Risks of tubo-ovarian abscess

in cases of endometrioma and

assisted reproductive technologies
are both under- and overreported
Claire Villette, M.D.,a Antoine Bourret, M.D.,a Pietro Santulli, M.D., Ph.D.,a,b,c Vanessa Gayet, M.D.,a
Charles Chapron, M.D.,a,b,c and Dominique de Ziegler, M.D.a
a
Department of Obstetrics, Gynecology, and Reproductive Medicine, Universite
Paris Descartes, Paris Sorbonne Cite

–
Assistance Publique Ho ^ pitaux de Paris, CHU Cochin; b Laboratoire d’Immunologie, Institut Cochin, Inserm Ua0af,

Paris Descartes, Sorbonne Paris Cite
Universite
partement de Ge

; and c De
netique, De
veloppement et Cancer, Institut

Paris Descartes, Sorbonne Paris Cite
Cochin, Inserm, Ua0af, Universite
, Paris, France

Objective: To study possible associations among endometriosis, pelvic infectious disease, and ART.
Design: Retrospective cohort analysis over 4 consecutive years, based on medical records and insurance coding in a tertiary endome-
triosis reference center.
Setting: Tertiary university-based reference center for endometriosis.
Patient(s): We retrieved all charts carrying the diagnoses infectious process and endometriosis in 2009–2012. Each chart was individ-
ually analyzed for categorization of the infectious episode and determining whether ART had been performed.
Main Outcome Measure(s): Hospitalization for acute infection in women with known endometriosis and possible past ART.
Intervention: Retrospective insurance codes–triggered chart analysis.
Result(s): Ten patients were admitted for an acute infection with fever, acute abdomen syndrome, elevated white blood cell count, and
adnexal mass. Three women had oocyte retrieval, and an endometrioma was present 16, 57, and 102 days earlier. In one patient, the
complication occurred 37 days after a cesarean section without prior ART. In the remaining six cases tubo-ovarian abscesses (TOAs)
occurred spontaneously in endometriosis women who never had ART. Medical treatment succeeded in only two patients, and the
remaining eight needed laparoscopic drainage. In 6 out of those 8 cases, laparoscopic drainage was a second-stage measure justified
by failure to respond to antibiotic therapy.
Conclusion(s): Our data indicate that some putative complications of ART and endometrioma may actually not be linked to ART, but
rather constitute sporadic occurrences in endometriosis. Furthermore, TOAs occurring in women with endometriosis are best treated by
early surgical drainage together with intravenous antibiotics. (Fertil SterilÒ 2016;106:410–5. Ó2016 by American Society for Repro-
ductive Medicine.)
Key Words: Endometriosis, endometriomas, pelvic inflammatory disease, tubo-ovarian abscess (TOA), ART
Discuss: You can discuss this article with its authors and with other ASRM members at http://fertstertforum.com/villettec-art-
endometrioma-infection/

M
anagement of endometrio- questioned whether endometriomas further hampering already com-
mas in women undergoing should be surgically removed before promised responses to controlled
assisted reproductive tech- ART (1, 2). Several recent reports ovarian stimulation (3–9). Further
nologies (ART) has changed. Over the suggest that surgery might not supporting surgical abstinence are
past decade, many publications have improve ART outcome, while risking reports indicating that obstetrical
complications are rare in women who
Received January 13, 2016; revised April 11, 2016; accepted April 12, 2016; published online May 10,
conceived with the use of ART while
2016. endometriomas were present (10, 11).
C.V. has nothing to disclose. A.B. has nothing to disclose. P.S. has nothing to disclose. V.G. has nothing Taken together, these reports led to
to disclose. C.C. has nothing to disclose. D.d.Z. has nothing to disclose.
Reprint requests: Dominique de Ziegler, M.D., Professor and Head, Reproductive Endocrinology and not systematically removing endometrio-
Infertility, Service de Gyne
cologie Obste

trique II, Groupe d’Ho ^ pitaux Paris Centre Cochin Broca mas before ART (3, 6, 12). As a result,
Ho^ tel Dieu, Ho ^ pital Cochin, 53, Avenue de l’Observatoire, 75014 Paris, France (E-mail:
ddeziegler@me.com). oocyte retrievals for ART are commonly
conducted while endometriomas—
Fertility and Sterility® Vol. 106, No. 2, August 2016 0015-0282/$36.00 possibly large, multiple, and bilateral—
Copyright ©2016 American Society for Reproductive Medicine, Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.fertnstert.2016.04.014 remain present.

