Detection of endometrial nerve fibres – a novel technique to diagnose endometriosis


Purpose: To detect the presence of nerve fibres by immunohistochemical staining method using anti-PGP9.5 antibody in the eutopic and ectopic endometrium of patients with endometriosis.

Methods: Patients presenting with pelvic pain and/or infertility were subjected to premenstrual endometrial biopsy during laparoscopy. Thirty patients diagnosed with endometriosis formed the study group and 30 patients with no histological/laparoscopic evidence of endometriosis formed the control group.

Endometrial biopsy tissue of all the 60 patients along with biopsy from endometriomas and peritoneal implants of confirmed cases of endometriosis (n = 30) were subjected to immunohistochemical staining to detect nerve fibres.

Results: Neural elements were observed in the eutopic endometrium of 24/30 cases (80%) in the study group and in none of the controls (p<0.01), while the sections from endometriomas and peritoneal implants demonstrated nerve fibres in all the 30 cases of confirmed endometriosis. The nerve fibres in the endometrial tissue were heterogeneously distributed and were sparse in number with an observed density of <1 nerve fibre/mm² while they were present in large numbers in the ectopic endometrium (10 ± 3 nerve fibre/mm²). Also, the presence of nerve fibres in the eutopic and ectopic endometrium positively correlated with the intensity of pain (p<0.001).

Conclusion: Eutopic endometrium exhibited positivity for nerve fibres in 24 out of 30 cases of proven endometriosis, though the density of nerve fibres was low as compared to that reported in a few studies.

J Endometr Pelvic Pain Disord 2013; 5(4): 144 - 150




Garima Yadav, Gita Radhakrishnan, Navjeevan Singh, A.G. Radhika

Article History


Financial Support: Financial support has been received from Intra Mural Research Grant, UCMS, Delhi, India, and the Indian Council of Medical Research, New Delhi, India.
Conflict of Interest: None.

This article is available as full text PDF.

Download any of the following attachments:


Endometriosis is a common gynaecological disorder that poses a dilemma in diagnosis because of nonspecific clinical presentation. In many women with endometriosis, no abnormality is detected during the clinical examination (1) while uterosacral tenderness or palpable ovarian cyst when detected may suggest endometriosis. Among the imaging studies for endometriosis, ultrasound (USG), computed tomography and magnetic resonance imaging (MRI) may be useful but with limited diagnostic importance in early stages of the disease. Several studies have been attempted to detect a serum marker for endometriosis, for instance measurement of CA-125 with only limited efficacy (2), and histological confirmation of laparoscopic-directed biopsy remains the gold standard for diagnosing endometriosis. As laparoscopy is an invasive procedure and is associated with several complications, diagnosing endometriosis by lesser invasive modalities would be a preferred option.

Although it is recognized that endometriosis is an inflammatory condition with the presence of multiple leucocytes of different types in both endometriotic plaques and eutopic endometrium (3), the mechanism of pain generation is not well understood. There is a possibility of local production of pain-stimulating molecules from the leucocytes, but limited information is available on pain-stimulation pathways for mechanism of sensory nerve activation in eutopic endometrium or endometriotic plaques (4).

Lessey et al (5) and Ota et al (6) have suggested that eutopic endometrium of women with endometriosis demonstrates multiple molecular and cellular abnormalities, based on their findings. Fraser et al (7) speculated that this might include increased expression of nerve growth factors and cytokines in the eutopic endometrium of patients with endometriosis. They demonstrated the presence of fine, unmyelinated sensory nerve fibres in the functional layer of endometrium, only in women with endometriosis, whereas Bokor et al (8) also found nerve fibres in the endometrial samples of women without endometriosis, albeit at a much lower density. These nerve fibres have also been observed in the ectopic plaques of peritoneal endometriosis.

Barcena de Arellano et al (9) have refuted the assumed neurotropic properties of eutopic endometrium and concluded that there was no significant difference with regards to the neuritic outgrowth of sensory ganglia in patients with or without endometriosis. Later on in 2013 (10), the same authors also found that nerve fibre density is altered even in the unaffected peritoneum close to peritoneal endometriosis with significantly elevated levels of nerve growth factor (NGF) and interleukin (IL)-1β expression in comparison to healthy peritoneum. In a recent study by Leslie et al (11) nerve fibres were detected in endometrial biopsies (EBs) of both healthy patients and patients with endometriosis.

