accepted January 22, 1991 Address reprint requests to Paul0 Rocha, Service d'Explorations Fonctionnelles, HBpital Ambroise Pa& 92 100 Boulogne, France.
Transseptal Catheterization in PMBV
Fig. 1. Group I: the relationship between the Brockenbrough needle and the aortic catheter observed in three fluoroscopic views: A) anterioposterior, B) 45" right oblique, and C) lateral. The aortic catheter pigtail is in contact with the aortic valve. The atrial septum puncture is performed probably at the foramen ovale level.
often falls into the foramen ovale, moving slightly to the left. For the last 110 patients (group II), the puncture site was lower than in the first group-at mid distance between the aortic valve and the diaphragm (Fig. 2). In both groups, the relationship between the needle and the pigtail aortic catheter was observed before atrial septal puncture in 3 fluoroscopic views: anteroposterior, 45" right anterior oblique (RAO), and lateral. In the 45" RAO view, the needle tip was rotated until it pointed Protocol Diagnostic catheterization and left ventriculography slightly towards the aortic root. In the lateral view, the were performed immediately before valvotomy . Trans- needle tip was pointed slightly towards the opposite side septal catheterization was performed when the mitral of the aortic root (Figs. 1, 2). In both views, the needle valve area, hemodynamically calculated by Gorlin's for- tip appeared between the two aortic catheter segments mula [ l l ] , was smaller than 1.5 cm2. Ten minutes after (ascending and descending aorta). The Mullins sheath PMBV, another hemodynamic study was carried out, was never pushed across the septum before it had been including a left ventriculography, a left atrial opacifica- ascertained that the pressure collected at the tip of the Brockenbrough needle matched left atrial pressure. Heption, and an oximetry run. arin (1 mg/kg) was infused IV after successful atrial septal puncture. The atrial septum was dilated with an 8 mm Transseptal Catheterization diameter balloon, prior to the passage of mitral balloons. In most cases, an 8 F Mullins sheath was advanced over a 0.032 guide wire placed in the superior vena cava, through a 14 F venous sheath inserted into the right fem- Mitral Dilation A tipped-balloon catheter (Critikon) was inserted oral vein. A Brockenbrough needle was then introduced into the Mullins sheath, without going beyond the through the Mullins sheath. The mitral valve was crossed sheath's extremity. The external needle index, parallel to with this balloon, previously inflated in the left atrium the needle curve, was oriented down at 45" between the with a diluted radiographic contrast medium. One 0.038 patient's legs. The set (needle plus Mullins sheath) was 2.6 m long J guide wire (USCI) was introduced into the pulled back from the superior vena cava into the right catheter (or two guidewires into the Mullins sheath) and atrium. For the first 80 patients (group I), the needle tip placed in the left ventricular apex, The catheter-balloons was placed immediately below the aortic valve, identi- for mitral dilation were pushed across the mitral valve fied by the pigtail tip of the diagnostic catheter pushed over this (or these) guidewire(s). One or more inflations against the aortic valve (Fig. 1). At this level, the needle were performed until the mitral waist disappeared.
valve was dilated with only one balloon; two balloons were used for all the others. Bifoil balloons were used for 57 of the 110 patients in group 11, but for none in group I. These discrepancies preclude the possibility of a comparison between groups. Nevertheless, the mitral valve area obtained after PMBV are presented here and the data for the two groups are examined.
Rocha et al.
Fig. 2. Same as in Figure 1: group II. In the 45” right oblique view, the needle tip is rotated until it points slightly towards the aortic root. In the lateral view, the needle tip is pointed slightly towards the opposite side of the aortic root. In the anteroposteriorview, the needle tip is placed at mid distance between the aortic valve and the diaphragm.
