Benign hysterectomies account for more than 400,000 surgeries annually, and the most common indications include leiomyomas and abnormal uterine bleeding.1 The route of hysterectomy is shifting from abdominal to laparoscopic-assisted approaches, which confers a more rapid recovery and fewer perioperative complications.1,2 Specimen morcellation is often necessary to perform a minimally invasive procedure.
Inadvertent morcellation of an unexpected malignancy is a surgical risk of hysterectomy and although the clinical significance is unclear, current practice supports surgical staging and appropriate subsequent treatment. Evidence suggests that morcellation of uterine sarcoma potentially upstages disease, which portends a poor prognosis.3,4 Uterine sarcomas are a rare group of heterogeneous and potentially aggressive tumors. Leiomyosarcoma is notable for frequent local recurrence, a 60% or less 5-year survival, and high metastatic potential.5 Among women undergoing a benign hysterectomy, the incidence of unexpected uterine sarcoma ranges from 0.09 to 0.49%6–11 and for unexpected endometrial cancer from 0.13 to 0.45%.8–11
Prior studies that evaluated the incidence of unexpected gynecologic malignancies reported small cohorts at single medical institutions with mixed results. Given the varying reports of incidence and clinical significance of unexpected cancer at the time of minimally invasive hysterectomy, we sought to use a prospectively collected population-based data set to define the incidence of unexpected uterine sarcoma and other gynecologic malignancies among women undergoing hysterectomy for a benign indication.
MATERIALS AND METHODS
We performed a data analysis of an existing all-payer, quality and safety surgical database maintained by the Michigan Surgical Quality Collaborative. The Collaborative is a statewide group of hospitals that voluntarily reports perioperative surgical outcomes and is funded by Blue Cross and Blue Shield of Michigan/Blue Care Network. Currently, 52 hospitals participate, and 60% are community-based. This represents approximately 30% of Michigan hospitals, although the majority of hospitals that do not participate have small bed sizes. At each hospital, a trained nurse researcher abstracts patient characteristics, perioperative processes of care, and 30-day postoperative outcomes. The first 25 surgical cases are collected every 8 days, resulting in approximately 25% of all hysterectomies performed at each institution included in the database. To reduce sampling error, a standardized data collection methodology is used and routinely validated through scheduled site visits, conference calls, and internal audits. Institutional review board approval for this analysis was obtained from the University of Michigan institutional review board–medical (HUM00073978).
All cases of hysterectomy performed for benign indications in the database from January 1, 2013, through December 8, 2013, were included in the analysis. Preoperative surgical indications for hysterectomy were abstracted from the medical record and benign indications included pelvic mass, family history of cancer, endometrial hyperplasia without atypia, uterovaginal prolapse, endometriosis, pelvic pain, abnormal uterine bleeding, or leiomyomas. Women with a preoperative surgical indication of cancer, cervical dysplasia, or endometrial hyperplasia with atypia and those who underwent an oncologic staging procedure were excluded from the benign indications and were considered separately. Cases with an obstetric indication for hysterectomy and those with missing data for preoperative indication were also excluded from this analysis.
Unexpected cancer was defined as malignancy confirmed on pathology specimen without a preoperative surgical indication for cancer or indication that raises the concern for cancer (cervical dysplasia and endometrial hyperplasia with atypia). Gynecologic cancers were grouped into uterine sarcoma, endometrial cancer, cervical cancer, ovarian, peritoneal, fallopian tube cancer, and metastatic cancer. The diagnosis of uterine sarcoma and the associated subtypes (leiomyosarcoma, carcinosarcoma, and adenosarcoma) were established based on the final pathology report abstracted from the clinical record. In clinical practice, pathologic diagnosis of uterine sarcoma is typically based on guidelines established by the World Health Organization and the College of American Pathologists.5
We reviewed patient and hospital demographics and detailed perioperative characteristics. Preoperative information included patient comorbidities, history of preoperative venous thromboembolism, and preoperative blood transfusion. Operative data comprised surgeon type (benign gynecologist or gynecologic oncologist), preoperative surgical indication for hysterectomy, mode of hysterectomy, specimen weight, estimated blood loss, operative time, and final pathology of surgical specimen. Postoperative details such as blood transfusion, venous thromboembolism, and length of stay were also recorded.