Nerve fibres were demonstrated using an antibody for polyclonal rabbit anti-PGP9.5, a highly specific pan-neuronal marker, and anti-human neural factor (NF), a highly specific marker for myelinated nerve fibres. PGP9.5 has only recently been evaluated in a few studies for detecting nerve fibres in the functional layer of endometrium (7, 8, 12) (Tab. I). It is highly expressed in vertebrate neurons and neuroendocrine cells where it constitutes 1%-5% of the total soluble proteins. Thus, the antibody displays a cytoplasmic staining pattern in several tumours with neural elements like carcinoid tumours, schwannomas, astrocytomas and neuroblastomas in a dilution of 1:100-1:200.


Study Number of cases Sensitivity Specificity Density of nerve fibres
Fraser et al, 2006 (7) 35 100% 100% 10 ± 7/mm2
Bokor et al, 2009 (8) 20 95% 100% 1.96 ± 2.73/mm2
Al-Jefout et al, 2009 (12) 64 83% 98% 2.7/mm2

If endometrial PGP9.5 in eutopic endometrium is confirmed, the diagnosis of endometriosis may easily be arrived at by performing a simple, less invasive EB, thereby overcoming the need for a laparoscopic procedure.

The present study was undertaken to evaluate the diagnostic accuracy of detection of nerve fibres in eutopic endometrium and biopsy from peritoneal implants and endometriomas in cases of endometriosis.


This study was conducted at the UCMS and GTB Hospital, a tertiary care institution in Delhi, India. A clearance from the ethical committee of the institute was obtained and written informed consent was taken from all the subjects. The study was conducted from November 2009 to April 2012.


Women in the reproductive age group of 15 to 45 years seeking treatment for chronic pelvic pain (described as pelvic pain of duration >6 months affecting the functional ability of a patient and necessitating medical treatment) and/or infertility (defined as failure to conceive after 1 year of unprotected intercourse) diagnosed to have endometriosis based on laparoscopic criteria followed by histological confirmation formed the study group (n = 30), while patients with either chronic pelvic pain or infertility with no visual or histological evidence of endometriosis formed the control group (n = 30). Patients who had a history of hormonal therapy in the preceding 3 months, acute pelvic inflammatory disease (PID), suspected pregnancy, suspected or diagnosed genital malignancy, undiagnosed vaginal bleeding, documented genital tuberculosis along with women in whom laparoscopy was contraindicated and those who were not willing for laparoscopy were excluded from the study.


Initial assessment

A detailed history regarding the chief complaint and associated pain (dysmenorrhoea, dyschezia, dyspareunia or generalized chronic pelvic pain), if present, was taken and patient’s pain intensity was scored 1-10 on a visual analogue scale (VAS) for pain.

Specific details in the examination were documented regarding the presence of abdominal tenderness, palpable abdominopelvic mass, restricted uterine mobility, palpable adnexal mass, nodules in pouch of Douglas (POD) and uterosacral tenderness on per-vaginal (P/V) and/or per-rectal (P/R) examination. Transvaginal sonography for detailed echotexture of uterus and adnexa was done. Routine haematological, biochemical and radiological investigations for fitness for anaesthesia were done prior to surgery.

Premenstrual EB with laparoscopy/hysterolaparoscopy was performed under general anaesthesia. EB in these patients was performed after laparoscopic evaluation, biopsy of suspected lesions and chromopertubation where indicated. Appropriate therapeutic surgical interventions were also carried out as indicated. The reference standard for diagnosing endometriosis was taken as laparoscopic visualization and histological confirmation of suspected lesions. Surgical staging of the disease as per the Revised American Fertility Scoring System (rAFS) (13) was also done.

