Dr. Devanshi Mishra is MBBS and is currently working as third-year Resident Surgical Officer, Department of O&G, Sultania Zanana Hospital, Gandhi Medical College Bhopal, MP, which is the oldest and largest teaching hospital of the city. She, in the capacity of selected team member from the institute, has participated in Quiz and won the gold medal at the National Conference of Obstetrics and Gynaecology, Basics to Recent and 27th State Conference of UPCOG 2015 held at GSVM Medical College Kanpur on 28 and 29 November 2015. She has five publications to her credit as a co-author. She is recipient of many certificates of merit during her undergraduate course. She has a keen interest in ‘high-risk pregnancy’.
Introduction Abnormal uterine bleeding (AUB) is the commonest menstrual problem during perimenopause. The International Federation of Gynaecology and Obstetrics working group on menstrual disorders has developed a classification system (PALM–COEIN) for causes of the AUB in non-gravid women. The present study was conducted with the aim to study the two components of this system in clinical practice in general and to establish a clinico-pathological correlation of AUB with context of PALM component in particular.
Materials and Methods Two hundred and thirty-six perimenopausal women (aged 40 years and above till 1 year beyond menopause) admitted with complaints of abnormal uterine bleeding were studied. After thorough history and examination, a clinical diagnosis was made as per PALM– COEIN classification. Relative contribution of various causes of PALM (structural) and COEIN (functional) components was analysed. After all indicated investigations, endometrial sampling and hysterectomy specimen were assessed by histology. A clinicopathological correlation was analysed statistically.
Result PALM and COEIN components contributed almost equally for AUB when assessed clinically. On the other hand, the histological examination revealed significantly more cases of PALM (structural or anatomical) component of AUB, i.e. 50.23 versus 63.98 % (p B 0.05) The difference was mainly attributed to the detection of more cases of AUB-M (malignancy and hyperplasia) in highly significant proportions (p B .01) and coexistent cases of AUB-A;L. AUB-L was the commonest (41.1 %) aetiology overall.
Conclusion The PALM–COEIN classification system should take into account both the clinical and histopathological diagnoses in women having AUB around perimenopause as the two diagnostic modalities are complementary to each other and clinical impression should be placed into proper perspective of this classification in order to optimise outcome.
Keywords : Abnormal, Uterine bleeding, Perimenopause, PALM–COEIN, FIGO, Histopathology
Abnormal uterine bleeding (AUB) is one of the common presenting complaints encountered by a gynaecologist. It is a significant cause of hysterectomy and thus is a major health problem [1]. AUB is also associated with significant social and physical morbidities in all societies and may be a reflection of serious underlying pathology [2].
AUB may be acute or chronic and is defined as bleeding from the uterine corpus that is abnormal in regularity, volume, frequency or duration and occurs in the absence of pregnancy [3, 4].
AUB is the commonest menstrual problem during perimenopause which is defined as the period of 2–8 years preceding menopause and 1 year after the final menses [5]. Follicular development at this time has been demonstrated to be erratic, with consequent variability in oestrogen levels and an increased percentage of anovulatory cycles making them more likely to experience abnormal uterine bleeding.
In addition to the erratic ovulation, there may be many structural or functional aetiologies for the AUB. The International Federation of Gynaecology and Obstetrics working group on menstrual disorders has recently developed a classification system (PALM–COEIN) for causes of the AUB in non-gravid women [6]. There are nine main categories, which are arranged according to the acronym PALM–COEIN: polyp; adenomyosis; leiomyoma; malignancy and hyperplasia; coagulopathy; ovulatory dysfunction; endometrial; iatrogenic; and not yet classified.
PALM side of the classification refers to structural causes that may be evaluated by imaging techniques and/or histopathology and the COEIN side by investigating the underlying medical disturbances. Perimenopausal women show a significant number of underlying organic pathology. The onus here in AUB management is to exclude complex endometrial hyperplasia and endometrial cancer. Evaluation of endometrium and/or organ histopathology have the dual advantage of finding an accurate reason causing the AUB and to rule out endometrial or other cancers or a potential for the cancer in future like endometrial hyperplasia with atypia.
A thorough histopathological work up and clinical correlation is mandatory as there is always a possibility of reallocation of category. Assessing a correlation can ascertain the degree of accuracy of clinical assignment to the category of AUB and may provide an insight as to when one must go for a pathological correlation particularly that of the PALM aspect of PALM–COEIN, whereas for COEIN (functional) aspect, the same is done using other investigations namely haematological and endocrinological work up.
