184 PUERPERAL INFECTION Cystoscopy has been mentioned frequently in the course of this paper as a necessary diagnostic procedure. This will very properly raise the question as to whether it is a practicable procedure in infants and young children. One can say without hesitation that cystoscopes are made of such a fine calibre that they can be safely passed on infants, without fear of damage to the parts. PUERPERAL INFECTION. By JOAN TAYLOR, B.SC., M.B., B.S.(LOND.), D.P.H. From the Obstetric Unit and Department of Bacteriology, University College Hospital, London. PUIERPERAL infection includes all abnormal conditions resulting from the entrance of organisms into the genital tract during labour or the puerperium. Puerperal infection has been recognized for many hundreds of years, and was men- tioned by both Hippocrates and Galen in their works. Later, in the seventeenth century, Willis, in Enlgland, wrote a treatise on the subject as it occurred at that time. About the middle of the nineteenth century interest in this subject was aroused, as at that time about 15 per cent. of all patients delivered in institutions died of this disease. Semmelweis, working in Vienna, showed that puerperal infection was identi- cal with wound infection. Pasteur cultivated organisms from cases, and later Lister showed the value of antiseptic methods in midwifery. From this short historical survey one realizes that it was not until the latter half of the nineteenth century that our present views on this subject began to emerge. BACTERIOLOGY. As early as the eighteenth century it was recognized that puerperal infection was con- 1Working during part of the time with the aid of a grant from the Medical Research Council. tagious. It must be remembered that at this time the death-rate from sepsis was i in 6 in institutions owing to the rapid spread of epidemics, which are controlled to-day by the universal use of antiseptic methods. The contagious nature of the disease was not urged strongly until Oliver Wendell Holmes read his classical paper at Boston in 1843. In this paper he showed that epidemiics could always be traced to lack of cleanliness on the part of the attendants in charge of cases. This work was not generally appreciated until as a result of the application of Lister's antiseptic methods there was a sudden fall in the number of cases infected. Much detailed bacteriology done during the whole of this present century goes to show that puerperal infection is due to the invasion of the genital tract by pyogenic organisms. As will be shown subsequently, any pyogenic organisms can give rise to infection. This is readily understood wlhen we consider that the large raw area in the uterus, in continu- ity, through the vagina and open cervix, with the external surface, is comparable with any raw area elsewhere in the body. Certain organisms, such as the gonococcus, are more likely to infect the uterus than a surface wound, but taking all the known pyogenic organisms, as any one may infect a surface wound, so may any one, or more than one, infect the uterus. Streptococci.-Streptococci may be classi- fied as follows : Streptococci.-Streptococci may be classified as follows:- Aerobic A ,naerobic haemolytic non-haemolytic Schotmuller, in 1903, first differentiated between hemolytic and non-haemolytic streptococci, by showing that when grown on blood-agar plates, certain strains were surrounded by a clear zone of haemolysis in which no red blood-cells were present, whereas other strains did not exhibit this property. The former are known as haxmo- lytic, and the latter as non-haemolytic, streptococci. copyright. on August 18, 2021 by guest. Protected by http://pmj.bmj.com/ Postgrad Med J: first published as 10.1136/pgmj.6.71.184 on 1 August 1931. Downloaded from
Transcript
184 PUERPERAL INFECTION
Cystoscopy has been mentioned frequentlyin the course of this paper as a necessarydiagnostic procedure. This will veryproperly raise the question as to whetherit is a practicable procedure in infants andyoung children. One can say withouthesitation that cystoscopes are made ofsuch a fine calibre that they can be safelypassed on infants, without fear of damageto the parts.
PUERPERAL INFECTION.By JOAN TAYLOR,
B.SC., M.B., B.S.(LOND.), D.P.H.
From the Obstetric Unit and Department ofBacteriology, University College Hospital, London.
PUIERPERAL infection includes all abnormalconditions resulting from the entrance oforganisms into the genital tract duringlabour or the puerperium.
