492 POSTGRADUATE MEDICAL JOURNAL October 1953

however. In the initial stages of treatment it isessential to estimate the blood electrolytes, par-ticularly chloride and potassium, at least weekly.There is a tendency to hyperchloraemia, soammonium chloride should not be used as anadjuvant to mercurial diuretics while resins arebeing given. Special care is necessary when thereis any evidence of renal damage. Treatmentshould always be started in hospital but can bemaintained after discharge.

In some of our cases we have noticed thatalthough mercurial diuretics produced little re-sponse before resin treatment, a good diuresisfollowed subsequent injections. Some patientswho improve very little with resin treatment alonemay improve when this is combined with a lowsodium diet. The best combination of sodiumintake, mercurial diuretics and resin treatment stillseems to be a matter of trial and error in re-fractory cases of heart failure and it must beadmitted that although the cation exchange resins

are a valuable new weapon in the control ofsodium metabolism in heart failure our use ofthem is still rather tentative and exploratory.

In covering so broad a field as the medicalmanagement of rheumatic heart disease it is notpossible to discuss details comprehensively, but Ihave tried to place some emphasis upon threeaspects of the approach to management; first,upon the paramount importance of regularmoderate exercise in maintaining general fitness,especially in the earlier stages of the disease;next, upon the need for constant vigilance in theanticipation and early recognition of complica-tions; and finally upon the necessity to search forand treat causes of deterioration not directly con-nected with the heart disease. It will be obviousthat in such an approach constant alert super-vision by the family doctor is far more likely toensure success than occasional survey by aconsultant.

DIAGNOSIS AND TREATMENT OFRHEUMATIC FEVER WITH SPECIALREFERENCE TO EARLY CARDITIS

By GERALD THOMAS, M.R.C.P.Canadian Red Cross Memorial Hospital, Taplow

Diagnosis of Rheumatic FeverThe diagnosis of rheumatic fever is seldom

wrongly made when a migratory polyarthritis andfever follow a sore throat. Occasionally it may bedifficult to distinguish between rheumatic feverand other conditions, notably Still's disease, osteo-myelitis, Henoch-Schonlein purpura, leukaemiaand poliomyelitis. In Still's disease there may beinvolvement of several joints at the onset and thejoint changes may be of such short duration as tosimulate rheumatic fever. Furthermore, peri-carditis may occur; friction may easily be mistakenfor heart murmurs and effusion for cardiac dilata-tion. Often in early cases of Still's disease acharacteristic rash consisting of discrete pinkmacules is present which enables the correctdiagnosis to be made. Furthermore there is fre-quently a persistent high swinging temperature andgeneralized glandular enlargement. Rarely thejoint involvement in rheumatic fever may be mostmarked in the hands, and these changes may per-sist for days or even weeks, suggesting a diagnosis

of Still's disease. In long-standing rheumaticfever also, there may be nodule formation over theknuckles simulating the fusiform fingers ofrheumatoid arthritis; rarely ulnar deviation mayresult as a residual deformity (Jaccoud's syndrome).The diagnosis between rheumatic fever and

osteomyelitis m*y be particularly difficult. 'Thepresenting symptom in osteomyelitis may be painin the joints above and below the site of the lesion;there may be pain in the shoulder and elbow fromosteomyelitis of the humerus. On the other hand,in rheumatic fever there may be severe involvementof one joint only, often a large one such as the hip,and the signs may persist in that joint alone forseveral days; the differential diagnosis is thenbetween rheumatic fever and suppurative arthritisor osteomyelitis. X-rays are of no value in diag-nosis at this stage as bone changes do not occuruntil later. Neither is the total or differentialwhite blood count, for there may be little or noleukocytosis in the first 48 hours of osteomyelitisand there is frequently a high white count in rheu-

Protected by copyright.

on Septem

ber 18, 2020 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.29.336.492 on 1 October 1953. D

ownloaded from

October 1953 'THOMAS: Diagnosis and Treatment of Rheumatic Fever 493

........l..

FIG. I.-Erythema marginatum.

matic fever. In cases where there is any doubt,full penicillin therapy and immobilization shouldbe instituted forthwith and the opinion of anorthopaedic surgeon sought.

