10.1.1 The Role of Minimally Invasive Surgery in the Treatment of Uterine Fibroids

Hysterectomies and myomectomies, when done via a minimally invasive surgery (MIS) approach, have been shown to have far fewer complications. The benefits of MIS are well documented and include faster postoperative recovery including shorter hospital stay and shorter recovery time at home leading to faster return to work, less pain, and reduced risk of perioperative complications. A Cochrane systematic review by Nieboer et al. in 2009 evaluated 27 randomized controlled trials comparing total abdominal hysterectomies, laparoscopic hysterectomies, and vaginal hysterectomies performed for benign causes and reported that vaginal hysterectomy has the best outcomes and should be performed when possible. The review also recommended that although laparoscopic hysterectomy may be associated with longer operating times and higher rates of urinary tract injury, laparoscopic hysterectomy has advantages over total abdominal hysterectomy, including faster return to normal activity, shorter duration of hospital stays, lower intraoperative blood loss, and fewer wound infections [4]. Furthermore, a retrospective cohort study of 465,798 women by Wiser et al. evaluated outcomes of women who underwent abdominal and laparoscopic hysterectomies, and found a three times higher mortality risk for abdominal hysterectomy compared to laparoscopic hysterectomy (0.03% vs. 0.01%, OR 0.48 (0.24–0.95), respectively) [5].

Following the results published by Nieboer et al., the American College of Obstetricians and Gynecologists (ACOG) published a committee opinion [6] supporting the recommendations made by the Cochrane review. The AAGL followed with a position statement recommending that most hysterectomies for benign disease should be performed either vaginally or laparoscopically and that continued efforts should be taken to facilitate these approaches. Moreover, surgeons without the training and skills required for the safe performance of vaginal hysterectomy or laparoscopic hysterectomy should request the aid of colleagues who do, or refer patients requiring hysterectomy to such individuals capable of performing the procedure via a minimally invasive route, for their surgical care [7].

Given such emphasis on minimally invasive hysterectomies, gynaecology practice has seen significant changes in trends of modes of hysterectomy. Among approximately 600,000 hysterectomies performed every year in the United States, 63% were performed via laparoscopy in 2012, compared to 30% in 2002 [7]. Symptomatic uterine leiomyomas are the indication for 210,000 hysterectomies and 50,000 myomectomies annually [8]. Thus, it comes as no surprise that 50,000–150,000 patients with large-size specimens undergo power morcellation to maintain a minimally invasive approach [9]. This highlights the importance of the morcellation procedure avoiding extension of small laparoscopic incisions in order to allow completion of minimally invasive procedures and to optimize perioperative outcomes by avoiding exploratory laparotomy and its inherent risks.

10.2.1 Timeline of the Debate

Since the initial use of electronic power morcellation, a variety of articles have been published referencing dissemination of benign and malignant tissue after uterine and fibroid morcellation, suggesting the possibility of worsening prognosis. In 1997, Schneider published the first case report of an undetected adenocarcinoma of the uterus after morcellation [14]. The article stated that malignant cells could have been detected via preoperative curettage, and recommended this diagnostic procedure be performed routinely prior to laparoscopic hysterectomy with planned morcellation. A more recent retrospective study by Siedman et al. in 2012, reviewing 1,091 cases of laparoscopic uterine morcellation, reported that the rate of unexpected sarcomas in women undergoing power morcellation was 0.09%, suggesting a nine times higher risk of spreading undetected cancer than previously suspected [15]. While these outcomes were acknowledged to be concerning, the gynaecologic community accepted them as a rare occurrence, with an overall balance favouring morcellation due to the markedly decreased perioperative morbidity and mortality rates of laparoscopic versus abdominal surgery.

However, the opposition to morcellation increased after the care of Amy Reed, MD, an anaesthesiologist at Beth Israel Hospital, in Boston, who underwent a laparoscopic hysterectomy in October 2013 for presumed benign uterine fibroids utilizing electronic power morcellation. The unfortunate final pathologic report was consistent with leiomyosarcoma. Given that electronic power morcellation was used during this procedure – thus leading to possible peritoneal dissemination of sarcomatous tissue – the possibility of increased staging and potentially worsening prognosis was a strong concern.

Within the same month, the Society of Gynecologic Oncology (SGO) released a position statement asserting that morcellation is generally contraindicated when there exists a high suspicion of malignancy or documented malignancy. The document also stated that less than 1 in 1,000 women undergoing hysterectomy for fibroids have an underlying malignancy. Further, the SGO acknowledged that while there is no reliable method to differentiate benign from malignant myomas before resection, leiomyosarcoma and endometrial stromal sarcoma have an extremely poor prognosis even when specimens are removed intact. They recommended informed consent to allow patients to make an informed decision to accept or decline the option of morcellation [16]. In an additional statement, the SGO added that hysterectomies are done to treat benign uterine fibroids in the majority of cases, and that since intracorporeal morcellation benefits thousands of women, it would be a disservice to deny these women this option [17].

