IMPACT OF THE PRESERVATION OF THE SUBVALVULAR APPARATUS OF THE ANTERIOR MITRAL LEAFLET ON THE LEFT VENTRICULAR FUNCTION AFTER MITRAL VALVE REPLACEMENT IN EARLY POST-OPERATIVE PERIOD

This research was conducted to evaluate the immediate and short-term improvements (6 months) in LV output following MVR with chordea tindinea preservation. Thirty patients who underwent MVR with complete or partial preservation of chordea tindinea were studied to obtain this objective, compared to other group of thirty patients with nearly same preoperative risks and severity of disease, without any type of subvalvular apparatus preservation of the AML. We excluded from the study patients with associated coronary heart disease requiring coronary artery bypass surgery, associated other valvular disease necessitating other valves replacement, Patients with infective endocarditis, and Redo heart surgery. Patients with mild functional tricuspid regurgitation were included in the study.

Mitral valve was inspected through a left atriotomy or transatrial incision.. A semi-elliptical-shaped piece of tissue was excised from the annulus of the anterior leaflet, leaving a 5--10-mm long rim of leaflet whose free edge remained attached to the primary and secondary chordae tendineae. The strip was detached only from the annulus at the anterolateral commissuresand reattached to the annulus beginning at the posteromedialcommissures in a counterclockwise fashion with mattresssutures that was also used for the valve replacement .Because the strip is not detached from the annulus at the posteromedialcommissures, no additional sutures were placed in the annulus. The strip was usually shorter than the annulus, soit was rotated in a posteromedial direction as it was sutured and,as a result, did not protrude into the left ventricular outflowtract.
When the leaflet was found to be thickened and calcified, it was divided into 2 to 5 chordal segments depending on the size of the vlavular leaflet. Each segment then was trimmed into chordal buttons and reattached to the annulus in an anatomic fashion. When the chordal buttons appeared to be excessive and couldn't be excised, a tonsil clamp was used to hold it on the atrial side of the annulus when the valve sutures were tied; this prevented it from protruding into the left ventricular outflow tract or interfering with prosthetic valve function.
The posterior leaflet, when pliable, was retained completely, together with the attached chordae tendineae. Redundant leaflet tissue was folded up into the annulus by placing the valve sutures through the annulus and bringing them through the leading edge of the leaflet tissue. Incisions or small wedge resections of the leaflet were performed if the posterior leaflet was thickened and fibrotic to allow implantation of a larger valve.
Data were statistically described in terms of range, mean  standard deviation ( SD), frequencies (number of cases) and relative frequencies (percentages) when appropriate. Comparison of quantitative variables over the study period was done using repeated measure analysis of variance (ANOVA) test with posthoc multiple 2-group comparisons. For comparing categorical data, Chi square ( 2 ) test was performed. Yates correction equation was used instead when the expected frequency is less than 5. A probability value (p value) less than 0.05 was considered statistically significant. All statistical calculations were done using computer programs Microsoft Excel version 7 (Microsoft Corporation, NY, USA) and SPSS (Statistical Package for the Social Science; SPSS Inc., Chicago, IL, USA) statistical program for Microsoft Windows.

Results:-Preoperative Assessment:
This study was conducted on 60 patients having rheumatic or degenerative mitral valve disease undergoing mitral valve replacement. All patients completed the study; there were no mortality among the patients.

Inotropic support: Group A:
Weaning of bypass was smooth and without inotropic support in 12 patients (40%), while in 18 patients (60%) a small dose of adrenaline (0.05 mic/kg/min) was used.

Postoperative Data:
All patients were transferred to the cardiothoracic ICU and mechanically ventilated.

Group A:
The mean mechanical ventilation time was 5.63 ±1.87 hours while the mean total stay in the ICU was 38.4 ± 10.63 hours.

Group B:
The mean mechanical ventilation time was 7.40±3.96 hours, (Fig. 6-4), while the mean total stay in the ICU was 44.62 ± 9.79 hours, (Table 7)

Postoperative complications: Bleeding:
Re-exploration for bleeding in the early postoperative hours was needed in 2 patients (6.7%) in each group.