Henoch-Schonlein purpura may cause jointchanges indistinguishable from those of rheu-matic fever, but the diagnosis can usually be madeon the appearance of the characteristic maculo-papular rash with central petechiae, most markedbelow the knees and over the buttocks. Petechiaemay, however, occur in rheumatic fever. Leu-kaemia may cause joint pain which, together withfever, tachycardia and anaemia, may lead to awrong diagnosis of rheumatic fever. Furthermore,there may be cardiac enlargement and systolic anddiastolic murmurs due to anaemia. Often thecorrect diagnosis is revealed by the blood picture,but if the disease happens to be in the aleukaemicphase, marrow puncture is necessary. Malignantdisease, e.g. neuroblastoma, with secondaries inthe bone may cause similar difficulty. In polio-myelitis there may be severe limb pains maximalin the joints and the case may be diagnosed asrheumatic. Not until paralysis develops is thetrue character of the disease revealed.Two rheumatic stigmata that are particularly

helpful in diagnosis are erythema marginatum andnodules. Erythema marginatum (Fig. i) may de-velop at any time from the onset, and its appear-ance makes a doubtful diagnosis of rheumaticfever certain. The rash may occur anywhere onthe trunk and limbs, rarely on the face. It isfleeting and may vary in extent and distributionfrom hour to hour. Nodules do not usually de-

velop until six to eight weeks after the onset ofrheumatic activity. They are not helpful, there-fore, in the diagnosis of early cases, but theirappearance later is undoubted evidence of recentactivity. They occur most often over the ole-cranon processes where they may be preceded bythickening of the subcutaneous tissue (pre-nodularthickening) for a few days or a week. Nodulesmay rarely occur in Still's disease but may be dis-tinguished from those of rheumatic fever by theirlarger size and longer duration, as well as histo-logically.An increased antistreptolysin o titre is often

mentioned as an important point in the diagnosis ofrheumatic fever. Although helpful it is not diag-nostic as it is raised after 8o per cent. of Group Astreptococcal infections. A raised antistrepto-lysin o titre soon after the onset of suspectedrheumatic fever means that there has been a recentstreptococcal infection, a point in favour of thediagnosis. In some cases of rheumatic fever,about 20 per cent., the antistreptolysin o titre isnot raised.The blood sedimentation rate is usually high at

the onset and then gradually falls over a period ofseveral weeks. Occasionally, however, it may re-turn to normal after two or three weeks. A normalblood sedimentation rate, therefore, soon after theonset of suspected rheumatic fever does not pre-clude this diagnosis.

Occasionally there may be no acute joint painat the onset but merely vague pains in the legs orin the arms and legs. Such limb pains are not un-common in normal children but when they are

Protected by copyright.

on Septem

ber 18, 2020 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.29.336.492 on 1 October 1953. D

ownloaded from

494 POSTGRADUATE MEDICAL JOURNAL October 1953

...:..

... .... ...W........

...

FIG. 2.-Disappearance of loud mitral systolic murmur over a period of six weeks. (I,F-low fre-quency; HF-high frequency.)

accompanied by other symptoms such as poorappetite, fatigue and loss of weight the po3sibilityof active rheumatism must be considered. Inmany such cases a diagnosis of innocent' growingpains ' is wrongly made, the patients are allowed tocontinue leading normal school lives and, whenseen some weeks or months later, have well-marked heart lesions. In many young childrenwho already have severe cardiac involvement whenadmitted to hospital, there is no history of previousacute joint pains, only transitory discomfort in thelimbs and symptoms of general ill health.

The Diagnosis of CarditisHaving made the diagnosis of rheumatic fever,

the question of heart involvement arises and mustbe considered from the following points of view.Are there or are there not any abnormal signs inthe heart? If there are, are they due to rheumaticor to congenital heart disease? If the former, arethey related to the present attack or to previousones?The possibility of associated congenital heart

lesions must always be borne in mind lest abnormalheart signs be mistakenly attributed to rheu-matism. Congenital lesions which may causedifficulty are atrial septal defect, ventricular septaldefect, patent ductus arteriosus and coarctation ofthe aorta. In atrial septal defect there may be anabnormal apex beat, a noticeable systolic murmurand sometimes a basal diastolic murmur. Thesystolic murmur of ventricular septal defect maybe well heard towards the apex, simulating amitral lesion; there may also be a mitral diastolicmurmur. The murmur of patent ductus arteriosusmay sound like that of rheumatic aortic disease,and a mitral diastolic murmur may be heard at theapex. In coarctation an aortic diastolic murmurmay be present and again there may be a mitraldiastolic murmur. At times it may be necessaryto perform cardiac catheterization to confirm thepresence or absence of congenital lesions.

