92

RECTAL LYMPHOGRANULOMA VENEREUMBy R. P. M. MILES, M.B., B.S., F.R.C.S.

Consultant Surgeon, Chichester; late Lecturer in Surgery, University College of the West Indies

Lymphogranuloma venereum is a diseaseof world-wide distribution, particularly amongcoloured peoples. It is prevalent in Europe, butfew cases have been reported in England, 363new casts being reported' between I95I and 1955(Galbraith, Graham-Stewart and Nicol, 1957). Itis very likely, however, that the incidence willincrease with the continuation of West Indianimmigration into this country.

Various other synonyms are in use for thisdisease- some picturesque, others frankly mis-leading. ' Climatic bubo' and 'Esthiomene'(from the Greek, meaning' eating') describe onlyindividual aspects of the disease. ' Nicholas-Favre' disease commemorates its French dis-coverers. It should not be confused with Granu-loma Inguinale (or Donovanosis), an ulceratinggranuloma of the skin caused by the specificDonovan body.The disease is caused by a virus of the lympho-

granuloma-psittacosis group, and there are threeclinical varieties:

i. The Inguinal (Climatic Bubo). This is anacute febrile disease affecting almost exclusivelymales, although purulent proctitis has been re-ported in males with the acute disease (Galbraith,Graham-Stewart and Nicol, I957).

2. The Genito-Ano-Rectal Syndrome (Esthio-mene). There is oedema, ulceration, and destruc-tion of the external genitals, with urinary andfaecal fistula formation. There is gross perinealscarring, with variable spread of the disease upinto the rectum.

3. The Ano-Rectal Variety (which is the mainsubject of this paper). This occurs predominantlyin women, and produces extensive- lesions in thepelvic colon, rectum, and anus. It may be com-bined with the full genito-ano-rectal syndrome,but more commonly occurs as a lesion confined tothe bowels and perineum.

It should be stressed that this disease is notconfined to the rectum, but is a procto-colitisaffecting the left colon in a large number of cases.It is generally conceded now that it is the com-monest, if not the only, cause of inflammatorystricture of the rectum. The aetiology is not

clearly understood, but it is reasonable to supposethat the rectal lesion is caused by direct invasionof the mucous membrane by the virus implantedin the anal canal through ' spill' during inter-course, although a more widely accepted theory inthe past has been that of the French workersBensaude and Lambling (1936), who stated thatin the female the infection entered through aprimary sore deep in the vagina, and spread viathe lymphatics to the perirectal tissues, causingstricturing of the rectum from without: in themale direct implantation in the rectum took placethrough the practice of sodomy. In support ofthis they quote their findings that in womenstricture is the predominant lesion, in men proc-titis alone. These views are discussed andcriticized elsewhere (Miles, I957). The incuba-tion period is unknown, but a long interval-possibly measured in years-may intervene be-tween exposure to infection and the onset ofrectal symptoms.

Clinical ManifestationsFemales are predominantly affected. In 9I

cases studied recently in the West Indies only sixwere males. The average age of patients in thisseries was 40. The ano-rectal variety is seen inthree phases, which overlap considerably:

Phase I: Proctitis.Phase 2: ' Prestricture.'Phase 3: Established stricture.The symptoms vary with the phase, but, in

general, the first complaint is of the passage ofcopious pus, mixed with blood, per rectum.There may be spurious diarrhoea or the passage ofnarrow pencil-like stools. Cramp-like abdominalpain may be present with distension suggestive ofchronic obstruction, and occasionally acute ob-struction is the presenting symptom. Thesesymptoms are often tolerated for months or years,and by the time advice is sought the patients areanaemic, debilitated, and often showing toxicmanifestations of the disease, such as skin ulcera-tion, iritis, and toxic arthritis of medium-sizedjoints. They may present with fistula-in-ano,ischio-rectal abscess, or recto-vaginal fistula.

