THE CANCER PROBLEM-WITH SPECIAL REFERENCE TO …pmj.bmj.com/content/postgradmedj/36/420/602.full.pdf · THE CANCER PROBLEM-WITH SPECIAL REFERENCE TO CANCER OF THE CERVIX ... part

602

THE CANCER PROBLEM-WITH SPECIALREFERENCE TO CANCER OF THE CERVIX

Prevalence, Special Surgical Technique and Evaluation of ResultsSUBODH MITRA, M.D.(BERLIN), F.R.C.S.(EDIN.), F.R.C.O.G., F.A.C.S., F.N.I.

Director, Chittaranjan National Cancer Research Centre, Calcutta

One of the most urgent tasks of the day is tofight cancer with knowledge. During the lasthalf-century, remarkable progress has been madein reducing the incidence and mortality rate ofsome chronic diseases like tuberculosis andnaturally the workers on cancer research feelthat in the not-too-distant future the nature ofcancer will be revealed, and the incidence and themortality rate controlled. As it stands today,except in undeveloped tropical countries, canceris No. 2 killer of human beings, No. i beingheart disease.

During the last 50 years, extensive and ex-haustive investigations have been carried out fromclinical and demographic points of view butfundamental researches through biological studieswere begun not more than a quarter of a centuryago. Research is always a co-operative enterprise.Ideas are born in individual minds but theirformulation is influenced by a wide variety ofcontacts and their translation into practicalreality is aided by the entire scientific community.

Prevalence of CancerAssessment of the prevalence of a disease is

not an easy task, specially for a vast country likeIndia, having a population of nearly four hundredmillions extending over about I.2 million squaremiles. Excepting a few principal cities, a greaterpart of the land is underdeveloped and the regis-tration for vital statistics is inadequate. Properassessment is hardly possible without samplingstudies of the whole population.The incidence of cancer is usually deduced

from hospital records, from admissions andautopsies as well as from total mortality figuresin general. Clinical assessment is often vitiatedby wide errors unless checked by biopsies whichare not always possible. Autopsies will, perhaps,give more correct information, but autopsieson every death can never be possible. Even ina well-organized country like Germany where

the post-mortem examination was obligatory inall the Statc hospitals, autopsies were made onan average of 4.3% of the total deaths of the land.7A certain amount of work on cancer mortality

in India has already been done. From theinitial subjective opinion that, as compared to theEuropean races, the Indian races were com-paratively less susceptible to carcinoma, there hasbeen a gradual shift; the analysis of hospital andautopsy records and the reported causes ofdeath in city population tend to show thatthe cancer mortality is not much lower in Indiathan in the West. But the picture is still far fromdefinite.Some investigations on these lines were made

by the author14 with the data from the city ofCalcutta. The prevalence has been broadlyequated to mortality from cancer. Before dealingwith the death records of Calcutta CorporationArea (on which this note mainly relies), a passingmention may as well be made of other relevantdata collected for this purpose. For example,out of a total of 9,130 deaths occurring among theinpatients of the Calcutta Medical College Hos-pital during 1946-50, only 249 deaths (or 2.7%)were caused by cancer; the annual rate of cancerdeaths in the Hospital actually varied between2.2% (1942) and 3.8% (I943) during this period.Autopsy records of the same Hospital for thesame period disclosed only ii cancer deaths outof 549 total cases, giving a cancer mortality ofabout 2%. Autopsy records of Rogers17 fromthe same Hospital showed a 4.59% cancer mor-tality. Data were also collected from a fewinsurance companies in India and the cancermortality was seen to vary between 2 and 3.5%among the assured. These total measures ofmortality are, however, of little validity or use forcomparative study; they are besides based onhighly select and rather scanty material. VishwaNath and Grewal20 from collected autopsy recordsof different parts of India and Burma, found the

copyright. on 12 June 2018 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

October I960 MITRA: The Cancer Problem-with Special Reference to Cancer of the Cervix 603

TABLE I

PROPORTION OF CANCER DEATHS IN 0OO DEATHS FROM ALL CAUSES(Calcutta Corporation Area, 1954 and 19"55)

Age in YearsCalendar Sex - --- AllYear 0-19 20-29 30-39 .40-49 50-59 Over 59 Ages

1954 *- .. Male 0.2 o.8 2.0 5.2 6.2 5.5 2.4(1I2.7) (2.2) (2.1I) (2.5) (2.8) (4-5) |(26.8)

1955 *. .. ,, 0. 1.7 3.0 5.7 6.8 7.1 2.9(13-4) (2.1) (2.I) (2.5) (2.8) (4-7) (27.6)1954 .. .. Female o.i I.I 4.0 8.8 9-5 4.5 2.3(12.2) (2.3) (I.5) (I.3) (I.5) (4.2) (23.0)

1955 * *0.1I . I 4.6 10.7 9.6 5-4 2.7(12.4) (I .9) (I*4) (1.z) (I.5) (4.2) (22.6)

Combined persons, 1954-55 O.1 I.I 3.2 6.9 7.6 5.7 2.6(50-4) (8.6) (7.2) (7.6) (8.6) (I7.6) (IOO.O)

TABLE 2

PROBABILITY OF DEATH FROM CANCER IN DIFFERENT AGE RANGES(Calcutta Corporation Area, 1954 and I955, and 1951 Census West Bengal Life Tables)

