Accepted Manuscript Recurrent postmenopausal bleeding: a prospective cohort study Paul P. Smith, Siobhan O’Connor, Janesh Gupta, T Justin Clark
PII:
S1553-4650(14)00202-7
DOI:
10.1016/j.jmig.2014.03.007
Reference:
JMIG 2273
To appear in:
The Journal of Minimally Invasive Gynecology
Received Date: 2 December 2013 Revised Date:
4 March 2014
Accepted Date: 6 March 2014
Please cite this article as: Smith PP, O’Connor S, Gupta J, Clark TJ, Recurrent postmenopausal bleeding: a prospective cohort study, The Journal of Minimally Invasive Gynecology (2014), doi: 10.1016/j.jmig.2014.03.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Smith
ACCEPTED MANUSCRIPT 1 2 3
RI PT
4 5 6 7
9
SC
8
Title: Recurrent postmenopausal bleeding: a prospective cohort study
11
M AN U
10
Author Details: Paul P Smith of Birmingham Women’s Hospital, Birmingham, UK Siobhan O’Connor of Birmingham Women’s Hospital, Birmingham, UK
13
Janesh Gupta of Birmingham Women’s Hospital, Birmingham, UK
14
T Justin Clark of Birmingham Women’s Hospital, Birmingham, UK
TE D
12
15
Corresponding Author: Paul P Smith of Birmingham Women’s Hospital, Birmingham, UK
17
E-mail: [email protected]
18
Phone number: +447932044361
20 21 22
AC C
19
EP
16
23 24 25 26 27
Page 1
Smith
ACCEPTED MANUSCRIPT 28
Condensation:
29
Women who present with recurrent postmenopausal bleeding appear less likely to have
30
malignant endometrial disease, but are more likely to have endometrial polyps.
31
RI PT
32 33 34
SC
35 36
M AN U
37 38 39 40 41
TE D
42 43
46 47 48 49
AC C
45
EP
44
50 51 52 53 54
Page 2
Smith
ACCEPTED MANUSCRIPT
Abstract
56
Study objective. The aim of this study was to estimate the prevalence of genital tract
57
pathologies in women presenting with initial and recurrent postmenopausal bleeding (PMB)
58
to help inform diagnostic pathways
59
Design. Prospective cohort study
60
Design classification. Canadian Task Force Classification: II-2
61
Setting. A large, urban teaching hospital in Birmingham, UK.
62
Patients. A total of 1938 consecutive women presented with postmenopausal bleeding, of
63
which, 106 (5%) women were investigated for a recurrent episode after previously having
64
normal investigations.
65
Interventions. All women underwent a pelvic examination and ultrasound scan. An
66
endometrial biopsy was performed when endometrial thickness was >4mm in women
67
presenting for the first time with PMB, with recourse to outpatient hysteroscopy following
68
correlation between clinical and pathological findings. All women had an endometrial biopsy
69
and outpatient hysteroscopy with a recurrent PMB presentation.
70
Measurements and main results. The chance of having endometrial cancer or hyperplasia
71
with atypia was significantly less in women who presented with recurrent PMB as compared
72
with those presenting with PMB for the first time (0%, 8% respectively, p=.002), but were
73
significantly more likely to have benign endometrial polyps compared with women presenting
74
with PMB for the first time (28%, 19% respectively, RR 1.47 [95% CI 1.07, 2.02], p=.02).
75
Conclusion. Recurrent PMB results in less likelihood of pre-malignant and malignant
76
endometrial disease but one in four women have endometrial polyps as a cause of PMB.
77
First line investigation for women with recurrent PMB should be with tests that have a high
78
accuracy for diagnosing focal pathologies such as outpatient hysteroscopy or saline infusion
79
sonography.
AC C
EP
TE D
M AN U
SC
RI PT
55
80 81
Keywords Postmenopausal bleeding, polyps, endometrial cancer, hysteroscopy.
Page 3
Smith
ACCEPTED MANUSCRIPT
Introduction
83
Postmenopausal bleeding (PMB) is a common condition affecting between 7-15% of the
84
postmenopausal population1. Prompt investigation is needed because there is a 3-10% risk
85
of endometrial cancer2. Guidelines3,4 recommend first line investigation with transvaginal
86
sonography (TVS) to measure endometrial thickness. The prevalence of malignancy is
87
reduced to 12
121
months.
122
cancer, hyperplasia or an untreated polyp during the first episode of bleeding.
RI PT
110
We prospectively collected standardised data on a specially
M AN U
SC
PMB was defined as an
TE D
Women were excluded from the recurrent cohort if they were diagnosed with
123
All women with PMB underwent TVS as a first-line screening test. Final diagnosis for PMB
125
was made using the following reference standards:
EP
124
1. Transvaginal sonography (TVS): if the endometrium was regular, ≤4 mm on TVS with
127
no evidence of fluid, the patient was considered to have an atrophic endometrium
128
and sampling was not performed. The patient was reassured after a lower genital
129
tract examination and discharged. A trained sonographer who was not blinded to the
130
clinical symptoms performed all TVS. Colour Doppler was used when endometrial
131
irregularities were seen.
AC C
126
132
2. Endometrial biopsy (EB): women who screened positive i.e. TVS with EEC >4 mm or
133
irregular endometrium (even if the EEC was ≤4mm), or incomplete visualisation of
134
the endometrium, underwent aspiration endometrial biopsy (EB). A senior pathologist
135
reviewed all histology and was not blinded to the clinical symptoms. Page 5
Smith
ACCEPTED MANUSCRIPT 136
3. Office
hysteroscopy
(OH)(under
conscious sedation):
where
an
outpatient
137
endometrial biopsy (EB) failed (defined as failure to instrument the uterus or a non-
138
diagnostic sample) an OH was arranged and a further biopsy taken unless the cavity
139
seemed atrophic on hysteroscopic visualisation.
140
visualisation was defined as a pale, avascular uterine surface with visible underlying
141
stroma with no endometrial tissue avulsed following contact with the tip of the
142
hysteroscope. OH was performed on all patients with a recurrent parsentation of
143
PMB.
144
fibroids) were suspected on TVS or EB reports. If confirmed on OH, they were
145
removed and sent for histological confirmation. All OH was performed by a senior
146
gynecologist or a specialist nurse who was not blinded to the clinical symptoms.
RI PT
Atrophy on hysteroscopic
M AN U
SC
An OH was also performed where focal pathology (polyps, submucous
147
For the purposes of this study we restricted final diagnosis to a single pathology. When
149
more than one endometrial histological diagnosis was made the patient was categorised
150
using the following hierarchy: Endometrial cancer / hyperplasia with atypia > hyperplasia
151
without atypia > polyps > infection > functional > atrophic. Cancer and hyperplasia with
152
atypia were included together because of the high rate of under-call and progression to
153
cancer where complex hyperplasia is found with cytological atypia13–15.