410 VOL. 106 NO. 2 / AUGUST 2016


This new clinical trend—undertaking ART while endome-
triomas are present—raised fears, however, that this might in-
crease the risk of infectious complications of ART. Several
case reports (13–17) indeed described sporadic cases of
abscesses, but without truly sizing the reality of this risk,
i.e., the incidence of serious post-ART complications.
Conversely, a prospective study of 214 ART cycles conducted
in women having one or several endometriomas present at the
time of oocyte retrieval reported no pelvic infection within
2 months after oocyte retrieval (18).
The discrepancy between isolated reports of infections in
women undergoing ART with endometriomas and a fairly
large systematic analysis finding no complications led us to
question whether certain infections might have eluded the
systematic analysis. For clarifying this issue—critical for an
adequate risk and safety assessment of ART in women with
endometriomas—we analyzed all acute infectious complica-
tions that occurred in women with endometriosis at our insti-
tution over 4 years. For this, we retrieved the charts of all
patients admitted over the 4-year interval that bore at least
the two diagnoses of endometriosis and infectious process.
Working from this list, we report here all cases of acute
tubo-ovarian abscesses (TOA) encountered in women with
endometriomas during the observation period, regardless of
whether ART had been performed or not.
MATERIALS AND METHODS
We conducted a retrospective analysis of all women who were
admitted at our institution with the two diagnoses of infectious
process and endometriosis over 4 consecutive years (January
2009–December 2012). At our institution, we conduct
200
surgical and 450 ART procedures annually in women with
endometriosis. All women admitted at our institution consent
to the anonymous use of their medical data for quality assess-
ment and/or publication. Therefore, no Institutional Review
Board approval was sought for this retrospective analysis, as
stipulated in our standard operation procedures which are re-
viewed yearly through our ISO certification process. All medi-
cal diagnoses and procedures (present and past) are coded for
each new and returning admission according to the Interna-
tional Classification of Diseases published by the World Health
Organization. With the use of this database, we retrieved all
charts of patients hospitalized during the defined interval,
which bore the diagnoses of: 1) ‘‘endometriosis’’; and 2)
‘‘salpingitis,’’ ‘‘oophoritis,’’ and/or ‘‘other pelvic inflammatory
disease (PID).’’ All charts were individually reviewed to retain
only those patients with known endometriomas diagnosed sur-
gically or based on standardized imaging procedures (ultra-
sound or magnetic resonance imaging) and had an
emergency admission for an acute infectious process. Patients
having a history of pelvic infection (former PID, appendicitis,
etc.) but not related to the actual hospitalization were therefore
excluded from the present analysis.
Demographic data collected in this chart review included
age, parity and gravidity, and infertility history. Endometri-
osis was staged with the use of the revised American Fertility
Society classification for endometriosis (rAFS), based on sur-
gical or imaging data. Presence of endometrioma and history
VOL. 106 NO. 2 / AUGUST 2016
Fertility and Sterility®
of surgical resection of endometrioma were noted. When re-
viewing the infertility history, we noted whether ART had
been performed at our institution or elsewhere. For patients
who had ART, the use of prophylactic antibiotics and/or
possible accidental puncture of endometrioma(s) during
oocyte retrieval were also noted when available. During the
hospital stay, the following parameters were followed: pelvic
pain, fever, metrorrhagia, white blood cell (WBC) count, ultra-
sound findings, computerized tomographic (CT) scan results,
antibiotic use, whether surgical drainage was performed or
not (noting the number of days after admission), surgical
findings, results of bacteriologic cultures, and duration of
hospitalization. Women whose hospitalization for acute in-
fectious process occurred <1 month after pelvic surgery for
endometriosis or delivery (vaginal or by cesarean section)
were arbitrarily excluded form analysis.
This retrospective study received no specific funding.
RESULTS
A total of 230 patients were admitted at our department be-
tween January 2009 and December 2012 with the dual diag-
nostic coding of endometriosis and one of the selected
infectious diagnoses. After a thorough chart review, 213 of
these 230 endometriosis cases were excluded because they
underwent scheduled surgery and simply had a former history
of pelvic infection that justified the diagnostic coding but was
unrelated to the actual hospitalization. Five more patients
were excluded because they had undergone surgery for endo-
metriosis <1 month before their admission for an acute infec-
tious process. One more patient was excluded because she had
a vaginal delivery 18 days before the admission for an infec-
tious complication. Finally, one more patient had to be
excluded because insufficient data were available. Ultimately
therefore, ten patients were retained for analysis, as illus-
trated in the patient flow chart (Fig. 1). All had endometriosis
diagnosed by means of surgery or standardized imaging
criteria with or without endometrioma and underwent emer-
gency hospitalization for a TOA.
The baseline characteristics of these ten patients are pre-
sented in Table 1. Median age was 33 (range 27–44) years.
Four women were infertile, four had children, and two were
nulliparous with no immediate desire for pregnancy. All
were classified to have endometriosis stage III or IV according
to the rAFS scoring system. Seven women had one or more
endometriomas and the remaining three had no cyst.
Details on the clinical course of each case are described in
Table 2. All ten patients reported pelvic pain and had fever
(38 C–39.5 C). Three of them reported metrorrhagia. WBC
count ranged from 10.8  109/L to 25  109/L. Pelvic ultra-
sound with or without CT scan performed on admission sup-
ported the diagnosis of TOA in all of the admitted women.
Eight of the ten patients were first treated medically (patients
1, 2, 3, 4, 5, 6, 8, and 9) with intravenous antibiotics. Despite
using two or three large-spectrum antibiotics, six of these
eight failed to improve, which prompted surgical exploration
2–9 days later. Patient 4 developed septic shock 2 days after
admission and onset of antibiotic therapy. In two patients, 7
and 10, physical examination at admission revealed high
411
ORIGINAL ARTICLE: ENDOMETRIOSIS

