Uterine factor infertility accounts for approximately 10–15% of infertile couples. Uterine anomalies may be either congenital or acquired [1]. Congenital uterine malformations vary from subtle anomalies such as arcuate uterus to a major anomaly such as Mullerian agenesis. Different classifications for uterine anomalies have been published such as those by the American Fertility Society (1988), European Society of Human Reproduction and Embryology (ESHRE) and the European Society of Gynaecologic Endoscopy (ESGE) in 2013, along with modifications such as American Society of Reproductive Medicine (ASRM) (2016) and Congenital Uterine Malformation by Experts (CUME) classification for arcuate and T-shaped uterus [2, 3, 4, 5, 6]. Acquired causes include myometrial lesions as myomas, adenomyosis, adenomyoma and sarcomas. Endometrium may be the cause of infertility as in cases of endometrial polyps, endometrial hyperplasia, chronic endometritis, and endometrial carcinoma. Specific entities in the acquired causes of uterine disease include intrauterine adhesions, abnormal caesarean scar, and scar niche [1].

Chronic endometritis (CE) is a chronic inflammation of the endometrium caused by bacterial infection of the cavity. The most common organisms responsible for this pathology are Enterococcus faecalis, Enterobacteriaceae, Gardnerella vaginalis, Streptococci, Staphylococcus species and Mycoplasma. Sexually transmitted infections, such as Chlamydia trachomatis, Ureaplasma urealyticum, and gonorrhoeae are potential causes of the infection [7, 8].

CE is asymptomatic in most cases, although it may present with pelvic pain, abnormal uterine bleeding, dyspareunia, and vaginal discharge [9]. While its actual prevalence in the general population is difficult to calculate, researchers estimate it as ranging from 0.8 to 19% [10]. Although CE is asymptomatic, it may account for lower pregnancy rates from both natural and in vitro fertilization conceptions; further, it may lead to poor obstetric outcome and neonatal complications [11, 12, 13, 14, 15, 16, 17]. CE has been diagnosed in 2.8–39% of women suffering from infertility [14, 18, 19]. However, prevalence may be as high as 60% in patients with recurrent pregnancy loss or 66% in patients with repeated implantation failure [20]. Diagnosis is difficult due to lack of specific symptoms as well as diagnostic ultrasound findings. Histopathology is diagnostic for CE by the presence of plasma cells in the endometrial stroma. Histopathology often has false positive results when the biopsy is taken in the late luteal or during menstruation [9].

Hysteroscopy is a diagnostic tool for CE [8, 20]. Hysteroscopic diagnosis depends on subjective findings observed such as stromal oedema, focal or diffuse hyperaemia, and/or the presence of micropolypi [20]. Treatment of CE may improve in vitro fertilization (IVF) success in women with recurrent implantation failure. Endometrial biopsy is recommended to confirm CE diagnosis before initiating IVF cycle [21].

The uterine cavity can be assessed through different techniques such as 3D transvaginal sonography, hysterosalpingogram, and hysteroscopy [22]. Hysterosalpingogram has low specificity, as well as high false-negative and false-positive rates [23]. Anomalies of the uterine cavity can be diagnosed by 3D ultrasound, a tool that exceeds diagnostic hysteroscopy, demonstrating accuracy as high as 90% in diagnosing uterine cavity abnormalities [24, 25]. Hysteroscopy enables visual assessment of the cervical canal and uterine cavity for any pathology such as intrauterine adhesions, polyps, grade zero myomas, chronic endometritis, or congenital uterine anomalies that may affect implantation. Hysteroscopy is also a good therapeutic tool in most conditions, while hysteroscopic guided biopsy is more informative about endometrial pathology [26, 27]. Two prospective and randomized controlled trials (RCTs) showed improvement in the outcome of assisted reproductive techniques (ART) cycles following office hysteroscopy [28, 29] despite lack of definitive pathological diagnosis [30]. On the other hand, a large, randomized study (INSIGHT) found no improvement in the live birth rate after office hysteroscopy before first IVF attempt in cases with normal transvaginal ultrasound [31]. Another study (TROPHY) found no improvement even in cases of previous ART failure [32].

The aim of this study was to test the accuracy of office hysteroscopy in the diagnosis of chronic endometritis as compared to histopathological diagnosis.

