The Intraoperative Complication Rate of

Nonobstetric Dilation and Curettage

Lukas Hefler, MD, Andrea Lemach, MD, Veronika Seebacher, MD, Stephan Polterauer, MD,
Clemens Tempfer, MD, and Alexander Reinthaller, MD

OBJECTIVE: To evaluate the intraoperative complication

rate of nonobstetric dilation and curettage (D&C) in a
large series of consecutive patients.
N ew technologies, such as transvaginal ultrasonog-
raphy and outpatient endometrial sampling,
have led to a decline in the performance of dilation
METHODS: In this retrospective study, 5,359 nonobstet- and curettage (D&C) procedures in the operating
ric D&Cs performed in 2,542 premenopausal and 2,817 room in the past decades.1,2
postmenopausal patients between October 1995 and A hospital D&C can be performed for both diag-
December 2006 were evaluated. Intraoperative proce- nostic and therapeutic indications. Indications for a
dure-associated complication rate and identification of diagnostic hospital D&C include patients with a nondi-
risk factors for the occurrence of complications were the agnostic office biopsy, patients with endometrial hyper-
main outcome measures. Univariable and multivariable plasia, and patients with cervical stenosis. In most cases,
analyses were performed. diagnostic hysteroscopy is performed before a diagnos-
RESULTS: A total of 103 (1.9%) intraoperative complica- tic D&C to obtain a visual image of the endometrial
tions were noted. Uterine perforation occurred in 50 cavity and to exclude focal disease.3 A therapeutic
cases (0.9%) (perforation site: fundus, nⴝ47; cervix, nⴝ3). indication for performing D&C is the temporary man-
Forty-two (0.8%) cases of false passage, seven cases agement of prolonged or excessive vaginal bleeding.
(0.1%) with severe hemorrhage, three cases of vaginal A number of studies have been published on com-
laceration, and one case of cervical laceration were plications from obstetric D&C procedures, but few data
noted. In a multivariable analysis, retroversion of the are available on nonobstetric D&C procedures.4 Com-
uterus (Pⴝ.008), postmenopausal status (Pⴝ.003), and plications include hemorrhage,5 uterine perforation,4
nulliparity (Pⴝ.03) were significantly associated with oc- infection,6 formation of intrauterine adhesions,7,8 text,
currence of intraoperative complications.
and trophoblast embolization.9 Although a number of
CONCLUSION: The overall complication rate of D&C is studies have been published on complications of preg-
low. A retroverted uterus, postmenopausal status, and nancy-related D&Cs, only one study reports on the
nulliparity are independent risk factors for intraoperative procedure-related morbidity of nonobstetric diagnostic
complications. and therapeutic D&Cs. In 1980, Ben-Baruch et al re-
(Obstet Gynecol 2009;113:1268–71)
ported on 3,299 patients undergoing both diagnostic
LEVEL OF EVIDENCE: III and therapeutic D&Cs not related to pregnancy. The
complication rate varied between 0.05% and 1.8% de-
pending on the indication for surgery.4
Preoperative counseling of patients, including an
From the Department of Obstetrics and Gynecology, Medical University of
Vienna, Vienna, Austria. explanation of risks and complications, is essential in
Supported by the Ludwig Boltzmann Institute of Gynecology and Gynecologic today’s medicine. The aim of the present study was to
Oncology, Vienna, Austria. provide data for the preoperative counseling of patients
Corresponding author: Lukas Hefler, MD, Department of Obstetrics & Gyne- regarding the intraoperative surgical complication rate
cology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 of nonobstetric inpatient D&Cs using a series of 5,359
Vienna, Austria; e-mail: lukas.hefler@meduniwien.ac.at.
consecutive patients.
Financial Disclosure
The authors did not report any potential conflicts of interest.
MATERIALS AND METHODS
© 2009 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins. This study was found to be exempt from ethical
ISSN: 0029-7844/09 approval by the Ethics Committee of the Medical