Endometrial samples of patients in both study and control groups and laparoscopic-directed samples of endometriotic implants of 30 patients with confirmed endometriosis were subjected to immunohistochemical staining by using the two-step technique in which after antigen retrieval (PGP9.5), primary antibody was applied (anti-PGP9.5; dilution 1:700) and incubated for 1 hour at room temperature, following which secondary (biotinylated Ab) and tertiary antibodies (streptavidin) were applied one after the other at an interval of 30 minutes. After an incubation of 30 minutes, staining was done with DAB chromogen followed by counter-staining with haemotoxylin. Sections were then air dried and mounted for evaluation for neural elements by a senior pathologist who was blinded to the study and control EBs.

Outcome measures

Primary outcome of interest was to evaluate the presence of nerve fibres in eutopic and ectopic endometrium recovered from both endometriosis and healthy patients.

Secondary outcome of interest was to evaluate the potential correlation between nerve fibre density in the eutopic endometrium and VAS score for pain or stage of endometriosis.

Statistical analysis

A 2 x 2 contingency table was used to find the sensitivity, specificity, positive predictive value and negative predictive value of the presenting complaints, clinical diagnosis and nerve fibres in premenstrual endometrial biopsy (PMEB) for diagnosing endometriosis. One-way ANOVA followed by Tukey’s test was used to find the association between VAS score, type of endometriotic lesion and presence or absence of nerve fibres in EB. Chi-square test was used to find the difference between the presenting complaints in the study and control groups along with association between VAS score and stage of endometriosis. Fisher exact test was used to assess the difference in the presence of nerve fibres in eutopic endometrium of the study and control groups.


Among 60 subjects included in the study, 32 patients (53.33%) had only infertility as the presenting complaint, 19 patients (31.67%) presented with chronic pelvic pain (CPP) while the remaining 9 patients (15%) had infertility with associated pain symptoms like dysmenorrhoea, dyspareunia, dyschezia, etc.

The study and control groups were similar in terms of age distribution (p = 0.06). However, there was significant difference in the presenting complaint between the study and control groups, with CPP seen in 50% (15 cases) of study group as compared to 13.33% (4 cases) of control group (p<0.001) (Tab. II).


Parameter Subcategory Study group Control group p value
aTwo-way ANOVA followed by Tukey's test.
bChi-square test.
cFisher's exact test.
dStudent's t test.
Age (mean ± SD) Infertility 28.27 ± 4.22 26.77 ± 4.169 0.06a
Chronic pelvic pain 34.47 ± 7.31 29.50 ± 5.07
Presenting complaint Infertility 15 (50%) 26 (86.67%) <0.001b
Chronic pelvic pain 15 (50%) 4 (13.33%)
Pain symptoms Dysmenorrhoea 23 (76.67%) 5 (16.67%) <0.001c
Dyspareunia 13 (43.33%) 3 (10%)
Dyschezia 7 (23.33%) 1 (3.33%)
>1 symptom 13 (43.33%) 3 (10%)
Nil 6 (20%) 25 (83.33%)
Menstrual cycle Regular 28 (93.33%) 29 (96.67%) >0.05b
Irregular 2 (6.67%) 1 (3.33%)
VAS score (mean ± SD) 5.60 ± 2.99 1.33 ± 2.845 <0.001d

Provisional diagnosis of endometriosis based on laparoscopic findings was made in 38 patients, of which biopsies from 30 cases (78.95%) were confirmed on histology. Most of these lesions were active implants of endometriosis. Eight patients who were suspected to have endometriosis on laparoscopy had histopathological diagnosis of corpus luteum cyst (n = 2), simple serous cyst (n = 1), fibrocollagenous tissue with chronic inflammation (n = 2), focal collection of haemosiderin, without endometrial glands and stroma (n = 3).

Nerve fibres in endometrial tissue

Neural elements stained with PGP9.5 antibody were found in 24 of 30 cases in the study group as against none in the control group. These neural elements were identified as either single cell positive staining or linear nerve fibres (Figs. 1 and 2). Spindle-shaped linear nerve fibres were seen in 7 of 24 cases of proven endometriosis.

Eutopic endometrium showing single cell positive staining in endometrial stroma. Immunoperoxidase method using diaminobenzidine (DAB chromogen) as chromogen and anti-PGP9.5 as primary antibody (x 400).

Eutopic endometrium showing linear nerve fibre in endometrial stroma.