copyright. on A

pril 24, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.35.400.92 on 1 F

ebruary 1959. Dow

nloaded from

February 1959 MILES: Rectal Lymphogranuloma Venereum 93

The Rectal LesionsThe perineum may show multiple sinuses

draining pus, with large condylomatous lesionssurrounding the anal orifice. A recto-vaginal fis-tula may be present, situated characteristically lowdown at the ano-rectal ring. There may bemerely button-holing of the skin or completeabsence of the recto-vaginal septum. However,there may be no external lesion at all, proctitisbeing the commonest mode of presentation. Onrectal examination the granular proctitis has atypical feel resembling morocco leather. Largerpolypoid granulations can often be felt, and thewall of the rectum is thickened and inelastic, sothat the finger appears to be gripped by it. Onwithdrawal of the finger a large amount of puspours forth. The rectal surface has a velvetyappearance, bright red, with darker areas ofpunctate haemorrhages. Small shallow ulcers areseen. In most cases the severe infection is limitedto the anus and rectum, but it is not uncommonto find the lesions extending above the full lengthof the sigmoidoscope.

In the prestricture phase the inflammatorychanges may be much less pronounced, butfibrosis has already occurred in the rectal wall.There is a definite feeling of gripping of the finger,and the granularity of the surface has becomefirmer and mammilated.The established rectal stricture is situated at

two distinct levels-2 cm. from the anal margin(the length of the distal phalanx of the indexfinger) and 5 to 7 cm. from it (the length of thedistal and middle phalanges). It may vary incalibre from a slight narrowing of the rectal wallto complete stenosis. In the recent case it has asoft consistency and is easily split by the examiningfinger, often with considerable haemorrhage. Inthe long-standing case the tissues have a leatheryconsistency which completely bars further entry ofthe finger. The stricture varies in length andshape and it is conyeenjent to describe them asbeing diaphragmatic, tubular or conical, althoughsome combination of these; is the most usual find-ing. The upward extension of the stricturecannot always be appreciated by the finger, manyextending well above the rectum itself.

Radiology has revealed much information aboutthe distribution of the lymphogranulomatouslesion in the bowel, Barium enema studies wereperformed on most of the cases of the series men-tioned above, and, in addition to helping to definethe shape of the stricture and its upward extent, itwas also possible to study the spread of the diseaseto the rest of the left colon. The pelvic colon wascommonly involved, showing narrowing and ashortening of its total length (Fig. I).The colon proximal to this showed varying

FIG. i.-Extreme narrowing of pelvic colon and rectum,Pericolic sinuses. Tubular colon proximal tostrictured area.

degrees of involvement, a ' saw tooth ' patternsuch as that seen in ulcerative colitis being commonthroughout the left colon. In some the colon haslost its haustrations and become tubular. Isolated'skip ' areas giving a ' string sign' as in regionalenteritis were a striking finding (Fig. 2). Pericolicsinus formation is common, but there, is.- notendency to form internal fistulae.The diagnosis is confirmed by the use of the Frei

intradermal test, which is highly specific but needsexperience in interpreting it, or by the lympho-granuloma complement fixation test, which hasthe advantage of being quantitative, being moresimply performed, and requiring only oneattendance by the patient.

TreatmentIn the acute disease the antibiotics have proved

themselves virucidal, and it is probable that thedisease can be eradicated if early treatment is madeavailable. In the chronic types, however-,; theireffect is much less dramatic. In proctitis ex-

copyright. on A

pril 24, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.35.400.92 on 1 F

ebruary 1959. Dow

nloaded from

94 POSTGRADUATE MEDICAL JOURNAL February x959

FIG. 2.-Stricturing of left colon. Extreme narrowingwith a ' string sign ' in descending colon.

cellent results are obtained with antibiotics, and aclinical cure can be anticipated in 8o per cent. ofcases. The symptoms and signs are completelyrelieved, but the blood remains sero-positive foran indefinite period. Within a few days of thecommencement of treatment the pain and tenes-mus cease and the discharge of pus and blobd ismuch less. Full dosage is required-28 grammesduring a 14-day course of treatment, and withthis dosage the three antibiotics used-chloram-phenicol, tetracycline, and erythromycin-gaveequally good results.