Age Range of Life (years)Present Age Sex . Total

(years) o-I9 20-29 30-39 40-49 50-59 Over 59

0. .... Male .001 .002 .001 .005 .oo6 .011 .026Female .ooo .001 .004 .010 .oo8 .oo8 .031

20 .. .. Male - .002 .004 .010 .012 .022 .050Female - .001 .009 .019 OI7 .015 o6 I

30 .. .. Male - .004 .011 .015 .025 .055Female .010 .021 .019 .oi8 .o68

40 .. .. Male |0- 1I 4 .oi8 .030 .o62Female - .o28 .024 .023 .075

50 .. .. Male - - .025 .040 .o65Female 3- - - .034 .032 .o666o .. .. Male - .o63 .063

Female - - - .050 .050

cancer mortality rate to be 4.2%. Khanolkar6had found 4.41% from the Bombay K.E.M.Hospital autopsy records. The percentagesgiven by both Vishwa Nath et al. and Khanolkarare to be considered as relative cancer mortalityfigures because they eliminated deaths due toinfectious and tropical diseases from the totalnumber of deaths.To enable better assessment and comparisons,

the total deaths registered with Calcutta Cor-poration were examined in detail. The proportionof uncertified deaths was very low, less than 2%of total deaths in Calcutta, whereas 98% ofdeaths were certified by qualified medical officers.Out of a total of 64,487 certified deaths (i.e.

32,113 in I954 and 32,374 in 1955), the proportionof cancer deaths came out as 2.4 and 2.8% re-spectively. The variation between the yearsappeared to be high for a cause like cancer, but

improved identification was probably responsiblefor the increased proportion in 1955. While theproportion of cancer deaths to all deaths generallyrose to a maximum in the 5o-6o age-group anddeclined after that, cancer mortality as such wassubstantially higher in the over-59 age-group(Table i).The information about the probability of death

from cancer at the various ages will be interesting.The probabilities found by fusing 1951 WestBengal Census Life Table'6 mortality with Cal-cutta Corporation cancer death proportions inage-groups have been worked out. (Table 2.)

People at large will be more interested in know-ing their own chances of escaping or dying ofcancer than in the mean annual mortality from it.The probability of death from cancer was of thelevel of 2.8% at birth, and this rose to the level of6.8% by the time age 40 was attained. The



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

604 POSTGRADUATE MEDICAL JOURNAL October I960

60

50 INDI* IML9I ANUtICAN MALE -

40

~30

XX,'/..20 ./

'10,-X

0 20 30 40 50 60 0Age it yers

FIG. i.-Percentage distribution of cancer deathsby age: comparison of Indian and Americanstatistics (males only).Indian males (Calcutta Corporation records,I954 and 1955).American males (Steiner, 1954).

probability of death from cancer was higher forthe female than for the male up to age 40; theposition was reversed by age 6o.The percentage distribution of cancer deaths

by age for the Calcutta Corporation area wascompared with similar distribution for theU.S.A. data given by Steiner.18 The comparisonis presented graphically in Figure i for males.The age distribution patterns of cancer showedgood agreement between the two countries, ex-cepting at very advanced ages. The age dis-tribution of cancer of the female genital organswas also similarly compared; the patterns of agedistribution of this particular site disclosedstrikingly close agreement, as Figure 2 demon-strates. The patterns could have such closeresemblance as the defects of diagnosis and re-porting, which pulled down the totals, werebasically independent of the age of the subject.

An Estimate of the Prevalence of Cancer inIndia

Since the extent of prevalence of cancer inIndia was not known. a dimensional estimatebased on the Calcutta investigation is attemptedbelow. A number of broad assumptions had tobe made and the estimate is somewhat speculative.Yet it is clear that in view of the grave defects inthe sphere of diagnosis and reporting, even a veryrough and tentative estimate indicating thelowest level of prevalence will be useful as astarting point.The first important assumption was that the

Calcutta Corporation cancer death proportionscould be taken as representative of the country as

60 -

50 INDIAN50- ~ AMERICANm

40

at ,1 2010 - -

0 30 40 50 60 70Age in years

FIG. 2.-Cancer of the female genitale or-gans. Percentage distribution of deathsby age: comparison of Indian andAmerican statistics.Indian data (Calcutta Corporation re-cords, 1954 and I955).American data (Steiner, 1954).

a whole. Other assumptions were about thecurrent level of total mortality in the countryand the distribution of life with cancer overduration since onset.

In any region, the community habits are notbasically different between the rural and the majorbulk of the urban sector; nor are any other sharpdifferences between the rural and the urbanenvironmental factors known which might intro-duce a significant disparity in carcinogenic dosage,except perhaps in relation to the cancer of thelung. Lung cancer is of very minor frequencyeven in the Calcutta Corporation area. TheCalcutta experience could therefore be repre-sentative of the country in this respect, particularlyas it is a big cosmopolitan city inhabited by peoplefrom many parts of India; and the migrantsusually carry their habit-patterns with them.

