circulation 1959 ellis 803 20

A Clinical Study of 1,000 Consecutive Casesof Mitral Stenosis Two to Nine Years after

Mitral ValvuloplastyBy LAURENCE B. ELLIS, M.D., DWIGHT E. HARKEN, M.D., AND

HARRISON BLACK, M.D.

A study is presented of 1,000 cases of predominant mitral stenosis operated by valvulo-plasty between 1949 and 1956. It is shown that the survival of these patients is betterthan would have been expected under medical management. Sixty-nine per cent of thesurvivors of the operation in groups II and III improved, and 55 per cent in group IV.Factors influencing the late results are discussed. After substantial improvement lastinga year or more, 228 of this series deteriorated; the factors affecting this deterioration are

discussed, of which mitral insufficiency,rheumatic fever are the most striking.

T HE present study is a report of theclinical results in 1,000 consecutive cases

with a preoperative diagnosis of predominantmitral stenosis on whom mitral valvuloplastywas performed between the years 1949 andMarch 1956. Ninety-two per cent of the opera-tions were performed by D.E.H. and the re-mainder by H.B. The follow-up of all ofthese patients has been carried out under thedirection of the cardiologist, L.B.E., who hasnad the final word in the preoperative classifi-cation of the patients and in the estimate ofthe degree of improvement that they have had.In 150 (table 1) mitral insufficiency of somedegree was suspected before operation; 121showed evidence of aortic stenosis or insuffi-ciency but this was not thought to be clinicallyimportant, and there were 6 who were believedto have tricuspid stenosis. No attempt wasmade to classify patients with tricuspid insuf-

From the Thorndike Memorial Laboratory, Secondand Fourth (Harvard) Medical Services, Boston CityHospital, the Surgical Service of the Peter BentBrigham Hospital, and the Departments of Medi-cine and Surgery, Harvard Medical School, Boston,Mass.

This material was presented in part at the ThirdWorld Congress of Cardiology, Brussels, September1958.The study was supported by grant no. 442 of the

National Heart Institute, U. S. Public Health Serv-ice. Physical facilities were generously provided bythe Boston Mutual Life Insurance Company.

an inadequate valvuloplasty, and recurrent

ficiency, which was presumably due in mostcases to functional dilatation of the annulus inpatients with a failing right ventricle. Patientsin whom the preoperative diagnosis of signifi-cant mitral stenosis was in doubt or who hadvery substantial amounts of associated valvu-lar disease, in whom exploratory cardiotomieswere carried out, are not included in thepresent series and will be the subject of aseparate report. No patient has been droppedfrom the series, even if the operative findingsdid not confirm tile preoperative diagnosis.The technic of operation has been described

elsewhere.1 2 The basic principles of the valvu-loplastic procedure have remained unchangedthroughout the series, although with increas-ing experience the correction of stenosis hasundoubtedly been improved. An initial groupof 11 patients operated on between the yearsof 1947 and 1949 by technics not strictly com-parable to the present one have not been in-cluded.The results in the first 500 patients in this

series followed for a shorter period of timeand a preliminary statement of some of theresults of the entire group of 1,000 have beenpreviously published.3-6The classification employed' roughly corre-

sponds to the functional classification of theNew York Heart Association, but it is de-signed to represent a more dynamic picture

Circulation, Volume XIX, June 1959803

by guest on May 24, 2015http://circ.ahajournals.org/Downloaded from

http://circ.ahajournals.org/

ELLIS, HARKEN, BLACK

of the course of the patient's illness. Thisseries included no patients in group I, that ispatients without significant symptoms. Nine-teen were in group II; these were patientssomewhat handicapped by symptoms fromtheir disease but who were able to carry ona sedentary occupation successfully and inwhom the condition was not progressive. In4 of these patients who did not have severesymptoms of cardiac disability the primaryindication for operation was one or mo e pe-ripheral emboli. Group III included patientssuffering mainly from pulmonary symptoms,which were usually progressive in nature, andwere sufficiently handicapped, so that ordinaryactivities were significantly limited. If thesepatients had had overt congestive failure, ithad been because of some unusual precipi-tating cause such as pregnancy or a seriousinfection and they had rapidly recovered fromit. There were 711 in this group. Group IVcomprised cardiac patients with more severesymptoms who were mostly cardiac invalidssuffering from chronic congestive failure orwho were maintained in a precarious state ofcompensation only by vigorous medical meansThere were 270 in this group.

In 58 of the patients included in this series,a second operation was performed by us formitral stenosis and in 1 patient 3 operationswere done. We originally operated on 49 ofthese patients, and in 10 the operation hadbeen previously performed elsewhere. In thesepatients the clinical evidence prior to the see-ond operation pointed toward significant mi-tral stenosis and the operation was performedwith a view toward carrying out a mitralvalvuloplasty. Five patients in this series havehad a second operation for mitral stenosisperformed by other surgeons. Five have beenreoperated on by us for mitral insufficiency,and in a an exploratory eardiotomy has beenperformed. In 2 a second operation for the cor-rection of aortic stenosis has been carried out.This series includes 17 patients who have beencounted twice and 1 who was counted 3 times.Hence, the number 1,000 refers to the num-ber of operations rather than the number of

patients. In the remaining 41 who were reoper-ated on, only 1 operation is included in thisseries, the other beincg done either before orafter this series of 1,000 or was carried outelsewhere.Twenty-one patients were operated on while

they were pregnant and there were 3 deathsamong them. Although this represents a spe-cial group, in a consideration of operativemortality, pregnancy does not alter the lateresults. In the follow-up studies these patientshave been included with the others. The placeof mitral surgery during pregnancy has beendiscussed elsewhere.7Females outnumbered nmales in this series

about 3 to 1, 78 per cent of the patients ingroup III being women and 74 per cent ofthe group IV patients. The average age ofthe patients in the series was 39.4 years(38.8 years for women and 40.5 years formen) (table 3).

Accuracy of tihe Preoperative Diagnosis ofMitral Stenosis awnd Miitral IisifficiencyJudged by Finidinigs at Operation

A preoperative diagnosis of mitral insuffi-cieney complicating mitral stenosis was notmade unless it was considered to be clinicallysignificant. This would correspomi(l to the find-ing of a moderate or marked mitral insuffi-ciency by the surgeon at operation. We have,therefore, compared the preoperative diagnosiswith the operative finding at operation. Whenithe preoperative diagnosis was pure mitralstenosis the findings were confirmed at opera-tion in the group III patients in 84 per centof those in atrial fibrillation and 94 per centof patients in normal sinus rhythm, and illgroup IV patients in 83 and 75 per cent re-spectively. However, if a preoperative diag-nosis of mitral stenosis and insufficiency wasmade, in only 56 per cent of the patients inboth groups was the diagnosis of moderateto severe insufficiency confirmed at operation.If the situation is considered from the oppositepoint of view, the findings are similar. Whenpure mitral stenosis was found at operationthe lesion had been correctly diagnosed pre-

