The study was a prospective cross-sectional study of all qualified fallopian tube specimen submitted to the histopathology laboratories of University of Nigeria Teaching Hospital (UNTH) Enugu and Enugu State University Teaching Hospital (ESUTH) from January 2012 to December 2014. In all, fallopian tubes from 120 subjects were studied. Seventy five cases were from UNTH and 45 came from ESUTH. Ethical clearance was obtained from University of Nigeria Teaching Hospital Health Research Ethics Committee (NHREC/05/01/2008B-FWA00002456-1RB00002323). All fallopian tubes resected in the course of surgery for gynaecological conditions of the cervix, uterine corpus or ovaries and obstetric emergencies like ruptured uterus were included as in these conditions the fallopian tube was not the primary site of the offending pathology and salpingectomy was thus incidental. Segmental salpingectomy tissues of bilateral tubal ligation and tubal ectopic pregnancies were excluded. Tubal ligation and ectopic pregnancy specimens fall outside the definition of incidental salpingectomy and were thus excluded. Also, tubal ligation specimens are incomplete segments of tubes which lack fimbriated end and so cannot be examined by the Sectioning and Extensively Examining the Fimbriated End (SEEFIM) sampling protocol adapted in this study.

Patient’s biodata and clinical information were extracted from the requisition forms. The primary lesion which prompted the surgery was grossly examined and sampled according to routine histopathological gross examination procedures. The fallopian tubes were sampled using SEE FIM protocol in order to ensure optimal histological evaluation [18].

As per the protocol the distal 2 cm (the fimbrial end) of the fallopian tube was amputated from the rest of the tube and sectioned longitudinally. The remainder of the tube was cut in cross sections (bread loafed) at 2-3 mm interval. The entire specimen was submitted for processing.

This sampling technique ensured that the surface epithelium of the fimbria is extensively exposed since in situ/early neoplastic epithelial lesion was most often seen in the fimbriae.

All the sections were stained with Haematoxylin and Eosin and the slides were studied by three of the investigators to arrive at the diagnoses. Lesions were diagnosed and classified based on the criteria established in two reference texts [19,20].

STIC is defined as focal replacement of normal fallopian tube epithelium by cells with malignant features [13,15,21].

Data obtained were analysed using frequency and cross tabulation procedures in IBM SPSS software version 20.0 for windows. Measures of central tendency and variation were calculated for numerical variables while frequency and proportions were calculated for categorical variables. The chi-square and Fisher’s-Exact tests were done as appropriate. An alpha level of 0.05 was used for all statistical tests.

Fallopian tubes from 120 female subjects were studied. The mean age of the women was 48.7±10.1 (range: 26-79) years. Majority (71%) of the patients were within the age group of 40 to 59 years [Table/Fig-1].

[Table/Fig-2] show the spectrum of histopathological diagnoses in the fallopian tubes. Sixty one out of the 120 women (61/120, 50.8%) had fallopian tubes within normal limits. Eighty pathological lesions were diagnosed in fallopian tubes of 59 women. This number represented 49.2% (59/120) of all cases studied. Mucosal epithelial proliferations consisting of moderate to marked epithelial hyperplasia together with atypical epithelial hyperplasia constituted 43.7% (35/80) of the lesions diagnosed. Fifteen cases of chronic salpingitis were seen representing 18.8% (15/80) of all lesions. The histo-morphological forms of salpingitis diagnosed included salpingitis with prominent lymphoid follicles as well as chronic follicular salpingitis [Table/Fig-3]. One case of Invasive serous adenocarcinoma of the fallopian tube was seen [Table/Fig-4]. This case clearly showed a point of transition from benign to malignant epithelium and associated small tumour deposits on the ovarian surface. Secondary involvement of the fallopian tube by carcinoma was detected in 3.3% (4/120) of the women studied. Reckoned as percentage of lesions diagnosed, secondary involvement or metastases constitutes 5% (4/80). The primary tumour in two of these cases was endometrial carcinoma and the remaining 2 cases had high grade serous adenocarcinoma of the ovary. The secondary involvement was present as tubal serosal implants or luminal tumour. Atypical epithelial hyperplasia was diagnosed in 6 (5%) of women studied. It constituted 7.5% (6/80) of all lesions diagnosed. None of the cases of atypical hyperplasia met the criteria for diagnosis of serous tubal intraepithelial carcinoma thus no case of STIC was diagnosed in this series.

