Abstract
Recurrent pregnancy loss (RPL) is defined as two or more failed clinical pregnancies before 20 weeks' gestation and may be caused by genetic, endocrinologic, anatomic and immunologic abnormalities. Anatomic uterine anomalies include congenital malformations (bicornuate, didelphic, septate and unicornuate uteri) and acquired defects (fibroids, adenomas, adhesions and polyps). Women with septate and bicornuate uteri, intrauterine adhesions, and some adenomas and fibroids are at increased risk of RPL. Data support surgical treatment of all of these lesions except bicornuate uteri. The role of polyps in RPL is unclear. Minimally invasive options for surgical correction of intrauterine lesions include hysteroscopy, laparoscopy with and without robotic assistance and minilaparotomy.
Keywords
Overview of recurrent pregnancy loss causes & cures
Recurrent pregnancy loss (RPL) is a devastating problem for couples desiring parenthood, and it is often a frustrating and confusing clinical challenge for their treating physician. RPL occurs in 2–4% of all reproductive aged couples [1]. The traditional definition of RPL included only couples with three or more consecutive miscarriages. The recent Committee Opinion released by the American Society for Reproductive Medicine classifies RPL as “a distinct disorder defined by two or more failed clinical pregnancies” [2]. This change in definition reflects several studies that have indicated that the risk of miscarriage after two successive losses is similar to the risk of miscarriage in a woman after three or four or more successive losses [3].
A miscarriage is any pregnancy loss before 20 gestational weeks. Miscarriages can be further divided into embryonic losses, which is the demise of a pregnancy before 10 weeks' gestation, and fetal losses, defined as the demise of a pregnancy after 10 weeks' gestation. Couples diagnosed with primary recurrent miscarriage have never had a pregnancy progress beyond 20 weeks' gestation, and those with secondary recurrent miscarriage have had at least one pregnancy progress beyond 20 weeks' gestation followed by two or more consecutive miscarriages. Couples with secondary RPL have identifiable problems just as frequently as those with primary RPL [3]. Therefore, both types of loss should be offered a thorough evaluation.
In this field, as in many fields of medicine, large prospective randomized trials are lacking. The opinions of clinical experts are in most cases the best available evidence. A complete evaluation of RPL includes chromosomal testing on both parents; a careful evaluation of the uterine cavity for congenital and acquired anomalies using pelvic sonohysterography, hysterosalpingography or hysteroscopy; immunological tests for lupus anticoagulant, anticardiolipin antibodies and beta-2 glycoprotein 1 antibodies; and an endocrine profile including blood tests for thyroid stimulating hormone, hemoglobin A1c, prolactin and midluteal progesterone (Table 1). A complete evaluation will result in the identification of a probable cause in 40–60% of patients [1,3]. The most common causes of RPL include both autoimmune factors and anatomical abnormalities [3].
Diagnosis and management of recurrent pregnancy loss.
PGD: Preimplantation genetic diagnosis; POC: Products of conception; TSH: Thyroid stimulating hormone
For those couples with genetic abnormalities, most frequently balanced translocations, genetic counseling is the standard of care. Some programs have recommended preimplantation genetic diagnosis after
Anatomical causes of RPLs
Uterine anomalies occur in about 19% of women with at least two or more consecutive miscarriages [3]. These include congenital malformations (most commonly bicornuate, didelphic, septate and unicornuate uteri) as well as acquired defects (fibroids, adenomas, adhesions and polyps). Congenital anomalies may arise from abnormalities in the development of the Müllerian ducts, the embryologic precursors to the uterus, such as defects in the formation of a duct (unicornuate uterus), the fusion of the ducts (didelphic and bicornuate uteri) or the regression of the tissue between the fused ducts (complete and partial septate uteri) [5]. Among women with RPL, septate uteri are the most prevalent of the congenital anomalies (Table 2). Furthermore, a septate uterus significantly increases the rate of miscarriage in women with RPL (86%) compared with that among women with unexplained RPL (72%) [6]. Bicornuate uteri are associated with a similar increase in miscarriage rate among women with RPL (86%), although bicornuate anomalies are much less common.