80)4



CLINICAL STUDY OF MITRAL STENOSIS

TABLE 1.-Preoperative Diagnosis

Group IIand III Group

Without associatedheart lesions 556 73 186

With aorticvalvular disease 77 19 21

With tricuspidstenosis 1 1 0

With coronarydisease 3 0 6

Totals 637 93 213

operatively in 94 per cent of thepatients and 84 per cent in thepatients, whereas when significantsufficiency was found by the surgeoi

tion a correct diagnosis had been xto operation in 42 per cent of groi

52 per cent of group IV patients.One of the patients who had be

in group III preoperatively had a

appeared normal at operation. In alIII patients were found to have2.0 cm.2 or more, prior to valvulithese 6 had some degree of mitral in.In 3 of these 15 the chief indicaticgery was a history of emboli. In 2:preoperative valve size was fromcm.2; and 18 of these had mitral in:Two patients in group IV he

mitral valves; 1 had a large atria]overlying the valve, and in the otheifailure was presumably due toatherosclerosis. Two other patientsIV had valve orifices of 2.0 cm.2 or

in 3 the preoperative valve size wa

to 1.9 cm.2 In all 5 of these patienttial mitral insufficiency was present

OPERATIVE MORTALITYThe over-all operative mortalit'

group has been previously reportgiven in table 2. Because of the

Iv

c. Q

ba

N . Total

48 863

TABLE 2.-Effect of Experience on OperationMortality

Serial Groups II and III Group IVno. of Mortality Mortality

patients Number (%) Number (%)

1-100 59 14 41 32101-500 296 4 104 24501-1000 375 0.8 125 19

4 121 group-II patients

these have been included with group-III pa-

4 6 tients in the subsequent analyses, and when-

ever the expression "group III" is used, it1 10 denotes group II and III patients. There was

57 1000 1 operative death in the group-II patients. Forthe purpose of this analysis "operative mortal-ity" denotes death during the operation or

group III during the period of the hospitalization whengroup IV the operation was performed.mitral in- The operative mortality according to age

n at opera- is given in table 3, and it will be seen thatnade prior there is no significant change in operativelp III and mortality depending upon age. It is a matter

of interest that none of the 42 patients over

.en classed 50 in group III died an operative death.valve that The operative mortality for males in group

1, 15 group III was 2.5 per cent and for females it was

valves of 3.3 per cent. Among the group-IV patients)plasty; of the mortality was somewhat higher in thesufficiency. male group, being 29.6 per cent as compared ton for sur- 21.6 in the females but this difference is not5 more the significant since the p valve is between 0.051.5 to 1.9 and 0.10.* These figures are for the entiresufficiency. 1,000 cases. Actually, the operative mortalityLd normal has fallen strikingly in the second 500 cases,

I thrombus particularly in group-Ill patients (table 2).r the heart

PERIPHERAL EMBOLIZATIONcoronary

oin group One hundred eighty-six of the 1,000 patientsmore, and had had one or more well-documented attacks

Ls from 1.5 of peripheral embolization prior to surgeryts substan- The time of onset of these episodes varied

from many years before surgery to a few days.The risk of developing an operative embolusin the last 500 patients operated upon was

y for thisbed and isvery small

*Unless otherwise noted, p values shown here andin the ensuing discussion are based on a contingencyX2 test.' Values of p less than .05 are significant.

805




TABLE 3.-Operative Mortality by Age

Groups II and III Group IVAge by No. of Mortality No. of Mortalitydecades operations (%) operations (%)

10-19 6 0.020-29 120 3.3 7 28.630-39 290 3.4 55 23.6

40-49 272 3.3 118 20.2

50-59 40 0.0 79 29.260-69 2 0.0 11 18.2

Total 730 3.1 270 23.6

2.1 per cent in group III and 8.0 per cent ingroup IV (table 4). The higher rate of op-erative embolization in the first 500 patientshas been discussed elsewhere.3 The figures forthe second 500 reflect more accurately ourcurrent experience. Only 1 of the 8 emboliin group III was fatal, but 8 of the 10 ingroup IV had a fatal outcome. Although thenumber of emboli occurring in fibrillating pa-tients was higher than in patients with normalrhythm, the differences are not statisticallysignificant. The frequency of embolization inpatients with normal rhythm is of interest.Of the entire group of 913 patients who

survived operation 25 have developed one ormore peripheral emboli after the operativeperiod to July 1, 1958. Most of the patientswho developed late peripheral emboli werefibrillating at the time of surgery and werepresumably fibrillating at the time of embo-lization. A few who were in normal rhythm atthe time of surgery were known to have de-veloped fibrillation prior to the occurrence oftheir late emboli. No relationship, however,could be ascertained between the occurrenceof late emboli and emboli occurring at opera-tion or a history of embolization occurringprior to surgery. The average duration of timesince operation in this group is almost exactly4 years and this therefore represents an ex-perience of about 3,600 patient years or an em-bolization rate of 0.7 per cent per year in thisgroup. More than half of the surviving pa-tients were in chronic atrial fibrillation. It is

TABLE 4.-Operative Embolization in Relation toRhythm in the Second Five Hundred Patients of theSeries