[Table/Fig-3]:

a) A photomicrograph of chronic salpingitis. The stroma of the fallopian tube is infiltrated with chronic inflammatory cells predominantly lymphocytes; b) A photomicrograph showing chronic xanthogranulomatous salpingitis. The stroma is infiltrated with foamy macrophages and plasma cells; c) A photomicrograph showing chronic follicular salpingitis. The plicae are fused to form a pseudoglandular pattern; d) A photomicrograph showing endometriosis in the fallopian tube. (H&E; original magnification 100X).

[Table/Fig-4]:

a) A photomicrograph showing secondary involvement of the fallopian tube by metastatic carcinoma; b) A high power photomicrograph showing primary high grade serous adenocarcinoma of the fallopian tube; c) A high power photomicrograph showing marked epithelial hyperplasia of the fallopian tube. Marked epithelial stratification is present but ciliation is retained precluding the diagnosis of serous tubal intraepithelial carcinoma (H&E stain; original magnification x400).

[Table/Fig-5,6] show the gynaecological tract lesions which prompted the surgical procedure in which the fallopian tubes were incidentally removed. A total of 154 pathologic lesions were seen in the 120 cases. Overall, 75.3% (116/154) of the gynaecological tract lesions were benign while 24.7% (38/154) were malignant. Categorisation of these lesions according to specific gynaecological organs shows that 103 (66.9%) were lesions of uterine corpus, 38 (24.7%) were lesions of the ovary and 13 (8.4%) were lesions of the cervix. Uterine leiomyoma was by far the most common disease leading to hysterectomy, accounting for majority (63/154, 41%) of the lesions provoking total abdominal hysterectomy with salpingo-oophorectomy.

Statistically significant association (Fisher’s-exact test; p<0.001) was demonstrated between sex cord stromal tumours of the ovary and moderate to marked epithelial hyperplasia of the fallopian tube. Out of 13 cases of sex cord stromal tumours of the ovary, moderate to marked epithelial hyperplasia was present in the accompanying fallopian tubes of 11 cases. Similarly, statistically significant relationship was seen to exist between endometrial hyperplasia and moderate to marked epithelial hyperplasia (Fisher’s-exact test; p=0.006). No statistically significant relationship existed between uterine leiomyoma and moderate to marked epithelial hyperplasia (chi-square: χ² (1)=0.27, p=0.60) or uterine leiomyoma and chronic salpingitis (chi-square: χ² (1)=0.39, p=0.53).

This present study was undertaken in a Nigerian population to ascertain the histopathological status of a common group or subset of fallopian tube specimen (incidental salpingectomies) in a histopathology laboratory and determine, among other concerns, if (STIC) occurred with any reasonable frequency in these tubes. The report by Piek JMJ et al., was crucial in its recommendation for closer scrutiny of the fallopian tube in ovarian cancer research [4]. In their study of occurrence of (pre)neoplastic lesions in overtly normal fallopian tubes from women predisposed to developing ovarian carcinoma, they found preponderance of dysplasia in the fallopian tubes and thus recommended that the fallopian tube should attract closer attention [4].

A major contributor to identification of dysplasia or STIC in the fallopian tube has been the SEE-FIM (Sectioning and Extensively Examining the Fimbriated End) protocol [18]. Authors do not know of any study emanating from Nigeria or Sub-Saharan Africa which utilised the SEE-FIM protocol. The only histopathological study of the fallopian tube in an African population, to the best of our knowledge, is by Izegbu MC et al., who retrospectively reviewed the slides of all fallopian tube specimens received at the Lagos State University Teaching Hospital over a 4 year period [22]. The tissues had been sampled by conventional method which entails examination of 1-3 sections per tube.

As shown in the result of our study, there were pathologic lesions in 49% (59/120) of the cases. Fifty one percent of the tubes were histologically within normal limits on H&E. This finding stands in sharp contrast to the report by Izegbu MC et al., who in a retrospective analysis of all fallopian tubes submitted to a teaching hospital over a 4-year period observed no pathologic lesion in the fallopian tubes of total abdominal hysterectomy with salpingectomy specimens (the category of specimen investigated in this present study) [22]. The disparity in finding may be explained by differences in study and sampling methods-theirs was a retrospective review of conventionally sampled fallopian tubes while ours is a prospective study employing extensive (SEE-FIM) sampling protocol. The findings of this current study thus, suggest that extensive sampling will reveal some identifiable pathological process in a good number of the incidental salpingectomy tissues. Fortunately, most of the identified lesions are not of malignant nature suggesting that the cost of such elaborate sampling protocol should be weighed against clinical benefit in any future institutional change in gross examination policy. However, we identified for cases of metastatic carcinoma of the fallopian tube which translated to fallopian tube involvement in 23.5% (4/17) of endometrial and ovarian carcinoma. This finding seems to reinforce an earlier recommendation that all tissues of both fallopian tubes in cases of ovarian, peritoneal and endometrial serous carcinomas be routinely submitted for histopathological analysis [19].