But owing to the location of specialised cancerinstitutions and hospital departments, and be-cause of the availability of advanced surgicalfacilities generally, more cancer cases come toCalcutta; the question therefore was whetherthis eventuality resulted in a higher proportionof cancer deaths in Calcutta Corporation area.Few cases died while on the operation table orwhile being treated actively and not able to moveaway. With no hope for recovery, people preferto return to their own home surroundings in theirlast days, specially in an affliction like cancerwhen there is nothing more to be done. Generalreasoning therefore warrants the assumption thatthe Calcutta Corporation deaths will not be in-flated by migrated cancer cases and informedclinical opinion is not to the contrary. We there-



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

October I960 MITRA: The Cancer Problem-wth Special Reference to Cancer of the Cervix 605

T'' v., .)'u:I2i\;PStI|t0. '

fler-lT (t)LSlEiS-~rsFIG. 3.-Survival with cancer. (Estimated on the basis of the author's

experience.)

fore consider the assumption valid that the cancerdeath proportions of the Corporation area holdgenerally for the country.Adopting the current national death rate as 20

per thousand broadly, the total annual deathsfrom all causes will number 7.6 millions in theestimated 1955 Indian population of 380 millions;applying the Corporation cancer death pro-portions, the estimated cancer deaths numberO.I98 or o.2 million roughly.The next problem was to relate the cancer

death proportion to the prevalence rate. Understationary conditions, the rate of incidence will beequal to the death rate plus the cure rate. Thecure rate in the case of cancer being small for thecountry as a whole, the cancer death rate may betaken as the rate of cancer incidence for thepurpose of this rough approximation; the rateof incidence will actually be slightly higher.The relationship between the rate of incidence

and the rate of prevalence could be discussed withadvantage on a theoretical plane at this stage. If ibe the annual rate of incidence and f(t) thefrequency distribution of cases living with cancerover duration t since onset, so that f(o) = i andf(ax)= o, then P the rate of prevalence is givenunder stationary conditions by the area integral,

00

P = i f f(t). dt.

0

While the mathematical relationship betweenincidence and prevalence given above is quitestraightforward, all its implications are not. Forexample, if cancer of different organs were con-sidered separately, it will be found that the organprevalence rates will not be proportional to therelative incidence rates; but will differ according

to the average duration of life with cancer for theparticular organs.The duration of life with cancer varies with the

site and type of cancer. But it is not possible tointroduce any refinement for this variation in thepresent attempt at first approximation. Theaverage duration of life with cancer even forall cancers taken together is not known firmly forany country, much less for India. A broadassumption was again made on the basis of ourexperience; and the shape of frequency distri-bution of life with cancer was taken as in Figure3; if such a distribution holds, the averageduration of life with cancer is about 3 years, andthe prevalence of cancer in India amounts too.6 million in round figures.

Personally I feel that the gaps in diagnosis andin reporting are very substantial and the pre-valence of cancer in India is at least three times asmuch as has been estimated, i.e. i.8 million. Thisstatement may be substantiated by the fact thatthough it is not definite that vulnerability to canceris the same in India as in say U.K., age-specificCalcutta-reported cancer death rates applied to theU.K. age structure produce an overall cancer deathproportion of 7.6% (of total deaths), whereas theactual reported proportion in U.K. (I953-54) is17.9%. On the other hand, age-specific cancerdeath rates of U.K. (I953-54) applied to theCalcutta age structure produced an overall cancerdeath proportion of 6.5%, instead of 2.6%, as hasbeen actually registered in I954 to 1955. Hencemy assumption that the prevalence of cancer inIndia is at least I.8 millions.

Prevalence of Cancer of the CervixAlthough the incidence of cancer does not vary

much in different countries of the world, it hasbeen well established that the anatomical dis-



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

6o6 POSTGRADUATE MEDICAL JOURNAL October I960

TABLE 3INCIDENCE OF CERVICAL CANCER ON ETHNOLOGICAL BASIS-CHITTARANJAN SEVA SADAN COLLEGE OF OBSTETRICS,

GYNAECOLOGY AND CHILD HEALTH

I940-50

Religious Group Total Number ofGynaecological Cases Cancer Cervix Cases Percentage*

Hindu .. .. 89,271 569 o.637 ± 0.03Muslim .. .. 4,556 20 0.439 + 0.10Others .. .. 550 4

94,377 593* Difference between the Hindu and Muslim is not statistically significant, the value being

O.I98 + O.IO2.

TABLE 4INCIDENCE OF CERVICAL CANCER ON ETHNOLOGICAL BASIS-CHITTARANJAN CANCER HOSPITAL (1950-55)

Total Gynaecological Cases Attended Cancer of the Cervix DetectedYear

Hindu Muslim Others Hindu Muslim Others

I950 .. .. 487 8 5 228 4 3(59. I) (50.0)

1951 .. .. 755 I9 9 396 IO 3(52.4) (52.6)1952 .. .. 797 25 I2 379 i6 7(47.6) (64-0)

1953 . 747 31 10 328 15 2

|(439) (48.4)I954 .. .. 794 37 9 341 25 3

(42.9) (67.5)I955 .. .. 908 56 15 369 32 II(40.6) (57.1)

Total .. 4,488 I76 6o 2,101 102 29(47.8 ± 0.74)% (56.7 ± 4.46)%

Difference between Hindus and Muslims is statistically significant at 5% level, the value being 8.9 ± 4.52.

7:'"

* t -.......S ....