A Safety Communication was issued in April of 2014 by the FDA, discouraging the use of laparoscopic electronic power morcellation during hysterectomy or myomectomy for uterine fibroids, based on their review of literature showing a prevalence of 1 in 352 of unsuspected uterine sarcoma and 1 in 498 of unsuspected leiomyosarcoma [18, 19]. Since then, the concern for dissemination of tumour cells at the time of electronic power morcellation has led many hospitals and hospital systems in the United States to ban the use of electronic power morcellation, thus by necessity leading to the conversion of several cases to a laparotomy. Nevertheless, the debate has remained contentious due to further review of literature, revealing a much smaller risk of sarcoma and the decreased risks of minimally invasive surgery as compared to laparotomy.

As a response to the evidence published in the FDA Safety Communication report, a number of opinion papers by several medical societies have been written recommending the use of electronic power morcellation in the appropriate patients only after proper informed consent. Other medical institutions issued new policies advocating the use of a morcellation bag to attempt containment of the specimen and decrease the risk of spread of pathology into the peritoneal cavity.

In response to the FDA statements and reaction by the medical community, the AAGL assembled a Task Force to evaluate electronic power morcellation for tissue extraction [20], recommending that electronic power morcellation should be avoided in postmenopausal patients, and in the setting of known malignant or pre-malignant conditions. The Task Force further recommended that consideration of electronic power morcellation should be granted for patients with reassuring preoperative evaluation, including cervical cytology and endometrial biopsy. Moreover, alternatives to morcellation, including laparotomy, should be considered in cases of a non-reassuring evaluation. Finally, the importance of informed consent and the further investigation of morcellation within the specimen retrieval bag containment system was stressed. Lastly, the group concluded that since there is no single method of tissue extraction that can protect all patients, all current methods should remain available.

An expert panel convened by the FDA met in July 2014 to discuss the options for morcellation including status quo, black box warning, and a ban on electronic power morcellation; no consensus was reached. During the meeting, and later published in the Journal of Minimally Invasive Gynecology (JMIG) [9], Dr Jubilee Brown stated that there is limited data available on the risk of undetected uterine malignancy at time of hysterectomy with power morcellation, estimates ranging from 1:360 to 1:7,400. Dr Brown cautioned against eliminating a beneficial technology based on scant and imprecise data regarding the risk of undetected uterine malignancy prior to hysterectomy with planned electronic power morcellation, and further emphasized the limited data on the risk of upstaging, or worsening prognosis, of a leiomyosarcoma after power morcellation. Dr Brown further stated that ‘the elimination of power morcellation would result in conversion to open procedures; the risks of power morcellation would have to exceed the benefits of MIS in order for that to be justified’.

Ultimately, the FDA updated their Safety Communication in November 2014, and issued a black box warning against power morcellation stating that the use of laparoscopic electronic power morcellators during fibroid surgery may spread cancer and decrease the long-term survival of patients [21]. The report also established contraindications for laparoscopic electronic power morcellation used for removal of uterine tissue with suspected or known malignancy as well as tissue containing suspected fibroids in peri- or postmenopausal women or women who are candidates for en bloc tissue removal (i.e. vaginal extraction or via mini-laparotomy). It is interesting that criteria for perimenopause were not defined.

Following these recommendations, in a study by Siedhoff et al., a decision tree model was created to evaluate outcomes of a hypothetical cohort of 100,000 premenopausal women undergoing hysterectomy for presumed fibroids. Outcomes included postoperative complications, incidence of leiomyosarcoma, death related to leiomyosarcoma, and procedure-related death. The analysis predicted fewer overall deaths with laparoscopic hysterectomy compared to abdominal hysterectomy (98 vs. 103 per 100,000). While more deaths from leiomyosarcoma followed laparoscopic hysterectomy (86 vs. 71 per 100,000), increased deaths related to the open procedure were predicted with abdominal hysterectomy (32 vs. 12 per 100,000). Morbidity results for the laparoscopic group included lower rates of transfusion (2,400 vs. 4,700 per 100,000), wound infection (1,500 vs. 6,300 per 100,000), venous thromboembolism (690 vs. 840 per 100,000), and incisional hernia (710 vs. 8,800 per 100,000), but a higher rate of vaginal cuff dehiscence (640 vs. 290 per 100,000) [22].

