aberrant glandular polypi ofthe uterine pathology
TRANSCRIPT
J. clin. Path. (1969), 22, 84-89
Aberrant glandular polypi of the uterine cervixassociated with contraceptive pills:
Pathology and pathogenesisA. D. T. GOVAN, W. P. BLACK, AND JESSIE L. SHARP
From the Research Department, Royal Maternity Hospital, Glasgow
SYNOPSIS Fifteen cases of aberrant polypoidal glandular hyperplasia of the cervix are described.The lesion was found in five pregnant patients. The remaining 10 patients had taken contraceptivepills over a considerable period of time.The lesion while histologically bizarre is benign, the changes being to a great extent cytoplasmic
and not nuclear. It is suggested that it is formed by the influence of steroids on a cervix which isalready pathological. It is further suggested that two changes are essential before steroids can havethis effect. Polypoidal lesions must already be present and must exhibit reserve cell hyperplasia.
Pincus (1956) appears to have been the first tosuggest that 'contraceptive pills' might have a con-siderable effect on cervical function. This wasreiterated by Greenblatt (1959) and in 1964 Zafiartudescribed a 'progestational-like' change in the cervixof women taking the 'pill'.
Recently Taylor, Irey, and Norris (1967) reportedon atypical endocervical hyperplasia in 13 patientstaking oral contraceptives. Since 1963 we have beencollecting material of a similar nature and up to datehave found 10 cases. As Taylor et a! (1967) havestated, the histological appearances of the lesion areso apparently bizarre that the possibility of adeno-carcinoma must be considered. In addition to these10 cases we have found similar lesions in fivepregnant patients.
CLINICAL FINDINGS
The clinical history of the non-pregnant patients isnot particularly striking on the whole and with sucha small number the significance of their symptoms isdoubtful. A summary of the main features is givenin the accompanying table. It will be seen from thisthat vaginal discharge was the commonest complaint.Abnormal bleeding was a feature in half of thesepatients and in three it was not associated withmenstruation.Received for publication 17 May 1968.
TABLESUMMARY OF CLINICAL FINDINGS
Case AgeNo. (yr)
2
3
456
78910
22 2 years2234
24
413827
45364444
Parity Complaint
8
4
36
326
Discharge and pruritisDischarge anddyspareuniaIntermenstrual bleedingand dyspareuniaDyspareuniaMenorrhagiaPost-coital bleedingand dischargeMenorrhagiaDischarge and pruritisDischargePost-coital bleedingand discharge
DurationofTherapy
2 years2 years
6 months
1 year2 years1 year
1 year
On examination a localized erosion was found innine patients. In the remaining patient, the youngestin the series (case 1 in the table) the whole cervixwas much enlarged, had a rough cauliflower ap-pearance, and bled on touch. The appearances weresuch that the possibility of a malignant tumour wassuspected. In the pregnant patients, the changes werefound in cervical polypi discovered at antenatalexamination.
TYPE OF PILL
All of the 'pills' used by these patients were of the84
on April 17, 2022 by guest. P
rotected by copyright.http://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.1.84 on 1 January 1969. Dow
nloaded from
Aberrant glandular polypi of the uterine cervix associated with contraceptive pills
'combined' type. Progestogens in the pills were ofthe nor-steroid variety and included ethynodiol di-acetate, norethisterone, and norethynodrel. Theaccompanying oestrogens were of two types,mestranol and ethynyl oestradiol. In all cases wherea clear history was obtained the pill had been takenfor a considerable time, the shortest period beingsix months.
MICROSCOPICAL FINDINGS
Although the lesions in all cases were very similarthe changes were most striking in the patient aged22, showing the diffuse lesion. In this instance thesurface of the cervix was covered by numerous poly-poid structures varying in size, the largest being0-25 cm in length (Fig. 1). These were covered in mostinstances by a single layer of cuboidal cells or a thin
layer of stratified epithelium. Beneath this there wasa honeycomb or cribriform syncytial mass of cells(Fig. 2). Closer examination of the underlying tissuerevealed that it consisted of tightly packed glandularacini. The striking feature was the great variety insize and shape of the glands and the character of theepithelium lining them. In a few the glands were oflarge adult type lined by a single layer of columnarmucus-secreting cells. Most, however, were ofirregular shape, lined again by a single layer of cellswhich were mostly flattened and elongated butoccasionally cuboidal (Fig. 3). At first sight thepresence of these flattened cells gave an impressionof considerable supporting stroma but in reality thestroma was confined to very narrow compressedstrands surrounding rather large masses of epithelialtissue. There was no evidence that individual glandswere invested with a stromal basement membrane asin normal glandular tissue. The structure seemed to
FIG. 1. FIG. 2.