1268 VOL. 113, NO. 6, JUNE 2009 OBSTETRICS & GYNECOLOGY

University of Vienna. Clinical data were obtained Table 1. Indications for Dilation and Curettage
retrospectively from files at the Medical University of Premenopausal Postmenopausal
Vienna, Department of Obstetrics and Gynecology. Patients Patients
Two investigators (A.L., V.S.) independently re- Postmenopausal — 1,637 (58.1)
corded the information and extracted the data. Any bleeding
equivocal data were analyzed together with the pri- Menorrhagia, 1,681 (66.1) —
mary investigator (L.H.). Patients undergoing preg- metrorrhagia
Cervical and 468 (18.4) 428 (15.2)
nancy-related D&C were excluded from the present
endometrial
study. A total of 5,359 consecutive patients underwent polyps
hospital D&C between October 1995 and December Endometrial 174 (6.9) 625 (22.2)
2006 and were included in the present study. hyperplasia and
mucometra
All D&Cs were performed as inpatient procedures Cervical dysplasia 112 (4.4) 71 (2.5)
under general anesthesia with the patient in the dorsal Fibroids 47 (1.9) 28 (1)
lithotomy position. A gynecologic examination was Cervical occlusion 16 (0.6) 18 (0.6)
performed first. After the examination, the perineum, Others 44 (1.7) 10 (0.4)
vagina, and cervix were cleansed with an aseptic Data are n (%).
solution. A single-tooth tenaculum was used to grasp
the anterior lip of the cervix. Traction was applied to tic regression analysis with the above risk factors as
the tenaculum to align the axis of the cervix and the independent variables and the occurrence of an intra-
uterine canal. The uterus was sounded to document operative complication as a dependent variable was
the size and confirm the position. After sounding the performed. P values ⬍.05 were considered statisti-
uterus, the cervix was dilated using Hegar dilators cally significant. We used the statistical software SPSS
(Rudolf Heintel GmbH, Vienna, Austria). The dilation 11.0 for Windows (SPSS, Inc., Chicago, IL) for statis-
was conducted to 6 mm, at which point the hystero- tical analysis.
scope was inserted comfortably. A diagnostic hysteros-
copy using a standard 30° 5-mm hysteroscope was RESULTS
performed. The uterine cavity was distended with nor- Indications for D&C and the histologic results broken
mal saline. Subsequently, an endocervical curettage was down by menopausal status are given in Tables 1 and
performed to avoid contamination of the histologic 2, respectively. In 4,213 (78.6%) cases, diagnostic
specimen with endometrial cells. Sharp curettes were hysteroscopy was performed before the D&C. A total
used for all procedures. Curettage was performed sys- of 1,146 (21.4%) patients underwent D&C without
tematically, beginning at the fundus. The entire surface prior diagnostic hysteroscopy. Menopausal status was
of the endometrium was sampled by moving around the known in all cases. Two thousand five hundred
uterus in a consistent and systematic fashion. forty-two (47.4%) and 2,817 (52.6%) patients were
When a perforation was suspected during sur- premenopausal and postmenopausal, respectively, at
gery, a diagnostic hysteroscopy was performed, the time of D&C. Mean age of the patients was 53.0
checking the site of perforation and the severity of (⫾13.2) years. Five hundred twenty-five (9.8%) pa-
hemorrhage. If no bleeding was detected, surgery tients were nulliparous at the time of D&C. Two
was stopped immediately. If any visceral trauma thousand seven hundred seventy-eight (51.7%) and
was suspected, further surgical exploration was
performed. Table 2. Histologic Results
Values are given as mean (standard deviation) for Premenopausal Postmenopausal
normally distributed values. Metric variables were Patients Patients
compared using Student t test. The position of the Benign endometrial 2,474 (97.3) 2,566 (91.1)
uterus (anteverted compared with retroverted uterus), histology
menopausal status (premenopausal compared with Complex atypical 23 (0.9) 34 (1.2)
postmenopausal), parity (at least one compared with 0 endometrial
hyperplasia
deliveries), performance of a diagnostic hysteroscopy Endometrial cancer 30 (1.2) 183 (6.5)
(no compared with yes), and surgeon’s experience Uterine sarcoma 0 (0.0) 12 (0.4)
(attending physician compared with resident) were Cervical cancer 4 (0.2) 8 (0.3)
assessed as risk factors for intraoperative complica- Sampling failure 11 (0.4) 14 (0.5)
tions using univariable ␹2 tests. A multivariable logis- Data are n (%).