Descriptive data were analyzed using χ2 or Fisher's exact test where applicable for categorical variables, and Student's t test, analysis of variance, or Wilcoxon rank-sum analysis for continuous variables as appropriate. Normally distributed data were reported as mean±standard deviation and nonparametric data were reported as median (range). Binomial 95% confidence intervals (CIs) were calculated using the Wilson interval. Data analyses were performed using Stata 13.1.
A total of 7,499 hysterectomies were performed during the specified time period. Thirty cases with missing preoperative indication data and eight nonobstetric cases were excluded. Of the remaining 7,461 women, 6,360 (85.24%) had a hysterectomy for benign indications. The incidence of unexpected gynecologic cancer was 2.7% (n=172) and included ovarian, peritoneal, and fallopian tube cancer (1.08% [n=69]), endometrial cancer (1.02% [n=65]), uterine sarcoma (0.22% [n=14]), metastatic cancer (0.20% [n=13]), and cervical cancer (0.17% [n=11; Fig. 1). The incidence of unexpected uterine sarcoma varies from 0.21 to 0.38% based on surgical indication (Table 1; Fig. 2). Additionally, unexpected endometrial cancer was identified in 38 (20.88%) women with an indication of hyperplasia with atypia and eight (6.84%) women with hyperplasia without atypia (Table 1; Fig. 2).
The demographics and clinical care of women who underwent hysterectomy for a benign indication and were confirmed to have benign disease on pathology are described subsequently. Benign pathology was confirmed in 6,188 (97.3%) women who had a hysterectomy for a benign indication. The most common indications for hysterectomy were abnormal uterine bleeding and leiomyomas (65.43% [n=4,054]). The majority of these women were non-Hispanic, white, and had private insurance. The average age was 47.0±10.76 years, and the average body mass index (BMI, calculated as weight (kg)/[height (m)]2) was 30.6±7.5 kg/m2 (Table 2). Nearly 63% underwent a laparoscopic or robotic approach, 25% abdominal, and the remainder were performed vaginally (Table 3).
The demographics of women with unexpected uterine sarcoma were compared with women who had a preoperative indication of cancer. The overall incidence of uterine sarcoma among all women undergoing hysterectomy in this data set was 0.32% (24/7,461). More than half of all sarcomas were unexpected (14/24) and were found in 14 of 6,360 women undergoing hysterectomy for reportedly benign indications (0.22%, 95% CI 0.001–0.004). Although women with sarcoma and a preoperative diagnosis of cancer were older than those with occult sarcoma (64.0±10.95 compared with 46.86±10.57 years, P≤.001), there was no difference in the mean age of women with unexpected sarcoma compared with those with benign pathology (46.86±10.57 compared with 47.0±10.76 years, P=.96). Furthermore, there was no difference in parity, ethnicity, race, insurance type, BMI, or surgical indication among women who were diagnosed with unexpected sarcoma compared with benign disease (Table 2).
No statistically significant clinical risk factors for unexpected uterine sarcoma were identified in this analysis. Compared with women with benign pathology, those with unexpected uterine sarcoma tended to have a higher prevalence of history of venous thromboembolism and a blood transfusion before hysterectomy, although this was not statistically significant (Table 3). The majority of women with unexpected uterine sarcoma had a surgical indication of abnormal uterine bleeding or leiomyomas, and the final pathology consistently included benign tissue in addition to uterine sarcoma (Table 4). Among women who had a hysterectomy for the indication of uterine leiomyomas (n=2,575), the incidence of unexpected uterine sarcoma was 0.27% (n=7). The incidence among women 55 years or older is 0.80% (n=2) and for women younger than 55 years 0.22% (n=5; Table 5).
A minority of surgeries with benign indications were performed by a gynecologic oncologist (7.4% [n=471]), and the most common indication was pelvic mass (n=278). Two cases with a diagnosis of unexpected uterine sarcoma were performed by a gynecologic oncologist and the preoperative surgical indications were abnormal uterine bleeding and pelvic mass and abnormal uterine bleeding and leiomyomas. Nearly 70% of the unexpected endometrial (n=46) and ovarian, peritoneal, and fallopian tube (n=47) cancer cases were performed by a gynecologic oncologist. Respectively, the most common preoperative indications were abnormal uterine bleeding and leiomyomas (60.87% [n=28]) and pelvic mass (93.62% [n=44]). Analysis of data assessing only hysterectomies performed by a benign gynecologist yielded similar results when compared with the primary analysis.