Data were entered and analysed using IBM SPSS software package version 21.0, by SPSS Inc. Chicago, IL, USA [34]. Qualitative data were described using numbers and percentages. Quantitative data were described using range (minimum and maximum), mean, standard deviation and median. Significance of the obtained results was judged at the 5% level.

The study was conducted on 220 women with complaints of primary infertility scheduled for the first attempt of ICSI. There was no drop out as the two procedures (office hysteroscopy and endometrial biopsy) were done in the same setting. Patient age ranged from 20–40 years, mean 29.99 $\pm$ SD 5.10 years. 174 cases (79.1%) were diagnosed as CE at hysteroscopy (Table 1), while 162 (73.6%) cases were positive at histopathology (Table 2). Fig. 1 shows hyperaemia which was found in 99.4% of the cases of CE, followed by oedema which was present in 74.7% of cases. Micropolypi were found in 58.6% of cases with CE. Fibrotic nodules and fibrosis or filmy adhesions were seen in only 4.6% and 2.9% of cases with CE, respectively. The different hysteroscopic findings are shown in Fig. 2. The sensitivity, specificity, and positive and negative predictive values of hysteroscopy were 93.83%, 62.07%, 87.36% and 78.26%, respectively (Table 3).

Fig. 1.

Distribution of the hysteroscopic findings in cases of chronic endometritis (n = 220).

Fig. 2.

Different appearance of chronic endometritis during office hysteroscopy. (a) hypervascularity, (b) micropolyp near right ostia, (c) endometrial oedema and hyperaemia, and (d) endometrial filmy adhesions separated by fluid gush.

IHC staining of the specimen for cases diagnosed as CE by histopathology showed that 16% of cases were negative for CD138 and 31.5% showed few CD138 cells while 52.5% showed considerable concentration of CD138 (Table 3). Repeating the endometrial biopsy after treatment showed that 22 cases (13.6%) still demonstrated chronic endometritis and 140 cases (86.4%) resolved (Table 3). The distribution of the studied cases according to IHC results is shown in (Table 4). It also shows the result of endometrial biopsy after treatment of endometritis. It was normal in 86.4% of treated cases and 13.6% in persistent cases.

Chronic endometritis is a pathology that is hardly diagnosed via non-invasive approaches. TVS and MRI cannot confirm the diagnosis, while CE can be suspected only in patients with uterine synechia, hydrometra or pyometra. CE is asymptomatic in many cases or may cause mild symptoms such as vaginal spotting, bleeding and vague pelvic pain. The causative pathogen was found to be non-gonococcal, non-chlamydial infections [35].

Cicinelli et al. [35] established diagnostic criteria for CE using hysteroscopy. Fluid hysteroscopy is more sensitive than CO${}_2$ hysteroscopy in the detection of CE.; saline does not affect endometrial vascularity and helps endometrial micropolyps to float. The authors’ diagnostic criteria for CE included: oedema of endometrial stroma, focal or diffuse endometrial hyperaemia, and micropolyps (1 mm in size). Stromal oedema plus hyperaemia is a predictor of inflammation. Micropolyps are small, vascularized ingrowths that are covered by inflamed endometrium, and it shows accumulation of inflammatory cells (lymphocytes, plasma cells, or eosinophilic granulocytes) and stromal oedema; there may be an expression of an active inflammatory process of the endometrium and of massive release of interleukins and certain growth factors [36].

Hysteroscopy has the advantage of direct inspection of the uterine cavity and the detection of signs of inflammation; it may serve as a first choice for diagnosis as it can be done as an outpatient procedure [37]. Unsuspected uterine pathologies have been diagnosed in 11 while 45% of women received hysteroscopy before IVF [38]. Reports have concluded that hysteroscopy in the cycle preceding IVF could be helpful for patients with recurrent implantation failure (RIF) [39]. One of the pathologies that cannot be diagnosed with ultrasound and HSG is chronic endometritis. The gold standard for CE diagnosis is the presence of plasma cells in the endometrial stroma [40]. However, leukocytes that may be found in the endometrium, especially in the luteal phase, must be taken into consideration. Additional research has confirmed that fluid hysteroscopy can diagnose CE reliably based on the presence of diagnostic signs such as micropolyps, stromal oedema and focal or diffuse hyperaemia [7, 35].