Puerperal infection has been recognizedfor many hundreds of years, and was men-tioned by both Hippocrates and Galen intheir works. Later, in the seventeenthcentury, Willis, in Enlgland, wrote atreatise on the subject as it occurred at thattime. About the middle of the nineteenthcentury interest in this subject was aroused,as at that time about 15 per cent. of allpatients delivered in institutions died of thisdisease. Semmelweis, working in Vienna,showed that puerperal infection was identi-cal with wound infection. Pasteur cultivatedorganisms from cases, and later Listershowed the value of antiseptic methods inmidwifery. From this short historicalsurvey one realizes that it was not until thelatter half of the nineteenth century that ourpresent views on this subject began toemerge.
BACTERIOLOGY.As early as the eighteenth century it was
recognized that puerperal infection was con-
1Working during part of the time with the aid ofa grant from the Medical Research Council.
tagious. It must be remembered that atthis time the death-rate from sepsis was i in6 in institutions owing to the rapid spreadof epidemics, which are controlled to-dayby the universal use of antiseptic methods.The contagious nature of the disease was noturged strongly until Oliver Wendell Holmesread his classical paper at Boston in 1843. Inthis paper he showed that epidemiics couldalways be traced to lack of cleanliness onthe part of the attendants in charge of cases.
This work was not generally appreciateduntil as a result of the application of Lister'santiseptic methods there was a sudden fallin the number of cases infected. Muchdetailed bacteriology done during the wholeof this present century goes to show thatpuerperal infection is due to the invasion ofthe genital tract by pyogenic organisms. Aswill be shown subsequently, any pyogenicorganisms can give rise to infection. Thisis readily understood wlhen we consider thatthe large raw area in the uterus, in continu-ity, through the vagina and open cervix,with the external surface, is comparable withany raw area elsewhere in the body. Certainorganisms, such as the gonococcus, are morelikely to infect the uterus than a surfacewound, but taking all the known pyogenicorganisms, as any one may infect a surfacewound, so may any one, or more than one,infect the uterus.
Streptococci.-Streptococci may be classi-fied as follows :
Streptococci.-Streptococci may be classified as follows:-Aerobic A ,naerobic
haemolytic non-haemolytic
Schotmuller, in 1903, first differentiatedbetween hemolytic and non-haemolyticstreptococci, by showing that when grownon blood-agar plates, certain strains weresurrounded by a clear zone of haemolysis inwhich no red blood-cells were present,whereas other strains did not exhibit thisproperty. The former are known as haxmo-lytic, and the latter as non-haemolytic,streptococci.
copyright. on A
ugust 18, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.6.71.184 on 1 A
Since then it has been the common beliefthat haemolytic streptococci gave rise tovirulent, and non-haemolytic to mildinfection.Though this view is fairly correct in the
main, a great deal of evidence has beencollected to show that non-hemolyticstreptococcal infection may be virulent,and haemolytic streptococcal infectionmay be very mild. In this hospital wehave had forty-eight cases of haemolyticstreptococcal infection of which none havedied, and the average duration of tempera-ture has been about eight days. Also it hasbeen shown that haemolytic streptococci havebeen present inside the uterus, and thepatient has had a perfectly normal puer-perium with no rise of temperature. Thefact still remains that puerperal sepsis, dueto the haemolytic streptococcus, has ahigher mortality-rate than that due to thenon-haemolytic streptococcus.The anaerobic streptococci can be grown
only in an atmosphere free from oxygen.These frequently infect the uterus, as a rulegiving rise to mild sepsis, which runs a longcourse; a temperature of Ioo0 to 102° F. overa period of several weeks being a commonresult. The serious cases due to theseorganisms may have throinbosis of thepelvic veins. Pure cultures have beenisolated frotn the uterus, thrombi and theblood of infected patients, thus showing thatthese micro-organisms are the cause of thedisease. Recent work at Queen Charlotte's[ i ] and at this Hospital [2] goes to show thatthe anaerobic streptococci are as importantas the haemolytic streptococci in causingmorbidity.
Staphylococci.-Staphylococcus pyogenesaureus is not uncommonly found as theinfecting organism; it may give rise toserious sepsis, causing the death of thepatient.
Staphylococcus pyogenes albus may causemild sepsis; in most cases the prognosis isgood.