Prevalence of uterine anomalies among 875 patients with recurrent pregnancy loss.
Five patients (0.6%) had both congenital and acquired anomalies. These were septum and adhesions, septum and fibroid(s), septum and polyp(s), bicornuate uterus and fibroid(s), and unicornuate uterus and polyp(s). Three patients had a T-shaped uterus.
Three patients (0.3%) had two acquired anomalies. These were fibroid(s) and polyp(s), adhesions and polyp(s), and adhesions and fibroid(s).
Data taken from [3].
Diagnosis of congenital uterine anomalies is most accurately done with MRI or with three-dimensional ultrasound. Sonohysterogram can show the contour of the uterine cavity but would not differentiate between septate and arcuate or bicornuate uteri unless the fundus was clearly viewed. Three-dimensional rendering can enhance the diagnostic accuracy of sonohysterogram to the level of MRI and is often the first diagnostic tool that is used [7]. The primary treatment options for women with congenital uterine anomalies are surgical repairs (Table 3). Although randomized controlled trials are lacking, septum resection has been reported to substantially reduce the rate of miscarriage in women with RPL [8]. One review found that miscarriage rates of 86% among women with a septate uterus decreased to 16% following hysteroscopic metroplasty [9]. There is no strong evidence that surgical repair reduces the miscarriage rate for women with RPL who have bicornuate uterine defects [8,10].
Minimally invasive surgical management of congenital and acquired anatomical uterine anomalies.
Adenomas may be treated the same as fibroids in this table
The above are general recommendations; specific patient characteristics must also be taken into account as part of appropriate surgical decision making
MIS: Minimally invasive surgical.
In contrast to congenital uterine defects, acquired uterine anomalies develop in response to hormonal or physical stimuli experienced after puberty; and in women with RPL, they are almost twice as prevalent as congenital anomalies (Table 2) [8]. Sonohysterogram is the best way to diagnose acquired anomalies as it allows for a full gynecologic ultrasound plus detailed intracavitary assessment of each wall of the endometrial lining independently. Intrauterine adhesions typically form after endometrial trauma, often as a result of curettage following spontaneous abortions, which may exacerbate the problem in women with RPL [8]. In one review, women with adhesions experienced a high rate of miscarriages (40%) compared with women who had surgical adhesiolysis (25%) [11]. Although surgical removal of adhesions is usually recommended for many women with RPL for whom no other causes are known, it is extremely important that the procedure itself does not create endometrial trauma that might trigger the development of further adhesions.
Fibroids and polyps are hormonally induced growths of either the myometrium or the endometrium, respectively. Although systematic reviews have reported increased rates of miscarriage in women with fibroids, research is lacking regarding the role of polyps in pregnancy loss [8]. There are multiple techniques that have been proposed to remove or destroy fibroids in women with RPL, including a variety of surgical methods. Two of three studies utilizing laparotomy for removal of leiomyomata reported significant declines in miscarriage rates following the procedure among women with RPL [8]. Generally, fibroids are more likely to contribute to RPL if they distort the endometrial cavity and/or are >6 cm [12]. Adenomas are overgrowths of the glandular endometrium within the muscular myometrium; if large, they can cause the same distortion of the endometrial cavity seen with fibroids and thus possibly contribute to RPL. For polyps, surgical removal is often considered for women with RPL if no other causes for pregnancy loss have been found.
MIS options for treatment of RPL
Retrospective studies provide some evidence that surgical management of certain congenital or acquired uterine defects decreases the overall rate of miscarriage, but the consensus given lack of randomized trials is that surgical correction should be considered in patients with RPL [2]. The vast majority of congenital and acquired uterine anomalies are amenable to correction by MIS options including hysteroscopic and laparoscopic resection (Table 3). Intrauterine defects such as polyps, septa, adhesions, and some fibroids and adenomas may be accessed via the cervical canal; so operative hysteroscopy has presented a way to simultaneously diagnose and treat these lesions without an epidermal incision. Laparoscopy, both with and without robotic assistance, provides a way to access and remove multiple and/or large fibroids and adenomas with as few as one incision [13]. These modalities allow gynecologic surgeons to correct defects in an outpatient surgical setting with all of its inherent benefits including rapid return to normal activity for the patient.