If it is thought that abnormal heart signs are dueto rheumatism, it must be d&cided whether theyare due to the present attack or to previous ones.The absence of a history of a previous attack,though helpful, is not conclusive, for there mayhave been unnoticed or unrecorded ones. Webelieve that a diagnosis of active carditis, i.e.carditis related to the present attack, can only bemade under the following circumstances:

i. When murmurs either develop or disappearunder observation or show a striking change inintensity.

2. When a significant change in heart sizeoccurs (more than i cm.).

3. In the presence of pericarditis or heartfailure.

The Diagnosis of Recent Carditis onAuscultationThe diagnosis of receint and active carditis can

most often be made on auscultation; on the de-velopment, disappearance or change in intensity oforganic murmurs. Organic murmurs-mitral oraortic diastolic murmurs or systolic murmursmaximal at the apex and filling systole-may de-velop within a short time of the onset of rheumaticfever, certainly within 24 hours, though occasion-ally not for several days or weeks. Their sub-sequent behaviour then varies from case to case;in some they may persist unchanged throughoutthe period of observation in hospital, in others theymay become louder or softer and they may evendisappear altogether. The diastolic murmurs ofrecent carditis are soft, short, localized and diffi-cult to hear. The mitral diastolic murmur is heardbest at the apex with the child lying on the leftside, using light pressure with the bell of thestethoscope. The basal diastolic murmur is bestheard to the left of the lower end of the sternumwith the child supine, on expiration, and usingfirm pressure with the diaphragm. Murmurs thatare already present when the patients are admitted

Protected by copyright.

on Septem

ber 18, 2020 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.29.336.492 on 1 October 1953. D

ownloaded from

October I THOMAS: Diagnosis and Treatment of Rheumatic Fever 495

...-:::i'.'.'.,

.......

, x t tF 7 7 -.::.

eSRS

41MII0 T;-

FIG. 3.-Marked diminution in intensity of basal diastolic murmur five days after admission. Furtherdiminution I5 days later.

may later show a significant change in intensity;only then can they be related with certainty to thepresent attack. Thus soft aortic and mitraldiastolic murmurs may increase in intensity orloud murmurs become softer and even disappear.Sometimes very loud murmurs may prove to bedue to the present attack as in the following twocases. A girl, aged nine, was admitted withinthree weeks of her first known attack. She hadthen a Grade III mitral systolic murmur withenlargement and paradoxical pulsation of the leftauricle. Six weeks later she had no significantmurmurs and screening showed no abnormality(Fig. 2). A boy aged eight, admitted within a fewdays of his first known attack, had a very loudbasal diastolic murmur with a marked thrill. Fivedays later the murmur was much softer and thethrill had disappeared (Fig. 3).

In most cases the diagnosis of active carditiscan be made on the auscultatory findings; in afew it is made on other criteria-pericarditis, asignificant change in heart size, or heart failure.Pericarditis occurs in about 15 per cent. of casesof rheumatic fever. Pericardial friction is usuallyeasy to diagnose, but may be mistaken for themurmurs of aortic disease; the electrocardiogrammay be normal in the absence of effusion. Peri-cardial effusion may be suspected clinically when arise in temperature and venous pressure occur,together with abnormal signs at the lung bases;abnormal cardiac dullness and altered heart signsappear later (Thomas, Besterman and Hollman,1953). The earliest radiological evidence ofeffusion is straightening of the left border; the

effects of changes in posture are of little value(Besterman and Thomas, 1953).

Cardiac enlargement is often quoted as an earlysign of rheumatic carditis, but in our experience itis absent in the majority of recent cases. In thosewith longstanding activity or who have had pre-vious attacks, however, some enlargement may bepresent on admission and subsequent changes inheart size may indicate further activity. In inter-preting films for changes in heart size the followingfactors must be borne in mind: the height of thediaphragms, the heart rate and the stage of thecardiac cycle-systole or diastole; an increase ordecrease of i cm. or more is thought to be signifi-cant. Sudden increase in heart size is rare save incases of established valve lesions and failure;many examples of so-called acute dilatation are dueto pericardial effusion.

Signs of failure in children with rheumatic heartlesions indicate active carditis.

Tachycardia has often been quoted as an im-portant sign of active carditis. We have found,however, that there is nothing remarkable aboutthe pulse rate in the majority of recent cases unlessthe attacks are particularly severe. T he pulse rateis seldom raised out of proportion to the tempera-ture and it returns to normal with the temperature;this applies to both the day and sleeping pulserates. Not infrequently there may be a sinusbradycardia.