Staphylococcus pyogenes citreus rarely
invades the uterus, when it does so theinfection is very mild.Pneumococcus.-This group of organisms
is rarely found as the cause of infection.Figures show it to account for about o05 percent. of morbidity. In this hospital, how-ever, during the last three and a half yearstype T pneumococcus was found as theinfection organism in two cases, both ofwhich died in a few days.Gonococcus.-As would be expected, the
gonococcus frequently invaded the genitaltract. 9 per cent. of febrile cases are due tothe gonococcus, of which the larger numberrecover. A few fatal cases have beenrecorded.
Bacillus diphtheriac.-A few cases of injec-tion with this organism have been recordedwhich recovered when treated with anti-diphtheritic serum.Gas Gangrene Group of Organisrs.-This
group contains spore-bearing anaerobicbacilli, B. welchii, Vibrion septique, B. cede-miatiens, and a number of less commonorganisms. B. welchii is by far the mostimportant in causing infection of the uterus.B. welchii is more commonly found in casesof sepsis following criminal abortion thanafter normal parturition; when the organismis present in pure culture, the prognosis isgood, but when found in mixed culture thevirulence of both organisms appears to beincreased and the prognosis is doubtful. Itmay be remembered that in surface woundinfection these anaerobic bacilli can invadethe tissues only when there is massivedamage which may be traumatic, chemicalor bacterial, thus making conditions suitablefor anaerobic growth. This fact may accountfor the increased virulence of B. welchiiwhen present in association with otherorganisms. A considerable number of casesof gas gangrene arising from the uterus havebeen described, and they have generally beendue to infection with B. welchii. They oftenshow a well-marked and easily identifiableclinical picture; the patient becomes rapidlyill, develops a brick-red colour known by the
copyright. on A
ugust 18, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.6.71.184 on 1 A
German writers as ictero-cyanosis, as itappears to be due to a combination of icteruswith cyanosis due to methlemoglobin forma-tion. The patient may have haemoglobinuria.In some cases the gas in the uterus is detec-table by percussion. The prognosis in thesecases is bad. Owing to the profoundtoxaemia death may ensue within three days,the gas gangrene sometimes spreadingthroughout the body.
B. typhosus.--This organism is not usuallythe cause of infection, it is probably presentas part of the general septicaemia in earlycases of typhoid fever.
B. coli.-Owing to the close proximity ofthe anus to the genital tract, contaminationof the vagina is likely to follow any pervaginam examination, so that it is not sur-prising to find B. coli infections are verycommon, but are rarely present in pureculture. The infection usually runs a benigncourse, most patients recovering, though afew fatal cases have been recorded.Bacillary.-Other bacilli lhave been found
as the infecting organisms, including B.pyocyaneus, which has been recorded ascausing some fatal cases.
Diphtheroids have been found in theuterus in febrile cases, but they are commonlypresent in normal cases, therefore the con-clusion is that this group is of little, if any,importance in causing puerperal infection.
PATHOLOGY.
The infection may be limited to a part ofthe genital tract, or may spread involvingthe vagina, uterus, Fallopian tubes, ovariesand neighbouring structures.
Endometritis.-The commonest infectionis one involving the uterus, giving rise toendometritis. The whole of the endometriummay be affected or only a small area ; thisdepends to some degree on the virulenceof the infecting organism.
Virulent organisms invade the uterine walland.lymphatic vessels, from there they passto the peritoneal surface-of the uterus, giving
rise to peritonitis. Furthermore organismsmay spread to blood-spaces, which are stillnumerous and large in the puerperal uterus,and eventually pass into the general blood-stream, giving rise to septicamia.The macroscopic appearance of the uterus
from such a case shows little except someperitonitis and a ragged endometrium.
Microscopically the uterus is lined by athin necrotic zone, deep to this a thin,irregular zone of leucocytes, which is notvery clearly defined. Organisms are presentthroughout the uterine wall.
Mild infections of the uterus show moremacroscopically, as there is usually a verydefinite local lesion. The uterus is bulky,containing foul-smelling material. Theinterior of the uterus is ragged and lined bynecrotic tissue. Microscopically the endo-metrium is necrotic to a certain definitedepth which varies in different cases; deepto this is a definite zone of leucocytic infil-tration, inside this zone the uterine wallappears to be fairly no-rmal. Organisms areseen in the necrotic area, but not in theuterine wall.