EP
TE D
148
AC C
154 155
The relative proportion of different pathologies in women with first and recurrent PMB was
156
compared using chi-square test or fisher’s exact test when the expected frequency was less
157
than five. For the purpose of working out relative risk, values of zero were replaced with 0.5.
158
Statistical analysis was performed using SPSSTM version 15.
159 160
Results
161
Page 6
Smith
ACCEPTED MANUSCRIPT Over the 53-month (just under 4.5 years) study period, 1832 women with a median duration
163
of follow up of 35 months (range 1–56 months), presented with PMB for the first time to the
164
Birmingham Women’s Hospital. Of this cohort, 106 (5%) women had recurrent PMB. The
165
median follow up in this subgroup of women with recurrent PMB was 12 months (range 1–45
166
months). All had previously been investigated for PMB and had endometrial cancer or
167
hyperplasia excluded. The mean age (62.9 vs 61.1 years), body mass index (BMI) (30.7 vs
168
31.8 kg/m2), EEC (8.3 vs 7.7 mm) and percentage taking HRT (11.7% vs 12.3%) was similar
169
in women with first or recurrent PMB.
SC
RI PT
162
170
Endometrial cancer or hyperplasia with atypia was diagnosed in 152/1832 (7.8%) women
172
with first presentation PMB but no cases were found in the 106 women with recurrent PMB.
173
In contrast, benign endometrial polyps were found to be significantly more prevalent in the
174
recurrent PMB group compared with first presentation PMB (30/106, 28.3% vs 349/1832,
175
19.6%, p=.02 respectively). The prevalence of all other endometrial pathologies was
176
equivalent between initial and recurrent PMB groups (Table 1). Because of a possible
177
association between HRT and the development of polyps16, we re-analysed the data after
178
removing the women taking HRT. Endometrial polyps remained more prevalent in women
179
with recurrent PMB, but the difference was no longer significant (25/94, 27% in recurrent
180
PMB presentation vs 313/1616 19%, in initial PMB presentation, p=.09). Removing the
181
women taking HRT had no effect on the significant decrease in cancer and atypia in
182
recurrent PMB presentation. No significant differences in the prevalence of endometrial
183
pathologies were noted in women with recurrent PMB when stratified into re-presentation
184
within and beyond 12 months of initial presentation (Table 2).
AC C
EP
TE D
M AN U
171
185 186 187
Discussion
Page 7
Smith
ACCEPTED MANUSCRIPT 188
In our study, 106 (6%) women had recurrent PMB over a 4.5 year period, but no cases of
189
endometrial cancer or hyperplasia with atypia were detected.
190
recurrent PMB had a significantly increased risk of having benign endometrial polyps
191
irrespective of the time to recurrence. The apparent increased prevalence of endometrial
192
polyps with recurrent PMB may reflect the higher accuracy of OH for detecting focal
193
pathologies,
194
selectively in women presenting with PMB for the first time. Alternatively, women with
195
recurrent PMB may have more frequently developed polyps de novo accounting for further
196
bleeding symptoms. In either case, the finding that benign endometrial polyps are more
197
prevalent has potential implications for how to best manage women with recurrent PMB.
198
With it’s focus on diagnosing endometrial cancer, the current consensus for working up
199
women with PMB using first-line TVS EEC measurement3,4,19, may be inappropriate in
200
women with recurrent PMB.
201
benign20,21, and the consequences of diagnosis less serious to health than endometrial
202
cancer, they are associated with abnormal uterine bleeding and removal by polypectomy
203
frequently resolves symptoms20,22. Thus, diagnostic pathways in recurrent PMB should be
204
developed with the aim of diagnosing benign as well as malignant endometrial disease.
RI PT
M AN U
SC
which was undertaken routinely in the recurrent PMB group, but only
TE D
Although the vast majority of endometrial polyps are
EP
205
17,18
However, women with
Figure 1 illustrates a potential testing pathway for women presenting with recurrent PMB
207
based upon the findings from our study. We have introduced saline infusion sonography
208
(SIS) along with OH to make are proposed pathway more generalisable. We propose that
209
OH or saline infusion sonography (SIS), should be incorporated because these tests are
210
more accurate in diagnosing discrete pathologies such as endometrial polyps18,23. Moreover,
211
structured interviews exploring women’ preferences in the evaluation of PMB have shown
212
that women are prepared to undergo more invasive hysteroscopy to evaluate for any
213
additional pathology, rather than adopt the currently recommended expectant management
214
after ultrasound24. It is possible that this preference for additional testing may be even more
AC C
206
Page 8
Smith
ACCEPTED MANUSCRIPT 215
likely with recurrent PMB because of increased anxiety associated with repeated symptoms,
216
even though paradoxically the risk of endometrial cancer appears to be low.
217 The low risk of serious endometrial disease in recurrent PMB after previous normal
219
investigation is reassuring and so reinvestigation of women who have already had
220
endometrial polyps excluded by OH or SIS may be unnecessary. However, if the nature of
221
bleeding has changed (ie increased quantity or persistence) or other relevant symptoms
222
such as pain have developed then it would seem prudent to undertake timely reinvestigation
223
to exclude endometrial cancer (Figure 1). We found no difference in the likelihood of
224
endometrial cancer or hyperplasia in women with recurrent PMB according to whether re-
225
presentation was within or beyond 12 months of initial negative testing. However, the
226
median duration of follow up in our series was only 12 months. Therefore a cautious
227
approach should be adopted, especially in women with recurrent symptoms and longer time
228
intervals from previously negative testing. For this reason the proposed diagnostic algorithm
229
recommends reinvestigation for all women with recurrent PMB after 12 months, regardless
230
of previous testing history (Figure 1).
TE D
M AN U
SC
RI PT
218
231
A strength of our study is that the data were collected prospectively, consecutively and in a
233
standardised fashion minimising bias from incomplete data. However, whilst the
234
generalisability of our findings are limited because they are derived from a single centre, we
235
believe that they are likely to mirror other centres because we adopted testing for PMB in
236
line with current guidelines3,4 from a large, unselected, general postmenopausal population.
237
Several other studies have estimated recurrence rates for PMB and evaluated the
238
prevalence of malignant although not benign uterine pathology9,11,25–27. One study evaluated
239
126 women presenting with recurrent PMB and reported an 8% recurrent PMB27 rate in
240
keeping with our observed estimate of 6%. In contrast to our findings, no significant
241
difference in the incidence of cancer or atypical hyperplasia were observed compared to
AC C
EP
232
Page 9
Smith
ACCEPTED MANUSCRIPT women presenting for the first time. However, initial diagnostic work up was not in line with
243
current recommendations3,4 such that two of the five women with endometrial cancer in the
244
recurrent group represented within a year but neither had undergone initial endometrial
245
sampling despite EEC>9mm. The likelihood of recurrent bleeding has been reported to be
246
as high as 33%11 and the rates of cancer and atypical hyperplasia up to 22.7% (15/66)9.