......'I,;@ *7., ~

3.'y'SS'

FIG. 4.-Comparative incidence of cancer in varioussites: Cancer Hospital of Calcutta, Calcutta, India,and the Royal Marsden Hospital, London, England.

tribution of cancer is not the same everywhere.The adjoining percentage distribution diagram,(Fig. 4) shows the comparative incidence of cancerin different sites of two institutions, namely, theChittaranjan Cancer Hospital of Calcutta and theRoyal Marsden Hospital of London. The pre-ponderance of oral, laryngeal and cervical cancer isvery definitely marked in India. During theperiod of five years from 1954 to 1958, 4,258female malignant cases were detected having cancerin different parts of the body of which 483 (I I *3%)were cancer of the breast, 2,420 (56.8%) of thefemale genitalia, 2,135 (50.I%) of the cervix, andonly 6i (I.4%) of the body of the uterus. Thus,for each cancer of the breast, there are fourcancers of the cervix and for each cancer of thebody of the uterus, there are 35 cases of cancer ofthe cervix.

Cancer of the cervix was believed to be rareamong the Mohammedan women but the experi-ence of the author is otherwise. The comparativeincidence of cancer of the cervix between the



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/


Hindus and the Mohammedans has been workedout by the author. Tables 3 and 4 indicate thatthe incidence is practically the same between thetwo communities, the difference noted not beingstatistically significant. This leads us to presumethat it is not the ethnological difference but thehabits and environments, social status and assaultson the cervix due to repeated child births whichare responsible for the frequency of cancer of thecervix in one or the other community.

Recently there has been a great move all overthe world to screen adult women free from anysuggestive symptoms of malignancy with a view tofinding out cases of early cancer. We havestarted similar type of work here at the ChittaranjanSeva Sadan and Chittaranjan Cancer Hospital witha view to finding out the incidence of unsuspectedcancer of the cervix.The number of women screened was 3,458, of

which 2,966 had some gynaecological symptomsnot suggestive of malignancy, 149 were withoutany symptom, 3 i6 and 27 being in the early andlate months of gestation respectively.

Altogether, 9,044 smears were taken, 6,9I6 beingprimary and the rest repeat.

Cancer of the cervix was detected by smeartechnique and confirmed by biopsy in i8 cases,having a prevalence rate of 0.52%, I2 of thembeing invasive and the rest intraepithelial. If onlythe invasive cases are taken into consideration, theprevalence rate becomes 0.34%. This percentagemeets with the general agreement with otherworkers in different parts of the globe.

Since the I958 Congress of the InternationalFederation of Obstetrics and Gynaecology, theclassification of cancer of the cervix has beenaltered. Stage 0, i.e. non-invasive intraepithelialcancer, is no longer classified under the generalclassification in cancer of the cervix. The classi-fication as modified and accepted by that meetingis as follows:

Stage I. The carcinoma is strictly confinedto the cervix.Group Ia-Preclinical carcinoma or less than i cm.

in diameter.Group Ib-A carcinoma i to 5 cm. in diameter.Group Ic-A carcinoma more than 5 cm. in

diameter. The growth has extendedto the corpus.

Stage II. The carcinoma involves the vaginabut not the lower third. The carcinoma infiltratesthe parametrium but has not reached the pelvicwall.Group IIa-Carcinoma encroaching upon the

vaginal wall, but not obviously in-filtrating the parametrium.

Group IIb-Carcinoma infiltrating the para-metrium.

Stage III. The carcinoma involves the lowerthird of the vagina. The carcinomatous infiltra-tion of the parametrium has extended into thepelvic wall, when on rectal examination the in-filtration feels firm and nodular and there is nosmooth ' cancer-free ' space between the tumourand the pelvic wall. A swelling on the pelvic wall,not attached to the tumour, should not be con-sidered in the staging.

Stage IV. The carcinoma involves the bladderor the rectum, or has extended outside the truepelvis, i.e. below the vaginal inlet or above thepelvic brim (distant metastases).Group IVa-Clinical invasion of the bladder or

rectum.Group IVb-Histologically proved invasion of the

bladder or rectum, ulceration orfistulae. Extension of the carcinomaoutside the true pelvis.

Only cases of invasive carcinoma should beincluded. The staging should be performed priorto the treatment and should not be changed. Itis desirable that the staging should be carried outunder anaesthesia. Findings at urography orvenography or at surgery should not influence thestaging.

I have already pointed out that cancer of thebody of the uterus is very rare in my series, theratio between the cancer of the body of the uterusto that of the cervix being i: 35 So far as myinformation goes, cancer of the body of the uterusis rare all over India, although in Western countriesand even in Japan the incidence is muchhigher.

In order to ascertain the incidence of unsus-pected cancer of the body of the uterus amongadult females above the age of 35, 2,037 patientsfrom the polyclinic of the Chittaranjan SevaSadan were screened by taking vaginal andintrauterine aspiration smears. Altogether 6,394smears were taken including repeat examinations.Smear was positive in ten cases and endometrialbiopsy proved cancer of the body of the uterus innine, thus giving a prevalence rate of o.44°%. Ithas been noted previously that screening of anunsuspected group for cancer of the cervix gaveus a prevalence of 0.34%, showing thereby thatamongst the unsuspected group with minimalsymptoms the prevalence rate of the cancer of thecervix and cancer of the body of the uterus isalmost the same. It will have to be worked outwhy at a later stage, cases of the cancer of the bodydo not present themselves for treatment. Itmight be that the growth of the cancer in the bodyof the uterus is less rapid than in the cervix andbefore the growth is well established, a goodnumber of these patients die of some otherdisease.



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

6o8 POSTGRADUATE MEDICAL JOURNAL October I960

Special Surgical TechniqueI present a new surgical operation for the

treatment of cancer of the cervix with the five-yearend-results.