In the established case of stricture the effect ofthe antibiotics must be mainly on the secondaryinfection, and probably as good an effect is pro-duced by the use of the unabsorbable sulphona-mides. There can be obviously no direct effecton the thick fibrous tissue of the stricture, but theresolution of the inflammatory oedema producesapparent widening of the bowel lumen, with con-siderable relief of symptoms. Many other typesof treatment have been tried for the established

stricture, most of them being in vogue before theaetiology of the disease was known. Theseincluded rapid and intermittent dilatation, short-wave therapy, diathermy and electro-coagulation,and internal proctotomy. Radical surgery for thecondition was first practised by Henri Hartmann(1930), who described his operation of intra-sphincteric amputation of the rectum performedon 47 cases of inflammatory stricture of the rectum.It is interesting that no mention is made in hispaper of the possible aetiology of the disease.

Surgical TreatmentMany cases can be kept comfortable by courses

of antibiotics and occasional dilatation, and willnot require surgery, but the disease in the chroniccases remains active and progressive. Morebowel is involved, toxic symptoms persist and thepatients remain chronically anaemic and under-weight. Malignant change to epidermoid and,occasionally, to adeno-carcinoma has been de-scribed in long-standing cases.

DilatationThis is the most enduring of palliative methods,

but care must be exercised in its use. Too forcefuldilatation causes sudden splitting of the stricturedarea, with haemorrhage and perforation of therectal wall. Very considerable fibrosis occursduring healing, and the lumen of the stricture isthereby diminished. In this connection mentionmight be made of the ' fibro-plastic diathesis' ofthe Negro (Matas, I896; Rosser, I923), wherebyall fibrous tissue growth is greatly exaggerated,e.g., in healing wounds, keloids, fibromata, andstrictures. It certainly makes all attempts atintermittent dilatation of strictures very uncertainof success.

ColostomyThis may be indicated in cases with obstruction,

whether further surgery is planned or not. Wheremajor surgery is planned a preliminary colostomyis nearly always indicated. Permanent colostomy,however, should be avoided as many of thesepatients are comparatively young women, and thedisease, if left in situ, may progress upward to in-volve the colostomy stoma. A transverse colos-tomy is advisable to avoid this possibility and alsoto allow adequate room for mobilizing bowel atany subsequent operation.

Radical SurgeryRemoval of the diseased rectum and colon has

been practised in many clinics dealing with largenumbers of coloured patients. Bacon (1941) ad-vocated a two-stage perineal resection with per-manent colostomy. Woods and Hanlon (I944)

copyright. on A

pril 24, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.35.400.92 on 1 F

ebruary 1959. Dow

nloaded from

February I959 MILES: Rectal Lytiphogranuloma Venereum 95

performed 35 recto-sigmoid resections with goodresults (and incidentally comment on the poorresults obtained in treating io8 cases by dilatationonly). Wright, Berg, Bolden and Freeman (I946)describe 26 cases treated by intra-sphinctericresection following the technique of Pauchet andHartmann. Breidenbach and Slattery (1948)described i8 cases treated by sacro-perinealexcision and reanastomosis of bowel to anus.The ideal type of operation is a restorative

resection with the colon sutured to the anal stump.However, most of the strictures are at too low alevel for this operation to be feasible, and in theauthor's experience the most generally useful isthe abdomino-anal pull through operation (Pauchetand Hartmann), in spite of the fact that rectalsensation is lost or severely impaired, and controlis not always perfect. The abdomen is exploredfirst to ascertain the extent of the disease. Whereskip areas of disease are present these may have tobe resected and end-to-end anastomosis per-formed. The bowel is mobilized to its fullestextent to ensure adequate length to reach the anus.The perineal part of the operation consists inremoval of the rectum after careful dissection ofthe sphincters. The preservation of the musclering is extremely important, and if this provesimpossible, owing to involvement by disease, theoperation is best abandoned.

In cases with advanced disease, with extensivefistula formation and complete destruction of therectal wall, an abdomino-perineal resection withpermanent colostomy is performed.

Pathological FeaturesThe naked eye appearance of the diseased bowel

is shown in Fig. 3-a specimen removed by ab-domino-perineal resection. The rectal surface isdenuded of epithelium and is covered withhaemorrhages. The strictured area extends intothe distal part of the pelvic colon, and the cutsurface shows the considerable thickening of thewall, this being most pronounced in the sub-mucosa. Also shown in this specimen is a recto-vaginal fistula and extensive condylomata forma-tion. Histologically, there is a non-specific in-flammatory reaction extending through all layers,but more pronounced near the inner coats. Themucosa is practically absent, being replaced bygranulation tissue, and the submucosa shows veryconsiderable thickening, with much fibrous tissue,obliterative endarteritis and aggregations of lym-phocytes.