The current study was conducted on 220 patients suffering from infertility, selected from a private infertility clinic. Patient age was 20–40 years, mean $\pm$ SD 29.99 $\pm$ 5.10 years. 174 cases (79.1%) were diagnosed as CE at hysteroscopy, and 162 (73.6%) were positive at histopathology. Following treatment of 162 CE cases, 140 cases (86.4%) resolved, while 22 cases (13.6%) still showing chronic endometritis. The prevalence of chronic endometritis at hysteroscopy was based on the presence of hyperaemia, oedema and, micropolyps. The criteria on histopathology was based on the presence of plasma cells. The sensitivity, specificity, and positive and negative predictive values of hysteroscopy were 93.83%, 62.07%, 87.36% and 78.26%, respectively.

These results are in line with those of Gebauer et al. [37] who found presence of micropolyps in 51 patients in their study group (83 patients) by hysteroscopy. Polyps were diagnosed by endometrial curettage alone in only 43% of these cases. Our results were also supported by Garuti et al. [41] who found that hysteroscopy had a high accuracy for the diagnosis of micropolyps (95.3% sensitivity, 95.4% specificity, 98.9% NPV, and 81.7% PPV). The results of the present study demonstrate that diagnosis of absence of inflammation at hysteroscopy is very reliable, while the detection of endometrial hyperaemia, oedema and micropolyps strongly suggests the presence of CE. Our study carries acceptable negative predictive value (78.26%) and a high positive predictive value (87.36%).

Cicinelli et al. [35] found that 88.6% of patients with positive findings at hysteroscopy were histologically positive for CE, but in our study, 93.48% of them were positive histologically. Our results disagree with those of other researchers who concluded that hysteroscopy is not useful for screening chronic endometritis in asymptomatic infertile women. The rationale for their conclusion assumed that incidence of chronic endometritis in the population is 10%, such that a negative predictive value of 89.1% added no additional tools for risk assessment for CE. However, these researchers used CO${}_2$ and not fluid hysteroscopy; furthermore, the prevalence of chronic endometritis in 10% of infertile women is significantly lower than that reported by most authors [14].

Micropolypi characterized by the presence of inflammatory cells and stromal oedema are likely the expression of an active inflammatory reaction and release of interleukins and certain growth factors [36]. Micropolyps were demonstrated in 58.5% of cases with CE. Other researchers demonstrate that there is a clear correlation between different signs of CE at hysteroscopy and severity of inflammation at histopathology. Our research reveals that there is a good corelation between hysteroscopic and histological evaluation of CE. Previous studies demonstrate that the presence of micropolyps is a reliable sign of inflammation and that this sign has a high positive predictive value [35]. The presence of micropolyps is proportionate to the severity of inflammation, to the severity of mucosal injury and to an abnormal stimulus in the endometrium. Leucocytic infiltration in chronic endometritis is not only a quantitative but also a qualitative variation in infiltration by leucocytes. It exists as an abnormal distribution of endometrial lymphocyte [42].

Micropolypi are minute in size; detection of these minor lesions is easier when saline is used for distention as it allows the micropolyps to float. On the contrary, when using CO${}_2$, micropolypi are flattened against the endometrial surface. The continuous washing of the uterine cavity keeps the subject visually clear in the presence of mucus or bleeding that are frequently present with endometrial inflammation [43]. Moreover, CO${}_2$ provokes an abrupt distention of the endometrial cavity and induces dilatation of capillaries. Petechiae and micro-haemorrhages may therefore occur within the endometrium, rendering diagnosis of endometritis less reliable [43].

Limitations of this study are that CE is a subtle pathology during hysteroscopic examination, and thus requires time to identify such cases. Some hysteroscopic pictures such as hyperaemia and oedema may be present in other pathologies like endometrial hyperplasia which may be seen in PCOS patients seeking fertility.

Office hysteroscopic evaluation is relatively accurate on its own but can be further confirmed by CD138 staining.

MA did the hysteroscopic examination and took the endometrial biopsy, and analysed and interpreted the patient data regarding the studied parameters. IG and MS contributed in the recruitment of the cases, as well as follow-up and data analysis. AA was responsible for histopathologic examination of the biopsies. MA and MS contributed in writing the manuscript. All authors read and approved the final manuscript and agree for publication.

This study was approved from the Ethics committee, faculty of medicine, Alexandria university (number 0304338). Informed consent was taken from all women recruited in the study, that they accept to go through the study and they agree publishing the medical data and the hysteroscopic images in medical journals.

The authors wish to acknowledge the gynaecology outpatient clinic staff for their assistance.

This research received no external funding.

The authors declare no conflict of interest.