This difference with respect to the presence of nerve fibres in the endometrial samples between the study and control groups was found to be statistically significant (p<0.01) (Tab. III). Nerve fibres in the eutopic endometrial tissue were heterogeneously distributed and were sparse in number with an observed density of <1 nerve fibre/mm².


Nerve fibres in EB Study Control p value (Fisher exact test)
Linear nerves present 7 0 0.01
Linear nerves absent 23 30
Neural element present (fibre + single cell) 24 0 <0.001
Neural element absent 6 30

Nerve fibres in ectopic endometrium

A large number of nerve fibres were detected in the peritoneal lesions of endometriosis (10 ± 3 nerve fibre/mm²) (Fig. 3).

Peritoneal biopsy from endometriotic implant showing multiple nerve fibres.

Diagnostic accuracy of the test

When linear nerve fibres along with single cell positivity was considered to be positive for neural element, the sensitivity and specificity of EB in diagnosing endometriosis was 80% and 100%, respectively. However, when only those EB tissues with linear nerve fibres were considered to be positive, the sensitivity decreased to 23.3% while the specificity of the test remained unchanged as none in the control group showed any kind of positive staining (Tab. IV).


Neural element detected No. of cases Sensitivity Specificity PPV NPV
NPV = negative predictive value; PPV = positive predictive value.
Any kind of neural element 24 80% 100% 100% 83.3%
Linear nerve fibres 7 23.3% 100% 100% 56.6%

Correlation of nerve fibres with pain intensity and stage of endometriosis

There was a significant difference in the VAS score depending on the type of endometriotic lesion, i.e. peritoneal lesions were associated with higher pain intensity as compared to the presence of ovarian endometrioma alone (p<0.001). However, pain intensity of the patient as evaluated by the VAS score was not found to have any correlation with the stage of endometriosis.

Among 30 patients of confirmed endometriosis, it was found that patients with any positive staining for neural elements had a significantly higher VAS score as compared to those with negative staining (p<0.001). However, there was no significant difference of VAS score within the subgroups of positive staining (Tab. V).


Absent neural element Single cell positivity Linear nerve fibres p valuea
aOne-way ANOVA followed by Tukey’s test.
VAS score (mean ± SD) 0.33 ± 0.816 (n = 6) 6.56 ± 0.922 (n = 17) 8.0 ± 2.19 (n = 7) <0.001


PGP9.5 antibody was used to detect nerve fibres in both eutopic and ectopic endometrium, after standardization of the procedure, a dilution of 1 in 700 and an optimal incubation period of 60 minutes at room temperature was arrived at, to overcome undesired background staining. There is no universal recommendation either for the dilution to be used or for the standard procedure for staining nerve fibres in the endometrium but in previous studies the dilution ranged from 1:500 to 1:1400.

The neural elements could be visualized either as scantily distributed single cell positivity or as linear nerve fibres in the eutopic endometrium (Figs. 1 and 2). However, the ectopic EB specimens revealed a large number of nerve fibres (Fig. 3).

It was found that among 30 EBs of patients with endometriosis, 23.33% (7/30) revealed definite linear spindle-shaped nerve fibres especially near the endometrial glands, whereas 56.67% (17/30) showed only single cell positivity. Though PGP9.5 is a highly specific pan-neuronal marker, it might show weak staining with epithelium, muscular tissue and connective tissue. While a reasoning can be put forward to consider these single cell positive staining to be nonspecific, due to the fact that such staining was not observed in any of the EB specimens of the control group, we considered them as positive for neural antigens.

Therefore, when single cell positivity in EB was considered to be positive for nerves, this test yielded a high sensitivity and specificity of 80% and 100%, respectively, and a positive predictive value and negative predictive value of 100% and 83.3%, respectively. Whereas when only the EBs with linear nerve fibres were taken as positive, the test gave a sensitivity and specificity of 23.3% and 100%, respectively, along with a positive predictive value and negative predictive value of 100% and 56.6%, respectively.