Groups II and III Group IV

~~~~ u~~~~~~~~~~ 0.4Occurrence of o o g °

operative emboli Z a z 0 P , Z zJ, P, T

Total operativeemboliNormal sinus

rhythm 211 4 1.9 27 1 3.7

Atrialfibrillation 164 4 2.4 98 9 9.1

Total 375 8 2.1 125 10 8.0

our opinion that valvuloplasty confers sub-stantial protection against peripheral emboli-zation in patients of this type.

SURVIVALIn spite of the great number of studies that

have been made over the years on the survivalof patients with mitral stenosis it is extremelydifficult to obtain data on medically treatedpatients that are comparable to this series.Many studies are statistically invalid or aremade on groups that are not easily compara-ble, or consider only the survival period ofthose who ultimately were known to be dead.Most studies based on autopsy statistics areretrospective, so that it is almost impossibleto determine when and to what degree thepatients became symptomatic. A recent studyhas been reported by Wilson and Lim9 onthe survival of patients followed into thethird, fourth, and fifth decades of life fromthe onset of rheumatic fever in childhood. Al-though this is an important study, it is diffi-cult to compare it with our own. Elsewherewe have commented5 on the studies made byGrant,10 Wilson and Greenwood,12 and Hamil-ton and Thompson."1 Vedoya, Nessi, and Men-delzon13 have also reported the ominous prog-nosis of patients with symptomatic mitralstenosis. Recently, Rowe et al.14 found that 40per cent of 250 patients with mitral stenosisfollowed 10 years were dead, and Donzelot et

806




al.15 have published results that have indi-cated that surgically treated patients have abetter prognosis than those medically fol-lowed. So far as these studies are comparableto ours, they indicate that survival undermedical therapy is less than for our surgicallytreated patients.As we have previously indicated,5 the series

most comparable to ours which has been medi-cally treated and followed is that by Olesen,16who studied a group of patients first observedbetween the years of 1933 and 1949 inCopenhagen. In this series 72 per cent of thepatients were females as compared to 77 pgercent in our group and the average age of hispatients was 41.5 years as compared to 39.4years in our series. He classified his patientsaccording to the grouping of the AmericanHeart Association and also subdivided theclass-2 and class-3 patients into those who werefibrillating and those in normal sinus rhythm.Our group-III patients would most closelyapproximate his class-3 patients and class-2patients with atrial fibrillation, and ourgroup-IV patients are comparable to his class-4 patients. We have therefore utilized his datato compare our group-III patients with hismale and female patients in class 3 and class2 in atrial fibrillation. Figures 1 and 2 showthe survival of our patients compare toOlesen's groups. Patients who have bee re-operated on and are counted twice in the seriesare counted only once in the calculation ofthe survival curves.* It will be seen that, in-cluding an operative mortality of 3 per cent,83 per cent of our group-Ill patients havesurvived over the period of observation up to7 years, and 71 per cent to 9 years, althoughthe numbers dealt with in the last 2 years aresmall. This survival is better than for the med-ically treated patients. In group IV 57 percent have survived up to 9 years, which in-clude an operative mortality of 24 per centfor the group as a whole. This survival isvastly better than for the medically treated

*The survival curves were calculated according tothe method of Berkson et al.`7

GROUPS AND E

EARSFOLLU 2 3 4 5 6 7 aOF ac6e6

PM7 FOLLOED 16 692 661 664 566 403 240 *04 24 2

GROUP X

A

WEARS FOLL D 2 3 4

no. OF OPERATEDm. FOLLO~ 265 202 192 172 141

5 6 7 9

100 68 29 14 3

FIG. 1 Top. Survival rates. A comparison of oper-

ated with medically treated patients.FIG. 2 Bottom. Survival rates. A comparison of

operated with medically treated patients.

patients of whom none was alive at 8 years.

The survival rates have also been analyzedaccording to age, sex, and rhythm. Total sur-

vival curves for each of these classes are notshown in detail. The survival rates of patientsat the end of 5 years are shown in table 5.The survival of female patients is somewhat

better than of males; those in normal rhythmdo better than fibrillating patients; andyounger patients in group III survive longerthan do those who are older. The only one

of the differences, however, which is statisti-cally significant is the sex difference in groupIII (p < .01).

LATE DEATHS

Ninety-five patients have died since opera-

tion. The causes of the deaths are given intable 6. Eleven died of conditions clearly un-

related to their heart disease. Two of these,however, were patients in which death mightbe considered related to the operative pro-cedure, including 1 death 21/2 months follow-ing surgery from hepatitis, which might havebeen homologous serum jaundice acquiredfrom a transfusion at the time of surgery, and

807




TABLE 5.-Survival Rates of Patients at the End ofFive Years in Relation to Age, Sex, and Rhythm

Groups II and III Group IV

e W.:. ce c.-

SexMale 91Female 312

RhythmNormal sinus 242Atrial fibrillation 161

Age by decades10-19 420-29 7330-39 17640-49 13450-59 1560-69 1

8090

9085

8395918764

100

2679

2679

32849214

5463

6560

43

60656054

1 death from a reaction occurring during thecourse of an intercostal block for treatmentof residual intercostal pain. There were 4sudden deaths that have been assumed to becardiac in origin. Patients developing cerebralvascular accidents, whether fatal or not, fol-lowing surgery have been considered to havehad these on the basis of emboli dislodgedfrom the heart, though some of these may

have been due to independent vascular diseaseof the brain. In the calculation of the survivalcurves all deaths have been included, whetheror not they were of cardiac origin.

FOLLOW-UP STUDIESFollow-up ProcedureNine hundred thirteen patients survived op-

eration. All but 2 have been followed for atleast 1 year or more, up to July 1, 1958, atwhich point the analyses were made. Of thepatients followed for at least 1 year, 1 has beenlost to follow-up for 5 years, 3 for 4 years, 1for 3 years, and 8 for 2 years. Ninety-threehave been delinquent in follow-up less than 2years, but most of these patients are merelyoverdue in their annual follow-up question-

TABLE 6.--Causes of Late Deaths

Diagnosis Number of deaths

Primarily cardiac 76

Heart failure 60

Sudden 4

Peripheral emboli 9

Pulmonary infarcts 2Pneumonia and heart failure 1

Noncardiac ] 1

Unknown 8

Total 95

naire by 1 or 2 months. In addition the 48 pa-tients who have had a second cardiac operationhave been dropped from the series as unim-proved at that point and hence are technically"lost to follow-up. " All of the remainder havebeen followed up to the latest anniversary oftheir operation. All patients have been fol-lowed by annual questionnaires so worded asto obtain not only the subjective opinion of thepatient concerning his improvement but toprovide evidence from which an estimate canbe made as to whether the patient is capable ofcarrying on with less disability than he hadprior to operation. In addition all other obtain-able data concerning these patients have beenutilized. These include personal examinationsof a large number of these patients, lettersfrom doctors, hospital reports, and so forth.In a previously reported study'8 a comparisonhas been made of the accuracy of method offollow-up with a follow-up examination madepersonally by one of us on a sample of 101patients from this group. It was found thatthe grading of patients by the questionnairemethod was stricter and somewhat lower thanthe grading of patients by personal examina-tion. The final estimate of the degree of im-provement of each of the patients has beenmade by one of us on the basis of all the in-formation obtainable. Very often this is lessthan the subjective opinion of the patient him-self. Some patients, of course, have proved tobe difficult to evaluate, particularly whenthey have had other diseases or noncardiac

808




symptoms. This is especially true of manypatients with neurotic symptoms, and of pa-tients handicapped by the residua of vascularaccidents occurring either before or duringsurgery. Patients who have been partially in-capacitated by vascular accidents occurring atthe time of surgery have been graded in ac-cordance with the handicap they have sufferedfrom the neurologic residua of their emboliirrespective of the cardiac status.

CRITERIA OF IMPROVEMENT