Electrocardiographic changes are few and in-frequent. The PR interval was prolonged in only35 out of 479 (7 per cent.) cases of chorea andrheumatic fever with carditis seen here over a

Protected by copyright.

on Septem

ber 18, 2020 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.29.336.492 on 1 October 1953. D

ownloaded from

POSTGRADUATE MEDICAL JOURNAL

period of four years. In the majority of these 35patients the PR was prolonged for only the firstmonth after admission and most of them had well-marked heart lesions. The QTc (QT interval

corrected for heart rate; QTc - QT ) has re-cently been stressed as a useful sign of activecarditis (Taran, I947; Abrahams, 1949). In arecent series of 58 cases with carditis seen in thisUnit, the QTc was raised in only 13. In 27 therewas certain clinical evidence of active carditis; theQTc was abnormal in nine of them (25 per cent.).The remaining i8 had carditis but there was nocertain evidence that the carditis was active; four(i8 per cent.) had a raised QTc. The measure-ment of the QTc is difficult except for someonepractised in it and errors are easily made. T wavechanges are rare except as a sequel to pericarditis.

TreatmentBed rest. Patients should b3 confined to bed

for three to four weeks after all signs of rheumaticactivity have disappeared; they should have hadnormal weight increments, no anaemia and normalblood sedimentation rates. An exception may bemade in the case of girls over the age of pubertywho may have raised blood sedimentation rates-20 to 30 mm. per hour Westergren-long after allother signs of activity have disappeared. Ifotherwise well they should be allowed up, whennot infrequently the sedimentation rates fall tolower levels. If there has been slight carditis-soft diastolic murmurs and no cardiac enlargment-I0 to I2 weeks in bed is probably wise, eventhough signs of activity may have disappearedmuch earlier. With more severe degrees ofcarditis, often associated with continued activity,longer periods of rest are necessary.

Drugs. Salicylates are useful in that they lowerthe temperature and ameliorate joint swellings andpain, but we have no evidence that they influencecarditis. They should be given in the maximumdosage that can be tolerated; mild tinnitus is notan indication to reduce the dose. Cortisone andACTH are similar in their effects to salicylates;there is yet no indication that they are any morebeneficial to the heart. Both drugs should begiven for periods of not less than six weeks andpreferably for 12. Their side effects are wellknown; both may cause sodium retention. Re-lapses not infrequently occur on stopping treat-ment, but usually subside spontaneously. Asuitable scheme of dosage for cortisone is 300mgm. for one day, 200 mgm. for four days, Ioomgm. for the remainder of the first three weeks,75 mgm. for two weeks and 50 mgm. thereafter;and for ACTH, 8o mgm. for four days, 6o mgm.

for three days, 40 mgm. for two weeks, 30 mgm.for two weeks and zo mam. thereafter.The anaemia of active rheumatism does not

respond to iron or other specific therapy, but thepos3ibility of an associated iron deficiency anaemiamust not be overlooked.

Streptococcal infection. Overt clinical tonsillitisshould be promptly treated with intramuscularpenicillin. Ideally, and in hospital, throat andnose swabs should be taken weekly and positivestreptococcal infections treated with penicillin.There is now a good deal of evidence that prophy-laxis against streptococcal infection is important(American Heart Association Statement, 1953).Penicillin is the drug of choicei but is expensive,and sulphonamides in the dose of i gm. daily areeffective. Agranulocytosis is uncommon. Shouldit occur, prophylaxis should be changed to oralpenicillin. Tonsillectomy is indicated only whenthere is chronic tonsillar infection; it should becarried out under penicillin cover

Heartfailure is uncommon except in cases withmarked valvular lesions and cardiac enlargement.Its treatment then differs in no way from that offailure in established heart disease; digitalis,mersalyl and low salt diet. Sodium retention,despite o.6 gm. per day sodium diets, occurs notinfrequently in patients with cardiac enlargementwho are being treated with ACTH or cortisone. Itusually responds to mersalyl and there is seldomneed to abandon hormone therapy. So-calledheart failure in early cases-manifested by a risein venous pressure, enlargement of the liver andincrease in respiration rate-is usually due to peri-cardial effusion; there is little or no response todigitalis, in contrast to the good response seen incases with more longstanding disease who havewell-marked valve lesions and cardiac enlargement.It may be difficult sometimes to distinguish be-tween the signs of failure and those of effusion,but in failure the temperature and blood sedi-mentation rate usually fall whereas in effusion theyrise; and signs at the lung bases, particularly atthe left and straightening of the left border of thecardiac silhouette, are common early on ineffusion. No specific therapy is indicated for peri-cardial effusion, though when there is markedfever and tachypnoea salicylates may be effective inreducing both, often without any change in size ofthe effusion. There is no point in aspirating thepericardial sac unless the effusion is massive.Moreover, paracentesis is frequently unsuccessfulas the fluid so quickly clots and loculates. Aspira-tion of an accompanying pleural effusion, if large,may be of benefit however.