FIGURE 1

Flow chart.
Villette. Endometriomas, ART, and infections. Fertil Steril 2016.

fever (>39 C) and abdominal guarding, which justified im- endometriomas was reported. In one patient, 4, the acute
mediate surgical exploration for pelvic drainage. Medical complication occurred 37 days after a cesarean section
treatment alone was successful in only two patients, 6 and without earlier ART. In the remaining six cases, the acute in-
8. The length of hospitalization ranged from 5 to 20 days. fectious process could not be tied to any incidental event and
No mortality was encountered among the patients. were therefore considered to have occurred spontaneously
Only three of these ten cases, patients 1, 2, and 3, had un- (patients 5, 6, 7, 8, 9, and 10).
dergone ART 16, 57, and 102 days before the acute infectious
episode. In two of these three women, prophylaxis with intra- DISCUSSION
venous antibiotics (cefazolin, 2 g) had been instituted at the
Our retrospective analysis identified ten endometriosis pa-
time of the oocyte retrieval and no accidental puncture of
tients over a 4-year period who were hospitalized for an acute
episode of TOA. Only three of these women had undergone
ART. In one of those three cases, the pelvic infection occurred
TABLE 1
>60 days after oocyte retrieval and was therefore not reported
Baseline characteristics (n [ 10).
to the supervising agency, in accordance with French rules
overseeing ART. In six of the ten women, the infectious pro-
Characteristic n cess could not be tied to any incidental event and was there-
Age (y), mean  SD 33  5 fore considered to be spontaneously occurring. Our findings
Previous pregnancy indicate, therefore, that infectious complications following
0 6
1 2 ART conducted while endometriomas are present are likely
2 2 to be either under- or overreported. Infectious complications
Infertility 4 occurring >2 months after ART are likely to not be reported
rAFS score as ART complications, thus accounting for a possible under-
I 0
II 0 reporting. Conversely, our observation that similar infections
III 3 can occur spontaneously in women with endometriomas sug-
IV 7 gests that some late cases may not be necessarily linked to the
Endometriomas
0 3
earlier ART. In those women, therefore, making an erroneous
Unilateral 4 link between spontaneously occurring TOAs and an earlier
Bilateral 3 ART procedure would lead to overreporting complications
History of surgical resection of endometriosis 5 conducted while endometriomas are present.
Note: rAFS ¼ revised American Fertility Society classification for endometriosis.
In women who had ART, the interval between oocyte
Villette. Endometriomas, ART, and infections. Fertil Steril 2016.
retrieval and the following infectious complication is highly