Patients have been classed as markedlyimproved, moderately improved, slightly im-proved, unchanged, and worse. Markedly im-proved patients are those who have gone up 2grades in the American Heart Associationclassification or the few group-II patients whohave lost all their symptoms. Thus patientswho were originally in group III are consid-ered markedly improved if they are now inclass 1 of the American Heart Associationclassification, and for the group-IV patientsan improvement into class 1 or 2 of the Ameri-can Heart Association classification wouldjustify a grading of markedly improved.Moderately improved patients are those whohave improved 1 grade in the American HeartAssociation classification. For the sake of theanalysis in the following discussion patientswho are moderately to markedly improvedhave been classed together as "significantlyimproved" or as "improved"; those who areonly slightly improved, unchanged, worse, ordead are considered as "unimproved."

CUMULATIVE IMPROVEMENT

Figure 3 shows the improvement of thepatients at each year of follow-up. This im-provement is calculated on the basis of thefollow-up of each patient at each anniversaryof his operation for the total period of follow-up. In a few patients where there has been ahiatus of more than 1 year between 2 follow-up reports, the state of improvement reportedat the first follow-up has been considered tohave been maintained to the year precedingthe next time that information is available by

GROUPS E AND m GROUP 17I00

_ .-_1_-1_ - .- - --- _- _ _.--11.

YEARS AFTER OPERATION YEARS AFTER OPERATION

705 694 602 432 271 113 29 205 195 166 131 90 41 16NUMBER OF PATIENTS NUMBER OF PATIENTS

FIG. 3. Status of operated patients at each year offollow-up. Solid, dead; Spotted, unimproved; Lined,improved.

a follow-up report, unless it is clear from theinformation on the last follow-up just whenany change in status occurred.The improvement in both groups III and

IV has tended to drop somewhat with suc-ceeding years of follow-up. To some extentthis may be due to the obvious fact that thosepatients who have been followed for thelongest time are also those who were operatedon early in our experience with this operation,and the operative attack on the valve mayhave been less adequate at that time. Whilethe quality of surgery undoubtedly improvedprogressively in this series no fundamentalchanges in technical principles were made un-til early in the second thousand operations.The tendency for the improvement to becomeless over the years represents at least in partthe inevitable ravages of the disease process.Residual damage to the cardiac muscle, pul-monary vasculature, liver, and so forth mayhave persisted and left its effect, particularlyin the group-IV patients. In addition, otherfactors which may be present are an operativefracture which was less than adequate; re-stenosis; mitral insufficiency; associated val-vular disease, and recurrent rheumatic ac-tivity. A more detailed analysis will be madelater in this report of the factors affecting thedeterioration of patients who previously havebeen significantly improved.

809




TABLE 7.-Status of 911 Patients at Last Follow-up

Groups II and III Group IVPer cent of Per cent of

Degree of No. of patients No. of patientsimprovement patients improved patients improved

Marked 327 46 74 36

Moderate 162 23 40 19

Slight 65 9 23 11

Unchanged 72 10 24 12

Worse 23 3 6 3

Late deaths 56 8 39 19

Totals 705 206

FACTORS INFLUENCING THE IMPROVEMENTOF PATIENTS FOLLOWING SURGERY

The factors that might have an effect on thepostoperative course of patients are so num-

erous that it is impossible to analyze them foreach year of follow-up. Therefore, the over-allimprovement status is given on the basis ofthe patients' condition at the last follow-up.

Sixty-nine per cent of group-II and IIIpatients and 55 per cent of the group-IVpatients have improved (table 7). In analyz-ing the effect of certain factors on the im-provement it is necessary to compare as ho-mogeneous groups as possible. Therefore statis-tical analyses have been made on groups inwhich all the patients have been followed for1 and also for 5 years, and the status at theend of the 1 and 5-year periods has been re-

corded. In table 8 the improvement at the endof 5 years is analyzed according to age, sex,

and rhythm. It will be seen that male andfemale patients in group III improved to ap-

proximately the same degree. Patients in thisgroup in normal rhythm were improved to a

greater extent than those who were fibrillating,since the p value for this correlation is be-tween 0.05 and 0.02. There is a tendency forpatients over 40 to do somewhat less well thanthe younger patients (p between 0.02 and0.01). There are more fibrillating patients inthe group over 40; nevertheless both age andrhythm play significant roles as the adjustedrates show.

In group IV significant differences due toage, sex, or rhythm are not apparent, althoughthe same tendencies are observed as in groupIII.The figures for improvement by age, sex, and

rhythm at the end of 1 year of follow-up showthe same tendencies that were observed at 5years, although differences become more strik-ing at the 5-year period.Of interest are the results in relation to the

findings at the time of operation. These in-clude the degree of mitral stenosis, the pres-ence and amount of mitral insufficiency, andthe adequacy of the valvuloplasty as esti-mated by the surgeon at the time of operation.There are a number of factors that affect theoperative results, some of which are withinthe control of the surgeon, depending on hisskill and experience, and others are inherentin the state of a valve as it is found at opera-tion. These include the degree of calcificationand its location, the rigidity of the leaflets,and the extent of fusion and shortening ofthe chordae tendineae. The quality of thevalvuloplasty is of the greatest importancein long-term results.With an increasing degree of mitral insuffi-

ciency, the results at the end of 5 years are

progressively worse (table 8). The degree ofinsufficiency was that amount which was esti-mated to be present by the surgeon aftercompletion of the valvuloplasty. In group III78 per cent of patients with pure mitral steno-sis were improved as compared to only 48 percent of those showing 2-plus or greater regur-gitation at the time of surgery. In group IVthe improvement rate dropped from 69 to 36per cent. These differences are all significant(p less than 0.01) except for the comparisonof patients with no versus 1-plus insufficiencyin group III, and between 1 plus versus 2 to3-plus insufficiency in group IV.

If the same correlations are made at theend of 1 year instead of 5, then the differencein the degree of improvement between thosewith no insufficiency and those with increasingdegrees of insufficiency is much less apparent.

810




TABLE 8.-Status at the End of One and of Five Years

Groups II and III Group IVOne year Five years One year Five years

O~~~~~~~~' o o0°8;a <'>,s5,}; %8 HE , 31 ° t'5°0 0 0 Q0 90 Q~~~~~

SexMaleFemale

RhythmNormal sinusAtrial fibrillation

Age by decades10-1920-2930-3940-4950-5960-69

Mitral insufficiencytNone

Preoperative valve sizet1.0 em.2 or less1.1 em.2 or more

151 83554 85

408297

8880

6 100116 91277 89262 7940 732 100

401 88 ,

174 86103 75

438240

86 /84

66205

158113

252115

_ 939

1585240

19852

6871

7860

100

81766311

78734a

50156

7463

41165

054294569

777251

845646

7271

8369

8074766362

778050

144 8442 52

5658.

*Adjusted rates calculated by indirect method.tIn a few patients clinical information regarding valve

lacking and these have been omitted from consideration.

It would appear, therefore, that the damaging outcomeeffect of mitral insufficiency takes time to the adjusappear. The mlWhen patients are studied in respect to anterior

their improvement in relation to the preopera- incision.;tive valve size, it is apparent that the patients velopedwho had tight mitral stenosis, that is an posteriorestimated valve area of 1 cm.2, or less, did this "mo:better at the end of 5 years than did patients sive) opewhose valve area was larger. This is true in fusion, o

both group III and group IV. Again this dif- quent miference was observed in patients when they another (

were studied at the end of 1 year in group IV operationbut was not apparent in group III. The pr

The effects of increasing degrees of mitral no statistinsufficiency and of a larger preoperative at the er

valve size in militating against a successful group IN

size or mitral insufficiency was

were additive, as demonstrated by3ted rates.iajority of these patients had onlyfusion bridge fracture or valvulotomeMore recently a technic has been de-for the adequate fracture of thefusion bridge as well. The effect ofre adequate" (certainly more exten-ration on sustained improvement, re-

)r indeed deterioration from subse-itral insufficiency will be discussed incommunication. The technic of thisi is discussed elsewhere.2-esence of associated valve disease hadtically significant effect on the resultsid of 5 years in either group III or

