The American




@ 37 2


Early Risks of Open Heart Surgery for Mitral Valve Disease


From the Department of Surgery, The University of Alabama Medical School and Medical Center, and the Veterans Administration Hospital, Birmingham, Ala. This study was supported in part by Program Project Grant HL 11,310 from the National Heart and Lung Institute, National Institutes of Health, Bethesda, Md. Manuscript accepted May 21, 1975. Address for reprints: John W. Kirklin, MD, Department of Surgery, University Station, Birmingham, Ala. 35294.

During 1972 and 1973, a total of 235 patients had open heart surgery for mitral valve disease unassociated with significant aortic or rheumatic tricuspid valve disease. Thirty-one underwent closed heart mitral commissurotomy, without mortality. Of the 204 patients undergoing open operation, 125 had sequential measurement of cardiac output and mixed venous oxygen pressure. The hospital mortality rate was 6.4 percent in the larger group of 204 patients and in the 125 with cardiac output measurements. The rate was greater in those with class IV disability (New York Heart Association criteria) preoperatively than in those with class Ill or II disability. The mean f standard deviation of the average cardiac index early postoperatively was 2.05 f 0.579 litersmin-‘em*. Cardiac index was lower in the patients who died early postoperatively than in those who did not. The probability of hospital death was a significant function of cardiac index. The predicted probability of death was 10 percent with an average cardiac index of and increased sharply with lower indexes. Car1.42 litersmin-‘*m-* diac index was lower early postoperatively than preoperatively, and was lower in patients in class IV than in those in class Ill. There was no significant difference in cardiac index between patients with mitral valve replacement and those with repair. A history of closed commissurotomy, age, duration of cardiopulmonary bypass, duration of cardiac ischemia and method of myocardial preservation did not significantly influence cardiac index or hospital mortality rate. There was no significant relation between mixed venous oxygen pressure and hospital death. Further improvement in results of mitral valve surgery requires adequate preservation of left ventricular performance before, during and after operation.

The results of mitral valve surgery are described by the long-term survival rate after operation and the functional result in surviving patients. Long-term survival rates are significantly affected by the hospital mortality rates. We have reviewed our recent experience to determine the current early risks of mitral valve surgery. We have stressed the early postoperative hemodynamic state since most deaths were found to be a result of inadequate cardiac performance.

February 1976

The American Journal of CARDIOLOGY

Volume 37






TABLE Operative


Lesion and Mortality

in 204 Patients with Open .____

Patients With MS Total no.

no. Who Died __

__..~_. Commissurotomy Repair Mitral valve replacement Mitral and tricuspid valve replacement Total MI

= mitral




0(&i 5(7.2%)


stenosis; TI

5(6.5%) = tricuspid

and Methods



Functional Class (NYHA) __~ II III IV Total

Class and Operative



in 204 Patients Total Hospital Deaths






7 152 45 204

3.4 74.5 22.1 100

0 2 11 13

0 1.3 24.4 6.4

Heart Association




= New York

February 1976

Total no.

:: : 55

The American Journal of CARDIOLOGY



Disease (1972-1973)

Patients With MS+MI+TI

no. Who Died

Total no.