In I931, when I submitted my five-yearstatistical end-results of cancer of the cervixtreated with radiation therapy, at the Third Inter-national Radiological Congress, I was very muchimpressed to find continental surgeons performingradical vaginal hysterectomies in similar cases.

Radiation therapy has already been accepted byus but surgical treatment has been reintroducedand is gaining popularity because of the limitationsand hazards of radiation therapy.

I was more or less satisfied with Schauta'soperation, which I had been doing for more than aquarter of a century; but my complacency had agood shaking when I presented the results of-Schauta's operation at the Royal Society ofMedicine in I947. The only criticism I had toface was that in Schauta's operation pelvic nodeswere ignored. Since then I have worked longhours in the Anatomical Dissection Hall and in thepost-mortem room with a view to making theSchauta's operation a complete one. During thisperiod I had an opportunity of making anatomicaldissections of the different layers of pelvic tissueswith Prof. Amreich of Vienna on differentoccasions.The operation starts with the extraperitoneal

dissection of pelvic nodes, ligation of ovarian anduterine vessels and partial mobilisation of para-metria and finally ends with the radical vaginalhysterectomy with massive removal of parametria.

This new operation has stood the test of timenot only in my hands but also in the hands ofother experts including my colleagues andassistants. It has been demonstrated in manyuniversity clinics here and abroad, and I had thegreatest satisfaction when in I952 I demonstratedthis operation at the Schauta Klinik of the Uni-versity of Vienna through the courtesy of theDirector, Professor Antoine. Incidentally, thepatient I operated on at the Schauta Klinik has-passed through a five-year check-up and is doingwell.The criticism has been made that my operation

requires three cuts.8 The two inguinal incisi'onscould have been reduced to one by making aPfannenstiel's incision but that would only cut-more healthy tissues unnecessarily.

In some of Stage II (vagina) cases, where thegrowth has extended down in the vagina and ofStage III cases where there is a metastatic isolatedgrowth in the lower part of the vagina, extensivedissection of the vagina will be necessary and thiscan better be done by the vaginal route. Ab-domino-perineal operation is not a new thing in

cancer surgery, especially for cancer of therectum; and lately for advanced cervical cancer inBrunschwig's exenteration operations. Even inWertheim's operation, in cases where the lowerthird of the vagina is involved in carcinoma,Wertheim used to advocate ' the pull-through '

method by extensive dissection of the vaginalintroitus and extraction of the specimen frombelow.21

It was stated some time back that surgery hasreached its limits in dealing with cancer of thecervix and there was no scope for further im-provement. This statement was made whenradiation therapy for cervical cancer was at the'height of its glory.

Curiously enough, the surgical treatment hasbeen reintroduced with great improvements. Ithas gained ground not only for the treatment ofStage I and II cases but also for Stage III andeven Stage IV cases.The liAmits of radical abdominal or radical

vaginal hysterectomies for cervical cancer havebeen extended today beyond those originally de-signed by Wertheim and Schauta. There arenewer approaches to make the Schauta's operationcomparable to Wertheim's and new addenda topelvic surgery, namely, the exenteration opera-tion for advanced cancer of the cervix.

It is not known whether the same difference ofopinion exists today about the radical abdominaland radical vaginal operations as when bothWertheim and Schauta did their operations inVienna, towards the end of the igth century, ascontemporary gynaecological surgeons. Theiroverall results did not show any remarkabledifference.

Since then, the radical operations of bothvarieties became popular all over Europe. Infact, in some of the clinics, one division used to doWertheim's operation and the other division theSchauta's operation. The end-results of boththese operations were almost identical.

In the United Kingdom, Victor Bonney, themaster-surgeon of the time, took up Wertheim'soperation. His technique was superb and end-results quite satisfactory, so that it became thetreatment of choice among British gynaecologists.The orthodoxy of some British gynaecologistssometimes becomes obvious. Dr. Aleck W.Bourne3 remarks: ' The Schauta extended vaginalhysterectomy will not be considered . . . since itis almost never done in this country'. Theresults of Walter Stoeckel,19 who was a contem-porary of Victor Bonney,2 indicate that the Britishgynaecologists would not have been much worseoff had they followed Schauta's radical hysterec-tomy.

During the last two decades, Dr. Meigs has



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/


revived Wertheim's operation and made itpopular among American gynaecologists. It isapparent that the influence of Ries and Clarkwas in the background. Before he took upWertheim's operation in real earnest, he consultedVictor Bonney of England and Warnekros ofDresden. Can it not be guessed that if, instead,he happened to consult Stoeckel in Berlin andAmreich in Vienna, Dr. Meigs might haveadopted the Schauta operation and through hisperfect surgical acumen and dexterity given us anequally good model series minus the compli-cations from his Wertheim series?

I take the liberty of making such a statementbecause of my continued practice of the Schautaoperation for the last 28 years.

While practising the Schauta operation, I hadto face criticistn from many quarters. Thepresent surgical approach, namely, ExtraperitonealPelvic Lymphadenectomy and Radical VaginalHysterectomy, is the outcome of the reactionswhich I had to face. It can be reasonably saidthat this new operation has by now proved itsworth as evidenced by its effective massive re-moval of the cancer-bearing tissues as well as bysatisfactory five-year end-results.