Other Complications:
In group A 4 patients (13.3%) had postoperative complications while in group B, 3 patients (10%) had postoperative complications These complications include:

Rhythm disturbance:
In group A, one patient had a nodal rhythm on the first 7 hours post operatively, then rapid AF which was then controlled by medical treatment in the ICU.
In group B, two patients suffered from rhythm disturbances, the first patient had a complete heart block and he was paced by a temporary pacemaker for 6 hours then he returned back into normal sinus rhythm, the other patient was in nodal rhythm for 3 hours then returned back to AF.

Superficial wound infection:
This occurred in two patients in group A, and one patient in group B and they all responded to medical treatment.

Bleeding due to excessive anticoagulation:
This occurred to one patient in group A in the form of excessive bleeding from the nose, the INR was 5.5. The anticoagulation was stopped for 3 days, also 2 units of fresh frozen plasma was transfused for the patient. Then the INR was adjusted and accordingly the anticoagulation was resumed.

Postoperative echocardiography:
In order to determine the role of the prosthesis, echocardiography was performed prior to discharge around the 7th postoperative day and 6 months postoperatively; it was normal in all cases. In group A, the mean gradient over the mitral prothesis was 3.734 ± 0.3075 mm Hg and in group B, 3.931 ± 0.291. Dimensions, EF and PASP have been calculated and the preoperative values have been compared.
In group B, there was a higher EF in both early and 6 months postoperative echo than in group A, and this is statistically important (P value = 0.001 and 0.00013).
Mean LVEDD decreased in group A from 6.08 ± 1.3 preoperatively to 6.04 ± 1.08 postoperatively on the 7th day, but the difference was not statistically important, P = 0.42. Mean LVEDD continued to decrease significantly to 5.89 ± 1.12, 6 months post-operatively, with a statistically non-significant P value of 0.064. Mean LVESD decreased marginally from 4.19 ± 1.1 preoperatively to 4.02 ± 0.96 preoperatively in the early postoperative period. The decline was also mild at 6 months postoperatively, with a non-significant decrease in mean LVESD to 3.87 ± 0.63 (P=0.31).
Mean LVEDD decreased in group B from 5.97 ± 1.24 preoperatively to 5.79 ± 1.13 postoperatively on the 7th day, but the difference was not statistically important, P = 0.08. The mean LVEDD continued to decrease significantly to 5.48 ± 1.9, 6 months after surgery, with a statistically significant P value of 0.042. Mean LVESD decreased marginally from 4.34 ± 0.83 preoperatively to 4.11 ± 0.91 preoperatively in the early postoperative period. The decline was noticeable after 6 months postoperatively, with a substantial decrease in mean LVESD to 3.68 ± 0.96 (P=0.05).