Epithelial hyperplasia (mucosal epithelial proliferation) of the fallopian tube which was a common finding in the present study had been the subject of a number of previous studies in the past and till date its clinical significance has remained unsettled with various researchers expressing varying opinions. An earlier work [23] concluded that it is a frequent finding of no clinical significance. Robey and Siva [24] found it to be associated with ovarian serous tumours of low malignant potential, a finding which could not be sustained by the work of Yanai I et al., [25]. However, Kurman RJ et al., identified a variant of tubal epithelial hyperplasia designated papillary tubal hyperplasia which showed strong association with borderline serous tumours of the ovary being present in 91% of fallopian tubes resected with such borderline tumours [26].

The studies by Yanai I et al., tried to standardise the diagnosis of epithelial proliferation of the fallopian tube, grading it into mild, moderate and marked (severe) categories by well-defined criteria [25]. They observed that mucosal epithelial proliferation of moderate to marked degree occurred in 25-40% of the tubes and with this came their recommendation that ‘only moderate to marked mucosal epithelial proliferation should be considered a diagnosable lesion in the future’.

The current study identified moderate to marked epithelial hyperplasia of the fallopian tube in 24.2% of the tubes, a figure which is close to that of Yanai I et al. Tubal epithelial proliferation has been anecdotally associated with salpingitis as well as endogenous and exogenous oestrogen [25]. However, none of the patient in our study had any history of hormonal drug intake. Further studies on the role of oestrogen and even multi-parity in the genesis of proliferative epithelial lesions of the fallopian tube is warranted as this is currently not well covered in literature.

Six cases of atypical epithelial hyperplasia were seen in the 120 subjects studied. In these cases, the tubal epithelium had some cytological atypia but did not meet the strict criteria required for diagnosis of STIC. Thus, no STIC was diagnosed in this study. This is not entirely surprising as STIC is more frequently seen in women at high risk for serous ovarian cancer. These are women with family history of breast and/or ovarian cancer or germline mutations of BRCA1 and BRCA2 genes. A systematic review of studies found that coexistence of STIC with high grade serous cancers ranged from 11% to 61% (mean: 31%, 95% CI: 17-46%) [27]. Meserve EEK et al., and Rabban JT et al., reported on the prevalence of serous tubal intraepithelial carcinoma in women without a genetic risk or history of high grade serous carcinoma [28,29]. In the two studies, the prevalence of STIC among this category of subjects was 0.1% (2/1747) and 0.8% (4/522) respectively. Present finding of no STIC was in the general population as our subjects were not stratified by presence or otherwise of increased risk of hereditary breast or ovarian cancer. Tang S et al., studied association of STIC with various conditions and reported no STIC in any of 74 cases of non-serous endometrial malignancies, 15 non-gynaecologic malignancies and 90 cases of benign conditions [12]. The small size of our study may have contributed to non-diagnosis of any STIC in our study. Also noteworthy is the report by Mahe E et al., that multiple deeper level sections of SEE-FIM blocks increased detection of STIC by 25% [30].

The study is limited by size. A larger sample size would have improved the study’s precision and external validity. Another limitation is that of denudation of focal areas of tubal epithelium, a known sequel of surgical and histopathological handling. This might have led to some degree of under diagnosis. In spite of these limitations, the study is relevant as it has started the process of filling the knowledge gap on histopathological profile of apparently normal fallopian tubes among Nigerian women and SSA population in general. The strength of this study, however, lies in its prospective design, with reduced likelihood of missing or incomplete data. Furthermore, the study utilized the SEE-FIM protocol which is a more extensive technique with higher sensitivity and specificity than the traditional or conventional sampling approach.

This study has shown that uterine leiomyoma is responsible for over half of the incidental salpingectomies in Enugu, Nigeria. With extensive sampling, an identifiable pathological process could be found in up to 49% of cases. Most of the lesions detected were mostly inconsequential. STIC, a lesion described in subjects at high risk of developing ovarian cancer seems to be of low prevalence in the general population, though studies employing larger sample size are needed to arrive at a more certain conclusion. Sex cord stromal tumour of the ovary showed strong association with epithelial hyperplasia of the fallopian tube.