I have a feeling that this operation can be donewith success and with sufficient security in alltypes of patients irrespective of age, weight andgeneral nutrition. There is only one proviso: thesurgeon must not only know the surgical anatomyof the pelvis but also be thoroughly trained inpelvic surgery. This kind of surgery is notmeant for occasional surgeons but for those whohave taken it as their life study, who are regularlyen'gaged in practising this very type of surgeryand who in their spare moments gather moreexperience of pelvic structures from the anatomyhall or the post-mortem room. Thus an operatoracquires a vivid clear cut mental picture of differentpelvic structures which he will have to handlewhile doing this operation. While operating, theoperator actually feels as if he is turning over the-'pages of the surgical anatomy and the completionof each operation is a reward by itself.

It is a moot question why I took up the surgicaltreatment in addition to radiation therapy. Tobegin with, I used to treat all cases with radio-therapy as was usually the custom during theearlier part of this century. My first five-yearend-results, though not very encouraging, werepresented at the Third International RadiologicalCongress in Paris in I93I. It was i6.6%.9 Afterthat I took up surgery along with radiotherapyfor the treatment of cancer of the cervix. Theresults of radiotherapy improved no doubt incourse of time, and with the addition of supervolttherapy, but the results' of surgical treatment were

definitely better than radiotherapy. Table 3shows the comparative five-year end-results ofStages I and II cases during the years 195o and195i. They are 47.4% and 65.6% for radio-therapy and surgery respectively (Table 5).

TABLE 5CANCER OF THE CERVIX, STAGES I AND IICOMPARATIVE FIVE-YEAR END-RESULTS

SurgicalRadiotherapy Treatment

Years Total TotalNo. Cure No. Cure

Treated Rate Treated Rate

1950and 1951 78 47.4% 32 65.6%

I have not yet been in a position to select mycases on the basis of radio-resistance and radiationresponse. We have tried to assess the radiationresponse of cervical cancer cases by studying thereaction of different types of cells (resting, mitotic,differentiating and degenerative) from 'youngtumour foci' after the first radiation exposure(7,ooor). Our results do not correspond withthose of Glucksmann and Spear.4 In our seriesthe radiation response has been found much thesame in all cases at the end of the first week afterradiation. The percentage count showed anincrease in the proportion of degenerating anddifferentiating cells and a correspondiing decreasein the proportion of mitotic and resting cellsuniformly in all cases.1' Hence we could notpredict radio-resistance after initial radiation inthe way suggested by Glucksmann and Spear.4Heyman5 does not believe that any group of

cervical cancer is specially radio-resistant but heconsiders that some tumours are more malignantthan others. This varying degree of malignancyis detected from clinical experience and cannot besubstantiated by microscopical evidence. Althoughvaluable contributions have been made byGlucksmann and Ruth Graham in this direction,the real clue to the varying degree of malignancyis yet to be found.Under the circumstances, the selection of my

cases for surgical treatment was based on oper-ability. I was not particular in selecting a modelseries of thin, comparatively young women withan ideal picture of health and having a smallcircumscribed growth in the cervix. My caseswere recruited both from thin and obese patients,and from Stages I, II and III groups. Mal-nutrition was rather a marked feature in most ofmy cases. Only those cases were taken up whowillingly sulzpitted themselves for operation, theremaining cases being treated by radiotherapy.



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

6io POSTGRADUATE MEDICAL JOURNAL October I 960

In I932, I started surgical treatment in additionto radiation therapy. Prior to this period,Wertheim's radical abdominal operation wastried with an enormous primary mortality. Thiswas partly due to the devitalised conditions ofpatients undergoing operation. Anaemia, lowblood volume and a hyposthenic condition wereusual in such cases. So, instead of taking up theradical abdominal operation, I preferred theradical vaginal method. Although antibioticsand sulfa drugs were not introduced early in theseries and blood for transfusion was not easilyavailable, I was much impressed with a remark-ably low primary mortality,10 which in the last105 cases was nil, although taking all the casestogethe'r it was 2.8%.12During the last io years, a total number of

2I6 cases of cancer of the cervix have beenoperated on by the new technique, although thetotal number of radical vaginal hysterectomies sofar done here and abroad amounts to 450. Out of216 cases dealt with by the new technique, I94cases were operated by the author and the restby his colleagues and assistants. When classifiedstage by stage, II4 belonged to Stage I, 8o toStage II and 22 to Stage III.The primary mortality of my new series was

3.6%. Unfortunately, we had a few accidentaldeaths due to malignant tertian malaria, haemolysisafter blood transfusion and opium poisoning. Ifthese are excluded, the corrected mortality be-comes I %. There was no primary mortality forthe last 55 cases.

Post-operative bladder troubles were remarkablyfew, the residual urine being reduced to about 2 OZ.within an average of IO-I2 days.

There was no ureteric fistula in this series. Twopatients had vesico-vaginal fistulae, both havinghad previous radium treatment in another hospital.

Associated Pelvic Lesions such as of tubo-ovarianmasses, ovarian or pelvic endometriosis or uterinetumours are said to cause difficulties in radicalvaginal operations. Out of 450 radical vaginalhysterectomies done by me including the newseries, tubo-ovarian masses were encountered inI8 cases, ovarian endometriosis in 5, uterinefibroids in 7, pyometra in I3 cases and uterinepregnancy in 4 cases. Table 6 gives the detailedlist of associated pelvic lesions. Fig. 5 shows acancer of the cervix Stage II with pregnancy ofi6 weeks' duration. There was no difficulty inremoving these associated secondaries along withthe malignant uterus by the vaginal route. Withmy new technique, removal of these associated,pelvic lesions has been greatly facilitated by theextraperitoneal ligature of ovarian and uterinevessels.