Hysteroscopy is the preferred modality for surgical treatment of intrauterine polyps, septa and adhesive disease (also known as either synechiae or Asherman's syndrome). Submucosal fibroids, those within the endometrial cavity, can be classified as type 0 (entirely in the uterine cavity), I (>50% in the cavity) or II (<50% in the cavity) based on their sonographic or MRI appearance [14]. Type 0 and I fibroids are those most amenable to hysteroscopic resection.
Hysteroscopic resection of a septum or of adhesions is short surgical procedure that is performed in most cases on an outpatient basis with rapid return to normal activity by the patient. The procedure has low associated morbidity and has been shown to significantly improve reproductive outcomes as outlined above [15]. It should be performed during the follicular phase to ensure that the patient is not pregnant and since a luteal endometrium could interfere with optimal visualization. In most instances, incision of the septum to the point of petechial hemorrhage suffices as opposed to removal since the lesion may retract after incision. Low pressure hysteroscopy can be used to aid in visualizing a septum to ascertain if resection is complete. This septotomy may be accomplished using laser, microscissors or electrocautery with excellent results noted with all instruments. The same instruments may be used for resection of intrauterine synechia and polyps.
Minimally invasive surgery is being used more frequently in the treatment of fibroids over the past decade. Laparoscopic removal is preferable for fibroids that would be difficult to extract hysteroscopically due to their size and/or position within the myometrium (i.e., intramural and subserosal fibroids). Several prospective randomized studies describe the advantages and disadvantages of laparoscopy compared with typical laparotomy for myomectomy in women of reproductive age. One reports that size, location and number of fibroids are not limiting factors in the hands of experienced surgeons [16]. The benefits include a shorter postoperative stay, better reproductive outcomes, decreased blood loss and postoperative complications including pain, and improved cosmesis compared with laparotomy [17,18]. Regardless of which method for myomectomy is selected, the removal of tissue from the muscular myometrium is required. This procedure is typically performed in women desiring future pregnancy, therefore a myomectomy requires meticulous tissue dissection and precise reconstruction to preserve as much muscle as possible and protect against uterine rupture during pregnancy. Many fibroids are amenable to laparoscopic myomectomy, with or without robotic assistance, in the hands of an experienced surgeon. Large retrospective studies following conventional and robot-assisted laparoscopic myomectomy (RALM) show a uterine rupture rate comparable to that of open myomectomy [19–21]. Alternative approaches to laparoscopic myomectomy including MRI-guided focused ultrasound ablation, uterine artery embolization [22], hysterectomy in extreme cases and expectant management are not recommended or possible in women planning to use their uterus for future childbearing.
The advanced surgical skills needed to properly perform a myomectomy have resulted in low adoption of the minimally invasive version of this common operation [23]. Robotic assistance, as an enabling tool, provides an MIS alternative with a higher adoption potential [24]. RALM, first described in 2004 [25], has rapidly increased in popularity. One study has compared RALM with open myomectomy and found decreased estimated blood loss, shorter hospital stays and fewer complications with RALM [26]. Retrospective studies have shown that the clinical results of conventional (without robotic assistance) and RALM are comparable [27–30]. The robot provides several advantages such as three-dimensional viewing of the surgical field, tremor reduction and improved surgeon ergonomics. Opponents of robotic surgery quote cost as a limiting factor. One study showed that in uterine cancer cases laparoscopy is less expensive than robotic-assisted laparoscopy but only when capital acquisition costs are included [31]. The same study showed cost neutralization with the robot since it decreases the rate of open hysterectomy cases [32].