FIGO recommends endometrial tissue testing as a firstline management in women of perimenopausal age group who have AUB [7, 8]. Histology clinches the diagnosis and guides the management plan. When planning hormone therapy, it is mandatory to rule out a precancerous neoplasia like suspicious hyperplasia or sub-clinical endometrial cancer. A histological assessment therefore remains the cornerstone in the current practice as it puts the clinical diagnosis in the accurate perspective and allows standardization of treatment.
The present study was conducted with the aim to study and analyse the structural (PALM) and the functional (COEIN) component of the PALM–COEIN system of AUB in perimenopausal age group women of our region. This was followed by the histopathological studies and correlation of cases wherever applicable particularly for the structural (PALM) component and for categories of AUBE or AUB-O of the COEIN (functional) aspect.
The present study was conducted at the tertiary centre, government medical college of Madhya Pradesh from 1 July 2014 to 30 June 2015. A total of 236 perimenopausal women (aged 40 years and above till 1 year beyond menopause) who were admitted with complaints of abnormal uterine bleeding comprised the study population. Women who were < 40 years of age and those beyond 1 year of menopause were excluded from the study.
The demographic details were noted, and a structured history of previous and current menstrual history, history of contraception use and medical/surgical history was followed by general, physical, systemic and gynaecological examination. On gynaecological examination, cervix (position of cervix, any erythematous lesion, hypertrophy, mobility, presence of polyp or ectopy), uterus (size, position, consistency, and mobility) and adnexae (any palpable enlarged lump, tenderness and mobility) were assessed. Clinical diagnosis and allocation to PALM–COEIN was done. A pelvic ultrasound to assess the uterus (uterine size, endometrial thickness, presence of endometrial polyp, adenomyosis or fibroids) and ovarian status (presence of any cyst, mass and its characteristics) was done. Endometrial biopsy and hysterectomy specimens (wherever applicable) were obtained and sent for histopathology. As per the histopathological findings, possible underlying causes were categorised. Clinical diagnosis was then correlated with histopathology-based final diagnosis.
For evaluation of the COEIN aspect, ovulatory dysfunction was defined as unpredictable timing and variable amount of bleeding, while endometrial disorders referred to cases when AUB occurred in line with predictable/ cyclic pattern. Iatrogenic category was categorised by the identification of hormone steroid intake during the preceding 3 months and/or onset of symptoms following contraceptive device or method. Following a thorough history and complete clinical examination, investigations including complete blood count, coagulation profile when applicable (for all previously known cases of defects of coagulation from younger age and AUB dating back from menarche), thyroid function test and blood sugar level estimations were done, and the results were correlated with the clinical allocation. Endometrial histology was correlated in cases of AUB-O and AUB-E with the clinical assignments.
Data were analysed by SPSS version 16, and descriptive statistics were presented as frequencies, percentages and bar charts. Z-test was applied to know the significance of the correlation.
Out of 1704 gynaecological admissions during the study period, there were 236 perimenopausal women who presented with the complaints of abnormal uterine bleeding and were included in the study.
The majority (97.46 %) of these women were less than50 years of age, and most of the study population lived inurban area (59.7 %). An overwhelming majority (50.80 %)women were grandmultipara and 78.01 % belonged topoorer sections of society, i.e. class III or less of modifiedPrasad classification. Most women experienced symptomsof abnormal uterine bleeding for a period of 6 months to1 year before seeking treatment (43.6 and 35.1 %,respectively).The most common presenting symptom in our study was heavy menstrual bleed (32.6 %) followed byintermenstrual heavy menstrual bleed in 28.3 % cases.
Table1shows distribution of cases as per clinicaldiagnosis. The PALM and COEIN components accountedfor 50.23 and 49.57 %, respectively. Leiomyoma (AUB-L)was assigned to be the major aetiology in 97/236 (41.1 %)in overall and 97/119 (81.51 %) in the structural group,whereas ovulatory disorders (AUB-O) were the proposedmajor contributor in the functional group accounting for88/236 (37.28) of overall and 88/117 (75.27 %) of the latergroup cases.