Discussion:-
Rushmer and his peers (25) showed in 1956 that the papillary muscles play an important role in the contraction of the left ventricle. Lillehei and his colleagues found that preservation of the chordae of the posterior mitral leaflet resulted in a decrease in mortality from 37 to 14 percent in MVR patients, referring to Rushmer's findings in 1964.In patients with mitral valve replacement after MVR without protection of chordae tendineae, a substantial reduction in LVEF was noted by most investigators. After traditional MVR, the decrease in LVEF was caused by several variables, such as decreased preload, increased afterload,Contractile control or disabled. Prelaod is reduced by elimination of the regurgitant volume after MVR for mitral regurgitation, whereas afterload is increased by the disappearance of the low impedance ejection root into the left atrium. Interruption of ventricular valve interaction with the shift in loading status (32) is thought to be the key factor responsible for decreased ejection efficiency.
David and co-workers reintroduced Lillehei's methodology in 1981 and Hetzer and his colleagues (14) in 1983(5). In a larger series of patients, they documented the beneficial effects of chordal preservation on the clinical status of the patients with respect to the need for catecholamines and other clinical parameters.
In 1988 Miki and his colleagues (19) published a new technique that allowed not only preservation of the posterior, but also of the anterior mitral leaflet.
Many surgeons continue to retain only the posterior leaflet with chordae tendineae because of concerns over greater technical complexity, longer operating time , potential interference with mechanical leaflet motion, need to undersize the mitral prosthesis, and the possibility of creating left ventricular outflow tract obstruction (LVOTO) (17) .
To overcome the limitation of bileaflet preservation, a variety of techniques have been introduced.These differprimarily in the location where the anterior leaflet chordaeare inserted in the mitral annulus. The tension of the preservedmain anterior leaflet chordae may act on the posterior annulus(Feike's technique), the trigonal area (Miki's technique),on the anterior annulus (Khonsari's I and II technique),or on a point half-way between these locations (Hetzer'stechnique) (4) .
This research was conducted to evaluate the immediate and short-term improvements (6 months) in LV output following MVR with chordea tindinea preservation. Thirty patients who underwent MVR with complete or partial preservation of chordea tindinea were studied to obtain this objective, compared to other group of thirty patients with nearly same preoperative risks and severity of disease, without any type of subvalvular apparatus preservation of the AML. We excluded from the study patients with associated coronary heart disease requiring coronary artery bypass surgery, associated other valvular disease necessitating other valves replacement, Patients with infective endocarditis, and Redo heart surgery. Patients with mild functional tricuspid regurgitation were included in the study.
The mean age in our sample was 31.53±6.68 and 31.47±5.60 years respectively. In our study the mean age is lower than in other studies since rheumatic affection is evident in Egypt in this age group. In 1996(21), Natsuaki and associates recorded a mean age of 55±10 in their study. In 1994, Okita and his colleagues(22) recorded a mean age of 52.6 years, despite the fact that the valve disease was only of rheumatic origin in their study. A mean age of 35.6± 19.0 was reported by Chowdhury and his colleagues in 2005(4); their research was performed in India and the valve pathology was rheumatic mitral disease.
In our research, females contributed to 58 percent of patients, which indicates that there is more female love. In the 1994 study of Okita and his peers, (22) women contributed 76 percent.
In Egypt, we have several patients with mitral stenosis with highly calcified leaflets and extreme pulmonary hypertension (PASP above 90 mmHg) in very late stages. These patients were removed from our study. In 2010, Zakai SB(35) and colleagues found that an important finding between the groups was that of PA pressures and LA scale.In the conservation classes, there was a substantial improvement relative to the resection group, whereas there was no statistically significant difference in the improvement of pulmonary artery pressure in our sample. (28) reported that preservation of the mitral subvalvular structures resulted in a significant reduction of arrhythmias in the postoperative course.In our study (in group B) AF existed preoperatively in 13 patients (43%); three of them were converted to SR before discharge. New onset AF occurred in one of patients, in whom re-conversion to sinus rhythm was successful (4) had a higher preoperative incidence of AF; 62.8% in the non-chordal group, 62 % in the posterior chordal-preservation group and 72 % in the total preservation group. Postoperatively the incidence of AF was 40%, 46 %, 42.4 % with no statistically significant difference (p = 0.61).They concluded that that chordopapillary preservation techniques did not affect the outcome of postoperative atrial fibrillation.
In our research, a technique used to maintain annulopapillary continuity was determined by the state of the subvalvular apparatus. We used the technique of cutting a semi-elliptical piece of tissue from the anterior leaflet when the annulus was not extremely thickened, which was the case in 6 patients (20 per cent). In 24 patients (80 percent), when the anterior leaflet was severely thickened and calcified, Based on the size of the valve leaflet, it was divided into 2 to 5 chordal parts. Each section is then trimmed into chordal buttons and anatomically connected to the annulus. All the native chordal structures of the anterior mitral leaflet were resected in group A patients because the subvalvular apparatus was markedly sick, so only posterior mitral leaflet preservation was performed.
Compared to other studies Hennein and colleagues in 1990 (12) reported a cross clamp time of 45 ± 10 min in the conventional group and 47 ± 11 in the preservation group, while in our study the mean crossclamp time in group A was 54.20±7.43 minutes, and it was 65.20±9.71 in group B. Gaiotto and his colleagues in 2007 (8) reported cross clamp time of 46 ± 12 min, they used intermittent antegrade cardioplegia with coronary reperfusion to reduce the cross clamp time. The other reason for the shorter cross clamp time in the two previous studies is that they were conducted on patients with isolated mitral regurgitation of degenerative pathology. Ghosh and his colleagues in 1992 (10) reported a median cross clamp time of 68 min range (39-110) in the conventional group and 57 min  in the preservation group, While in our study ,in the preservation group, the mean crossclamp time was more (p< 0.05).
One argument against preservation of the chordae of the anterior leaflet was that only undersized valve prosthesis could be implanted. However,in group B,we have implanted 31mm in 3 patients (10%), 29 mm in 7 patients (23%), 27mm in 20 patients (67%), and 25 mm wasn't used in this group(while in the non preservation group it was used once). In our study the mean BSA was 1.5 ± 0.14 m 2 , thus implantation of 27 mm prosthesis was acceptable according to the investigations of Rowlatt and his colleagues in 1963 and King and his colleagues in 1985.
Furthermore several homodynamic evaluations of the St. Jude cardiac valve prosthesis have shown that there is no significant difference of pressure gradients between the 27 mm and 29 mm prosthesis at rest and on exercise (9,13) . This means that chordal preservation of the anterior mitral leaflet in MVR is not necessarily associated with implantation of undersized mitral valve prosthesis.
Comparing with other studies, Chowdhury and his colleagues implanted 29 to 33 mm prosthesis in 85.6 % of patients with total chordal preservation, confirming that preservation of the anterior leaflet doesn't preclude implantation of large prosthesis.
In our study weaning of bypass was smooth and without inotropic support in 2 patients (6.7%), while in 26 patients (86.7%) a small dose of adrenaline (0.05 mic/kg/min) was used, while in 2 patients (6.7%) larger dose of adrenaline (0.1 mic/kg/min)was used in the preservation group ,and, weaning of bypass was smooth and without inotropic support in 12 patients (40%), while in 18 patients (60%) a small dose of adrenaline (0.05 mic/kg/min) was used in the non preservation group. Chowdhury and his colleagues in 2005 reported that 75.8% of patients in the conventional group required postoperative inotropic support while 22.5% of patients in the preservation group needed inotropic support. This could be due to less bypass time in the preservation group, while in our study there was significant increase in bypass time in the preservation group more than the conventional group.
In our study there was no mortality. Okita and his colleagues in 1994 (22) Tarelli and his colleagues in 1994 (31) and Kayagioglu and his colleagues in 2003 (15) had also no mortalities in their study. Hennein and his colleagues in 1990 (12) had four perioperative deaths for the entire study population (6%), all the result of low-output cardiac failure and all occurring in patients in whom the chordae had been excised at operation. All late deaths occurred in the group of patients undergoing conventional replacement, occurring at an average of 37 ± 29 months postoperatively. (20) reported that there was a survival benefit for patients undergoing leaflet preservation, either partial or total. Survival in the combined leaflet group was 92% versus 80% for total excision at 5 years, p = 0.001.