Given the potential heightened concern for malignancy associated with the preoperative indications of a pelvic mass (n=585) and family history of cancer (n=384), we performed a subanalysis excluding these two indications, resulting in a sample of 5,391 women. The subanalysis resulted in a lower overall incidence of occult cancer, 1.56% (n=84). The largest differences were noted both in the lower incidence of occult ovarian, peritoneal, and fallopian tube (n=9 [0.17%]) and metastatic (n=3 [0.06%]) cancers. The incidence of uterine sarcoma (n=12 [0.22%]) remained unchanged and the incidences of cervical (n=10 [0.19%]), and endometrial cancers (n=50 [0.93%]) did not appreciably change.
We examined data gathered from a large sample of Michigan hospitals and identified a 2.7% incidence of unexpected gynecologic malignancy after benign hysterectomy. The incidence of uterine sarcoma was 0.22% and the incidence of endometrial cancer was 1.02%. When the preoperative indications of pelvic mass and family history of cancer were excluded from the analysis, the decrease in unexpected malignancy incidence (1.56%) was primarily associated with the decreased incidence of ovarian, peritoneal, and fallopian tube cancer.
The 1.02% incidence of unexpected endometrial cancer is higher than previously published estimates ranging from 0.13 to 0.45%,8–11 which may be attributed to higher risk patients or differences in preoperative evaluation practices. Theben et al8 report that more than 85% of patients undergoing benign hysterectomy had cervical cancer screening and an ultrasonogram or dilation and curettage. Furthermore, Ouldamer et al10 described that all patients underwent cervical cancer screening and pelvic ultrasonography with endometrial biopsy as indicated before benign hysterectomy. This study confirmed risk factors associated with endometrial cancer including older age, higher BMI, lower parity, and comorbidities such as hypertension, diabetes, and sleep apnea.
Despite a comparatively low incidence, unexpected uterine sarcoma has received significant public attention. The 0.22% incidence of unexpected sarcoma is consistent with the 0.09–0.49% reported range.6–11 The 0.27% incidence among women with a surgical indication of leiomyomas is similar to the two prior studies that found a 0.23 and 0.49% incidence of unexpected sarcoma.6,7 The risk of uterine sarcoma is thought to increase with age, but given the infrequent occurrence, the current sample was likely underpowered to detect any statistically significant age-related difference (Table 3). A recent American College of Obstetricians and Gynecologists report notes that statistical analysis of studies that examine the incidence of uterine sarcoma is within the same 95% CI. Nonetheless, the variation in incidence may not be widely applicable given the small sample sizes, rarity of disease, and variability of age.12
There were no reliable predictors of unexpected uterine sarcoma, although women with unexpected uterine sarcoma were more likely to have a preoperative blood transfusion and history of venous thromboembolism. Specimen weight was larger among women with unexpected sarcoma, but no weight-predicted sarcoma and specimen weights varied with half less than 350 g. A prior study indicated black women were more likely to be diagnosed with sarcoma.13 We identified no significant difference in race among women diagnosed with uterine sarcoma, although the current sample is likely underpowered given the rarity of sarcoma.
The incidence of unexpected cervical cancer is not well established. Unexpected cancers are not surprising given the knowledge that not all cervical cancers are diagnosed with screening. For example, Chapman et al14 reported that among women diagnosed with occult cervical cancer at the time of benign hysterectomy, 27% had a normal preoperative pap smear.
The major limitation in this analysis is the lack of available preoperative testing information such as cervical cancer screening, ultrasonography, and endometrial biopsy, which should all be performed before hysterectomy as clinically indicated. Additionally, the final pathology of sarcoma recorded did not consistently differentiate among the subtypes of uterine sarcoma.
A unique strength of this study is the large sample size acquired from multiple community and academic institutions. The diversity of Michigan hospitals provides data that are generalizable to other settings. The variables were detailed and robust, especially compared with other previously published literature. Furthermore, Michigan Surgical Quality Collaborative safeguards the integrity of its prospectively collected data with site visits, evaluation of nurse researchers, and quarterly conferences to review data collection.
