BACKGROUND Uterine cancer is the most common gynaecological cancer in the United Kingdom
with approximately 9,400 new cases diagnosed each year, over 95% arising from the
endometrium. Almost all endometrial cancers are diagnosed after the menopause, with most
being detected after a woman presents with bleeding. Any woman presenting to her general
practitioner or accident and emergency department with post-menopausal bleeding in the
United Kingdom should receive an urgent two week wait referral to rapid access
gynaecology services. At this appointment women will typically undergo a transvaginal
ultrasound scan to assess the gynaecological anatomy, in particular the endometrium. If
the endometrium is deemed to be thickened, at more than 4 or 5mm depending on local
guidance, sampling is required to rule out endometrial cancer or its precursor,
endometrial hyperplasia.
There are two well established routes for obtaining an endometrial biopsy. These are
'blind' office endometrial sampling and hysteroscopy. If ultrasound has shown any focal
pathology such as an endometrial polyp or a submucosal fibroid, then the patient must
undergo hysteroscopy so this can be directly visualised and confidently removed. Whilst a
highly effective procedure, its primary drawbacks are higher costs, a larger side-effect
profile and potentially greater pain than the alternative, blind office endometrial
sampling.
If a patient has a uniformly thickened endometrium without any suspicion of focal
endometrial pathology they can undergo blind office sampling of the endometrium. This is
performed with a variety of thin, specially designed plastic biopsy devices that obtain
the sample via localised trauma and vacuum suction. This procedure is less invasive,
quicker and simpler to perform than a hysteroscopy, however it has a higher failure rate
with roughly 4 in 10 patient not having an adequate sample obtained. This is due to
either the endometrial sampler not being successfully introduced into the endometrial
cavity, or when successfully introduced, the sample obtained being inadequate for
histopathological assessment. These patients are then committed to a hysteroscopy so
their endometrium can be safely assessed.
PROPOSED STUDY It is theorised that by using ultrasound guidance, the failure rate of
office endometrial biopsies could be reduced. We propose to test in an adequately sized
randomised controlled trial, the hypothesis that Pipelle® (Pipelle de Cornier)
endometrial biopsies, performed under transabdominal ultrasound guidance, will have a
higher success rate than blind biopsies. Potential benefits to patients of positive
findings in this study include a reduced number of investigations, reductions in pain and
increases in patient satisfaction, and a reduction in time from presentation to
definitive treatment.
STUDY DESIGN This will be a single centre, prospective, non-blinded randomised controlled
trial carried out in the outpatient gynaecological oncology department at Queen
Charlotte's and Chelsea Hospital, part of Imperial College Healthcare NHS Trust. The
trial will be carried out by gynaecology doctors of registrar and consultant grade who
are deemed competent in gynaecological ultrasound and endometrial sampling. Both the
doctor and patient will be aware of the study arm allocated. The trial will aim to have
the same number of participants in each arm. The study arms are: (i) Blind Endometrial
Biopsy and (ii) Ultrasound Guided Endometrial Biopsy. 92 patients will be enrolled, with
46 in each arm.
Once randomised and ready for an endometrial biopsy to be performed, those allocated to
the intervention arm will have a transabdominal ultrasound performed by another
gynaecologist in the research team while the Pipelle® biopsy is performed. During the
procedure, the research doctor will, to the best of their ability, ensure the Pipelle® is
introduced to the level of the fundus and that a sample representative of the endometrium
is obtained.
As the patient leaves the clinic room, they will be asked to completed a printed Visual
Analogue Scale questionnaire to assess anxiety, pain during and after the procedure, and
overall patient satisfaction.
The study endpoint will be whichever is the later of: (i) the histopathology for the
final participant being reported with an appropriate management plan made, or (ii) when
the final patient has an attempted Pipelle® endometrial biopsy if this was a failed
attempt.
DATA COLLECTION Demographic patient data collected will include age, BMI, years
post-menopause and parity. Pre-procedural data will record history of gynaecological
surgery including minor procedures, risk factors for endometrial cancer, medical history,
and transvaginal ultrasound scan findings. Intra-procedural data collected for both
groups will record whether access to the endometrial cavity was obtained and if not,
whether this was due to cervical stenosis or other factors. The contents of the Pipelle®
aspiration obtained will be visually assessed by the doctor and if deemed to be an
adequate sample this will be sent on for histopathological assessment. In addition, the
intervention arm will have intra-procedural data collected to assess whether the Pipelle®
was clearly visualised on transabdominal ultrasound and seen at the fundus.
Post-procedural data for both groups will record the number of samples deemed adequate
for assessment by the histopathology department and for those not deemed adequate, the
specified reason. The histology results will be recorded and the proportion of patients
with benign pathology, endometrial hyperplasia (with and without atypia) and endometrial
cancer recorded. Any complications occurring during or post procedure will be recorded.
Patients will be monitored for repeat attendances during the study, to ensure that any
patient later diagnosed with higher order pathology has this recorded.
WITHDRAWAL CRITERIA Patients enrolled in the trial will be free to withdraw consent at
any point without giving reason for doing so. If this occurs their care will be returned
to standard practice. If the doctor performing the procedure feels at any point it is not
safe to proceed, or it is no longer in the patient's best interests, then the procedure
will be abandoned and a discussion had as to how best investigate or treat their
post-menopausal bleeding. Any decision and reason to withdraw from the study will be
recorded.
ASSESSMENT AND FOLLOW-UP Follow up for all trial participants will be the same as for
patients undergoing standard assessment of post-menopausal bleeding. Histopathology
results will be reviewed within two weeks of the sample being taken and if benign the
patient will be discharged from the clinic. Their GP will receive a letter stating that
their investigations are normal and their involvement in the trial has come to an end. If
they have further episodes of post-menopausal bleeding their GP will be asked to refer
them back to the rapid access clinic on a suspected cancer pathway. If non-benign
pathology is found, then the patient will be invited for a further appointment and follow
up will be determined as per the relevant protocol. If the laboratory reports that the
sample taken is not adequate for histopathological analysis the patient will be contacted
and recommended to undergo a hysteroscopy for further assessment.
ETHICS APPROVAL The Study Coordination Centre has obtained approval from the
London-London Bridge Research Ethics Committee (REC) and Health Research Authority (HRA).
The study will be conducted in accordance with the recommendations for physicians
involved in research on human subjects adopted by the 18th World Medical Assembly,
Helsinki 1964 and later revisions.