412 VOL. 106 NO. 2 / AUGUST 2016

VOL. 106 NO. 2 / AUGUST 2016

TABLE 2

Case descriptions.
No. of days between
WBC Drainage, beginning ATB
Patient Context Symptoms (109/L) Ultrasound findings 1st intent Drainage, 2nd intent and drainage Surgical findings Germs
1 16 d after Pelvic pain, fever 25 5 cm solid unilateral No Laparoscopic 7 Ovarian abscess 0
oocyte retrieval adnexal mass with OMA
2 57 d after Pelvic pain, fever 14.1 4 cm solid unilateral No Laparoscopic 3 Ovarian abscess E. coli
oocyte retrieval adnexal mass with OMA
3 102 d after Pelvic pain, fever, 13 4 cm solid unilateral No Laparoscopic, 9 Ovarian abscess with Streptococcus B
oocyte retrieval metrorraghia adnexal mass laparoconversion OMA, appendicular
plastron
4 37 d after Pelvic pain, fever 11.9 8 cm unilateral No Laparoscopic 2, septic shock Peritonitis, ovarian E. coli
cesarean section adnexal mass abscess, pyosalpynx
5 0 Pelvic pain, fever 20.5 Not available No CT-guided drainage 6 Peritonitis, retro-bladder No germ found
(2nd intent), abscess, retro-uterine
laparoscopic abscess, ovarian
(3rd intent) abscess with OMA
6 0 Pelvic pain, fever, 14.5 8 cm solid unilateral No No — — E. coli
metrorraghia adnexal mass
7 0 Pelvic pain, fever 13.7 16 cm solid unilateral Laparoscopic — 0 Peritonitis, ovarian Streptococcus B
adnexal mass, free abscess with OMA
fluid in the pelvis
8 0 Pelvic pain, fever 25 8 cm solid bilateral No No — — 0
adnexal mass
9 0 Pelvic pain, fever, 10.8 6 cm solid unilateral No Laparoscopic 5 Peritonitis, bilateral 0
metrorraghia adnexal mass pyosalpinx, ovarian
abscess with OMA
10 0 Pelvic pain, fever 15.6 5 cm solid unilateral Laparoscopic — 0 Peritonitis, pyosalpinx, Streptococcus B
adnexal mass, free salpingectomy
fluid in the pelvis