Fraser et al reported small sensory C nerve fibres in the eutopic endometrium of all the 35 cases of endometriosis and in none of the EB specimens of patients without endometriosis (7). In this study, the nerve density in the endometrial curettage was calculated using the criteria elucidated by Fraser et al and the density of nerve fibres detected was <1 nerve fibre/mm², as compared to a 10 ± 7/mm² (mean density ± SD) in their study (7).

Further, in a study by Bokor et al of secretory phase endometrium samples (n = 40), obtained from women with laparoscopically/histologically confirmed minimal-mild endometriosis (n = 20) and from women with a normal pelvis (n = 20), the density of small nerve fibres was found to be 14 times higher in the endometrium of women with endometriosis (1.96 ± 2.73) when compared with women without endometriosis (0.14 ± 0.46, p<0.0001) (8). Evers and van Steirteghem (14) reviewed the above two studies and suggested that this test may do extremely well in tertiary care centres where a high prevalence of patients with endometriosis are seen, but at the same time the limited understanding of this disease warrants the degree of effort and expenditure in its diagnosis.

Al-Jefout et al (12) demonstrated a mean nerve fibre density of 2.7 nerves/mm² ( ± 3.5 SD) in the functional layer of endometrium of 63 of 64 women confirmed to have endometriosis on laparoscopy. The specificity and sensitivity were 83% and 98%, respectively, positive predictive value was 91% and negative predictive value was 96%.

None of the above studies have described the various patterns of staining of neural elements, which we describe as either single cell positivity (17 cases) or linear nerve fibres (7 cases).

In contrast to the sparse density of nerve fibres in endometrial specimens, all the peritoneal biopsies showed the presence of large number of fine nerves irrespective of the stage of disease, the type of endometriotic lesion or the pain score. This finding is in cohesion with the findings of Barcena et al (10), who suggested that peritoneal implants are associated with increased expression of nerve growth factors and interleukins as compared to the eutopic endometrium and this could explain the development of endometriosis and pain mechanism in this disease.

The presence of peritoneal endometriosis was associated with a higher pain score irrespective of the presence of ovarian endometriosis, and the difference in VAS score was statistically significant (p<0.001). However, no significant association was found between VAS score for pain and stage of endometriosis according to the rAFS score (p>0.05). Chapron et al (15) also described that pain associated with endometriosis seldom correlated with the extent of disease.

The mean VAS score was significantly higher in patients whose endometrium was positive for nerves and the number of patients exhibiting nerve fibres in eutopic endometrium was also significantly higher when the VAS score was ≥6 (p<0.001). Similarly, the association of pain and nerve fibres in eutopic endometrium has been reported by Zhang et al (16). This was also seen in patients with pain due to adenomyosis, uterine fibroids, or endometriosis with adenomyosis.

Thus, the present study demonstrates a possible utility of a novel diagnostic test for endometriosis, by evaluating the endometrial tissue for the presence of nerve fibres using anti-PGP9.5 for immunohistochemical staining. EB is certainly a less invasive procedure as compared to laparoscopy, which however is the gold standard procedure, but because of the invasive nature, it is often postponed by both clinicians and patients alike. Besides being an invasive procedure with occasional complications, reservations to resorting to an early laparoscopy may be many, especially in young, unmarried girls.

Before the lesions are large enough to be picked up by imaging modalities like USG and MRI, a simple EB, if proven to be useful, can help initiate early and specific therapy. Benefits of laparoscopy cannot, however, be undermined as besides being diagnostic, definite operative procedures can be undertaken in the same sitting.

Attempts to diagnose endometriosis by documenting nerve fibres in EB are in fact in its conceptual stage and only a few hundreds of patients have been studied the world over. This study, though performed in just 30 cases of endometriosis, is possibly the first of its kind to be done in India and this approach seems to hold promise.

As with any immunohistochemical stain, the economics of the procedure need to be addressed as well. The test appears to be cost-effective, as 1 mL of concentrated antibody can be used to stain around 1500 slides, and after considering the cost of other facilities (technical staff, buffers, incubators, etc.), it would cost around INR 500 ($8) for one patient, which is much cheaper as compared to the cost of a diagnostic laparoscopy procedure.