VOL. 113, NO. 6, JUNE 2009 Hefler et al Complication Rate of Dilation and Curettage 1269
Table 3. Summary of Intraoperative found. Patient 2 was a 67-year-old women who was to
Complications undergo diagnostic hysteroscopy and D&C for post-
Patients menopausal bleeding and an endometrial hyperplasia
of 14 mm. During dilation of the cervix with Hegar’s
Total number of complications 103 (1.9)
Uterine perforation 50 (0.9) dilators, a perforation in the cervix occurred. Subse-
False passage 42 (0.8) quently, hysteroscopy was performed. Visibility was
Severe hemorrhage 7 (0.1) poor because of bleeding. It appeared that perforation
Vaginal laceration 3 (⬍0.1) to the right parametrium near the uterine vessel and
Cervical laceration 1 (⬍0.1) the ureter occurred. Therefore, emergency laparot-
D&C, dilation and curettage. omy and hysterectomy were performed.
Data are n (% of all dilation and curettage procedures).
In a univariable analysis, the position of the
uterus (anteverted: 1.8% compared with retroverted:
2,581 (48.3%) surgeries were done by residents and 3.6% uterus, P⫽.01), menopausal status (premeno-
attending physicians, respectively. pausal: 1.2% compared with postmenopausal: 2.6%,
The types of intraoperative complications are P⬍.001), parity (at least one: 1.7% deliveries com-
shown in Table 3. Demographic characteristics of pared with 0: 3.4%, P⫽.01), the performance of a
patients with and without intraoperative complica- diagnostic hysteroscopy (no: 0.8% compared with
tions are shown in Table 4. In cases of uterine yes: 2.2%, P⫽.002), but not the surgeon’s experience
perforation, the perforation site was the fundus and (resident: 2.0% compared with attending physician:
the cervix in 47 and three cases, respectively. Uterine 1.8%, P⫽.6) were associated with the occurrence of
perforation was done with Hegar dilators (n⫽27), the intraoperative complications. The data generated
curette (n⫽15), the hysteroscope (n⫽4), the sounding from a multivariable regression analysis are shown in
probe (n⫽2), and a grasping forceps (n⫽3). The Table 5.
operative consequences in cases with surgical compli-
cation were as follows: abortion of D&C (n⫽18), DISCUSSION
laparoscopy (n⫽2), laparotomy (n⫽1), and hysterec- Although a D&C is a diagnostic and therapeutic
tomy (n⫽1). The two most notable cases were those surgical procedure used frequently throughout the
undergoing laparotomy (patient 1) and hysterectomy world, few data are available on intraoperative surgi-
(patient 2). Patient 1 was a 37-year-old woman under- cal complication rates.
going diagnostic hysteroscopy and D&C for pro- The only data on surgical complications of D&Cs
longed abnormal bleeding. During D&C, perforation date back to 1980. Interestingly, more studies have
occurred with the sharp curette at the tubal ostium. been published on pregnancy-associated curettages
Hysteroscopy was performed for suspected adhesion and operative/diagnostic hysteroscopies than on non-
between the small bowel and the uterus with bowel obstetric D&Cs.4,5,9 –11 We performed a chart review
perforation. Subsequently, diagnostic laparoscopy and extracted data for 5,359 consecutive patients
and laparotomy were performed. No bowel perfora- undergoing D&C in a large teaching hospital.
tion was noted, but a lost intrauterine device intraab- The overall complication rate was relatively low
dominally densely adherent with the uterus was (1.9%). We cannot exclude, however, any unrecog-
nized intraoperative complication. Most of these com-

Table 4. Demographic Characteristics of Patients

With and Without Intraoperative Table 5. Multivariable Regression Analysis
Complications
P OR 95% CI
Patients Without Patients With
Position of the uterus (retroverted .01 1.6 1.01–2.2
Intraoperative Intraoperative
vs anteverted)
Complications Complications
Menopausal status (postmenopausal .004 2.7 1.4–5.2
Patients 5,256 103 vs premenopausal)
Age 52.8 (⫾13.0) 59.2 (⫾15.3) No. of deliveries (nulliparous vs .04 1.9 1.04–3.7
Nulliparous women 507 18 parous women)
Postmenopausal women 2,744 73 Performance of a diagnostic .2 0.4 0.1–1.6
Total pregnancies 2.3 (⫾2.4) 2.0 (⫾1.8) hysteroscopy (no vs yes)
Diagnostic hysteroscopy 4,119 94 Surgeon’s experience (attending .8 0.9 0.5–1.8
Retroverted uterus 452 17 physician vs resident)
Data are n or mean (⫾standard deviation). OR, odds ratio; CI, confidence interval.