In response to the updated FDA statement, a Leiomyoma Morcellation Review Group expressed disagreement with the methodology used by the FDA to calculate the prevalence of leiomyosarcoma among women with suspected benign fibroids and which led to their severely restrictive recommendations against the use of electronic power morcellators [8]. Based on their review, the FDA estimated that 1 out of 458 women undergoing surgery for presumed fibroids would have an occult leiomyosarcoma. To perform their search, the FDA used the words ‘uterine cancer’ and ‘hysterectomy or myomectomy’. Since the search required to look for ‘uterine cancer’, studies where cancer was not diagnosed or discussed were not identified. Out of nine studies, eight were retrospective, one was a non-peer-reviewed letter to the editor, and another was an abstract from an unpublished study. In addition, three of the leiomyosarcoma cases identified by the FDA did not meet pathologic criteria for cancer, and would not be classified as benign atypical leiomyomas. Excluding the data from the non-peer-reviewed letter and the three cases of atypical leiomyomas, the Leiomyoma Morcellation Review Group identified eight cases of leiomyosarcoma among 12,402 women who underwent surgery for suspected leiomyomas, with a prevalence of 1 in 1,550 (0.064%) [8]. The review group described two additional studies by Pritts and Bojahr who did not limit their methodologies to literature written in English, which reported a prevalence of 1 in 1,960 (0.051%) [23] and 2 in 8,720 (0.023%) [24], respectively.

On the basis of the above review of the literature, the review group asked the FDA to modify its guidance and to promote patient participation in the decision-making process by promoting informed consent that would empower patients to make informed decisions regarding the use of morcellation in their medical care. Lastly, the group suggested the following clinical recommendations:

Interestingly, the authors of this study were disappointed that the review group’s recommendations did not include the use of containment systems. Not surprisingly, following the FDA statement recommending against morcellation in perimenopausal women, and despite the declarations by the Leiomyoma Morcellation Review Group and articles by Siedhoff [22], Pritts [23], and Bojahr [24], the most recent trends in the mode of hysterectomy across the country have been observed to favour a decrease in laparoscopic hysterectomies or myomectomies and an increase in abdominal hysterectomies or myomectomies. A retrospective cohort study by Harris et al. reported their trends in mode of hysterectomy, including abdominal, laparoscopic, and vaginal, as well as postoperative complications from the 15 months preceding the initial FDA safety communication discouraging the use of power morcellation through the 8 months following the FDA statement. A decrease in laparoscopic hysterectomy rates by 4.1% (p = 0.005) was noted, and abdominal and vaginal hysterectomies increased by 1.7% (p = 0.112) and 2.4% (p = 0.012), respectively. A significant rise in major surgical complications excluding blood transfusions was noted, with an overall increment of 2.2–2.8% (p = 0.015) and an increased rate of hospital readmission within 30 days from 3.4% to 4.2% (p = 0.025) [25].

Another retrospective analysis by Barron et al. demonstrated a significant decrease in the proportion of minimally invasive hysterectomies and myomectomies following the initial FDA statement. Proportions of minimally invasive surgery cases including laparoscopic, robotic-assisted, and vaginal cases were compared between an 8-month period before and after the FDA statement. The study found an overall downtrend in minimally invasive hysterectomies after the FDA warning statement, from 85.7% before to 79.9% (1,451/1,694 vs. 1,350/1,690; p = 0.001) after the statement. When oncology cases were excluded, a downtrend was again observed, from 90.2% before to 81.5% (985/1,092 vs. 834/1,023; p = 0.001) after the statement. A similar decline was seen in myomectomies, with 62.7% procedures prior to the FDA statement versus 43.7% (64/102 vs. 38/87; p = 0.009) in the 8 months afterwards [26].

Lastly, a more recent retrospective review by Wright et al. in 2016 evaluated trends of abdominal, minimally invasive, and vaginal hysterectomies before and after the FDA advisory [27]. Between the last quarter of 2013 and the first quarter of 2015, the study showed an increase of abdominal hysterectomy rates from 27.1% to 31.8% (p = 0.004), and a decline in minimally invasive hysterectomy rates from 59.7% to 56.2% (p < 0.001), as well as a declining trend in vaginal hysterectomy. Additionally, a marked decline was noted in electronic power morcellation use within the minimally invasive group; a remarkable downtrend from 13.5% in the first quarter of 2013 to 2.8% in the first quarter of 2015 (p < 0.001). The study once again showed the markedly lower risk of complications in the minimally invasive surgery group.

The studies above illustrate the increasing trends of abdominal procedures versus laparoscopic procedures, despite an increase in complication rates. Furthermore, these results underscore the need for further research to investigate methods to improve data collection on postoperative diagnosis of occult uterine malignancy, the diagnosis of pathology pre-operatively, and safe ways to perform morcellation in an effort to improve the safety of methods for specimen retrieval.