Muthialu and his colleagues in 2005
Left ventricular ejection fraction (LVEF) usually falls after conventional MVR for chronic MR. the falls has been explained by the increased afterload produced by a competent mitral valve (3,23) . Several reports indicate that EF does not change significantly when chronic MR is corrected by mitral repair techniques.These observation suggest that left ventricular function may be enhanced when the mitral valve is not excised during correction of MR (6,7) .
In our study the EF has declined in both groups at 7 th day postoperative, in group A, EF declined from 61.8 ± 7.9 to 43.6 ± 5.9, and, in group B, EF declined from 60.4 ± 5.9 to 48.1 ± 4.2 .Then EF improved with time to be 58.4 ± 6.1 in group A, and significantly improved in group B to be 64.53 ± 5.2, at 6 months postoperatively.
Straub and his colleagues in 1996 (28) reported that in the chordal preservation group the EF remained almost unchanged 7 days after operation (from 44.4 ± 14.0 % to 42.7±8.7%) and increased to 54.2 ± 11.2 % 3 months postoperatively (P ≤ 0.05). In contrast, the EF in the conventional group decreased from 40.2 ± 12.7% preoperative to 32.7± 8.4 % 7 days postoperatively (P≤0.05) and recovered after 3 months postoperatively to 48.1 ± 12.4 %. (8) performed MVR with preservation of the chordea tindinea in patients with end stage dilated cardiomyopathy. They reported an improvement in LVEF (P =0.008) at the 3 rd postoperative month. (4) reported that the EF continued to decline in the nonchordal group and returned to preoperative levels after initial decline in the chordal preservation group. At the end of 4 years, the fractional change of the EF was statistically significant only in the chordal preservation group. (1) reported that the EF, in his patients who underwent MVR with preservation of the chordea tindinea, was maintained without deterioration in the immediate postoperative period.