Fertility and Sterility®

Note: ATB ¼ antibiotic; OMA ¼ endometrioma; WBC ¼ white blood cell count.
Villette. Endometriomas, ART, and infections. Fertil Steril 2016.
413
ORIGINAL ARTICLE: ENDOMETRIOSIS
variable, ranging from 16 to 102 days in our study and from 4
to 120 days in other case reports (13–17).
Our finding of spontaneously occurring acute pelvic in-
fections in women bearing endometriomas has been reported
by others (19–23). Chen et al. (24) indicated that patients with
advanced-stage endometriosis (III and IV) had a higher risk of
developing TOA than those without endometriosis. Practi-
cally, the finding of spontaneously occurring acute pelvic
infection in women with endometriomas points to the possi-
bility that late-occurring complications of ART may not be
truly linked to ART.
The main strength of the present study is the long obser-
vation period (4 years) and the systematic character of our re-
view. The long observation period maximizes the chances that
all late-occurring complications of ART occurring in endome-
triosis patients are reported. This contrasts with other studies,
which had much shorter observation periods, thereby risking
underestimating the true incidence of post-ART infections in
women with endometriomas.
Our study suffers, however, from the limitations of its
retrospective design and the relatively small number of cases
observed. Moreover, our study design could not account for
ART complications that might have been treated at different
institutions. This, however, was likely compensated for by
women having had ART elsewhere but seeking emergency
treatment at our institution.
Several factors could account for the fact that endometri-
osis women might be more prone to infection (14, 17, 24): 1)
Endometriosis is associated with alterations of the immune
system that could enhance vulnerability to infection; 2) blood
accumulations contained in endometriomas or elsewhere in
the peritoneal cavity constitute an ideal culture medium for
bacterial growth and are therefore susceptible to fueling any
infectious process; and 3) hypothetic fragility of the cyst wall
may promote bacterial diffusion and infection of the cyst
content.
When undertaking ART in women with known endome-
triomas, one must recognize the risk of infection and counsel
patients accordingly. Moreover, one should deploy preventive
measures during the oocyte retrieval even if their efficacy has
not been formally proven. These include vaginal prepping
with Betadine followed by profuse vaginal irrigation with
normal saline solution (25, 26) and prophylactic antibiotic
with the use of 1 g intravenous cefazolin starting just before
oocyte retrieval. Avoiding puncturing the endometrioma has
been recommended (27). Finally, patients should be counseled
to promptly report to the treating institution in case of pelvic
pain and/or fever. Despite all these prophylactic measures,
some patients will develop TOAs, however, as shown by the
present report.
TOA in women suffering from endometriosis is a severe
complication that requires prompt and accurate diagnosis
and adequate therapeutic measures. In our series, only two
of the ten patients were successfully treated by means of
prompt commencement of vigorous medical treatment,
namely, intravenous multiple antibiotic therapy. Conversely,
eight women had a laparoscopic drainage accompanying the
intravenous antibiotic therapy, which was conducted second-
arily in six of these eight women because they had failed to
414
improve. One of the latter underwent a septic shock and
needed to be admitted to the intensive care unit. The possible
impact of TOA on ovarian reserve and further fecundity is a
valid concern. The retrospective nature of our study did not
permit assessing this parameter, however, because seven of
the ten women were not actively attempting to conceive.
One of the three women whose TOA occurred in the aftermath
of ART conceived in a later ART attempt.
A recent retrospective study (28) compared 148 women
hospitalized for PID with or without TOA and with or without
endometriosis. The authors reported that the characteristics of
PID were more severe in women with endometriosis and more
prone to resist to conventional antibiotic therapy. Recent
French guidelines on PID (29) recommend drainage for TOA
with a sonographic mass >3 cm, in association with broad-
spectrum antibiotic. Drainage can be performed during lapa-
roscopy or be guided by sonography or CT. Our own data
support this view, indicating that it is not appropriate to
wait for lack of clinical improvement and/or appearance signs
of peritonitis for draining TOA. This rule should be even
strictly approved in women with endometriosis.
In our series, only two patients had first-intent laparos-
copies, which were motivated by compromised general condi-
tion. Conversely, laparoscopic drainage of TOAs was secondly
decided 2–9 days later in six of eight women because they failed
to improve. Retrospectively, our data indicate that it would
have been sounder to opt up front for laparoscopic drainage
on diagnosing TOA, as recommended by others (30). Laparos-
copies may, however, be technically difficult in certain cases
owing to the presence of severe pelvic adhesions secondary
to endometriosis and past surgery for endometriosis. These
technical difficulties constitute reasons for not systematically
attempting to remove all endometriotic lesions (22). An alter-
nate option for draining pelvic abscess in case of TOA is to pro-
ceed transvaginally under ultrasound guidance, with success
rates of 93% reported in nonendometriosis women (31, 32).
CONCLUSION
Our 4-year review of infectious processes in women with endo-
metriosis indicated that such complications can occur long af-
ter ART, or even spontaneously. This indicates that infectious
complications of ART conducted in women with endometriosis
can be either under- or overreported. Indeed, certain late occur-
ring complications—not necessarily recorded—can actually be
spontaneously occurring and not ART related.
Our data also confirmed the prevailing concept that TOAs
occurring in women with endometriosis—spontaneously or
after ART—are best treated by surgical drainage. An early
drainage in conjunction with intravenous antibiotics consti-
tutes the best assurance that serious complications of TOA,
such as septic shock, are avoided. The option of nonsurgical
ultrasound-guided drainage still awaits to be validated in
endometriosis patients.
REFERENCES
1. Brosens I, Gordts S, Puttemans P, Benagiano G. Pathophysiology proposed
as the basis for modern management of the ovarian endometrioma. Reprod
Biomed Online 2014;28:232–8.
VOL. 106 NO. 2 / AUGUST 2016