In total 22 (9.3 %) women were obese, 15 (6.3 %) werehypertensive, nine (3.8 %) had thyroid abnormalities, andonly four (1.6 %) had diabetes mellitus (Table2).
Table3shows the histopathology changes in abovewomen. The majority had secretory changes in 98(41.52 %) and proliferative in 88 (37.28 %). This wasfollowed by simple adenomatous hyperplasia as the thirdcommon finding in 21 (8.9 %) women who were withoutatypia in 19/21 (90.47 %) instances (Table4).
On histopathology-based diagnosis (Table5), ThePALM component turned out to be accounting for 151/236(63.98 %) cases of AUB which was 32 (13.74 %) cases
more than those assigned by clinical criteria, whereasAUB-O and AUB-E classes of COEIN component onlycould be evaluated histologically and constituted 85/236(36.01 %) of overall AUB cases. The difference was sig-nificant statistically (pB0.05) on clinical and the diag-nostic correlation (Table6). Values did not differsignificantly in cases of AUB-L and AUB-O. On the otherhand, histopathology could diagnose more cases in com-parison with clinical-based diagnosis in the categories ofAUB-A (8.47 vs. 3.81 %), AUB-M (10.16 vs. 2.54 %) andcases having both adenomyosis and leiomyoma (4.23 % vs. nil). The difference was significant statistically in all three.The only instance where clinical diagnosis was ascribed tosignificantly more number of cases than those confirmed byhistology was in AUB-E (12.28 vs. 5.5 %).
The PALM–COEIN classification has an advantage ofconsideration of the entire range of possible aetiologies butshould be followed by further investigation to arrive at amore accurate and consistent diagnosis in perimenopausalgroup of women so as to rule out organic diseases partic-ularly precancerous lesions and cancers. The demographicprofile and the pattern of menstrual complaints were inaccordance with other researchers [9–13]. Chronic anovu-lation is a predominant phenomenon in perimenopausewhich is associated with an irregular and unpre-dictable pattern of bleeding that varies in amount, durationand character. In our study the PALM and COEIN com-ponents contributed almost equally for AUB when assessedclinically with AUB-L being the major contributor inPALM group. Leiomyomas are known be predominant inthe age group presently studied. In addition, 9.3 % womenwere obese. Obesity by increasing the overall lifetimeexposure to oestrogen by peripheral aromatisation ofadrenal androgens increases the incidence of polyps,leiomyomas and endometrial carcinoma (relative risk3–10 %). The risk of leiomyomas is seen to be increasingby 21 % for each 10-kg increase in body weight [14,15].Obesity has proved to be a main predisposing factor forAUB [16].
Although hysteroscopy and directed biopsy is the goldstandard in diagnostic work up of AUB, endometrialsampling is still the most common available practice inpublic hospitals. Histopathological pattern of endometriumin women with AUB is quite variable depending upon age,parity, and ethnicity. Endometrial hyperplasia was presentin 8.9 %of our cases, out of which most were simpleadenomatous hyperplasia without atypia (91.3 % cases ofendometrial hyperplasia) The incidence of endometrialhyperplasia is grossly variable, yet incidence of endome-trial carcinoma is small in all cited studies [9,10,17,18].
Leiomyoma as the leading cause of AUB in peri-menopause is also noted by various researchers[11,12,17]. Age is the most important risk factor, withlifetime risk in women over the age of 45 years to be morethan 60 %. Higher association of AUB is seen with sub-mucosal type, compared with intramural and subseroustype [15].
In perimenopausal years, ovulatory disorders are com-mon due to derangements in the hypothalamo–pituitary–ovarian axis resulting in derangements of follicular
maturation, ovulation or corpus luteum formation, andanovulatory cycles are most frequent, and chronic anovu-lation is associated with an irregular and unpre-dictable pattern of bleeding. This explains why ovulatorydisorders were found to be the second most common causeof AUB in this study and most other studies.
The other important cause of AUB was AUB-M, i.e.malignancy and hyperplasia. The unopposed oestrogenicaction on the endometrium in the anovular cycles foundin perimenopausal women predisposes them to develophyperplasia and eventually endometrial carcinoma. In thepresent study, endometrial hyperplasia accounted for8.9 % cases and adenocarcinoma for 1.2 % cases. Theaverage age for women with endometrial carcinoma is61 years, but 5–30 % cases occur in premenopausalwoman [19].