V.

811

5057

6551

6759525633

2664

2070

32244183

322422

6117

694636

614550

6124




IMPROVEMENT IN RELATION TO

POSTOPERATIVE COMPLICATIONFollowing surgery many of the patients

complained in varying degrees of symptomsnot directly related to the status of their car-diac compensation. Some of these symptomscomprise a syndrome that has been reportedunder the term of "postoperative" or "post-commissurotomy syndrome," and is charac-terized chiefly by exacerbation of pain in thechest of varying types, with fever. There maybe evidence of pericarditis, pleuritis, andpneumonitis in varying degrees. In our ex-perience this situation has been benign andself-limited, lasting a week or 2 and has notappeared to be affected particularly by thetype of therapy given, such as acetylsalicylicacid, penicillin, or adrenal steroids. The syn-drome is considered by some to be an activa-tion of rheumatic infection. In our experience,however, clear-cut evidence of rheumatic feveroccurs in only a minority of patients. The syn-drome tends to be recurrent and may appearfor the first time several years after surgeryand recurrences may occur over several years.The relation, therefore, to the surgical pro-cedure is obscure. It has been pointed out byIto, Engle, and Goldberg17 that this syn-drome also occurs in patients with nonrheu-matic heart disease who have had intracardiacsurgery involving opening of the pericardiumand it is their opinion that this may repre-sent recurring nonspecific pericarditis. Wehave no evidence bearing on this point, exceptthat we have no patients with pericardial ef-fusions in the postoperative period or later.

Since most of these patients have returnedto their homes and have been observed bytheir local doctors, and only occasionally per-sonally by us, the reported incidence of thissyndrome cannot be accurate, and may in-clude some cases of pneumonia or other res-piratory infections as well as pulmonary in-farctions.The incidence of the postoperative syn-

drome occurring in these patients was 30.8per cent (30.4 per cent for group III and 32.2per cent for group IV). This is comparable

to that previously reported for the first 500patients of this series followed for a shortertime3 as well as reports of others. It is ofinterest, however, that the over-all improve-ment of patients suffering from the postopera-tive syndrome is essentially the same as forthe groups as a whole (68 per cent in groupIII and 51 per cent in group IV).Many patients complain also of vague

joint pains. Whether the incidence of suchjoint pains is any higher in this group ofpatients than it would be in any other care-fully followed group of middle-aged peoplewith chronic disease is uncertain. We havedivided our patients into those who have suf-fered from arthralgias only, and those whohad a clear-cut arthritis, sometimes of therheumatoid type, but with no clear evidenceof rheumatic fever. A history of arthralgia orof arthritis did not affect the results strik-ingly, but in both groups III and IV the pa-tients in whom rheumatic fever was diagnoseddid poorly. This will be discussed in a latersection.

IMPROVEMENT IN RELATION TO THE PRESENCEOF ASCHOFF BODIES IN THE BIOPSYOF THE LEFT ATRIAL APPENDAGE

In a previous study20' 21 the relationship ofthe presence of Aschoff bodies in biopsies ofthe left atrial appendages to the clinical find-ings was reported. The present study confirmsthe previous one. The biopsy reports utilizedin the present study were routine reports re-ceived from the pathologic laboratories of theseveral hospitals concerned. No attempt wasmade to have the microscopic findings re-viewed by a single pathologist as was the casein the earlier study. A total of 632 biopsyreports are available from this series of 1,000cases. In the group-III patients 43 per centwere positive. In group IV the incidence ofpositive biopsies was 20 per cent. Table 9shows that the incidence of positive biopsiesis greatest in the younger age groups and fallsprogressively with increasing age but is still21 per cent in the 50 to 59 decade group ingroup III and 18 per cent in group IV. The

812




TABLE 9.-Incidence of the Finding of Aschoff Bodies in 632 Atrial Biopsies

No. of Per cent with Improved at last follow-uppatients positive biopsies Positive biopsies Negative biopsies

RhythmNormal sinus 328 54 74 72Atrial fibrillation 304 18 63 64

Group III Group IVImprove- Improve- Improve- Improve-

Per cent ment ment Per cent ment mentwith % % with % %

No. of positive positive negative No. of positive positive negativepatients biopsies biopsies biopsies patients biopsies biopsies biopsies

Age by decades10-19 5 60 100 50 -20-29 79 63 79 79 4 50 -30-39 200 46 74 75 35 26 75 6140-49 181 35 65 68 68 16 60 5850-59 24 21 75 63 33 18 80 6269-69 1 100 2 50

Total 490 43 142 20

positive biopsies are much more likely to befound in any age group in the patients withnormal sinus rhythm than in patients withatrial fibrillation. Both of these findings con-firm our previous report. When patients arecompared by age groups, the percentage im-provement of those with positive biopsies wasthe same as of those with negative biopsies.When the incidence of positive biopsies is

correlated with the occurrence of the variouspostoperative complications, no relation is evi-dent. The percentage of positive biopsies inpatients later developing the postoperativesyndrome was nearly the same as for thegroup as a whole (40 per cent in group III,15 per cent in group IV). Patients who sub-sequently developed definite rheumatic fevertended to have a slightly but not impressivelyhigher incidence of positive biopsies (56 percent).

FACTORS INVOLVED IN THE DETERIORATIONOF PATIENTS WHO HAVE IMPROVED

FOLLOWING MITRAL VALVULOPLASTY

Two hundred twenty-eight of the patientsin this study have become worse after havingbeen significantly improved, that is "mark-edly" or "moderately," for at least 1 year

after valvuloplasty. For the sake of this anal-ysis, the definition of "deterioration" is thatthey have slipped by at least 1 class, accordingto the American Heart Association classifica-tion. Sixty-two of these patients slipped onlyfrom "markedly" to "moderately" improved,and hence would still be classed in the "im-proved" category. The remaining 166 deter-iorated from either being originally "mark-edly " or " moderately " improved into the"unimproved" classification; that is, they arenow either only " slightly " improved, theircondition is unchanged as compared to thepreoperative state, they are worse, or dead.Of these patients, 48 have died since operation.and 45 have been reoperated on for mitral val-vular disease. These patients were not allpersonally observed by the authors at the timeof their deterioration; evidence for their deter-ioration was obtained in some from their an-swers to annual questionnaires or from lettersfrom their physicians.

There are a number of factors that mayhave been of importance in these 228 patientsin producing the deterioration of these pa-tients. Many of these occurred in combina-tion. In 56 mitral insufficiency was present,either alone or in combination with other fae-

813




TABLE 10.-Influence of Significant Mitral Insuffi-ciency Found or Produced at Operation*

Patients followed All patients atfive years latest follow-up

Per cent Per centwith withmitral mitral

Status of No. of insuffi- No. of insuffi-patients patients ciency patients ciency

Patients who de-teriorated aftersubstantial im-provement 89 30 222 22

Patients who main-tained improve-ment 196 12 531 12

Patients whofailed toimprove 43 28 113 35

*Patients in whom there were no adequate records ofthe presence or absence of mitral insufficiency havebeen omitted from consideration.

tors, at the time of surgery. This mitral insuffi-ciency was of moderate to marked extent, andwas either present prior to fracture of thevalve or was produced at the time of surgery.From this particular study, it is, of course,impossible to say to what extent mitral in-sufficiency may have developed or increasedboth in patients in whom it had been previ-ously recognized and in those in whom it hadnot been suspected.

In table 10 there is shown a comparison of 3groups of patients all of whom have been fol-lowed for 5 years and thus can be comparedon a statistical basis. It will be seen that pa-tients from the present group who deterior-ated after substantial improvement as well aspatients who failed to improve at all exhib-ited mitral insufficiency at the time of opera-tion much more often than did patients whoshowed a maintained itilpiov(ement (p lessthan 0.01). There is no significant differencebetween the patients who deteriorated andthose who failed to improve.The incidence of mitral insufficiency in these