The low incidence of unexpected uterine sarcoma established in this study is valuable information to inform gynecologists in counseling women for fibroid-related symptoms. The benefit associated with minimally invasive surgery combined with the low risk of unexpected sarcoma outweighs the elevated risks that accompany open abdominal surgery.15
Comprehensive patient counseling should address the usual risks of surgery, the risk of unexpected malignancy, and the possibility of a poorer prognosis should an unidentified malignancy be morcellated. Although a complete preoperative evaluation and proper patient selection may decrease the risk of unexpected malignancy, it should be noted that no reliable markers distinguish uterine sarcoma from benign disease. In this setting, we should support technology to limit dissemination of unexpected malignancy with morcellation while preserving the benefits of minimally invasive surgery. Given proper counseling, a minimally invasive approach should remain a viable option for women seeking a hysterectomy.
1. Wright JD, Herzog TJ, Tsui J, Ananth CV, Lewin SN, Lu YS, et al.. Nationwide trends in the performance of inpatient hysterectomy in the United States. Obstet Gynecol 2013;122:233–41.
2. Nieboer T, Johnson N. Surgical approach to hysterectomy for benign gynaecological disease. The Cochrane Database of Systematic Reviews 2009, Issue 3. Art. No.: CD003677. DOI: 10.1002/14651858.CD003677.pub4.
3. Einstein MH, Barakat RR, Chi DS, Sonoda Y, Alektiar KM, Hensley ML, et al.. Management of uterine malignancy found incidentally after supracervical hysterectomy or uterine morcellation for presumed benign disease. Int J Gynecol Cancer 2008;18:1065–70.
4. Oduyebo T, Rauh-Hain AJ, Meserve EE, Seidman MA, Hinchcliff E, George S, et al.. The value of re-exploration in patients with inadvertently morcellated uterine sarcoma. Gynecol Oncol 2014;132:360–5.
5. DiSaia PJ, Creaseman W. Sarcoma of the uterus. In: Clinical gynecologic oncology. 8th ed. Maryland Heights (MO): Mosby; 2012.
6. Leibsohn S, d'Ablaing G, Mishell DR Jr, Schlaerth JB. Leiomyosarcoma in a series of hysterectomies performed for presumed uterine leiomyomas. Am J Obstet Gynecol 1990;162:968–74.
7. Parker WH, Fu YS, Berek JS. Uterine sarcoma in patients operated on for presumed leiomyoma and rapidly growing leiomyoma. Obstet Gynecol 1994;83:414–8.
8. Theben JU, Schellong AR, Altgassen C, Kelling K, Schneider S, Große-Drieling D. Unexpected malignancies after laparoscopic-assisted supracervical hysterectomies (LASH): an analysis of 1,584 LASH cases. Arch Gynecol Obstet 2013;287:455–62.
9. Takamizawa S, Minakami H, Usui R, Noguchi S, Ohwada M, Suzuki M, et al.. Risk of complications and uterine malignancies in women undergoing hysterectomy for presumed benign leiomyomas. Gynecol Obstet Invest 1999;48:193–6.
10. Ouldamer L, Rossard L, Arbion F, Marret H, Body G. Risk of incidental finding of endometrial cancer at the time of hysterectomy for benign condition. J Minim Invasive Gynecol 2014;21:131–5.
11. Rowland M, Lesnock J, Edwards R, Richard S, Zorn K, Sukumvanich P, et al.. Occult uterine cancer in patients undergoing laparoscopic hysterectomy with morcellation. Gynecol Oncol 2012;127(suppl):S29.
12. American College of Obstetricians and Gynecologists. Power morcellation and occult malignancy in gynecologic surgery: a special report. Washington, DC: American College of Obstetricians and Gynecologists; 2014:p. 1–6.
13. Brooks SE, Zhan M, Cote T, Baquet CR. Surveillance, epidemiology, and end results analysis of 2677 cases of uterine sarcoma 1989-1999. Gynecol Oncol 2004;93:204–8.
14. Chapman JA, Mannel RS, DiSaia PJ, Walker JL, Berman ML. Surgical treatment of unexpected invasive cervical cancer found at total hysterectomy. Obstet Gynecol 1992;80:931–4.
15. AAGL statement to the FDA on power morcellation. 2014. Available at: http://www.aagl.org/aaglnews/aagl-statement-to-the-fda-on-power-morcellation/
. Retrieved November 7, 2014.