::: . 8 8




for Mitral



Case material: During the years 1972 and 1973, a total of 235 patients were operated on at the University of Alabama Hospital in Birmingham for mitral valve disease unassociated with significant aortic or rheumatic tricuspid valve disease. Patients with tricuspid incompetence considered to be secondary to the hemodynamic effects of their mitral valve disease were not excluded. Two hundred four patients had open heart surgery, and 31 underwent closed mitra1 commissurotomy. There were no deaths among the patients with closed commissurotomy, and we will not discuss this group further. Among the 204 patients having an open operation, there were 123 women and 81 men with an average age of 49 years (range 8 to 78). Sixty-four patients had mitral stenosis, 77 mitral incompetence and 55 combined mitral stenosis and incompetence. Eight patients had severe secondary tricuspid incompetence. One hundred forty-nine patients had rheumatic valvular disease, 31 had rupture of chordae tendineae, 17 had “floppy” valve and 7 had valvular lesions secondary to bacterial endocarditis as determined by the history and examination of the valve at the time of operation. Thirty-five patients had one or more previous mitral commissurotomies. Preoperatively, 7 patients (3.4 percent) were in functional class II (New York Heart Association classification), 152 (74.5 percent) were in class III and 45 (22.1 percent) in class IV. Among the I73 patients who had valve replacement, 113 had a Starr-Edwards composite seat prosthesis, 44 a Braunwald-Cutter prosthesis and 16 a Bjork-Shiley prosthesis (Table I). One hundred twenty-five of the 204 patients had sequential measurements of cardiac output and mixed venous oxygen pressure in the first 24 to 72 hours postoperatively. Thirty-one of these patients had mitral stenosis, 53 mitral incompetence and 37 combined mitral stenosis and incompetence. Four of the patients in this group had severe secondary tricuspid incompetence. The distribution of the





MS = mitral




no. Who Died


1 KG.%)

. .





Patients With MS+MI

Patients With MI Total no.

Heart Operations

severe enough

Total no. Who Died




no. Who Died O(O%)




165 8

10(6.1%) 3(37.5%)




to warrant


on tricuspid


type of valvular lesion did not differ significantly in these 125 patients from that in the total group of 204 (P = 0.6) or in the remaining 79 (P = 0.06). Eighty-nine patients had rheumatic valvular disease, 21 had rupture of chordae tendineae, 9 had a “floppy” valve without evidence of ruptured chordae and 6 had lesions caused by bacterial endocard&is. Six patients (4.8 percent) were in functional class II preoperatively, 91 (72.8 percent) in class III and 28 (22.4 percent) in class IV. Twenty-four of these patients had a closed mitral commissurotomy 4 to 15 years before operation, and one had plication of the posterior leaflet of the mitral valve 1 year preoperatively. Sixty-two of these 125 patients had cardiac catheterization before operation. Systolic pulmonary arterial pressure was between 50 and 70 mm Hg in 16 of these and 70 mm Hg or greater in 17. Operative


During the 2 year period, two types

of myocardial preservation were used. In 90 patients the operation was performed with profound cooling to a myocardial temperature of about 22’C by the perfusate and one period of ischemic arrest produced by clamping the aorta; the body temperature was 28’ C. The total cardiopulmonary bypass time ranged from 33 to 95 minutes (mean 56). The duration of ischemic arrest ranged from 23 to 45 minutes (mean 32). In 35 patients the operation was performed with moderate cardiac hypothermia (32” C) and intermittent ischemic arrest for two to three periods of approximately 15 minutes separated by 3 minute periods of coronary perfusion. The ventricles were kept beating (not fibrillating) during the periods of coronary perfusion. Cardiopulmonary bypass time ranged from 45 to 111 minutes (mean 65), and the total ischemic time ranged from 25 to 57 minutes (mean 36). Ten of the 125 patients had mitral commissurotomy, 1 mitral plication and anuloplasty, 110 mitral valve replacement and 4 tricuspid valve replacement in addition to replacement of the mitral valve. Among the 114 patients who had valve replacement, a Starr-Edwards composite seat prosthesis was used in 82 patients, a Braunwald-Cutter prosthesis in 22 and a Bjork-Shiley prosthesis in 10. Postoperative protocol: Postoperatively, left atrial, right atria1 and radial arterial pressures were routinely measured. In some patients pulmonary arterial pressure was also measured. Blood was infused to keep the mean left atria1 pressure at about 14 mm Hg (pressure in the right atrium was usually less than in the left). All patients with systemic arterial hypertension early postoperatively were given drugs to reduce arterial blood pressure, Patients with tachyarrhythmias were treated appropriately. Fifty patients received digoxin early postoperatively to treat a rapid ventricular response to atria1 fibrillation. Sixty-two patients received catecholamine support, generally for “low

Volume 37


TABLE Operative


III Procedure

and Mortality

in 125 Patients

Early Postoperative

Patients With MI

Patients With MS

Repair Commissurotomy Mitral valve replacement Mitral and tricuspid valve replacement Total


Total no.

no. Who Died

Total no.