Alsaddique in 2007
In the perioperaitve stage LVEDD has been shown to be the most reliable index of LV function. A reduction in LVEDD has been found uniformly to correlate well with the level of clinical improvement after valve surgery (24,29) .
The mean LVEDD decreased in our sample in group A from 6.08 ± 1.3 preoperatively to 6.04 ± 1.08 postoperatively at day 7, but the difference was not statistically important, P = 0.42. Mean LVEDD continued to decrease significantly to 5.89 ± 1.12, 6 months post-operatively, with a statistically non-significant P value of 0.064. Mean LVESD decreased marginally from 4.19 ± 1.1 preoperatively to 4.02 ± 0.96 preoperatively in the early postoperative period. As the mean LVESD was not substantially reduced to 3.87 ± 0.63 (P=0.31), the reduction was also mild at 6 months postoperatively. Mean LVEDD decreased in group B from 5.97 ± 1.24 preoperatively to 5.79 ± 1.13 postoperatively on the 7th day, but the difference was not statistically important, P = 0.08. The mean LVEDD continued to decrease significantly to 5.48 ± 1.9, 6 months after surgery, with a statistically significant P value of 0.042. Mean LVESD decreased marginally from 4.34 ± 0.83 preoperatively to 4.11 ± 0.91 preoperatively in the early postoperative period. The decline was noticeable after 6 months postoperatively, with a substantial decrease in mean LVESD to 3.68 ± 0.96 (P=0.05). (15) reported that the LVEDD and LVESD decrease in the preservation group and increased in the conventional group postoperatively but the changes were statistically insignificant. EF decrease slightly postoperative in patients with preserved chordae, however it decreased significantly in patients with conventional MVR. (8) reported a reduction in the LVEDD (p = 0.038) and LVESD (p = 0.008) in patients with end stage cardiomyopathy who underwent MVR with preservation of the chordea tindinea. (4) reported that the left ventricular end-systolic volume (LVESV) decreased slightly from the preoperative level in the total excision group in the immediate postoperative period. Although there was gradual improvement at 1 to 4 years of follow up, the improvement was not statistically significant. The remaining groups (the posterior preservation and the total preservation) demonstrated statistically significant reduction in LVESV in the immediate as well as the late postoperative period. The total chordal group demonstrated greater fractional change as compared to the posterior chordal and the non chordal group. The left ventricular enddiastolic volume (LVEDV) decreased by comparable degrees in all three groups in the immediate postoperative period and on follow up. Only the total chordal group had statistically significant percentage reduction of LVEDV at 4 years of follow up.

Chowdhury and his colleagues in 2005
Fractional shortening was not universally used by most authors. Muthialu and his colleagues in 2005 (20) reported that the FS was 34% preoperatively vs. 26% postoperatively in the conventional group, and 31 % vs. 29 % in the preservation group, p=0.06.

Conclusion:-
We concluded that total chordal preservation is possible in the majority of patients undergoing MVR for rheumatic heart disease, as this study showed significant improvement in the LV function in the patients in which the AML was preserved.