These are the two extreme pathologicalfindings and individual cases fall into gradesbetween these two extremes.Another common finding is due to small
collections of organisms multiplying in theuterine wall and giving rise to scatteredabscesses.
Should the uterus be infected by the gasgangrene group of organisms, collections ofgas in the uterine wall are commonly found.American figures from post-mortems showthe following as the cause of death:-
Sections through the parametrium aboutthis time show a large increase in the numberof phagocytic cells in this region.
During the first few days of thepuerperium there is considerable prolifera-tion of the connective tissue in the uterus;there is also a proliferation of the tissuephagocytes, which help to dispose of theproducts of involution. Should anyorganisms be present these cells are able toingest and prevent their multiplying andinvading the uterus or surrounding tissues.Any organisms which may get beyond theuterus are phagocytic and rendered innocu-ous by the reticulo-endothelial system.
It is a well-known fact that when fresh, un-diluted blood is mixed with organisms thenallowed to stand, the mixture becomes sterilewithin a short time.
Recent work in Edinburgh shows thatthis anti-bacterial power of the blood ismuch increased during pregnancy, reachinga maximum during labour, and falling tonormal within the first ten days of thepuerperium. The importance of this pro-perty needs no further emphasis. Should itbe necessary to take blood-cultures in thepuerperium, the necessity for diluting theblood as soon as possible is obvious to all.
IETIOLOGY.
To recapitulate, puerperal infection is awound infection resulting from the intro-duction of pyogenic organisms into thegenital tract before, during or immediatelyafter labour. The infection is usually adirect infection from without and the com-monest mode of entry is on the hands,instruments or objects which come intocontact with the genital tract.
Modes of External Infection.The most usual way in ·which organisms
reach tlhe uterus is on the hands of theobstetrician or midwife. At the presenttime a number of individual cases andepidemics have been traced to whitlows and
septic foci on the hands of the attendants,thus showing that there is still need forimprovement in surgical hygiene. It is un-necessary to discuss the value of thesterilization of all instruments and dressingsduring delivery.
Droplet Infection.Air infection is a bad term, as organisms
are not free in the air but are usually presenton a nidus of dust or moisture. The termdroplet-infection suggests this nidus, &c., aswill be shown later, in puerperal infectionthe nidus is probably moisture rather thandust.The first time that the throats of students
was suggested as a possible source of infec-tion was in I1924 by two American workers.In I927 an epidemic of puerperal sepsis dueto the haemolytic streptococcus broke out inSloan Hospital, New York. Throat swabswere taken from all members of the nursingand medical staff and those which grewhaemolytic streptococci were sent off dutyuntil negative swabs were obtained. Eventu-ally no fresh cases of this infection arose inconnection with this epidemic. These factsmade one feel that there was some founda-tion for the theory that organisms present inthe throats of attendants might infect thepatient. It was in University CollegeHospital [3] in 1929 that the first case ofpuerperal infectionl was traced to an organ-ism present in the throat of the medicalattendant. The patient was a "district"case and had long labour. Two unsuccess-ful attempts at forceps delivery were madebefore she was finally admitted to hospital.The child was delivered with forceps afterconsiderable difficulty. The patient died onthe fifth day of the puerperium. Type i
pneumococci were isolated from the uterusand blood-cultures but not from the sputum.Mouth and throat secretions were examinedfrom all attendants and pneumococci wereisolated from the attendants who had madethe second attempt at forceps during deliveryin the patient's home.
copyright. on A
ugust 18, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.6.71.184 on 1 A
The following figures were published in areport from the Department of Health forScotland [4]:-13 cases of puerperal infection due to haemolytic
streptococci :-Source: 11 cases due to nose or throat of doctor.
1 case ,, septic finger of doctor.1 ,, , septic finger of patient.
2 cases of abortion infected with haemolytic strepto-cocci :
Source: 1 case due to nose of patient.1 ,, throat of doctor.
6 cases of B. coli infection :-Source: 3 due to faeces or intestine.
2 ,, urine.1 ,, septicaemia (? original lesion).