DiscussionMore study is required before the real mode of

formation of the striking bowel lesions found in

FIG. 3.-Specimen removed by abdomino-perinealresection.

lymphogranuloma venereum is known. It is byno means clear how the lesions actually commence,but it would not seem to be as a direct result of thevirus. It is more reasonable to suppose that thelesions represent the reaction of the bowel tocertain stimuli, e.g., an allergic or anaphylacticphenomenon. The general naked eye, pathologi-cal, and radiological appearances have manyfeatures common to the other granulomatousaffections of the bowel, conditions to which we

copyright. on A

pril 24, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.35.400.92 on 1 F

ebruary 1959. Dow

nloaded from

96 POSTGRADUATE MEDICAL JOURNAL February 1959

are forced to give names which mask our ignoranceof their true nature-regional enteritis and colitis,ulcerative colitis, hyperplastic tuberculosis (Lumb,195I). (It must be admitted that the latter may beoccasionally a genuine entity.) In bacillary andamoebic dysentery a very similar non-specificbowel lesion is found. It is reasonable to supposethat the bowel reacts in a general non-specificmanner to the various toxins or antigenic sub-stances produced in a group of conditions, somewithout any specific aetiological agent, others withsuch an agent, one of which is the virus of lympho-granuloma venereum.

Summaryi. Lymphogranuloma venereum is a disease of

world-wide distribution.2. The ano-rectal variety of the disease produces

extensive bowel lesions and is the main cause ofinflammatory stricture of the rectum.

3. The treatment of these is by means of anti-

biotics in the milder, earlier lesions, and by surgeryin the later, severe cases of stricturing.

4. The disease bears many similarities to theother non-specific granulomata that affect thelarge bowel.

AcknowledgementFig. 3 is reproduced by kind permission of the

Editor of the British Journal of Surgery.

BIBLIOGRAPHYBACON, H. E. (I94I), Sth. med. J7. (Nashville), 34, 31.BENSAUDE, R., and LAMBLING, A. (1936), Proc. roy. Soc. Med.,

24,144'.BREIDENBACH, L., and SLATTERY, L. R. (1948), Ann. Surg.,

128, 1079.GALBRAITH, H. J. B., GRAHAM-STEWART, C. W., NICOL,

C. S. (1957), Brit. med. J., ii, 1402.HARTMANN, H. (I930), Bull. Soc. nat. Chir., 56, 855.LUMB, G. (I95I), Brit. _J. Surg., 39, 233.MATAS, R. (I896), Tram. Amer. surg. Ass., X4.MILES, R. P. M. (I957), Brit. J. Surg., 45, I8o.ROSSER, C. (1923), Amer. J. Surg., 37, 265.WOODS, F. M., and HANLON, C. R. (I944), Ann. Surg., 120, 598.WRIGHT, L. T., BERG, B. N., BOLDEN, J. V., and FREEMAN,

W. A. (1946), Surg. Gynec. Obstet., 82, 449.

.X ... .. ....... ...:i

Where Medical Practitionersfind Specialists in .0 ..

ss.fi..-.,

NON-CANCELLABLE /- 6/tSICKNESS ACCJDENT /INSURANCE* LIFE ASSURANCE* PERSONAL PENSION POLICIES N

When you are BUYING A NEW CAR ask for details X

..--. -~~. --.......... ...--sof the HIRE PURCHASE SCHEME of the MEDICALSICKNESS FINANCE CORPORATION LTD.

-..-.

Please write for particulars, mentioning this advertisement.

gg 0init0; ~~~~............ ,.10 ..0.

................................... '.....''-'""''> ',..,,.....

3 CAVENDISH SQUARE LONDON W.iI. -.Te ephone: LAN hom 0341)

copyright. on A

pril 24, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.35.400.92 on 1 F

ebruary 1959. Dow

nloaded from