Financial Support: Financial support has been received from Intra Mural Research Grant, UCMS, Delhi, India, and the Indian Council of Medical Research, New Delhi, India.
Conflict of Interest: None.
  • 1. Berek JS Endometriosis. In: Berek JS ed. Berek and Novak’s gynecology New Delhi Wolters Kluwer India 2009 1137 1145 Google Scholar
  • 2. O’Shaughnessy A.,Check JH.,Nowroozi K.,Lurie D. CA 125 levels measured in different phases of the menstrual cycle in screening for endometriosis. Obstet Gynecol 1993; 81: 99-103 Google Scholar
  • 3. Khan KN.,Masuzaki H.,Fujishita A.,Kitajima M.,Sekine I.,Ishimaru T. Differential macrophage infiltration in early and advanced endometriosis and adjacent peritoneum. Fertil Steril 2004; 81: 652-661 Google Scholar
  • 4. Lu VinhPhuc Tran.,Tokushige N.,Berbic M.,Markham R.,Fraser IS. Macrophages and nerve fibers in peritoneal endometriosis. Hum Reprod 2009; 24: 835-841 Google Scholar
  • 5. Lessey BA.,Castelbaum AJ.,Sawin SW. Aberrant integrin expression in the endometrium of women with endometriosis. J Clin Endocrinol Metab 1994; 79: 643-649 Google Scholar
  • 6. Ota H.,Tanaka T. Integrin adhesion molecules in the endometrial glandular epithelium in patients with endometriosis or adenomyosis. J Obstet Gynaecol Res 1997; 23: 485-491 Google Scholar
  • 7. Tokushige N.,Markham R.,Russell P.,Fraser IS. High density of small nerve fibres in the functional layer of the endometrium in women with endometriosis. Hum Reprod 2006; 21: 782-787 Google Scholar
  • 8. Bokor A.,Kyama CM.,Vercruysse L. Density of small diameter sensory nerve fibres in endometrium: a semi-invasive diagnostic test for minimal to mild endometriosis. Hum Reprod 2009; 24: 3025-3032 Google Scholar
  • 9. Barcena de Arellano ML.,Arnold J.,Sacher F. Eutopic endometrium from women with endometriosis does not exhibit neurotrophic properties. J Neuroimmunol 2012; 249: 49-55 Google Scholar
  • 10. Arnold J.,Vercellino GF.,Chiantera V.,Schneider A.,Mechsner S.,Barcena de Arellano ML. Neuroimmunomodulatory alterations in non-lesional peritoneum close to peritoneal endometriosis. Neuroimmunomodulation 2013; 20: 9-18 Google Scholar
  • 11. Leslie C.,Ma T.,McElhinney B.,Leake R.,Stewart CJ. Is the detection of endometrial nerve fibers useful in the diagnosis of endometriosis? Int J Gynecol Pathol 2013; 32: 149-155 Google Scholar
  • 12. Al-Jefout M.,Dezarnaulds G.,Cooper M. Diagnosis of endometriosis by detection of nerve fibres in an endometrial biopsy: a double blind study. Hum Reprod 2009; 24: 3019-3024 Google Scholar
  • 13. Revised American Society for Reproductive Medicine. Fertil Steril 1997; 67: 817-821 Google Scholar
  • 14. Evers JL.,Van Steirteghem AC. All that glistens is not gold. Hum Reprod 2009; 24: 2972-2973 Google Scholar
  • 15. Chapron C.,Fauconnier A.,Dubuisson JB.,Barakat H.,Vieira M.,Bréart G. Deep infiltrating endometriosis: relation between severity of dysmenorrhoea and extent of disease. Hum Reprod 2003; 18: 760-766 Google Scholar
  • 16. Zhang X.,Lu B.,Huang X.,Xu H.,Zhou C.,Lin J. Endometrial nerve fibers in women with endometriosis, adenomyosis, and uterine fibroids. Fertil Steril 2009; 92: 1799-1801 Google Scholar



  • Department of Obstetrics and Gynaecology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Delhi - India
  • Department of Pathology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Delhi - India

Article usage statistics

The blue line displays unique views in the time frame indicated.
The yellow line displays unique downloads.
Views and downloads are counted only once per session.

No supplementary material is available for this article.