The identification of these organisms wasdone serologically.From these it will be seen that the infect-
ing organism probably came from the noseor throat of the attendant in 72 per cent. ofcases infected with the haemolytic strepto-coccus. It is obvious that masks should beworn during the delivery of every patient.The efficiency of masks is open to muchdiscussion owing to the fact that a thickmask is tiring to wear for any length of time.In this hospital a mask consisting of eightthicknesses of gauze is used, as damp masksare not as efficient as dry it is necessary tochange during a long delivery.
Blood-borne Infection.The incidence of this type of infection is
not well-known, as the primary disease isoften the more important. It is possiblethat true uterine sepsis may occur secondaryto septic foci, such as bad teeth, but theproof that infection takes place via the blood-stream is lacking.
A utogenous Infection.By autogenous infection one means infec-
tion of the genital tract or any part of it byorganisms present at delivery. The commonfinding is for organisms present in the vaginabefore or during labour to be carried intothe uterus. It may be as well to discuss herethe organisms commonly present in thevagina before and after delivery in normal
cases. A series of 250 cases had vaginalswabs taken on admission before a pervaginam examination had been made andagain on the third day of the puerperium.
Before delivery After deliveryD6derlein's bacillus ... 38'6 per cent. 4'2 per cent.Diphtheroids ... ... 21-0 , 49'8Coliform bacillus ... 24'0 , 46'7Mouth streptococcus 22'4 ,, 22-0Faecal ,, . 147 . 19'6Hoemolytic ,, ... 15 , 2-7Anaerobic ... 0 , 4*2
The figures for the incidence of theanaerobic streptococci are incorrect owingto lack of experience in factors necessary fortheir growth. There is a large increase inthe incidence of coliform bacilli afterdelivery,which suggests faecal contaminationof the vagina. The slight increase in theincidence of haemolytic streptococci is oflittle importance, as the actual number ofcases was small. No case with haemolyticstreptococci present in the vagina beforedelivery became infected.Twenty-one cases of sepsis in which the
bacteria present in the vagina before deliveryhad been fully worked out, showed thatautogenous infection had occurred in four.Two cases of B. coli infection had B. coli inthe vagina before delivery, likewise twocases of non-hzemolytic streptococcal infec-tion. These cases were very mild.
It is well known that the uterus of pregnantand non-pregnant women is sterile. Innormal cases immediately after delivery theuterus is sterile, about the third day of thepuerperium organisms can be isolated fromit in 50 per cent. of noymal cases showing nosigns of infection.
Frequency.Before the use of antiseptic methods the
mortality from puerperal infection in insti-tutions was io to 15 per cent. of all casesdelivered, these figures have dropped too'I5 per cent. As the present figures are sosmall it is common to use a morbidity rate,which attempts to give the number of casesinfected. The British Medical Association
copyright. on A
ugust 18, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.6.71.184 on 1 A
standard of morbidity is: " A temperatureabove Ioo° Fahrenheit, occurring on anytwo occasions between the end of the firstthirty-six hours and the end of the eighthday, recorded on a 4-minute thermometer,which must be placed in the mouth for aperiod not under three minutes."On looking through past records it is dis-
appointing to find that the incidence ofpuerperal infection is about the same as itwas forty years ago. This fact suggests thatthere must be a source of infection whichwas not known. It is possible that the useof masks by all attendants present at deliverywill cause a fall in the morbidity rate.
SYMPTOMS AND SIGNS.
The most important sign of sepsis is anincreased temperature or pulse or both.Very often there is little else of note, though,as one would expect, the patient usuallycomplains of headache with the raisedtemperature.When the infection is virulent the lochia
are usually decreased and the pad is practi-cally odourless. The uterus does notinvolute or else involutes slowly.With less virulent infections, where tlle
uterus is lined with necrotic material, thepads are stinking, there is an increase in thelochia. The uterus is bulky and tender.From these two extremes it will be realized
that stinking lochia is a favourable signrather than otherwise. The lochia maybecome infected in the vagina, causing theearly morning pad to have an unpleasantodour, this is found in many normalpuerperia and is not therefore a sign ofuterine sepsis.