After care. At the time of discharge fromhospital advice is given on any limitations ofphysical activity that may be necessary, on the need

496 O)ctober 1953P

rotected by copyright. on S

eptember 18, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.29.336.492 on 1 O

ctober 1953. Dow

nloaded from

October 1953 SOMERVILLE: Mitral Stenosis: Selection of Cases for Mitral Valvotomy 497

for prompt treatment of sore throats or recurrencesof limb or joint pains and on the importance ofpenicillin cover for dental extractions if heartlesions persist. If sulphonamide prophylaxis hasbeen started in hospital it should be continued forat least another two years. Patients should attendafter-care clinics at regular intervals. Our own areseen at the third, sixth and twelfth month after dis-charge and then once each year. These examina-tions are of value for the following reasons. First,to establish as far as possible whether there hasbeen any fresh rheumatic activity; the occurrenceof sore throats or limb and joint pains is noted; thehaemoglobin, blood sedimentation rate and weightare recorded and a search is made for nodules. Ifthere is any question of present activity the patientsare re-admitted for assessment and treatment.

Secondly, to ensure an adequate check on anychange in the cardiac status they may haveoccurred; at follow-up clinical and radiologicalsigns in the heart may be either more or lessmarked than they were in hospital and furtheradvice on physical activity must be given. Thirdly,to re-emphasize to the patients or their parents thecontinued need for sulphonamide prophylaxis andprompt treatment of sore throats or limb or jointpains should they occur.

BIBLIOGRAPHYABRAHAMS, D. G. (I949), Brit. Heart 7., II, 342.'American Heart Association Statement' (1953), Lancet, i, 285.BESTERMAN, E. M. M., and THOMAS, G. T. (I953), Brit.

Heart J., I5, 113.TARAN, L., and SZILAGYI, N. (I947), Amer. Heart 3'., 33, 14.THOMAS, G. T., BESTERMAN, E. M. M., and HOLLMAN

A. (1953), Brit. Heart3'., 31, 29.

MITRAL STENOSIS:SELECTION OF CASES FORMITRAL VALVOTOMY

By WALTER SOMERVILLE, M.D., M.R.C.P.Cardiologist, Thoracic Surgical Unit, Harefield Hospital. Assistant, Department of Cardiology,

The Middlesex Hospital. Chief Assistant, National Heart Hospital.

When it became apparent that the stenosedmitral valve could be treated by surgery, cliniciansimmediately were faced with the problem of de-ciding which patients would benefit by operation.The more obvious indications were predicted fromthe abnormal anatomy and physiology. Otherswere arrived at in time by the expedient of trialand error. A number of important points are stillsub judice.

It was soon evident that not every person withmitral stenosis was suitable for operation. Someof the earlier cases were failures partly because ofthe newness of the technique of operating insidethe heart and partly because of clinical featureswhich today would have contraindicated operation.In each of the first four cases, all fatal, reported byBailey and his colleagues (I950), one or more ofthe following features were present: A very largeheart, mitral incompetence, advanced cardiacfailure, gross left atrial enlargement and bronchiec-tasis. The unsuitability of each of these findingswill be referred to later.The broad principles for selection laid down by

the earlier workers in this field (Bailey, et al.,

I950; Harken, et al., 1950; Baker, et al., 1950)were applied to our first patients (Bedford, et al.,1953). With experience, criteria were modifiedslightly, mainly towards including patients withfeatures which heretofore would have been re-garded as unfavourable or frank contraindications.The current basis for selection, influenced to someextent by discussion with others interested in thesubject, but mainly by our observation and ex-perience, is in close accord with the views expressedrecently by Baker and his associates (1952). Theterm ' mitral valvotomy' refers to splitting of themitral valve commissures by finger or knife; it issynonymous with 'valvulotomy' and 'com-missurotomy ' used by other writers.

SymptomsThe main indication for mitral valvotomy is

breathlessness attributable to mitral stenosis. Thisfact needs emphasis, for patients with mitralstenosis may be breathless from other causes suchas severe associated aortic valve disease or chroniclung disease.

Protected by copyright.

on Septem

ber 18, 2020 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.29.336.492 on 1 October 1953. D

ownloaded from