FIG. 1. Low-power view of a polyp from a patient on 'contraceptive pill'. x 90.FIG. 2. Higher magnification of part of the polyp showing the cribiform pattern with immature acinar structures.The overlying epithelium is still well differentiated. x 250.
85
on April 17, 2022 by guest. P
rotected by copyright.http://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.1.84 on 1 January 1969. Dow
nloaded from
A. D. T. Govan, W. P. Black, and Jessie L. Sharp
FIG. 3. FIG. 4.
FIG. 3. This demonstrates the varying morphology of the gland epithelium. Despite cytoplasmic variation the nucleiare fairly uniform and there is no mitotic activity. x 300.FIG. 4. A peculiar variation of the polypoid structure. In this case the stroma is unduly prominent due to hyalineswelling. x 120.
be irregularly lobular or alveolar, the alveoli beingfilled by epithelial cells showing gland lumina.An apparent exception to this was found in one
of the polypi (Fig. 4). The stroma was more abundantand formed an interlacing network enclosingirregular spaces. At low magnification the structurewas not unlike lymphangiectasis. The increase instromal tissue was contributed to by the fact that ithad undergone hyaline degeneration. The spaces inthis tissue were lined in most instances by a layer ofpeg-shaped cells with pyknotic nuclei and remaininglumina were filled with cellular debris. In a few spacesrecognizable gland acini were present. It wasapparent from this that the structure was essentiallythe same as in other polypi and the differences weredue to degeneration. A further variant was found inone of the patients exhibiting a smaller lesion. In thiscase the polyp was covered by a fairly thick layer of
stratified squamous epithelium. On top of this, how-ever, there was a trellis-like structure of glandulartissue (Fig. 5). Despite the fact that this layer was ofconsiderable depth, there was no supporting stromabetween the acini.
In all of these lesions the changes were confined tothe polypi and there was no evidence of extensioninto the stroma at their bases. Although there wasvery great variation in cell configuration and someof the nuclei were dark and irregular, close inspec-tion showed that these changes were of degenerativetype. The dark nuclei were obviously pyknotic andwhere healthy nuclei could be seen they wereremarkably uniform in character and innocent inappearance.
ASSOCIATED CHANGES At the junction of thepolypi with the adjacent mucosa, lesser degrees of
86
on April 17, 2022 by guest. P
rotected by copyright.http://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.1.84 on 1 January 1969. Dow
nloaded from
Aberrant glandular polypi of the uterine cervix associated with contraceptive pills
FIG. 5. FIG. 6.
FIG. 5. In this instance the proliferated cells have erected a trellis work on the surface of stratified squamousepithelium. x 120.FIG. 6. Proliferated reserve cells adjacent to the base of one of the polypi. A slight attempt at acinar formation ispresent. x 300.
the above changes could be seen. In these areasthe mucosa showed a proliferation of the reservecells beneath the columnar epithelium and in amongthe former cells gland spaces had appeared. Figure6 demonstrates a slight or early change of thisnature; Fig. 7 represents a complete transformationto gland tissue.
Elsewhere in the cervical tissue numerous'polypoiderosions' were present. These tiny polypi did notshow any glandular proliferation but it was notice-able that they also did not show any proliferation ofreserve cells. There was, however, both in thesepolypoid erosions and in the aberrant polypi, amarked inflammatory reaction with many plasmacells.
PREGNANCY CASES
anaesthetic revealed a polyp projecting from thecervical os.
In essence the lesion when seen under the micro-scope was the same as that found in the non-pregnantpatients. There was the same epithelial hyperplasiawith irregular gland formation. Fewer degenerativechanges were seen and the nuclei were only occa-sionally pyknotic. Reserve cell hyperplasia was amarked feature (Fig. 8).