Mean and standard deviations were determined for quantitative variables. Intragroup and intergroup comparisons were analysed using a paired and an unpaired Student t-test. A probability p value of < 0.05 was considered significant. RESULTS
In group I (80 patients), PMBV could not be performed on three patients (in 1 it was impossible to cross the atrial septum with the 8 mm balloon, in 1 there was a stroke during left atrium manipulation, and in 1 a tamponade). In group 11, all procedures were completed. In group I, there were 1 aortic root puncture and 2 pericardial punctures during transeptal catheterization, with no tamponades. In group 11, all atrial septal punctures were successful, even for two patients with scoliosis and two patients with abdominal vascular distortion. A tamponade was detected immediately after PMBV in four group I patients and in one patient from group 11. In group I, the mitral valve area increased from 1.08 0.31 to 2.19 2 0.67 cm’ (p < 0.001). In group 11, the mitral valve area increased from 0.99 & 0.25 to 2.08 -+ 0.58 cm2 (p < 0.001). Mitral valve area enlargement was similar in both groups (NS).
DISCUSSION PMBV Failures
PMBV failures were not necessarily related to transseptal catheterization and if repeated today, the proce-
dure would probably be successful. PMBV could have been concluded by the Babic technique  for the patient whose atrial septum could not be crossed with the balloon. Today, PMBV would not be interrupted for the patient who developed a stroke during left atrial manipulation, if long ressucitation procedures were not needed. As for the patient whose case was complicated by a tamponade before mitral dilatation, PMBV could be performed after pericardium drainage. Atrial Septa1 Puncture
At the beginning of our experience (group I), there were 3 errors in septal puncture. In these 3 cases, the Brockenbrough needle, momentarily released in order to perform the radiologic arc manipulations needed for fluoroscopic guidance, moved spontaneously. From then on, the needle was carefully stabilized by the operator during fluoroscopic incidence shifts, and atrial septal punctures have since reached the left atrium without fail. These three false moves did not result in cardiac tamponades, probably because patients did not receive anticoagulants before septal puncture, and also because the larger segment of the Mullins sheath extremity was never pushed across the septum before it was ascertained that the pressure collected at the tip of the Brockenbrough needle was, in fact, left atrial pressure. Searching for the atrial septal puncture site under 3 fluoroscopic views has proved to be an effective procedure: it correctly guided atrial septal puncture for all patients. Fluoroscopic needle observation appears particularly useful when abdominal anomalies prevent a strict parallelism between the needle tip curve and the exterior
Transeptal Catheterization in PMBV
Fig. 3. A) Transseptal puncture level around the foramen ovale (group I). When the balloons are inflated, the catheter's acute angle often allows the balloons to slide back into the lett atrium. B) Transseptal puncture level below the foramen ovale (group 11). The distance between the atrial septum and the mitral valve
is shorter and the catheter curve is smooth. This tends to prevent the balloon from sliding back during inflation. In both images, an additional 0.038 J guide wire is seen across the mitral valve; it can be used if greater balloon diameter is necessary.
index during needle rotations. This was the case for 2 patients, one with a left inferior vena cava and the other an 8-month-pregnant woman. For these patients, the outer needle index was at 2 o'clock and 1 o'clock, respectively, on a cross-sectional plane perpendicular to the operating table, instead of at 4 o'clock or 5 o'clock, as usual. Fluoroscopic observation was also useful for 2 additional patients with scoliosis. For them, after placement of the 2 diagnostic catheters-one in the pulmonary artery, the other in the left ventricle-the radiologic arc was turned around the patients. The anteroposterior "zero point" was determined at the arc angle reproducing the usual anteroposterior catheter display (20" and 30" left anterior oblique, respectively). The septal puncture level was determined at this view (anteroposterior 20" (or equivalent), and then confirmed at 45" RAO 30") and at 90" 20" (or 30") for the lateral view. In all 4 patients, the needle reached the left atrium on the first attempt.
the left atrium. Nevertheless, results were similar in both groups. In fact, final mitral valve area depends more on mitral valve anatomy and on balloon diameter than on the balloon approach technique [ 13-15].