247
These differences may be explained by differences in patient selection and referral criteria,
248
variations in initial diagnostic work-up and the sample size and duration of follow-up. Closer
249
analysis of these studies revealed that significant pathology usually occurred 12 months
250
after initial investigation9,10,27. When cancer or atypical hyperplasia did present within 12
251
months, endometrial sampling was either not normal, inadequate or not performed at first
252
presentation9,27. It is for this reason that in our suggested pathway (figure 1) all patients with
253
recurrent PMB are investigated with OH or SIS with an attempt at biopsy irrespective of the
254
EEC and that a full review of previous investigations is undertaken. Larger studies are
255
needed to externally validate our findings before we can make robust clinical
256
recommendations.
TE D
M AN U
SC
RI PT
242
257
Although we did not follow up apparently asymptomatic women, it is unlikely that that our low
259
prevalence of serious endometrial disease in women with recurrent PMB reflects high rates
260
of loss to follow up. This is because PMB is an alarming symptom so women are unlikely to
261
ignore it and general practitioners are well educated in the need to rapidly refer.
262
Furthermore, postmenopausal women are not a particularly itinerant population and our
263
hospital is the largest provider of gynaecological care in Birmingham. Another potential
264
criticism of our study is the lack of a common reference standard for diagnosing endometrial
265
disease. Our study was pragmatic, with data collected within a clinical setting such that the
266
indication for further testing with OH or SIS, after the standard use of TVS and EB if
267
EEC≥4mm3,4, was based upon clinico-pathological correlation in the absence of uniform
268
guidance. We also restricted diagnosis to a single pathology so that we may have
AC C
EP
258
Page 10
Smith
ACCEPTED MANUSCRIPT 269
underestimated the prevalence of some pathologies. However, in the presence of multiple
270
endometrial pathologies, we based our final diagnosis according to the most significant
271
pathology and so the clinical implications of our simplification are likely to be negligible.
RI PT
272
Conclusions
274
At least one in twenty women with PMB will have recurrent symptoms. Compared to women
275
presenting for the first time with PMB, the risk of endometrial cancer in recurrent PMB in
276
women who have had previously normal investigations in line with current diagnostic
277
guidelines3,4, appears to be much lower but the risk of having benign endometrial polyps is
278
significantly higher. All women with recurrent PMB, irrespective of when they re-present after
279
their initial negative investigations, should undergo a test with a high sensitivity for focal
280
pathology such as OH or SIS.
281
Disclosure of interests all authors declare no conflict of interests
TE D
282
M AN U
SC
273
283
Contribution to authorship PS produced the original idea, performed the analyses and
285
drafted the article. SOC collected all data. TJC contributed to interpretation. TJC, SOC &
286
JKG revised the article for important intellectual content.
EP
284
AC C
287 288
Details of ethics approval Data collection and storage satisfied local and national
289
regulations. Ethical approval was not required.
290 291
Funding none
292 293
References
294 295
1.
Astrup K, Olivarius N de F. Frequency of spontaneously occurring postmenopausal bleeding in the general population. Acta Obstet Gynecol Scand. 2004 Feb;83(2):203–7.
Page 11
Smith
ACCEPTED MANUSCRIPT 2.
Bachmann LM, ter Riet G, Clark TJ, Gupta JK, Khan KS. Probability analysis for diagnosis of endometrial hyperplasia and cancer in postmenopausal bleeding: an approach for a rational diagnostic workup. Acta Obstet Gynecol Scand. 2003 Jun;82(6):564–9.
300 301
3.
NVOG (Dutch Society of Obsterics and Gynaecology). NVOG Guideline: Abnormal Vaginal Bleeding During the Postmenopause. NVOG; 1997.
302 303 304
4.
Scottish Intercollegiate Guidelines Network. Investigation of postmenopausal bleeding 1st edn [Internet]. SIGN, Royal College of Physicians; 2002. Available from: www.sign.ac.uk
305 306 307
5.
Gupta JK, Chien PFW, Voit D, Clark TJ, Khan KS. Ultrasonographic endometrial thickness for diagnosing endometrial pathology in women with postmenopausal bleeding: a meta-analysis. Acta Obstet Gynecol Scand. 2002 Sep;81(9):799–816.
308 309 310
6.
Smith-Bindman R, Kerlikowske K, Feldstein VA, Subak L, Scheidler J, Segal M, et al. Endovaginal ultrasound to exclude endometrial cancer and other endometrial abnormalities. JAMA J Am Med Assoc. 1998 Nov 4;280(17):1510–7.
311 312 313
7.
Tabor A, Watt HC, Wald NJ. Endometrial thickness as a test for endometrial cancer in women with postmenopausal vaginal bleeding. Obstet Gynecol. 2002 Apr;99(4):663– 70.
314 315
8.
Spencer CP, Whitehead MI. Endometrial assessment re-visited. Br J Obstet Gynaecol. 1999 Jul;106(7):623–32.
316 317 318 319
9.
Gull B, Karlsson B, Milsom I, Granberg S. Can ultrasound replace dilation and curettage? A longitudinal evaluation of postmenopausal bleeding and transvaginal sonographic measurement of the endometrium as predictors of endometrial cancer. Am J Obstet Gynecol. 2003 Feb;188(2):401–8.
320 321 322 323
10. Timmermans A, van Doorn LC, Opmeer BC, Kroeks MVAM, Duk MJ, Bouwmeester AM, et al. Follow-up of women after a first episode of postmenopausal bleeding and endometrial thickness greater than 4 millimeters. Obstet Gynecol. 2008 Jan;111(1):137–43.
324 325 326
11. Feldman S, Shapter A, Welch WR, Berkowitz RS. Two-year follow-up of 263 patients with post/perimenopausal vaginal bleeding and negative initial biopsy. Gynecol Oncol. 1994 Oct;55(1):56–9.
327 328 329
12. Gimpelson RJ, Rappold HO. A comparative study between panoramic hysteroscopy with directed biopsies and dilatation and curettage. A review of 276 cases. Am J Obstet Gynecol. 1988 Mar;158(3 Pt 1):489–92.
330 331 332
13. Lacey JV Jr, Ioffe OB, Ronnett BM, Rush BB, Richesson DA, Chatterjee N, et al. Endometrial carcinoma risk among women diagnosed with endometrial hyperplasia: the 34-year experience in a large health plan. Br J Cancer. 2008 Jan 15;98(1):45–53.