TABLE 6PELVIC LESIONS ASSOCIATED WITH CERVICAL CANCER

CASES OPERATED

Total No. of Operations Associated Pelvic Lesions

Radical vaginal hys- Multiple fibroids .. 7terectomies (new Tubo-ovarian masses.. i8series included) .. 450 Ovarian endometriosis 5

Hydrosalpinx .. .. 3Pyometra .. I 3Dermoid cyst .. IOvarian cyst .. ..Other conditions .. 7Pregnancy .. .. 4

Pelvic lymph nodes are removed as far as practi-cable-mostly by the extraperitoneal route andsome (paraureteric nodes) by the vaginal route.It is not possible to remove all the lymph nodes,big or small, from the pelvis. That is why,perhaps, Wertheim used to remove only thepalpable nodes and Schauta did not attempt toremove any. Amreich,' quoting a study of pelvicnodes by Riffenstuhl, has shown that there are atleast two inaccessible groups of glands in con-nection with the cervix, namely, (a) glands situatedat the most lateral part of Mackenrodt's ligament,the connecting lymph vessels accompanying theinferior gluteal artery and the pudendal vessels,and (b) pararectal glands which remain under therectal fascia and whose lymph channels accompanythe haemorrhoidal vessels. These two groupscannot possibly be removed by any operation, norcould they be effectively sterilized by radiation.I had difficulties in two cases, each belonging toStage I. In both these cases, none of the glandsremoved showed any metastatic cancer cells; butwithin six months after the radical operation, anodular growth was found in each case in thepelvis in the neighbourhood of the ischial spine.This might be from one of the inaccessible groupsof lymph nodes cited by Amreich.

Systematic removal of all the detectable lymphnodes was done in each of the whole series of 2I6cases. Table 7 gives the incidence of cancer-positive nodes. In 22.6% of the total number ofcases, positive nodes were detected, which breaksup into 15.8% in Stage I, 27.5% in Stage II and41% in Stage III.

Further analysis shows that the internal iliacand obturator nodes were most commonlyaffected. Recently I have started removing thelower lumbar group of glands (para-aortic) byextending upwards the left extraperitoneal in-cision. No evidence of metastases in this para-aortic group of glands was found in 25 cases so faroperated on, except in two cases, belonging one toStage III and another to Stage II. The lattercase, although it belonged to the clinical Stage II,



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

October I960 MITRA: The Cancer Problem-with Special Reference to Cancer of the Cervix 6i

. -.-=/1-

t\y SK

p7"d:'ji1..................t..

FIG. 5.-Cancer of the cervix associated with pregnancy of i6 weeks' dura-tion. Removed by radical vaginal hysterectomy.

had nine positive nodes out of I3 glandsremoved.

TABLE 7INCIDENCE OF CANCER-POSITIVE NODES

Cancer-Stages No. of Positive Percentage

Cases Glands

I, II and III .. 2I6 49 22.6I ..1 .. 14 i8 I5.8II .. .. 8o 22 27.5III .. 22 9 41 .0

Parametrial Infiltration.-It has been found byserial sections of the parametrium that only in32 out of 2i6 cases were the parametria infiltratedwith cancer celsk.4ither in the proximal, middleor in the distal thi-rd from the uterine side. Thusparametria were found infiltrated with cancer cellsin 14.7% of the total number of cases operatedon while the nodal involvement was found in

22.6% in the same series (Table 8). Furtheranalysis of our data indicates that in only 14 caseswere both the parametrial and the pelvic nodesinfiltrated simultaneously. This shows that what-ever may be the clinical staging, one is not justi-fied in limiting the surgery in any stage of cancerof the cervix except, of course, in Stage 0 whichhas not been included in this series.

Evaluation of Five-Year End-ResultsThe acid test of a new technique is shown not

only by the extensive and successful removal of the

TABLE 8PARAMETRIUM INFILTRATION AND PELVIC NODE

METASTASIS

No. of Cases Parametrium Pelvic NodeOperated Infiltrated Metastasis

zI6 32 49(14-7%) (22.6%)



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

6I2z POSTGRADUATE MEDICAL JOURNAL October I960

TABLE 9

FIVE-YEAR END-RESULTS OF CANCER OF THE CERvix TREATED SURGICALLY BY MITRA'STECHNIQUE

Five-yearSalvage

Year under Review Total No. Stages Lost Sight ofNo. Percentage

1949-53 '57 , II and III 35 6|.4 5.5%38 I 35 65912 II7 58.3f

_3 .8

TABLE IO

FIvE-YEAR END-RESULTS OF CANCER OF THE CERVIXTREATED SURGICALLY BY MITRA'S TECHNIQUE

I ear 1iIVE-YEAR SALVAGEunder Total StagesReview No. No. Percentage

1949-5I 20 I, II and III 14 70i6 I andII I2 75

cancer bearing tissues but also by the evaluation offive-year end-results. The materials at hand arenot large but may be taken to be sufficientlyrepresentative. During the years I949 to 1953,57 cases belonging to Stages I, II and III wereoperated on by the new technique, yielding anoverall five-year cure rate of 6 .4%. By breakingdown the results, 64% has been salvaged fromStages I and II cases (Table 9). Stage I results of1952 have been vitiated by a few accidental deathsby malignant tertian malaria, opium poisoning andhaemolysis after transfusion of blood.