In cases of infection due to the gono-coccus the first five to seven days of thepuerperium are usually normal, then thetemperature rises to about I00oo F., remainsup for about a week, then drops. There areoften no symptoms whatsoever. In spite ofthis the patient is often left, with a chroniccervicitis and possibly other chronic in-
fections as well. There will be furthersymptoms should the infection spread tothe broad ligament, tubes, &c.The appearance of the patient is much
the same as the appearance in any otherinfection elsewhere in the body.
In cases of haemolytic streptococcal in-fection the patient looks hectic, the pulse isfast and easily compressed. Rigors mayoccur in infection with any virulent organ-ism, they are common with haemolyticstreptococcus and B. coli.One has to be reserved in giving a prog-
nosis in cases infected with haemolyticstreptococcus, as they are very liable torelapse and have rigors even after some daysof normal temperature.
DIAGNOSIS.
In a well-conducted hospital puerperalsepsis is a rare disease and the same shouldbe true of general practice. A careful generalexamination of each patient is necessarybefore presuming her to be a case of puer-peral infection. Of 125 morbid cases onlyforty-eight were sliown to be due to genitalsepsis, therefore only one third of patientswith temperature are due to'sepsis in thishospital. In genital sepsis one includestemperature due to septic tears of theperineum. It must not be forgotten that atear with necrotic material on the surfaceand surrounding cedema is very likely tocause some rise of temperature.
Lochia.
It is not justifiable to state that a givenpatient is certainly septic until intra-uterinecultures have been made. In practice it isusually impossible to do this, but in thatcase one presumes a patient to be septic.In this hospital cultures are taken byFoulerton and Bonney's method, asfollows :"The apparatus consisted of a cotton-
copyright. on A
ugust 18, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.6.71.184 on 1 A
wool swab on a long flexible metal rodinside a curved glass tube closed at the endby a gauze plug attached to a string looselywound round the outside of the tube in itsentire length, and at the opposite end by acotton-wool plug through which the handleof the swab projects. The whole is sterilizedin a copper box. The external genitals werecarefully swabbed with antiseptics, thecervix brought into view through a speculumand then swabbed with dry sterile gauze.The glass tube was then passed through thecervical canal into the uterus and the gauzeplug dislodged, when the swab was pro-jected into the uterine cavity, withdrawninto the bent tube, removed and placed ina sterile test-tube."As few uteri are completely sterile after
delivery it is important to make cultures onsolid media in order to get some idea as tothe number of organisms present. Incultures from a septic case the growth isusually confluent, in non-infected cases theremay be a few colonies.The finding of haemolytic streptococci in
the uterus does not necessarily mean a badprognosis. The Americans show that ofeighteen patients infected with haemolyticstreptococci only one dies.
Blood-culture.
Where possible it is wise to take a blood-culture from all patients with a temperatureof over IoI° Fahrenheit. About io c.c.of blood should be taken, citrated, thendiluted with broth and incubated.One positive blood-culture is not of any
great significance, but, repeated positivesusually mean a bad prognosis. The bloodshould be taken while the temperature isrising, or, according to the textbooks,during a rigor, as it is often positive only atthese times.Where positive blood-cultures are found
together with evidence of thrombosis of thepelvic veins the prognosis is bad.
TREATMENT.The treatment will be discussed from a
bacteriological point of view, omitting allsurgical procedures.The most important point is prophylactic
treatment. To recapitulate, puerperal in-fecti6n is a wound infection, and so is dueto the introduction of organisms. From thisit follows that surgical cleanliness on thepart of the attendants is the most importantpoint. Attendanlts conducting a case shouldwear sterile gloves and rinse them inantiseptics at frequent intervals.From the figures on the number of cases
infected by nose and throat carriers ofhaemolytic streptococci it. is obvious thatmasks are always necessary.
It is most unwise to do any high vaginaldouching or swabbing with antisepticsbefore delivery, as it has been found thatat this time there are very few organismspresent there and douching increases ratherthan diminishes the number. It is importantto wash the vulva with soap and waterand then with some antiseptic before doingper vaginam examinations as one does notwish to introduce more faecal organisms thannecessary. In this hospital chloramineT-i in Ioo solution--is used, as it isthought to be more efficacious on mucoussurface than other antiseptics. From thefact that autogenous infection is rare andmild, ordinary surgical cleanliness.is all thatis needed.