333 334 335
14. Clark TJ, Neelakantan D, Gupta JK. The management of endometrial hyperplasia: an evaluation of current practice. Eur J Obstet Gynecol Reprod Biol. 2006 Apr 1;125(2):259–64.
AC C
EP
TE D
M AN U
SC
RI PT
296 297 298 299
Page 12
Smith
ACCEPTED MANUSCRIPT 15. Gallos ID, Krishan P, Shehmar M, Ganesan R, Gupta JK. Relapse of endometrial hyperplasia after conservative treatment: a cohort study with long-term follow-up. Hum Reprod Oxf Engl. 2013 May;28(5):1231–6.
339 340 341
16. Indraccolo U, Di Iorio R, Matteo M, Corona G, Greco P, Indraccolo SR. The pathogenesis of endometrial polyps: a systematic semi-quantitative review. Eur J Gynaecol Oncol. 2013;34(1):5–22.
342 343 344
17. Svirsky R, Smorgick N, Rozowski U, Sagiv R, Feingold M, Halperin R, et al. Can we rely on blind endometrial biopsy for detection of focal intrauterine pathology? Am J Obstet Gynecol. 2008;199(2):115–e1–3.
345 346 347
18. Van Dongen H, de Kroon CD, Jacobi CE, Trimbos JB, Jansen FW. Diagnostic hysteroscopy in abnormal uterine bleeding: a systematic review and meta-analysis. BJOG Int J Obstet Gynaecol. 2007 Jun;114(6):664–75.
348 349 350
19. Clark TJ, Barton PM, Coomarasamy A, Gupta JK, Khan KS. Investigating postmenopausal bleeding for endometrial cancer: cost-effectiveness of initial diagnostic strategies. BJOG Int J Obstet Gynaecol. 2006 May;113(5):502–10.
351 352
20. Bakour SH, Khan KS, Gupta JK. The risk of premalignant and malignant pathology in endometrial polyps. Acta Obstet Gynecol Scand. 2000 Apr;79(4):317–20.
353 354
21. Lieng M, Istre O, Qvigstad E. Treatment of endometrial polyps: a systematic review. Acta Obstet Gynecol Scand. 2010 Aug;89(8):992–1002.
355 356
22. Nathani F, Clark TJ. Uterine polypectomy in the management of abnormal uterine bleeding: A systematic review. J Minim Invasive Gynecol. 2006 Aug;13(4):260–8.
357 358 359
23. De Kroon CD, de Bock GH, Dieben SWM, Jansen FW. Saline contrast hysterosonography in abnormal uterine bleeding: a systematic review and metaanalysis. BJOG Int J Obstet Gynaecol. 2003 Oct;110(10):938–47.
360 361 362
24. Timmermans A, Opmeer BC, Veersema S, Mol BWJ. Patients’ preferences in the evaluation of postmenopausal bleeding. BJOG Int J Obstet Gynaecol. 2007 Sep;114(9):1146–9.
363 364
25. Fung Kee Fung M, Burnett M, Faught W. Does persistent postmenopausal bleeding justify hysterectomy? Eur J Gynaecol Oncol. 1997;18(1):26–8.
365 366 367 368
26. Van Doorn HC, Timmermans A, Opmeer BC, Kruitwagen RFMP, Dijkhuizen FPHLJ, Kooi GS, et al. What is the recurrence rate of postmenopausal bleeding in women who have a thin endometrium during a first episode of postmenopausal bleeding? Acta Obstet Gynecol Scand. 2008;87(1):89–93.
369 370 371
27. Ronghe R, Gaudoin M. Women with recurrent postmenopausal bleeding should be reinvestigated but are not more likely to have endometrial cancer. Menopause Int. 2010 Mar;16(1):9–11.
AC C
EP
TE D
M AN U
SC
RI PT
336 337 338
372 373
Page 13
ACCEPTED MANUSCRIPT
Recurrent PMB presentation 0 (0%) 30 (28%) 7 (7%)
Relative risk 2 0.42 1.47 1.24
152 (8%) 348 (19%) 866 (47%) 1832
0 (0%) 18 (17%) 51 (48%) 106
0.06 0.93 1.01
2
Confidence Interval 0.03-6.86 1.07-2.02 0.59-2.59
P value 0.3 0.02 0.6
0.00-0.90 0.62-1.42 0.82-1.24
0.002 0.7 0.9
1
RI PT
Infection Polyps Hyperplasia Cancer and Atypia 3 Functional Atrophy Total
Initial PMB presentation 20 (1%) 349 (19%) 97 (5%)
1
chi square, presented to one significant figure For the purpose of working out relative risk values of 0 were replaced with 0.5. 3 Functional = included histology described as weakly proliferative and weakly secretary PMB = postmenopausal bleeding
SC
2
1
P value
0 (0%) 17 (32%) 3 (6%) 0 (0%) 6 (11%) 27 (51%) 53
No value 0.5 0.7 No value 0.2 0.2
TE D
0 (0%) 13 (25%) 4 (8%) 0 (0%) 12 (23%) 24 (45%) 53
1
Recurrent PMB >12 months 21.8 (13-45)
EP
Average follow-up time in months (range) Infection Polyps Hyperplasia Cancer and Atypia 2 Functional Atrophy Total
Recurrent PMB ≤12 months 7.5 (2-12)
M AN U
Table 1. Comparison of endometrial pathology rates between women with PMB presenting for the first time or recurrently after previously normal investigations.
AC C
chi square, presented to one significant figure Functional = included histology described as weakly proliferative and weakly secretary PMB = postmenopausal bleeding
2
Table 2. Comparison of endometrial pathology rates stratified by time to representation in women with recurrent PMB.
ACCEPTED MANUSCRIPT
Did the woman’s last workup for PMB include hysteroscopy or SIS?
Yes
Were the woman’s last investigations >12 months ago?
M AN U
Perform hysteroscopy or SIS +/- EB
SC
No
RI PT
Recurrent PMB
No
TE D
No further investigation required unless a change in symptoms*.
Yes Perform hysteroscopy or SIS +/- EB
EP
EB = Endometrial Biopsy; PMB = Postmenopausal Bleeding; SIS = Saline infusion Sonography; * increased quantity or persistence of bleeding, pain, weight loss, bowel or
AC C
urinary symptoms.
Figure 1. Proposed diagnostic pathway for recurrent PMB
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Smith
ACCEPTED MANUSCRIPT 1
Precis
2 Women who present with recurrent postmenopausal bleeding appear less likely to have pre-
4
malignant and malignant endometrial disease, but are more likely to have endometrial
5
polyps.