Five-year end-results were satisfactory in thefirst series covering the period I949 to I95I,yielding 75% five-year cure in Stages I and II(Table io).Comparing the results of the present series

treated by the new technique with those of myprevious cases treated by Schauta's operation, itcan be seen that the results of the present series(6i.40%) are better than those of the older one(44R5/%) (Table ii).

Five-year end-results have also been calculatedon the basis of cancer-positive and cancer-negative

pelvic nodes in the new series. There was 73.8%five-year cure rate in gland-negative cases against26% in gland-positive ones (Table I2). In theprevious series of 1949-5I, five-year salvage ofgland-negative cases was 84.6%.13 Unfortunately,the percentage of gland-negative cases has beenvitiated by a new accidental death in 1952.Prognosis in gland-metastasis cases is unfavour-able. But the result that even in cases withmetastases 26% had a five-year cure, whenregional lymph nodes were sufficiently excised,encourages the surgeons and shows the im-portance of lymphadenectomy.

Supervolt radiotherapy is given as a routinetreatment to these cases where there is parametrialor nodal metastasis.

ConclusionThe rational conclusion I can draw from my

experience is that this new operation satisfies allthe essential conditions for the radical surgery ofcancer of the cervix. This operation can be donewith reasonable safety even in the aged, com-paratively bad surgical risk cases and in heavycorpulent patients. Besides, there is an ad-ditional advantage of the rehabilitation of theprolapsed bladder after the extensive removal ofthe vagina. There are fewer post-operative com-plications, both immediate and remote. There isno ureteric fistula, and bladder troubles areminimal. The end-results are satisfactory.Stump carcinoma, vaginal carcinoma and cancer

TABLE I I

COMPARATIVE FIVE-YEAR END-RESULTS OF CANCER OF THE CERVIX TREATED BY MITRA'S TECHNIQUEAND BY THE SCHAUTA OPERATION

Fivc-yearTotal No. Salvage

Years under Review Type of Operatlon and Stages Lost Sight ofNo. Percentage

I943-51 *. .. Schauta technique 74 33 44.5I, II and III

I949-53 ..I. Mitra technique |,7 35 61.4 5.5%______________ _____________ I, IIand III _ _ _ _ _ _ _ _ _ _ _ _



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/


TABLE 12FIVE-YEAR END-RESULTS IN CANCER-POSITIVE LYMPH NODES

Pelvic Nodes

Type of Negative Positive

Year Operation Five-year Salvage Five-year SalvageTotal TotalNo. No. Percentage No. No. Percentage

1949-53 * Mitra technique 42 31 73.8 15 4 26

of the body of the uterus can be electively operatedon by this new technique.

REFERENCESI. AMREICH, A. J. (1955), 'Klinik und operative Behandlung

des Uterus Karzinoms: Biologie und Pathologie des Weibes'.Vienna: Urban.

2. BONNEY, V. (1932), Amer. 7. Obstet. Gynec., 30, 8i5.3. BOURNE, A. W. (I958), 'Recent Advances in Obstetrics and

Gynaecology'. London: Churchill.4. GLUCKSMANN, A. (194i), Brit. J. Radiol., I4, I87.S. HEYMAN, J. (i95o), 'Modern Trends in Obstetrics and

Gynaecology'. London: Butterworth.6. KHANOLKAR, V. R. (1945), Indian J. med. Res., 33, 2.7. LUBARSCH, 0. (I9I8), cited by Koujitzecy, Munch. med.

Wschr.8. MEIGS, J. V. (I954), ' Surgical Treatment of Cancer of the

Cervix'. New York: (Grune & Stratton.9. MITRA, S. (I932), Brit. J. Radiol (N.S.), 58I.

so. MITRA, S. (1939), J. Obstet. Gynaec. Brit. Emp., 45, 6, 1003.i T. MITRA, S. (X954), Brit. J. Cancer, 8, 107.I2. MITRA, S. (i955),.J. Obstet. Gynaec. Brit. Emp., 62, 6.13. MITRA, S. (1957), The Congress Volume of the ist Asiatic

Congress of Obstetrics and Gynaecology, Tokyo, Japan.14. MITRA, S. (1959), Bull. Wid Hlth Org., Article 42, Nov. *II.IS. MITRA, S. (I96o), ' Mitra Operation for Cancer of the Cervix

Springfield: Charles C. Thomas, Publisher.i6. REGISTRAR-GENERAL, 'India and ex-Officio Census

Commissioner for India: Census of India, 195I '. Delhi:Manager of Publications.

I7. ROGERS, L. (1925), Glasg. med. J7., 103, T.i8. STEINER, P. E. (1954), 'Cancer: Race and Geography'.

Baltimore: The Williams & Wilkins Co.I9. STOECKEL, W. (1947), 'Lehrbuch der Gynakologie'.

Leipzig: Hirzel.20. VISHWA NATH (I935), IndianJ_. med. Res., 23, I49.21. WERNER, P. (sI94), ' D.e Wertheimsche Radikaloperation bei

Carcinoma Colli Uteri '. Vienna: Urban.

DI



j.bmj.com

/P


j.36.420.602 on 1 October 1960. D

ownloaded from

http://pmj.bmj.com/

Documents

THE CANCER PROBLEM-WITH SPECIAL REFERENCE TO …pmj.bmj.com/content/postgradmedj/36/420/602.full.pdf · THE CANCER PROBLEM-WITH SPECIAL REFERENCE TO CANCER OF THE CERVIX ... part