10 21’

O(O%) 1(4.80/o)

.1 52

0(&i 4(7.7%)



. 31

Ml = mitral incompetence;



no. Who Died


MS = mitral stenosis: TI = tricuspid


Patients With MS+MI+TI

Patients With MS+MI Total no.

no. Who Died

Total no.

::: 37

::: 2(5.4%)





no. Who Died

.. .

Total no. Who Died


101 110







z1 00 7(6.4%) 1(25.0%) 8(6.4%)


cardiac output.” Eighty percent of the patients had assisted ventilation with a volume-controlled ventilator for at least 20 hours after operation. Hemodynamic measurements: Cardiac output was determined by indicator-dilution technique, using in each instance the mean of duplicate or triplicate measurements. One milliliter of indicator solution (5 mg) was injected rapidly by hand through the catheter into the right atrium. Blood was simultaneously withdrawn at a rate of 7.64 ml/ min from the pulmonary arterial catheter through a Waters cuvette densitometer (no. G250) by a constant-rate pump (model 901, Harvard Apparatus). Five to 7 ml of blood was withdrawn for each indicator-dilution curve, and the blood was not reinfused. Cardiac output calculations were performed on a digital computer using the adaptation of the Stewart-Hamilton formula devised by Sekelj et a1.l Cardiac output data were normalized by body surface area and expressed as cardiac index. Mixed venous blood samples were obtained from the pulmonary arterial or right atria1 catheter in heparinized syringes and analyzed on automated assemblies (models 313 and 182, Instrumentation Laboratories, Inc.) for pH, partial pressure of oxygen (POs) and hemoglobin and oxygen saturation. The first cardiac output measurement was obtained within 3 hours of the patient’s arrival in the cardiac intensive care unit and thereafter usually every 4 to 6 hours for the first 24 to 72 hours, depending on the patient’s condition. In a few patients only a single measurement was obtained, generally because of a highly satisfactory hemodynamic state. Statistical analysis: This study was a retrospective one, made by reviewing in detail the hospital records of each patient. The hemodynamic data were analyzed for means and standard deviations. The relation between the preoperative cardiac index (obtained at cardiac catheterization by the Fick method) and the average postoperative cardiac index was determined by regression analysis. A paired t test was also employed to yield the average difference and its change between pre- and postoperative measurements. The interrelation of simultaneously measured cardiac index and mixed venous oxygen pressure was determined by weighted least squares regression in which the number of observations taken was used to weight the average overall observations in a given patient. Probability (P) of hospital death in relation to average postoperative cardiac index (CI) was generated by nonlinear weighted least squares fitting of the logistic model*: P(DeatqC1)


= l/(1 + eZ)

where z = -a - b(ln CI); e = base of natural logarithm; a = intercept; and b = regression coefficient on the natural logarithm (In) of CI.

The average value of all postoperative measurements of cardiac index for each patient was used and each value weighted by the factor [P.( 1 - P)]-‘. Test of the regression coefficient b for significance was performed using the t test for which the ratio of b to its standard deviation was used as t. A chi-square test for lack of fit of the model was performed. Factors influencing the level of postoperative cardiac output were investigated by multiple stepwise linear regression.3 Risk factors influencing hospital mortality were studied by multiple nonlinear regression to the logistic model presented for cardiac index.

Results Hospital mortality: The hospital mortality rate in the total group of 204 patients was 6.4 percent (Table I). This rate was significantly higher in patients with class IV disability than in those with class III disability (P

Early risks of open heart surgery for mitral valve disease.

During 1972 and 1973, a total of 235 patients had open heart surgery for mitral valve disease unassociated with significant aortic or rheumatic tricus...
998KB Sizes 0 Downloads 0 Views