RI PT
3
6 7
AC C
EP
TE D
M AN U
SC
8
Page 1
PII:
S1553-4650(14)00202-7
DOI:
10.1016/j.jmig.2014.03.007
Reference:
JMIG 2273
To appear in:
The Journal of Minimally Invasive Gynecology
Received Date: 2 December 2013 Revised Date:
4 March 2014
Accepted Date: 6 March 2014
Please cite this article as: Smith PP, O’Connor S, Gupta J, Clark TJ, Recurrent postmenopausal bleeding: a prospective cohort study, The Journal of Minimally Invasive Gynecology (2014), doi: 10.1016/j.jmig.2014.03.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Smith
ACCEPTED MANUSCRIPT 1 2 3
RI PT
4 5 6 7
9
SC
8
Title: Recurrent postmenopausal bleeding: a prospective cohort study
11
M AN U
10
Author Details: Paul P Smith of Birmingham Women’s Hospital, Birmingham, UK Siobhan O’Connor of Birmingham Women’s Hospital, Birmingham, UK
13
Janesh Gupta of Birmingham Women’s Hospital, Birmingham, UK
14
T Justin Clark of Birmingham Women’s Hospital, Birmingham, UK
TE D
12
15
Corresponding Author: Paul P Smith of Birmingham Women’s Hospital, Birmingham, UK
17
E-mail: [email protected]
18
Phone number: +447932044361
20 21 22
AC C
19
EP
16
23 24 25 26 27
Page 1
Smith
ACCEPTED MANUSCRIPT 28
Condensation:
29
Women who present with recurrent postmenopausal bleeding appear less likely to have
30
malignant endometrial disease, but are more likely to have endometrial polyps.
31
RI PT
32 33 34
SC
35 36
M AN U
37 38 39 40 41
TE D
42 43
46 47 48 49
AC C
45
EP
44
50 51 52 53 54
Page 2
Smith
ACCEPTED MANUSCRIPT
Abstract
56
Study objective. The aim of this study was to estimate the prevalence of genital tract
57
pathologies in women presenting with initial and recurrent postmenopausal bleeding (PMB)
58
to help inform diagnostic pathways
59
Design. Prospective cohort study
60
Design classification. Canadian Task Force Classification: II-2
61
Setting. A large, urban teaching hospital in Birmingham, UK.
62
Patients. A total of 1938 consecutive women presented with postmenopausal bleeding, of
63
which, 106 (5%) women were investigated for a recurrent episode after previously having
64
normal investigations.
65
Interventions. All women underwent a pelvic examination and ultrasound scan. An
66
endometrial biopsy was performed when endometrial thickness was >4mm in women
67
presenting for the first time with PMB, with recourse to outpatient hysteroscopy following
68
correlation between clinical and pathological findings. All women had an endometrial biopsy
69
and outpatient hysteroscopy with a recurrent PMB presentation.
70
Measurements and main results. The chance of having endometrial cancer or hyperplasia
71
with atypia was significantly less in women who presented with recurrent PMB as compared
72
with those presenting with PMB for the first time (0%, 8% respectively, p=.002), but were
73
significantly more likely to have benign endometrial polyps compared with women presenting
74
with PMB for the first time (28%, 19% respectively, RR 1.47 [95% CI 1.07, 2.02], p=.02).
75
Conclusion. Recurrent PMB results in less likelihood of pre-malignant and malignant
76
endometrial disease but one in four women have endometrial polyps as a cause of PMB.
77
First line investigation for women with recurrent PMB should be with tests that have a high
78
accuracy for diagnosing focal pathologies such as outpatient hysteroscopy or saline infusion
79
sonography.
AC C
EP
TE D
M AN U
SC
RI PT
55
80 81
Keywords Postmenopausal bleeding, polyps, endometrial cancer, hysteroscopy.
Page 3
Smith
ACCEPTED MANUSCRIPT
Introduction
83
Postmenopausal bleeding (PMB) is a common condition affecting between 7-15% of the
84
postmenopausal population1. Prompt investigation is needed because there is a 3-10% risk
85
of endometrial cancer2. Guidelines3,4 recommend first line investigation with transvaginal
86
sonography (TVS) to measure endometrial thickness. The prevalence of malignancy is
87
reduced to 12
121
months.
122
cancer, hyperplasia or an untreated polyp during the first episode of bleeding.
RI PT
110
We prospectively collected standardised data on a specially
M AN U
SC
PMB was defined as an
TE D
Women were excluded from the recurrent cohort if they were diagnosed with
123
All women with PMB underwent TVS as a first-line screening test. Final diagnosis for PMB
125
was made using the following reference standards:
EP
124
1. Transvaginal sonography (TVS): if the endometrium was regular, ≤4 mm on TVS with
127
no evidence of fluid, the patient was considered to have an atrophic endometrium
128
and sampling was not performed. The patient was reassured after a lower genital
129
tract examination and discharged. A trained sonographer who was not blinded to the
130
clinical symptoms performed all TVS. Colour Doppler was used when endometrial
131
irregularities were seen.
AC C
126
132
2. Endometrial biopsy (EB): women who screened positive i.e. TVS with EEC >4 mm or
133
irregular endometrium (even if the EEC was ≤4mm), or incomplete visualisation of
134
the endometrium, underwent aspiration endometrial biopsy (EB). A senior pathologist
135
reviewed all histology and was not blinded to the clinical symptoms. Page 5
Smith
ACCEPTED MANUSCRIPT 136
3. Office
hysteroscopy
(OH)(under
conscious sedation):
where
an
outpatient
137
endometrial biopsy (EB) failed (defined as failure to instrument the uterus or a non-
138
diagnostic sample) an OH was arranged and a further biopsy taken unless the cavity
139
seemed atrophic on hysteroscopic visualisation.
140
visualisation was defined as a pale, avascular uterine surface with visible underlying
141
stroma with no endometrial tissue avulsed following contact with the tip of the
142
hysteroscope. OH was performed on all patients with a recurrent parsentation of
143
PMB.
144
fibroids) were suspected on TVS or EB reports. If confirmed on OH, they were
145
removed and sent for histological confirmation. All OH was performed by a senior
146
gynecologist or a specialist nurse who was not blinded to the clinical symptoms.
RI PT
Atrophy on hysteroscopic
M AN U
SC
An OH was also performed where focal pathology (polyps, submucous
147
For the purposes of this study we restricted final diagnosis to a single pathology. When
149
more than one endometrial histological diagnosis was made the patient was categorised
150
using the following hierarchy: Endometrial cancer / hyperplasia with atypia > hyperplasia
151
without atypia > polyps > infection > functional > atrophic. Cancer and hyperplasia with
152
atypia were included together because of the high rate of under-call and progression to
153
cancer where complex hyperplasia is found with cytological atypia13–15.