It is probable that infection usually occursduring parturition. As cases may be infectedduring the puerperium, bed pans should besterilized before use; where this is impossible,separate ones should be kept for each patientand sterilized between cases.Serum.-Experimental findings go to show
that haemolytic streptococci may varyserologically, as their particular proteinmake-up may differ, but the toxin producedis the same. The amount of toxin producedby different strains varies. As the scarletstreptococcus forms more toxin than strainsisolated from other diseases it is able to
copyright. on A
ugust 18, 2021 by guest. Protected by
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.6.71.184 on 1 A
produce a more potent antiserum wheninjected into an animal. These anti-haemolytic streptococcal sera are anti-toxic,their anti-bacterial powers, however, areprobably not great.
In cases of puerperal sepsis due to thehaemolytic streptococcus it is the organismitself which causes the disease, the strepto-coccus multiplies in the uterus, invades theuterine muscle and gets into the blood-stream, the whole picture is one of invasionby the streptococcus itself. This explainswhy the use of anti-streptococcal serum hasnot been very satisfactory. The cases ofpuerperal fever in which the lesion can beascribed to the actual toxin of the strepto-coccus are very few. There was a caseadmitted to this hospital in January whichhad haetnolytic streptococci in tlhe uterusand a scarlet rash, her symptoms wereapparently due to the toxin of the strepto-coccus and she was improved by anti-streptococcal serum. The vast majority ofcases are due to invasion by the haemolyticstreptococcus, and the anti-streptococcalserum does not appear to have any verygreat effect on this property. Nevertheless,as our knowledge stands at present, in badcases of infection serum is indicated.The usual dose is 20-30 c.c., either
subcutaneously, intramuscularly or intra-venously. In bad cases the later methodshould always be used. Great care shouldbe exercised in the giving of antistrepto-coccal serum, particularly if the intravenousroute is used as the volume introduced isconsiderable. Therefore in giving anti-streptococcal serum one is giving a greatdeal of a foreign protein and is much morelikely to get toxic symptoms due to thehorse-serum itself. One must emphasizethe importance of testing the sensitivity ofany patient to the serum before giving a fulldose. Should collapse occur after the serum,o'I c.c. of adrenalin made up to i c.c. withsterile water and injected intravenously canbe relied upon to relieve drgent symptomsof an anaphylactic type. It is necessary to
have syringe and tourniquet ready in caseof accidents.
It is not justifiable to give serum unlessnecessary, as the operation itself is nevercompletely without risk, also it is unfair tosensitize a patient to horse-serum unlessabsolutely necessary.
Blood-transfusions should be given as a
prophylactic to cases in which there hasbeen ante-partum or post-partum haemor-rhage and one suspects that the patient maybecome infected. Cases of long-continuedfever usually show a lesser or greater degreeof anaemia, and blood-transfusions in thesecases are often beneficial.The use of drugs intravenously has always
been disappointing, arsenic and mercurycompounds and dyes have been tried, butwithout any encouraging results.
Vaccines have been used prophylacticallyand in the treatment of the acute disease.Figures do not suggest any good resultsfrom such treatment. There is no veryconvincing evidence from experimentalwork on animals that haemolytic strepto-coccal vaccines increase the immunity tothis organism.The introduction of sterile glycerine into
the uterus is often beneficial [5]. A rubbercatheter is introduced into the uterine cavitywith strict aseptic precautions; to thecatheter is attached a syringe containing10-20 c.c. of sterile glycerine, which is theninjected slowly. The glycerine promotes aflow of tissue fluid owing to its hygroscopicproperty, thus giving a freer uterine drainage.Experience in this hospital suggests thatthis treatment gives good results in cases ofpatients with persistent fever due to infectionlocalized to uterus and adnexa.
REFERENCES.[1] COLEBRIOOK, L. Brit. Med. Journ., 1930, ii, 134.[2] TAYLOR, J. and WRIGHT, W. D. Journ. Obstet.
and Gyncec. Brit. Emp., 1930, xxxvii, 213.[3] NIXON, J. and WRIGHT, H. D. (1929). Lancet, i,
1242.[4] SMITH, J. " Causation and Source of Infection
in Puerperal Fever," Dept. of Health forScotland. H.M.S.O., 1931.