Lower Atrial Septum Puncture Advantages
It seemed to us that crossing the mitral valve was easier and faster in the second group with the lower puncture site (below the foramen ovale); unfortunately, the procedure was not timed for all patients. Moreover, mitral balloons seemed more stable during inflations in group 11. Usually, balloons are stable during the first inflation, but not during subsequent inflations, because balloons no longer have the back-up of the stenotic mitral valve. In group 11, the distance between the septal penetration site and the mitral valve was shorter (Fig. 3). The balloon, thus, had less propensity to slide back into
Five cases of tamponade after PMBV are reported in this study. We have two reasons to believe that these tamponades were not related to transseptal catheterization but were caused by left ventricular perforation during PMBV (4times in group I by the straight tip of the balloon catheter and once in group I1 by a stiffer guide wire). First, tamponades were always detected after valvotomy (when recording of hemodynamic data was performed to evaluate valvotomy effectiveness) long after transseptal catheterization was completed. Second, despite pericardiocentesis in the catheterization laboratory, hemopericardium recurred and a surgical myocardial suture was always necessary. The perforation was never located at the atrial level, but rather in the left ventricle apex, site of the highest left ventricular contact with the catheter tip.
Atrial septal puncture for transseptal catheterization is always guided correctly by prior three view fluoroscopic observation of the relationship between the Brockenbrough needle tip and the catheter placed in the aorta. Fluoroscopic guidance was effective, at both puncture levels: below or at the foramen ovalis. This guidance was
Rocha et al.
particularly helpful for patients with light scoliosis or in cases of vena cava distortion. REFERENCES 1. Cope C: Technique for transeptal catheterization of left atrium: Preliminary report. J Thorac Surg 37:482-486, 1959. 2. Ross JJ, Braunwald E, Morrow AG: Transseptal left atrial puncture, new technique for the measurement of left atrial pressure in man. Am J Cardiol 3:653-655, 1959. 3. Singleton RT, Scherlis L: Transseptal catheterization of the left heart: Observations in 56 patients. Am Heart J 60:879-885, 1960. 4. Braunwald E: Transeptal left heart catheterization. Circ 37 (SUPPI111):74-79, 1968. 5 . Adrouny ZA, Sutherland DW, Griswold HE, Ritzmann LW: Complications with transseptal left heart catheterization. Am Heart J 65:327-333, 1963. 6. Inoue K, Miyamoto N, Owaki T, Nakamura T, Kitamura F: Clinical application of transvenous mitral commissurotomy by a new balloon catheter. J Thorac Cardiovasc Surg 87:394-402, 1984. 7. Block P, Palacios I: Comparison of hemodynamic results of anterograde versus retrograde percutaneous balloon aortic valvuloplasty. Am J Cardiol 60:659-662, 1987. 8. Mullins CE: Transseptal left heart catheterisation: Experience
with a new technique in 520 pediatric and adult patients. Pediatr Cardiol 4:239-246, 1983. O’Keefe JH Jr, Vlietstra RE, Hanley PC, Seward JB: Revival of the transseptal approach for catheterization of the left atrium and ventricle. Mayo Clin Proc 60:790-795, 1985. Croft CH, Lipscomb K: Modified technique of transseptal left heart catheterization. J Am Coll Cardiol 5:904-910, 1985. Gorlin R, Gorlin SG: Hydraulic formula for calculation of the area of the stenotic mitral valve, other cardiac values, and central circulatory shunts. Am Heart J 41:l-25, 1951. Babic UU, Pejcic P, Djurisic Z, Vucinic M, Grujicic SM: Percutaneous transarterial balloon valvuloplasty for mitral valve stenosis. Am J Cardiol 57:llOl-1105, 1986. Wilkins G, Weyman AE, Abascal V, Block P, Palacios I: Percutaneous balloon dilatation of the mitral valve: An analysis of echocardiographic variables related to outcome and the mechanism of dilatation. Br Heart J 60:299-308, 1988. Vahanian A, Michel PL, Cornier B, Vitoux B, Michel X, Slama M, Sarano LE, Trabelsi S, Ben Ismail M, Acar J: Results of percutaneous mitral commissurotomy in 200 patients. Am J Cardiol 63:847-852, 1989. Henmann HC, Wilkins GT, Abascal VM, Weyman AE, Block PC, Palacios I: Percutaneous balloon mitral valvotomy for patients with mitral stenosis. Analysis of factors influencing early results. J Thor Cardiovasc Surg 96:33-38, 1989.