EP
TE D
148
AC C
154 155
The relative proportion of different pathologies in women with first and recurrent PMB was
156
compared using chi-square test or fisher’s exact test when the expected frequency was less
157
than five. For the purpose of working out relative risk, values of zero were replaced with 0.5.
158
Statistical analysis was performed using SPSSTM version 15.
159 160
Results
161
Page 6
Smith
ACCEPTED MANUSCRIPT Over the 53-month (just under 4.5 years) study period, 1832 women with a median duration
163
of follow up of 35 months (range 1–56 months), presented with PMB for the first time to the
164
Birmingham Women’s Hospital. Of this cohort, 106 (5%) women had recurrent PMB. The
165
median follow up in this subgroup of women with recurrent PMB was 12 months (range 1–45
166
months). All had previously been investigated for PMB and had endometrial cancer or
167
hyperplasia excluded. The mean age (62.9 vs 61.1 years), body mass index (BMI) (30.7 vs
168
31.8 kg/m2), EEC (8.3 vs 7.7 mm) and percentage taking HRT (11.7% vs 12.3%) was similar
169
in women with first or recurrent PMB.
SC
RI PT
162
170
Endometrial cancer or hyperplasia with atypia was diagnosed in 152/1832 (7.8%) women
172
with first presentation PMB but no cases were found in the 106 women with recurrent PMB.
173
In contrast, benign endometrial polyps were found to be significantly more prevalent in the
174
recurrent PMB group compared with first presentation PMB (30/106, 28.3% vs 349/1832,
175
19.6%, p=.02 respectively). The prevalence of all other endometrial pathologies was
176
equivalent between initial and recurrent PMB groups (Table 1). Because of a possible
177
association between HRT and the development of polyps16, we re-analysed the data after
178
removing the women taking HRT. Endometrial polyps remained more prevalent in women
179
with recurrent PMB, but the difference was no longer significant (25/94, 27% in recurrent
180
PMB presentation vs 313/1616 19%, in initial PMB presentation, p=.09). Removing the
181
women taking HRT had no effect on the significant decrease in cancer and atypia in
182
recurrent PMB presentation. No significant differences in the prevalence of endometrial
183
pathologies were noted in women with recurrent PMB when stratified into re-presentation
184
within and beyond 12 months of initial presentation (Table 2).
AC C
EP
TE D
M AN U
171
185 186 187
Discussion
Page 7
Smith
ACCEPTED MANUSCRIPT 188
In our study, 106 (6%) women had recurrent PMB over a 4.5 year period, but no cases of
189
endometrial cancer or hyperplasia with atypia were detected.
190
recurrent PMB had a significantly increased risk of having benign endometrial polyps
191
irrespective of the time to recurrence. The apparent increased prevalence of endometrial
192
polyps with recurrent PMB may reflect the higher accuracy of OH for detecting focal
193
pathologies,
194
selectively in women presenting with PMB for the first time. Alternatively, women with
195
recurrent PMB may have more frequently developed polyps de novo accounting for further
196
bleeding symptoms. In either case, the finding that benign endometrial polyps are more
197
prevalent has potential implications for how to best manage women with recurrent PMB.
198
With it’s focus on diagnosing endometrial cancer, the current consensus for working up
199
women with PMB using first-line TVS EEC measurement3,4,19, may be inappropriate in
200
women with recurrent PMB.
201
benign20,21, and the consequences of diagnosis less serious to health than endometrial
202
cancer, they are associated with abnormal uterine bleeding and removal by polypectomy
203
frequently resolves symptoms20,22. Thus, diagnostic pathways in recurrent PMB should be
204
developed with the aim of diagnosing benign as well as malignant endometrial disease.
RI PT
M AN U
SC
which was undertaken routinely in the recurrent PMB group, but only
TE D
Although the vast majority of endometrial polyps are
EP
205
17,18
However, women with
Figure 1 illustrates a potential testing pathway for women presenting with recurrent PMB
207
based upon the findings from our study. We have introduced saline infusion sonography
208
(SIS) along with OH to make are proposed pathway more generalisable. We propose that
209
OH or saline infusion sonography (SIS), should be incorporated because these tests are
210
more accurate in diagnosing discrete pathologies such as endometrial polyps18,23. Moreover,
211
structured interviews exploring women’ preferences in the evaluation of PMB have shown
212
that women are prepared to undergo more invasive hysteroscopy to evaluate for any
213
additional pathology, rather than adopt the currently recommended expectant management
214
after ultrasound24. It is possible that this preference for additional testing may be even more
AC C
206
Page 8
Smith
ACCEPTED MANUSCRIPT 215
likely with recurrent PMB because of increased anxiety associated with repeated symptoms,
216
even though paradoxically the risk of endometrial cancer appears to be low.
217 The low risk of serious endometrial disease in recurrent PMB after previous normal
219
investigation is reassuring and so reinvestigation of women who have already had
220
endometrial polyps excluded by OH or SIS may be unnecessary. However, if the nature of
221
bleeding has changed (ie increased quantity or persistence) or other relevant symptoms
222
such as pain have developed then it would seem prudent to undertake timely reinvestigation
223
to exclude endometrial cancer (Figure 1). We found no difference in the likelihood of
224
endometrial cancer or hyperplasia in women with recurrent PMB according to whether re-
225
presentation was within or beyond 12 months of initial negative testing. However, the
226
median duration of follow up in our series was only 12 months. Therefore a cautious
227
approach should be adopted, especially in women with recurrent symptoms and longer time
228
intervals from previously negative testing. For this reason the proposed diagnostic algorithm
229
recommends reinvestigation for all women with recurrent PMB after 12 months, regardless
230
of previous testing history (Figure 1).
TE D
M AN U
SC
RI PT
218
231
A strength of our study is that the data were collected prospectively, consecutively and in a
233
standardised fashion minimising bias from incomplete data. However, whilst the
234
generalisability of our findings are limited because they are derived from a single centre, we
235
believe that they are likely to mirror other centres because we adopted testing for PMB in
236
line with current guidelines3,4 from a large, unselected, general postmenopausal population.
237
Several other studies have estimated recurrence rates for PMB and evaluated the
238
prevalence of malignant although not benign uterine pathology9,11,25–27. One study evaluated
239
126 women presenting with recurrent PMB and reported an 8% recurrent PMB27 rate in
240
keeping with our observed estimate of 6%. In contrast to our findings, no significant
241
difference in the incidence of cancer or atypical hyperplasia were observed compared to
AC C
EP
232
Page 9
Smith
ACCEPTED MANUSCRIPT women presenting for the first time. However, initial diagnostic work up was not in line with
243
current recommendations3,4 such that two of the five women with endometrial cancer in the
244
recurrent group represented within a year but neither had undergone initial endometrial
245
sampling despite EEC>9mm. The likelihood of recurrent bleeding has been reported to be
246
as high as 33%11 and the rates of cancer and atypical hyperplasia up to 22.7% (15/66)9.