2. Jadoul P, Kitajima M, Donnez O, Squifflet J, Donnez J. Surgical treatment of
ovarian endometriomas: state of the art? Fertil Steril 2012;98:556–63.
3. Garcia-Velasco JA, Somigliana E. Management of endometriomas in
women requiring IVF: to touch or not to touch. Hum Reprod 2009;24:
496–501.
4. Somigliana E, Arnoldi M, Benaglia L, Iemmello R, Nicolosi AE, Ragni G. IVF-
ICSI outcome in women operated on for bilateral endometriomas. Hum Re-
prod 2008;23:1526–30.
5. Celik HG, Dogan E, Okyay E, et al. Effect of laparoscopic excision of endo-
metriomas on ovarian reserve: serial changes in the serum antimullerian hor-
mone levels. Fertil Steril 2012;97:1472–8.
6. Ruiz-Flores FJ, Garcia-Velasco JA. Is there a benefit for surgery in
endometrioma-associated infertility? Curr Opin Obstet Gynecol 2012;
24:136–40.
7. Raffi F, Metwally M, Amer S. The impact of excision of ovarian endome-
trioma on ovarian reserve: a systematic review and meta-analysis. J Clin En-
docrinol Metab 2012;97:3146–54.
8. Busacca M, Vignali M. Endometrioma excision and ovarian reserve: a
dangerous relation. J Minim Invasive Gynecol 2009;16:142–8.
9. Somigliana E, Berlanda N, Benaglia L, Vigano P, Vercellini P, Fedele L. Sur-
gical excision of endometriomas and ovarian reserve: a systematic review
on serum antimullerian hormone level modifications. Fertil Steril 2012;
98:1531–8.
10. Benaglia L, Bermejo A, Somigliana E, et al. Pregnancy outcome in women
with endometriomas achieving pregnancy through IVF. Hum Reprod
2012;27:1663–7.
11. Vigano P, Corti L, Berlanda N. Beyond infertility: obstetrical and postpartum
complications associated with endometriosis and adenomyosis. Fertil Steril
2015;104:802–12.
12. Practice Committee of the American Society for Reproductive M. Endome-
triosis and infertility: a committee opinion. Fertil Steril 2012;98:591–8.
13. Wei CF, Chen SC. Pelvic abscess after ultrasound-guided aspiration of endo-
metrioma: a case report. Zhonghua Yi Xue Za Zhi (Taipei) 1998;61:603–7.
14. Younis JS, Ezra Y, Laufer N, Ohel G. Late manifestation of pelvic abscess
following oocyte retrieval, for in vitro fertilization, in patients with severe endo-
metriosis and ovarian endometriomata. J Assist Reprod Genet 1997;14:343–6.
15. Moini A, Riazi K, Amid V, et al. Endometriosis may contribute to oocyte
retrieval-induced pelvic inflammatory disease: report of eight cases. J Assist
Reprod Genet 2005;22:307–9.
16. Matsunaga Y, Fukushima K, Nozaki M, et al. A case of pregnancy compli-
cated by the development of a tubo-ovarian abscess following in vitro fertil-
ization and embryo transfer. Am J Perinatol 2003;20:277–82.