It must be emphasized that the lesions describedhave not been seen in non-pregnant patients otherthan those taking contraceptive pills. In the case ofpregnant patients aberrant glandular proliferationhas only been found in association with polypi inwhich a marked degree of reserve cell proliferationwas apparent.
All five patients had complained of vaginal discharge DISCUSSIONand in three it was bloodstained. Examination under This lesion raises several problems. The first and
87
Aw
on April 17, 2022 by guest. P
rotected by copyright.http://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.1.84 on 1 January 1969. Dow
nloaded from
A. D. T. Govan, W. P. Black, and Jessie L. Sharp
*'u.. . .
...5;~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~.........
-..:..e'Wt w%kS S 4t.......
.:o" ..... . !o
FIG. 7. FIG. 8.
FIG. 7. Another area adjacent to the base of a polyp showing complete transformation of the reserve cell layerto acinar structures. x 300.FIG. 8. Part of the cervix from a pregnant patient showing a markedproliferation of reserve cells. x 300.
most important is whether it is malignant. Althoughthere is undoubted proliferative activity and theappearances are bizarre, there is no evidence ofstromal invasion. Many of the abnormal cytologicalchanges appear to be the result of degeneration. Afurther possibility is that this represents a pre-invasive stage of tumour formation but again thecriteria of malignancy are lacking. So far, none ofthe patients has had any recurrence but the timeinterval, a maximum of four years, is short.
It is difficult to define a clear pathogenesis for thechanges observed but a clue may be found in thetissues immediately adjacent to the lesion. Twopoints are worth noting. Where these polypi unitewith the mucosa there is marked proliferation ofreserve cells. Among these, attempts at gland forma-tion with varying degrees of success are apparent.The second point is that the lesions are polypoidal
and show plasma cell infiltration. In all of thesepatients, in addition to the lesion described, therewere areas of polypoid erosion with plasma cellreaction. No sign of reserve cell hyperplasia wasevident, however, in these ordinary erosions andglandular proliferation was absent. The possibilityexists, therefore, that these lesions are most likely todevelop in polypi already present and in which thereis reserve cell hyperplasia. It is to be noted that pro-liferation of reserve cells is a common finding inpregnancy, especially in polypi. In addition, Nesbittand Hellman (1952) and Carrow and Greene (1951)describe adenomatous glandular proliferation inpregnancy which resembles the changes describedabove. Reserve cell hyperplasia in relation to theadenomatous change is apparent in their illustra-tions.Whether these lesions are directly related to the
88
on April 17, 2022 by guest. P
rotected by copyright.http://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.1.84 on 1 January 1969. Dow
nloaded from
Aberrant glandular polypi of the uterine cervix associated with contraceptive pills
action of oral contraceptive tablets cannot be provedbut would seem almost certain. Very few studies ofthe human cervix have been made in patientsreceiving this therapy. Maqueo, Azuela, Calderon,and Goldzieher (1966) studied cervical biopsies from147 patients undergoing treatment with thesepreparations. Although they describe hypersecretionand glandular hyperplasia, no mention is made ofaberrant glandular changes. This would tend tostrengthen the idea that the lesions we have describedonly affect the cervix which is already pathological.The findings of Nesbitt and Hellman (1952) andCarrow and Greene (1951) in biopsies of cervix
showing aberrant glandular hyperplasia in pregnantpatients would suggest that steroid substances aredirectly related to these changes.
REFERENCES
Carrow, L. A., and Greene, R. R. (1951). Amer. J. Obstet. Gynec.,61, 237.
Greenblatt, R. B. (1959). Fed. Proc., 18, 1055.Maqueo, M., Azuela, J. C., Calderon, J. J., and Goldzieher, J. W.
(1966). Amer. J. Obstet. Gynec., 96, 994.Nesbitt, R. E. L., Jr, and Hellman, L. M. (1952). Surg. Gynec. Obstet.,
94, 10.Pincus, G. (1956). Acta endocr. (Kbh.), suppl., 28, 18.Taylor, H. B., Irey, N. S., and Norris, H. J. (1967). J. Amer. med.
Ass., 202, 637.Zaniartu, J. (1964). Int. J. Fertil., 9, 225.
89
on April 17, 2022 by guest. P
rotected by copyright.http://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.1.84 on 1 January 1969. Dow
nloaded from