247
These differences may be explained by differences in patient selection and referral criteria,
248
variations in initial diagnostic work-up and the sample size and duration of follow-up. Closer
249
analysis of these studies revealed that significant pathology usually occurred 12 months
250
after initial investigation9,10,27. When cancer or atypical hyperplasia did present within 12
251
months, endometrial sampling was either not normal, inadequate or not performed at first
252
presentation9,27. It is for this reason that in our suggested pathway (figure 1) all patients with
253
recurrent PMB are investigated with OH or SIS with an attempt at biopsy irrespective of the
254
EEC and that a full review of previous investigations is undertaken. Larger studies are
255
needed to externally validate our findings before we can make robust clinical
256
recommendations.
TE D
M AN U
SC
RI PT
242
257
Although we did not follow up apparently asymptomatic women, it is unlikely that that our low
259
prevalence of serious endometrial disease in women with recurrent PMB reflects high rates
260
of loss to follow up. This is because PMB is an alarming symptom so women are unlikely to
261
ignore it and general practitioners are well educated in the need to rapidly refer.
262
Furthermore, postmenopausal women are not a particularly itinerant population and our
263
hospital is the largest provider of gynaecological care in Birmingham. Another potential
264
criticism of our study is the lack of a common reference standard for diagnosing endometrial
265
disease. Our study was pragmatic, with data collected within a clinical setting such that the
266
indication for further testing with OH or SIS, after the standard use of TVS and EB if
267
EEC≥4mm3,4, was based upon clinico-pathological correlation in the absence of uniform
268
guidance. We also restricted diagnosis to a single pathology so that we may have
AC C
EP
258
Page 10
Smith
ACCEPTED MANUSCRIPT 269
underestimated the prevalence of some pathologies. However, in the presence of multiple
270
endometrial pathologies, we based our final diagnosis according to the most significant
271
pathology and so the clinical implications of our simplification are likely to be negligible.
RI PT
272
Conclusions
274
At least one in twenty women with PMB will have recurrent symptoms. Compared to women
275
presenting for the first time with PMB, the risk of endometrial cancer in recurrent PMB in
276
women who have had previously normal investigations in line with current diagnostic
277
guidelines3,4, appears to be much lower but the risk of having benign endometrial polyps is
278
significantly higher. All women with recurrent PMB, irrespective of when they re-present after
279
their initial negative investigations, should undergo a test with a high sensitivity for focal
280
pathology such as OH or SIS.
281
Disclosure of interests all authors declare no conflict of interests
TE D
282
M AN U
SC
273
283
Contribution to authorship PS produced the original idea, performed the analyses and
285
drafted the article. SOC collected all data. TJC contributed to interpretation. TJC, SOC &
286
JKG revised the article for important intellectual content.
EP
284
AC C
287 288
Details of ethics approval Data collection and storage satisfied local and national
289
regulations. Ethical approval was not required.
290 291
Funding none
292 293
References
294 295
1.
Astrup K, Olivarius N de F. Frequency of spontaneously occurring postmenopausal bleeding in the general population. Acta Obstet Gynecol Scand. 2004 Feb;83(2):203–7.
Page 11
Smith
ACCEPTED MANUSCRIPT 2.
Bachmann LM, ter Riet G, Clark TJ, Gupta JK, Khan KS. Probability analysis for diagnosis of endometrial hyperplasia and cancer in postmenopausal bleeding: an approach for a rational diagnostic workup. Acta Obstet Gynecol Scand. 2003 Jun;82(6):564–9.
300 301
3.
NVOG (Dutch Society of Obsterics and Gynaecology). NVOG Guideline: Abnormal Vaginal Bleeding During the Postmenopause. NVOG; 1997.
302 303 304
4.
Scottish Intercollegiate Guidelines Network. Investigation of postmenopausal bleeding 1st edn [Internet]. SIGN, Royal College of Physicians; 2002. Available from: www.sign.ac.uk
305 306 307
5.
Gupta JK, Chien PFW, Voit D, Clark TJ, Khan KS. Ultrasonographic endometrial thickness for diagnosing endometrial pathology in women with postmenopausal bleeding: a meta-analysis. Acta Obstet Gynecol Scand. 2002 Sep;81(9):799–816.
308 309 310
6.
Smith-Bindman R, Kerlikowske K, Feldstein VA, Subak L, Scheidler J, Segal M, et al. Endovaginal ultrasound to exclude endometrial cancer and other endometrial abnormalities. JAMA J Am Med Assoc. 1998 Nov 4;280(17):1510–7.
311 312 313
7.
Tabor A, Watt HC, Wald NJ. Endometrial thickness as a test for endometrial cancer in women with postmenopausal vaginal bleeding. Obstet Gynecol. 2002 Apr;99(4):663– 70.
314 315
8.
Spencer CP, Whitehead MI. Endometrial assessment re-visited. Br J Obstet Gynaecol. 1999 Jul;106(7):623–32.
316 317 318 319
9.
Gull B, Karlsson B, Milsom I, Granberg S. Can ultrasound replace dilation and curettage? A longitudinal evaluation of postmenopausal bleeding and transvaginal sonographic measurement of the endometrium as predictors of endometrial cancer. Am J Obstet Gynecol. 2003 Feb;188(2):401–8.
320 321 322 323
10. Timmermans A, van Doorn LC, Opmeer BC, Kroeks MVAM, Duk MJ, Bouwmeester AM, et al. Follow-up of women after a first episode of postmenopausal bleeding and endometrial thickness greater than 4 millimeters. Obstet Gynecol. 2008 Jan;111(1):137–43.
324 325 326
11. Feldman S, Shapter A, Welch WR, Berkowitz RS. Two-year follow-up of 263 patients with post/perimenopausal vaginal bleeding and negative initial biopsy. Gynecol Oncol. 1994 Oct;55(1):56–9.
327 328 329
12. Gimpelson RJ, Rappold HO. A comparative study between panoramic hysteroscopy with directed biopsies and dilatation and curettage. A review of 276 cases. Am J Obstet Gynecol. 1988 Mar;158(3 Pt 1):489–92.
330 331 332
13. Lacey JV Jr, Ioffe OB, Ronnett BM, Rush BB, Richesson DA, Chatterjee N, et al. Endometrial carcinoma risk among women diagnosed with endometrial hyperplasia: the 34-year experience in a large health plan. Br J Cancer. 2008 Jan 15;98(1):45–53.