VOL. 106 NO. 2 / AUGUST 2016
Fertility and Sterility®
17. Padilla SL. Ovarian abscess following puncture of an endometrioma during
ultrasound-guided oocyte retrieval. Hum Reprod 1993;8:1282–3.
18. Benaglia L, Somigliana E, Iemmello R, Colpi E, Nicolosi AE, Ragni G. Endome-
trioma and oocyte retrieval-induced pelvic abscess: a clinical concern or an
exceptional complication? Fertil Steril 2008;89:1263–6.
19. El-Toukhy T, Hanna L. Spontaneous ovarian abscess associated with an en-
dometrioma. J Obstet Gynaecol 2007;27:328–9.
20. Kavoussi SK, Pearlman MD, Burke WM, Lebovic DI. Endometrioma compli-
cated by tubo-ovarian abscess in a woman with bacterial vaginosis. Infect
Dis Obstet Gynecol 2006;2006:84140.
21. Kubota T, Ishi K, Takeuchi H. A study of tubo-ovarian and ovarian ab-
scesses, with a focus on cases with endometrioma. J Obstet Gynaecol
Res 1997;23:421–6.
22. Mokdad C, Rozsnayi F, Delaunay F, et al. [Surgical treatment of tubo-ovarian
abscess occurring in deep endometriosis]. Gynecol Obstet Fertil 2010;38:
490–5. French.
23. Lipscomb GH, Ling FW, Photopulos GJ. Ovarian abscess arising within an en-
dometrioma. Obstet Gynecol 1991;78:951–4.
24. Chen MJ, Yang JH, Yang YS, Ho HN. Increased occurrence of tubo-ovarian
abscesses in women with stage III and IV endometriosis. Fertil Steril 2004;
82:498–9.
25. Hannoun A, Awwad J, Zreik T, Ghaziri G, Abu-Musa A. Effect of betadine
vaginal preparation during oocyte aspiration in in vitro fertilization cycles
on pregnancy outcome. Gynecol Obstet Invest 2008;66:274–8.
26. van Os HC, Roozenburg BJ, Janssen-Caspers HA, et al. Vaginal disinfection
with povidon iodine and the outcome of in-vitro fertilization. Hum Reprod
1992;7:349–50.
27. Bennett SJ, Waterstone JJ, Cheng WC, Parsons J. Complications of transva-
ginal ultrasound-directed follicle aspiration: a review of 2670 consecutive
procedures. J Assist Reprod Genet 1993;10:72–7.
28. Elizur SE, Lebovitz O, Weintraub AY, et al. Pelvic inflammatory disease in
women with endometriosis is more severe than in those without. Aust N Z
J Obstet Gynaecol 2014;54:162–5.
29. [No author] [Pelvic inflammatory diseases]. J Gynecol Obstet Biol Reprod
2012;41:930–4. French.
30. Rosen M, Breitkopf D, Waud K. Tubo-ovarian abscess management options
for women who desire fertility. Obstet Gynecol Surv 2009;64:681–9.
31. Gjelland K, Ekerhovd E, Granberg S. Transvaginal ultrasound-guided aspira-
tion for treatment of tubo-ovarian abscess: a study of 302 cases. Am J Obstet
Gynecol 2005;193:1323–30.
32. Granberg S, Gjelland K, Ekerhovd E. The management of pelvic abscess. Best
Pract Res Clin Obstet Gynaecol 2009;23:667–78.
415