3 groups has also been compared among all ofthe patients at the time of their latest follow-up and a trend similar to that shown in the 5-year group is apparent. The relatively lower

TABLE 11.-Influence of an Unsatisfactory Correctionof Mitral Stenosis*


Per cent Per centwith with

unsatisfac- unsatisfac-Status of No. of tory cor- No. of tory cor-patients patients rection patients rection

Patients who de-teriorated aftersubstantialimprovement 93 23' 226 17

Patients whomaintainedimprovement 193 9 513 7

Patients whofailed to improve 73 15 123 9

*Patients in whom it was impossible to assess ade-quately the quality of the operative procedure havebeen omitted from consideration.

incidence of mitral insufficiency in the dete-riorated group may possibly be due to the factthat mitral insufficiency takes time to produceits ravaging effect and the follow-up figuresfor the entire group include many patientsfollowed less than 5 years.At the time of operation, the surgeon was

frequently unable to accomplish a fully satis-factory correction of the stenosis. This failuremay have been due to conditions beyond hiscontrol, such as marked rigidity of the valvecusps, or widespread calcification or extremeshortening and fusion of the chordae tendi-neae. It is to be expected that with increasingexperience a surgeon will become more skillfulin effecting the maximum amount of correc-tion while avoiding the production of insuffi-ciency. There were 38 patients in the group of228 who deteriorated, in whom an adequatecorrection of mitral stenosis was not effectedat the time of operation, and 66 more in whomthe correction was less than satisfactory. Intable 11 a comparison is made of groups ofpatients followed for 5 years. Both the pa-tients who deteriorated after substantial im-provement and those who failed to improve atall were found to have an unsatisfactory cor-rection of their stenosis in a much higher per-

814




centage than did those who have shown main-tained improvement (p less than .01 for both).The percentage of unsatisfactory correctionswas somewhat higher in those who deterioratedthan in those who failed to improve at all,but the difference is not significant (p between.30 and .20). However it is suggestive that ifthe operations consisted chiefly of dilatation ofthe valve or temporary mobilization of thecusps, the patient may do well for a while.

This table also shows the results at thelatest follow-up of all patients. Although thesame trends are evident, the percentage figuresare lower in all categories. This is due to thefact that the over-all follow-up also includespatients operated on in the last half of the se-ries, who have been followed less than 5 years.The number of unsatisfactory operations per-formed in these recently operated patientswas lower than in the earlier half of the series.Indeed the number of unsatisfactory opera-tions in the earlier group is higher than itappears, for the surgeon must have at timesfailed to recognize an unsatisfactory valvulop-lasty. With greater experience the surgeonbecomes more exacting in his acceptance of"an adequate operation."The role of rheumatic fever in causing de-

terioration in patients is even more striking.There were 38 patients in the group of 228who had rheumatic fever sometime after sur-gery. Of those followed for 5 years, 19 per centdeteriorated after substantial improvement(table 12). In contrast, only 1.5 per cent ofthe patients showing maintained improvementand 9 per cent of those who failed to improveat all gave evidence of rheumatic fever. Thedifferences between patients showing main-tained improvement and the other 2 groupsare significant (p less than 0.01) but the dif-ference between deteriorated patients andthose who failed to improve is not statisticallysignificant (p 0.10 to 0.05).

It is of course often difficult to make aclear-cut diagnosis of rheumatic fever inadults, and most of these patients were notpersonally seen by us. A diagnosis of rheu-matic fever was only accepted in these pa-

TABLE 12.-Influence of Rheumatic Fever OccurringSince Operation


Per cent Per centwith with

Status of No. of rheumatic No. of rheumaticpatients patients fever patients fever

Patients who de-teriorated aftersubstantialimprovement 95 19 228 17

Patients whomaintainedimprovement 193 1.5 541 2.2

Patients whofailed to improve 73 9 142 14

tients, however, if they had reasonably con-

vincing evidence of it. Patients who sufferedfrom one or more attacks of the postoperativesyndrome, and those who complained merelyof vague joint pains did not show any unusualtendency to deteriorate.

Associated aortic valve disease occurred inthe group of 228 patients who did poorly afterinitial improvement in 10 per cent, which isabout the same percentage as in the series as

a whole (12 per cent). Other causes possiblyinvolved in the deterioration were peripheralemboli resulting in death or disability in 8patients and subacute bacterial endocarditis in3. In 62, there were no obvious factors explain-ing their deterioration. In some of these cases,the deterioration was unquestionably relatedto noncardiac causes, such as emotional statessometimes associated with the menopause, or

other neurotic factors.What does this study of patients who de-

teriorated reveal? Firstly, it shows that re-

current rheumatic fever often leads to deteri-oration of the cardiac status in these patientsjust as it does in rheumatic cardiac patientsin general.

Secondly, it shows that mitral insufficiencyis an important cause of relapse and that inmany cases this incompetence may be pro-duced or increased by the operative procedure.In another study we observed that mitral in-

815



ELLIS, I1ARKEN, 1LACK

sufficiency which was thought to be minimalor even not to exist at all at a first operationhad become a major factor at the time of re-operation. If this is true also in regard tothe entire group of 228 who deteriorated, it isevident that mitral inisufficienc(y may wellplay an even more important role than wasindicated by the 22 per cent incidence foundin this croup at surgery. The implication ofthis findingg in connection with patients whoare being considered for re-operation for mi-tral steiiosis is obvious, as, well as for surgeryiii nmitral stenosis in ge(nieral. It is evidentthat signiificanit mitral insufficiency may beproduced by the surgeon at operation and thismay constitute a warning agrainist too exteii-sive or poorly directed fracture of valves ilian effort to produce a complete relief of thestenosis. This also means that patients whohave deteriorated and are beimig consideredfor re-operation inust be scrutinized very (are-fully for the presence of initral inisufficienecybefore a second operation for mitral stenosis,since the major problem may really be mitralincompetence rather than stenosis and the con-ventional operation will not suffice.

Thirdly, this study shows how- important isthe quality of the technical performance. Notonly must the production of mitral insuffi-ciency be avoided but inadequate surgery forthe relief of stenosis must also be avoided sincethe patients with unsatisfactorily performedmitral valvuloplasties deteriorate in a muchhigher percentage than do those inl whom agood job was done.

Fourthly, the factor of predominant myo-cardial failure may well exist in many ofthese patients.

Finally, it calan be said that although re-stenosis of the mitral valve does occur, it isonly one of several factors that are of impor-tance in the deterioration of patients and sig-nificant re-stenosis requniriiig re-operation doesnot commonly occur if the original operationwas adequately done. It is impossible, on thebasis of these studies to give accurate figuresas to the rate of occurrence of re-stenosis.There are too many variable factors.

DisculssioNA great matiy papers have appeared con-