Risks of laparoscopic myomectomy with or without robotic assistance include those inherent in all surgeries plus, most pertinent in the RPL population, postoperative adhesion formation in the abdomen and pelvis including the uterus, incomplete removal or recurrence of fibroids, and uterine rupture prior to or during labor. The rate of complications is similar for conventional and RALM [29]. Patients should be made aware of the possible future need for a scheduled cesarean section in order to decrease the risk for uterine rupture in subsequent pregnancies depending on the size, number and depth of the fibroids to be removed. Most RALMs are outpatient procedures, and the patients go home 3–5 h after their surgery is completed.
Minilaparotomy has gained popularity as a minimally invasive alternative not only to traditional open abdominal myomectomy but also to laparoscopic myomectomy [33,34]. In the hands of well-trained surgeons, minilaparotomy can be performed safely and effectively as a same-day procedure. Using appropriate patient selection criteria (body mass index less than 30 kg/m2, less than five fibroids total, largest fibroid <7 cm), our surgical unit averages an operating time of less than 100 min, blood loss of approximately 125 cc and total outpatient stay of 3.5 h [33]. Generally, a Pfannenstiel incision of 3–4 cm in length is made approximately 2–3 cm above the pubic symphysis. Skin and subcutaneous tissues are opened horizontally, and the fascia and parietal peritoneum are opened vertically. During the procedure, packing of bowel and retractors other than sponges are avoided. A dilute solution of Pitressin is injected under the uterine serosal surface before each fibroid is excised; blood vessels are coagulated as they are encountered. The fibroid beds are sutured with interrupted 3–0 Vicryl sutures, and the serosa is closed with 4–0 Vicryl in a baseball stitch. After hemostasis is obtained, the pelvis is irrigated with copious amounts of normal saline. The fascia is closed with 0 Vicryl, and the skin is closed with subcuticular stitches of 4–0 Vicryl. Steristrips and a large bandage are placed over the incision, and a vertical pressure dressing is applied until the patient is discharged home. No patients in our practice have required additional pain medications, blood transfusions or overnight hospital admissions. All patients have been fully ambulatory by their postoperative visit at 1 week. No special equipment is required other that standard laparotomy instrument sets that are available in any surgical suite.
Conclusion
Anatomic abnormalities, both acquired and congenital, account for about 20% of the explainable causes of RPL. Minimally invasive surgery is suitable for correction of the majority of these abnormalities. In general, pregnancy rates are significantly improved after surgical correction
Future perspective
As RPL affects a large number of couples, it will remain an important research focus. Specifically, investigations into the utility of preimplantation genetic diagnosis for prevention of miscarriage will continue and likely reveal a subset of patients who will benefit from this technology. Additionally, further improvements in minimally invasive surgery such as new laparoscopic and hysteroscopic instrumentation will evolve and allow surgeons to provide even more precise operative treatment for uterine anomalies contributing to RPL.
Executive summary
Recurrent pregnancy loss (RPL) is defined as two or more failed clinical pregnancies before 20 weeks' gestation.
RPL may be caused by genetic, endocrinologic, anatomic and immunologic abnormalities with the last two being most common.
Anatomic uterine anomalies include congenital malformations (bicornuate, didelphic, septate and unicornuate uteri) as well as acquired defects (fibroids, adenomas, adhesions and polyps).
Women with septate and bicornuate uteri, intrauterine adhesions, and some adenomas and fibroids are at increased risk of RPL.
Data support surgical treatment of all of these lesions except bicornuate uteri.
The role of polyps in RPL is unclear.
Minimally invasive options for surgical correction of intrauterine lesions include hysteroscopy (for septa, polyps, adhesions, and some fibroids and adenomas), laparoscopy (for some fibroids and adenomas) with and without robotic assistance, and minilaparotomy (for some fibroids and adenomas).
Investigations into the utility of preimplantation genetic diagnosis for prevention of miscarriage will continue and likely reveal a subset of patients who will benefit from this technology.
Financial & competing interests disclosure
AP Bailey discloses a consulting relationship with Omni-Guide, Inc. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.