333 334 335
14. Clark TJ, Neelakantan D, Gupta JK. The management of endometrial hyperplasia: an evaluation of current practice. Eur J Obstet Gynecol Reprod Biol. 2006 Apr 1;125(2):259–64.
AC C
EP
TE D
M AN U
SC
RI PT
296 297 298 299
Page 12
Smith
ACCEPTED MANUSCRIPT 15. Gallos ID, Krishan P, Shehmar M, Ganesan R, Gupta JK. Relapse of endometrial hyperplasia after conservative treatment: a cohort study with long-term follow-up. Hum Reprod Oxf Engl. 2013 May;28(5):1231–6.
339 340 341
16. Indraccolo U, Di Iorio R, Matteo M, Corona G, Greco P, Indraccolo SR. The pathogenesis of endometrial polyps: a systematic semi-quantitative review. Eur J Gynaecol Oncol. 2013;34(1):5–22.
342 343 344
17. Svirsky R, Smorgick N, Rozowski U, Sagiv R, Feingold M, Halperin R, et al. Can we rely on blind endometrial biopsy for detection of focal intrauterine pathology? Am J Obstet Gynecol. 2008;199(2):115–e1–3.
345 346 347
18. Van Dongen H, de Kroon CD, Jacobi CE, Trimbos JB, Jansen FW. Diagnostic hysteroscopy in abnormal uterine bleeding: a systematic review and meta-analysis. BJOG Int J Obstet Gynaecol. 2007 Jun;114(6):664–75.
348 349 350
19. Clark TJ, Barton PM, Coomarasamy A, Gupta JK, Khan KS. Investigating postmenopausal bleeding for endometrial cancer: cost-effectiveness of initial diagnostic strategies. BJOG Int J Obstet Gynaecol. 2006 May;113(5):502–10.
351 352
20. Bakour SH, Khan KS, Gupta JK. The risk of premalignant and malignant pathology in endometrial polyps. Acta Obstet Gynecol Scand. 2000 Apr;79(4):317–20.
353 354
21. Lieng M, Istre O, Qvigstad E. Treatment of endometrial polyps: a systematic review. Acta Obstet Gynecol Scand. 2010 Aug;89(8):992–1002.
355 356
22. Nathani F, Clark TJ. Uterine polypectomy in the management of abnormal uterine bleeding: A systematic review. J Minim Invasive Gynecol. 2006 Aug;13(4):260–8.
357 358 359
23. De Kroon CD, de Bock GH, Dieben SWM, Jansen FW. Saline contrast hysterosonography in abnormal uterine bleeding: a systematic review and metaanalysis. BJOG Int J Obstet Gynaecol. 2003 Oct;110(10):938–47.
360 361 362
24. Timmermans A, Opmeer BC, Veersema S, Mol BWJ. Patients’ preferences in the evaluation of postmenopausal bleeding. BJOG Int J Obstet Gynaecol. 2007 Sep;114(9):1146–9.
363 364
25. Fung Kee Fung M, Burnett M, Faught W. Does persistent postmenopausal bleeding justify hysterectomy? Eur J Gynaecol Oncol. 1997;18(1):26–8.
365 366 367 368
26. Van Doorn HC, Timmermans A, Opmeer BC, Kruitwagen RFMP, Dijkhuizen FPHLJ, Kooi GS, et al. What is the recurrence rate of postmenopausal bleeding in women who have a thin endometrium during a first episode of postmenopausal bleeding? Acta Obstet Gynecol Scand. 2008;87(1):89–93.
369 370 371
27. Ronghe R, Gaudoin M. Women with recurrent postmenopausal bleeding should be reinvestigated but are not more likely to have endometrial cancer. Menopause Int. 2010 Mar;16(1):9–11.
AC C
EP
TE D
M AN U
SC
RI PT
336 337 338
372 373
Page 13
ACCEPTED MANUSCRIPT
Recurrent PMB presentation 0 (0%) 30 (28%) 7 (7%)
Relative risk 2 0.42 1.47 1.24
152 (8%) 348 (19%) 866 (47%) 1832
0 (0%) 18 (17%) 51 (48%) 106
0.06 0.93 1.01
2
Confidence Interval 0.03-6.86 1.07-2.02 0.59-2.59
P value 0.3 0.02 0.6
0.00-0.90 0.62-1.42 0.82-1.24
0.002 0.7 0.9
1
RI PT
Infection Polyps Hyperplasia Cancer and Atypia 3 Functional Atrophy Total
Initial PMB presentation 20 (1%) 349 (19%) 97 (5%)
1
chi square, presented to one significant figure For the purpose of working out relative risk values of 0 were replaced with 0.5. 3 Functional = included histology described as weakly proliferative and weakly secretary PMB = postmenopausal bleeding
SC
2
1
P value
0 (0%) 17 (32%) 3 (6%) 0 (0%) 6 (11%) 27 (51%) 53
No value 0.5 0.7 No value 0.2 0.2
TE D
0 (0%) 13 (25%) 4 (8%) 0 (0%) 12 (23%) 24 (45%) 53
1
Recurrent PMB >12 months 21.8 (13-45)
EP
Average follow-up time in months (range) Infection Polyps Hyperplasia Cancer and Atypia 2 Functional Atrophy Total
Recurrent PMB ≤12 months 7.5 (2-12)
M AN U
Table 1. Comparison of endometrial pathology rates between women with PMB presenting for the first time or recurrently after previously normal investigations.
AC C
chi square, presented to one significant figure Functional = included histology described as weakly proliferative and weakly secretary PMB = postmenopausal bleeding
2
Table 2. Comparison of endometrial pathology rates stratified by time to representation in women with recurrent PMB.
ACCEPTED MANUSCRIPT
Did the woman’s last workup for PMB include hysteroscopy or SIS?
Yes
Were the woman’s last investigations >12 months ago?
M AN U
Perform hysteroscopy or SIS +/- EB
SC
No
RI PT
Recurrent PMB
No
TE D
No further investigation required unless a change in symptoms*.
Yes Perform hysteroscopy or SIS +/- EB
EP
EB = Endometrial Biopsy; PMB = Postmenopausal Bleeding; SIS = Saline infusion Sonography; * increased quantity or persistence of bleeding, pain, weight loss, bowel or
AC C
urinary symptoms.
Figure 1. Proposed diagnostic pathway for recurrent PMB
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Smith
ACCEPTED MANUSCRIPT 1
Precis
2 Women who present with recurrent postmenopausal bleeding appear less likely to have pre-
4
malignant and malignant endometrial disease, but are more likely to have endometrial
5
polyps.
RI PT
3
6 7
AC C
EP
TE D
M AN U
SC
8
Page 1