cerning the clinical results in patients oper-ated on for mitral stenosis. Although for thlemost part the degree of improvement rel)ortedin these various articles is inl general agree-ment with our findings, in the majority of re-ports the length of follow-up has beei rela-tivelv short. How-ever, the reports by ljikofland Uricchio22 have dealt with lpatients fol-lowed for periods up to 8 years, and Glovter etal.23 reported a series of 50 patients followed5 or more years. It has nlow been well (1em-onstrated that patients with mitral stenosiscan be operated uipoit with anl acceptable oper-ative mortality and that they are improvedin the vast majority of instances. It is becomn-ing apparent what factors are affecting thisimprovement, both immediately and over along period of time. As we have pointed outpreviously, this is still a palliative ol)eratioualthough it is often life-saving and may befollowed by extraordinary improvement thatmay persist for a long time. However, factorsremain which lead in a substantial number ofthe patients to a gradual recurrence of diffi-culty in many of those who have improved.and militate against improvement in others.To some extent this has to do with the state ofthe valve itself. Even under the best of cir-cumstances the valve remains in a scarredcondition which may be conducive to recur-rence of stenosis or to fixation of the leaflets,so that mitral incompetence results. The pa-tient still remains a rheumatic subject, onein whom recurrence of rheumatic activity isever possible. Studies such as ours have beenof relatively negative value so far as elucidat-ing the knotty prbblem of persistent or recur-rent rheumatic activity in adults with chronicrheumatic heart disease. It is apparent thatfor the most part the criteria for rheumaticactivity are not clear-cut and only in a minor-itv of instances clan a definite (liagliosis bemade. Wheni undoubted rheumatic fever doestake place after operation, however, the likeli-hood of cardiac detem-ioration is greatly en-haneed.

816




The role of mitral insufficiency, either pres-ent prior to surgery or developing at the timeof or subsequent to the operation, as an im-portant factor leading to deterioration is be-coming ever more apparent. Although truere-stenosis of the mitral valve may occur insome patients, this is not common in our ex-perience nor in that of most observers. Baileyand his group24 noted the increasing fre-quency of re-stenosis in the patients they arefollowing over prolonged periods of time. Itis apparent that symptoms of cardiac disabil-ity not infrequently recur in patients in whomthe original operation was not fully satisfac-tory, either due to inexperience on the partof the surgeon or to other conditions. In anyconsideration of the problem of "re-stenosis,"it is important to distinguish such recurrencesafter unsatisfactorily performed valvuloplas-ties from true re-fusion of the commissuresafter a completely satisfactory mobilization ofthe leaflets.

Myocardial failure is undoubtedly an im-portant element in the poor results of someof the patients or in the deterioration ofothers. This has been pointed out by us andhas been emphasized by Harvey and associ-ates.25 Whether myocardial failure followsfrom prolonged mechanical strain on theheart due to valvular defects or whether it isthe result of the residual damaging effect ofprevious rheumatic myocarditis, or even ofpersistent rheumatic activity has not beensatisfactorily elucidated, nor is it possible attimes to make a definite distinction betweenmyocardial failure and the mechanical factorsresulting from valvular defects. Even withthe aid of catheterization findings obtained byexperienced investigators a definite diagnosismay not be made. Hemodynamic formulaehave their limitations, not the least being thefact that the application of such formulaemust be with data obtained only on isolatedoccasions and under the most abnormal cir-cumstances, namely the situation of cardiaccatheterization itself. We have observed on

a number of occasions patients who, fromcatheterization findings, have what one would

call physiologic left ventricular failure al-though the patients had no symptoms or signsof clinical heart failure. Conversely, it isoften possible for a patient to have manifesta-tions of congestion without exhibiting hemo-dynamic manifestations of what one callsheart failure, namely an elevated left or rightventricular diastolic pressure.

It has been suggested25' 26 that hemody-namic studies made before and after the ad-ministration of a parenteral digitalis orstrophanthin preparation may help to distin-guish the myocardial failure from the mechan-ical effects of the mitral block. Such testsare difficult to carry out and only valid in thehands of the most experienced investigatorsand hence could never have wide applications,even for such limited value as they may have.Most important of all, the quantitative evalu-ation of the relative degrees of stenosis versusregurgitation at a valve has as yet been ac-complished only very imperfectly by catheteri-zation or any other technics. A careful clini-cal evaluation is still as reliable as any of thespecial laboratory technics.

SUMMARY

A report has been made of the clinical re-sults in 1,000 consecutive cases with a pre-operative diagnosis of mitral stenosis thatunderwent valvuloplasty between 1949 and1956. All but 2 of the 913 survivors of theoperation have been followed at least 1 yearand most of them up to the latest anniversaryof their operation from 2 to 8 years ago.When the preoperative clinical diagnosis

was pure mitral stenosis, the surgeon con-firmed the diagnosis at operation in about 90per cent of cases, but when mitral stenosis andinsufficiency was diagnosed preoperatively,operation revealed significant insufficiency inonly about one half of the cases.The incidence of operative embolization in

the second 500 of this series has been 2.1 percent in group III and 8.0 per cent in groupIV. In only 25 have peripheral emboli oc-curred since operation.

817




Including operative mortality, 83 per centof group-III patients and 57 per cent ofgroup-IV patients have survived 7 years.These figures are much higher than reportedsurvival rates of medically treated patients.The effects of sex, rhythm, and age on sur-vival have been discussed.The over-all percentage of patients signifi-

cantly improved by operation has been 69 percent in group III and 55 per cent in group IV.The percentage improved has tended to dropsomewhat with succeeding years of follow-up.

In group III, fibrillating patients and thoseover 40 years of age did less well than thosein normal rhythm or under 40; there was,however, no difference due to sex. In groupIV significant differences due to rhythm, age,or sex were not apparent.With increasing degrees of mitral insuffi-

ciency the results were progressively poorerand the differences were more striking at theend of 5 years than at the end of 1 year. Pa-tients with preoperative tight mitral stenosis(1.0 cm.2 or less) did better than those whosestenosis was less severe. The presence of as-sociated valve disease (usually aortic) of milddegree did not affect the ultimate outcome.The "postoperative" or "postcommissurot-

omy" syndrome occurred in an estimated 30.8per cent of patients, but its presence did nothave any bearing on the results.

Aschoff bodies were described in the biop-sies of atrial appendages in 43 per cent ofgroup-III patients and in 20 per cent ofgroup-IV patients. They were present in ahigher percentage of patients in normalrhythm than those in fibrillation, and the in-cidence decreased progressively with age. Thepresence of a positive biopsy had no relationto the ultimate results or to the occurrence ofthe postoperative syndrome.A group of 228 patients who deteriorated

after substantial improvement persisting atleast 1 year were analyzed. Factors that werefound in a significantly higher percentagethan in patients who maintained their im-provement were mitral insufficiency, an unsat-isfactorily performed valvuloplasty, and clear-cut rheumatic fever occurring since operation.

SUMMARIO IN INTERLINGUAEs reportate le resultatos clinic in 1.000

casos consecutive in que le diagnose preopera-tori de stenosis mitral esseva sequite per val-vuloplastia, effectuate inter 1946 e 1956. Con2 exceptiones, omne le 913 superviventes deloperation esseva tenite sub observation du-rante al minus un anno. In le majoritate delcasos, le observation esseva extendite usqueal plus recente anniversario del operationeffectuate inter 2 e 8 annos retro.Quando le diagnose clinic preoperatori es-

seva pur stenosis mitral, le chirurgo confirma-va le diagnose al operation in circa 90 procento del casos, sed inter le casos in que ste-nosis e insufficientia mitral esseva diagnosti-cate ante le operation, le incidentia de insuf-ficientia significative constatate per le chirurgoesseva solmente circa 50 pro cento.Le incidentia de embolisation operatori inter

le secunde 500 casos del serie esseva 2,1 procento in gruppo III e 8,0 pro cento in gruppoIV. Embolos peripheric ha occurrite depositle operation in solmente 25 casos.Prendente in consideration le mortalitate

operatori, 83 pro cento de patientes de gruppoIII e 57 pro cento de patientes de gruppo IVha supervivite 7 annos. Iste cifras es multoplus alte que illos reportate pro le superviven-tia de patientes tractate per mesuras medical.Le effectos de sexo, rhythmo, e etate super leprognose es discutite.Le procentage total del patientes significa-

tivemente meliorate per le operation esseva 69pro cento in gruppo III e 55 pro cento ingruppo IV. Iste procentages monstrava untendentia descendente in le curso del annospostoperatori.

In gruppo III, patientes con fibrillation e