Clinico-pathological correlation of different componentsof PALM side and AUB-O along with AUB-E categoriesof the COEIN revealed significantly more cases to havestructural causes (PALM) of AUB on histopathologicalbasis in comparison with clinical assignment of the PALMcomponent.
On analysis of various categories, in AUB-P (polyp) thedifference in clinical and histopathological diagnosis wasnot significant (p[.05). Most of the cases were cervicalpolyps in present study which could be diagnosed clinicallyby per speculum examination. This observation differsfrom others [12] who found the difference to be highlysignificant in case of polyps. The variation may be attrib-uted to greater number of endometrial polyps in the laterstudy. In present study too, the histopathology identifiedhigher number of polyps although not to significantproportions.
In AUB-A (adenomyosis) the difference in clinical andhistopathological diagnosis was significant, (p\.05). Thisis due to the fact that symptoms and signs of adenomyosis and leiomyoma can be so similar that it can be impossibleto differentiate them clinically [15,20]. This very reasonexplains the difference in clinical and histopathologicaldiagnosis of a combination of AUB-A;L (adenomyosis andleiomyoma) which was highly significant (p\.01). Thisfinding emphasises the importance of histological exami-nation as a complementary diagnostic tool in PALMcomponent of AUB. Our observation is in accordance withothers [8,12].
In AUB-L (Leiomyoma) the difference in clinical andhistopathological diagnosis was not significant (p[.05).The explanation may be that most symptomatic fibroidscan be easily diagnosed by history and clinical pelvicexamination.
In AUB-M (malignancy and hyperplasia) the differencein clinical and histopathological diagnosis was highly sig-nificant (p\.01). This is due to the fact that the clinicalpicture including menstrual history is not specific and thatbimanual examination reveals an ordinary small uterus thatshows no obvious departure from the normal senile one inmost cases. Similar observation was made by others also[21]. Although clinically indistinguishable from non-ma-lignant causes, the genital malignancies have a protractedcourse and grim prognosis. Early detection and promptmanagement may lead to a better outcome in all thesewomen. The significant difference in clinical andhistopathological diagnosis in cases of genital malignan-cies and hyperplasia reiterates the complementary role ofthe two modalities where a case of AUB is provisionallyclassified to one category, but after histopathology it maybe reclassified, and in the process, a correct diagnosis ismade so that the woman is benefitted. As the experience ofthe clinician improve in context of PALM–COEIN classi-fication system both in clinical and in histopathologicaldiagnoses, there will be improved outcomes in womenhealthcare system.
In AUB-O (ovulatory disorders) the difference in clini-cal and histopathological diagnosis was not significant(p[.05). This is due to the fact that perimenopausalwomen have more anovulatory cycles. In the majority ofwomen with true anovulatory bleeding, the menstrual his-tory alone can establish the diagnosis with sufficient con-fidence that treatment can begin without additional labevaluation or imaging. In frequent, irregular, unpre-dictable menstrual bleeding that varies in amount, durationand character and is not preceded by any recognisable orconsistent pattern of premenstrual molimina or accompa-nied by any visible or palpable genital tract abnormality isnot difficult to interpret. Conversely, regular monthlyperiods that are heavy or prolonged are more likely relatedto an anatomical cause or a bleeding disorder than toanovulation.
In AUB-E (endometrial disorders) the difference inclinical and histopathological diagnosis was significant(p\.05), with the clinically assigned cases being higher innumber than those detected by histopathology. This may bebecause most women in this category tend to have nodefinable cause of AUB. AUB-E is presently reserved as adiagnosis of exclusion among other causes of AUB andmay represent a primary endometrial disorder. Most AUB-E cases appear to be due to disturbances of metabolicmolecular pathways such as those involving tissue fibri-nolytic activity, prostaglandins and other inflammatory orvasoactive mediators. The specific routine tissue assayswhich are not available at present may lead to negativehistopathology in some cases. If available, these sophisti-cated tests may have a potential in order to establish aclearer diagnosis in the future. So far no such validatedtests are available for clinical use, to attribute AUB-E asthe primary cause of a woman’s symptoms, so one has torule out all other causes of AUB in clinical examinationfollowed by a histological confirmation. In the presentstudy there were significantly greater number of casesassigned to AUB-E on clinical ground can be justified bythis arbitrary approach. If the histological confirmation isnot there, the final classification as per pathological diag-nosis may be to any other category it belonged to.