1270 Hefler et al Complication Rate of Dilation and Curettage OBSTETRICS & GYNECOLOGY
plications were related to myometrial violation (false REFERENCES
passage or perforation). Of note, our results defined 1. Gull B, Karlsson B, Milsom I, Granberg S. Can ultrasound
certain patients at a higher risk for intraoperative replace dilation and curettage? A longitudinal evaluation of
postmenopausal bleeding and transvaginal sonographic mea-
complications during D&C, such as older patients, surement of the endometrium as predictors of endometrial
patients with a retroverted uterus, and nulliparous pa- cancer. Am J Obstet Gynecol 2003;188:401–8.
tients. Our data do not include whether parous pa- 2. Gorlero F, Nicoletti L, Lijoi D, Ferrero S, Pullè A, Ragni N.
tients had cesarean deliveries or vaginal deliveries. Endometrial directed biopsy during sonohysterography using
Furthermore, no information was available on the NiGo device: prospective study in women with abnormal
uterine bleeding. Fertil Steril 2008;89:984–90.
whether patients had prior D&Cs for obstetric reasons
such as elective abortions. Whether or not diagnostic 3. Valle RF, Sciarra JJ. Intrauterine adhesions: hysteroscopic
diagnosis, classification, treatment, and reproductive outcome.
hysteroscopy was performed before the D&C did not Am J Obstet Gynecol 1988;158:1459–70.
affect the complication rate. Our analysis showed that 4. Ben-Baruch G, Menczer J, Shalev J, Romem Y, Serr DM.
residents did equally well regarding complication Uterine perforation during curettage: perforation rates and
rates as did attending physicians. Of note, we have postperforation management. Isr J Med Sci 1980;16:821–4.
investigated only intraoperative surgical complica- 5. Lowensohn RI, Hibbard LT. Laceration of the ascending
tions. Other possible surgery-related complications, branch of the uterine artery: a complication of therapeutic
abortion. Am J Obstet Gynecol 1974;118:36–8.
such as infection, pulmonary emboli, and thrombosis,
and other complications potentially related to anes- 6. Sacks PC, Tchabo JG. Incidence of bacteremia at dilation and
curettage. J Reprod Med 1992;37:331–4.
thesia or lithotomy position, were beyond the scope
7. Broome JD, Vancaillie TG. Fluoroscopically guided hystero-
of the present study. scopic division of adhesions in severe Asherman syndrome.
Our data compare favorably with a previous Obstet Gynecol 1999;93:1041–3.
smaller series published by Ben-Baruch et al.4 In this 8. Twiggs LB, Phillips GL. Documentation of subclinical tropho-
series, uterine perforation occurred in 0.16% of D&Cs. blastic embolization with invasive cardiac monitoring in a
woman with a molar pregnancy. A case report. J Reprod Med
The most common site of perforation was the uterine
1986;31:277–9.
fundus, and the instrument most often involved was a
9. Cohle SD, Petty CS. Sudden death caused by embolization of
sharp curette. The perforation rates in D&Cs performed trophoblast from hydatidiform mole. J Forensic Sci 1985;30:
for intrauterine adhesions and postmenopausal bleeding 1279–83.
were 1.8% and 0.2%, respectively.4 10. Park TK, Flock M, Schulz KF, Grimes DA. Preventing febrile
In a large teaching hospital, retroverted uterus, complications of suction curettage abortion. Am J Obstet
Gynecol 1985;152:252–5.
postmenopausal status, and nulliparity were indepen-
dent risk factors for intraoperative complications of 11. Di Spiezio Sardo A, Taylor A, Tsirkas P, Mastrogamvrakis G,
Sharma M, Magos A. Hysteroscopy: a technique for all?
D&C. These data can be used for preoperative patient Analysis of 5,000 outpatient hysteroscopies. Fertil Steril
counseling. 2008;89:438–43.

VOL. 113, NO. 6, JUNE 2009 Hefler et al Complication Rate of Dilation and Curettage 1271