patientes de plus que 40 annos de etate habevaun prognose minus favorabile que patientescon rhythmo normal e patientes de minus que40 annos de etate. Tamen, nulle differentiaesseva constatate secundo le sexo. In gruppoIV, nulle significative differentias esseva no-tate secundo rhythmo o etate o sexo.

Parallel al augmento del grado de insuffici-entia mitral, le resultatos se pejorava progres-sivemente. Le differentias essvea plus frap-

818




pante al fin de 5 annos que al fin de 1 anno.Patientes con tense stenosis mitral ante le ope-ration (1 cm2 o minus) reageva melio quepatientes in qui le stenosis esseva minus sever.Le associate presentia de morbo valvular(usualmente aortic), si illo esseva de levegrados, non afficeva le resultato final.Le syndrome "postoperatori" o "posteom-

missurotomic" occurreva estimatemente in30,8 pro cento del patientes, sed su presentiao absentia no influentiava le resultato final.

Corpores de Aschoff esseva describite in lebiopsias de appendages atrial in 43 pro eentodel patientes de gruppo III e in 20 pro centodel patientes de gruppo IV. Illos esseva pre-sente in plus alte procentages in patientes conrhythmo normal que in patientes con fibrilla-tion. Lor incidentia descendeva progressive-mente con le avantiamento del etate. lie pre-sentia de un biopsia positive habeva nullerelation con le resultato final o con le occur-rentia o non-occurrentia del syndrome post-operatori.

Esseva analysate le casos de un gruppo de228 patientes in qui marcate grados de deteri-oration se declarava post un melioration sub-stantial de un duration de al minus 1 anno.Factores que se trovava in iste casos in pro-centages significativemente plus alte que incasos in que le melioration se manteneva in-clude insufficientia mitral, un valvuloplastianon effectuate satisfactorimente, e le occur-rentia de clar episodios de febre rheumaticdepost le operation.

ACKNOWLEDGMENTWe are greatly indebted to Dr. Mindel Sheps of

the Department of Preventive Medicine, HarvardMedical School, for her valuable advice and as-sistance in the statistical evaluation of the results.The collection of follow-up information and com-pilation of statistics were largely performed byMrs. Georgiana, Mr. Arthur Spiro, and Mrs.Eleanor Angelokas.

REFERENCES1. HARKEN, D. E., ELLIS, L. B., DEXTER, L.,

FARRAND, R. E., AND DICKSON, J. F.: Theresponsibility of the physician in the selec-tion of patients with mitral stenosis forsurgical treatment. Circulation 5: 349, 1952.

2. -, AND BLACK, H.: Improved valvuloplastyfor mitral stenosis with a discussion ofmultivalvular disease. New England J. Med.253: 669, 1955.

3. ELLIS, L. B., AND HARKEN, D. E.: The clinicalresults in the first five hundred patients withmitral stenosis undergoing valvuloplasty.Circulation 11: 637, 1955.

4. -, AND -: Factors influencing the late re-sults of mitral valvuloplasty for mitralstenosis. Ann. Int. Med. 43: 133, 1955.

5. -, ABELMANN, W. H., AND HARKEN, D. E.:Selection of patients for mitral and aorticvalvuloplasty. Circulation 15: 924, 1957.

6. HARKEN, D. E.: Editorial. The surgical treat-ment of acquired valvular disease. Circula-tion 18: 1, 1958.

7. TAYLOR, W. J., BLACK, H., THROWER, W. B.,AND HARKEN, D. E.: Valvuloplasty formitral stenosis during pregnancy. J. A.M. A. 166: 1013, 1958.

8. SNEDECOR, G. W.: Statistical Methods Ap-plied to Experiments in Agriculture andBiology. Ed. 4. Ames, Iowa, The Iowa StateCollege Press 1946.

9. WILSON, M. G., AND LIm, W. N.: The naturalhistory of rheumatic heart disease in thethird, fourth, and fifth decades of life. I.Prognosis with special reference to survivor-ship. Circulation 16: 700, 1957.

10. GRANT, R. T.: After histories for ten years ofone thousand men suffering from heart dis-ease. Heart 16: 275, 1933.

11 WILSON, J. K., AND GREENWOOD, W. F.: Thenatural history of mitral stenosis. Canad.M. A. J. 71: 323, 1954.

12. HAMILTON, B. E., AND THOMSON, K. J.: TheHeart in Pregnancy and the Child BearingAge. Boston, Little, Brown & Company,1941, p. 402.

13. VEDOYA, R., NESSI, C. T., AND MENDELZON, J.:Prognosis of mitral stenosis. Second WorldCongress of Cardiology and the 27th An-nual Scientific Sessions of the AmericanHeart Association, Sept. 12-17, 1954, pp.119-120.

14. ROWE, J., BLAND, E. F., SPRAGUE, H. B., ANDWHITE, P. D.: The course of mitral stenosiswithout surgery: then and twenty year per-spectives. Third World Congress of Cardiol-ogy, Brussels, September 1958. Abstracts ofCommunication, p. 263.

15. DONZELOT, E., HEIM DEBALSAC, R. H.,SAMUEL, P., AND BEYDA, E.: Life expecta-tion of patients with mitral stenosis withand without operation. Brit. Heart J. 19:555, 1957.

16. OLESEN, K. H.: Mitral Stenosis. A follow-upof 351 Patients. Copenhagen, Ejnar Munks-gaards Forlag, 1955, p. 228.

819




17. BERKSON, J., AND GAGE, R. P.: Calculation ofsurvival rates for cancer. Proc. Staff Meet.,Mayo Clin. 25: 270, 1950.

18. OTTO, J. F., HUTCHESON, J. M., JR., ABEL-MANN, W. H., HARKEN, D. E., GARY, J. E.,AND ELLIS, L. B.: Clinical observations be-fore and after mitral valvuloplasty. Physi-cal, radiologic and electrocardiographicchanges. New England J. Med. 253: 995-1005, 1955.

19. ITo, T., ENGLE, M. E., AND GOLDBERG, H. P.:Postpericardiotomy syndrome following sur-gery for nonrheumatic heart disease. Circu-lation 17: 549, 1958.

20. DECKER, J. P., HAWN, C., VAN, Z., AND ROB-BINS, S. L.: Rheumatic "activity" as judgedby presence of Aschoff bodies in auricularappendages of patients with mitral stenosis.I. Anatomical aspects. Circulation 8: 161,1953.

2.1. MONEELY, W. F., ELLIS, L. B., AND HARKEN,D. E.: Rheumatic "activity" as judged bythe presence of Aschoff bodies in auricularappendages of patients with mitral stenosis.II. Clinical aspects. Circulation 8: 337, 1953.

22. LIKOFF, W., AND URICCHIO, J. F.: Results ofmitral commissurotomy. Clinical status oftwo hundred patients five to eight yearsafter operation. J.A.M.A. 166: 737, 1958.

23. GLOVER, R. P., O'NEILL, J. J. E., AND JANTON,0. H.: An analysis of fifty patients treatedby mitral commissurotomy five or moreyears ago. J. Thoracic Surg. 30: 436, 1955.

24. BAILEY, C. P., GOLDBERG, H., AND MORSE, D.P.: Recurrent mitral stenosis. Diagnosis bycatheterization of the left side of the heart.J.A.M.A. 163: 1576, 1957.

25. HARVEY, R. M., FERRER, M. I., SAMET, P.,BADER, R. E., COURNAND, A., AND RICHARDS,D. W.: Mechanical and myocardial factorsin rheumatic heart disease with mitral steno-sis. Circulation 11: 531, 1955.

26. Yu, P. N., NYE, R. E., JR., LOVEJOY, F. W.,JR., MACIAS, J. DEJ., SCHREINER, B. F.,AND Lux, J. J.: Studies of pulmonary hy-pertension. VIII. Effects of acetyl strophan-thidin on pulmonary circulation in patientswith cardiac failure and mitral stenosis.Am. Heart J. 54: 235, 1957.

Medical EponymsB.y ROBERT W. BUCK, M.D.

Babinski's Phenomenon. Joseph Francois Felix Babinski (1857-1932) described thissign in the Comptes rendus hebdomadaires des Seances et Me'moires de la Socie'te de Bio-logie 48: 207-208 (Feb. 22) 1896. The title of the article from which the following quo-tation is taken is "A Plantar Cutaneous Reflex in Certain Organic Affections of theCentral Nervous System" (Sur le Reflexe Cutane Plantaire dans Certaines AffectionsOrganiques du Systeme Nerveux Central):

"I have observed in a certain number of cases of hemiplegia or monoplegia involvingthe leg which were associated with an organic affection of the central nervous system, adisturbance in the plantar reflex of which I here present a short description: On thehealthy side, pricking of the sole of the foot causes a flexion of the thigh on the pelvis,the leg on the thigh, the foot on the leg, and the great toe on the metatarsus. This occurssimilarly in normal patients. On the paralyzed side, a similar stimulus also gives riseto a flexion of the thigh on the pelvis, the leg on the thigh, the foot on the leg, but thegreat toe, instead of being flexed, is extended on the metatarsus."

820



LAURENCE B. ELLIS, DWIGHT E. HARKEN and HARRISON BLACKYears after Mitral Valvuloplasty

A Clinical Study of 1,000 Consecutive Cases of Mitral Stenosis Two to Nine

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1959 American Heart Association, Inc. All rights reserved.

75231is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TXCirculation

doi: 10.1161/01.CIR.19.6.8031959;19:803-820Circulation.

http://circ.ahajournals.org/content/19/6/803located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer

of the Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle columnClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:


http://circ.ahajournals.org/content/19/6/803

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


circulation 1959 ellis 803 20

Documents

tricuspid stenosis

correction of stenosis

mitral valvuloplastywas

evidence of aortic stenosis

present study

ofthese patients

tile preoperative diagnosis

clinical study