Pelvic masses during pregnancy have historically posed diagnostic and therapeutic dilemmas for the obstetrician. Maternal mortality rates with surgery were prohibitive through the early 20th century. For example, among 720 pregnant women with surgically treated adnexal masses reviewed by McKerron (1906), the maternal mortality rate was 21 percent, and the fetal mortality rate was 50 percent.

These high mortality rates with surgical treatment led to further decades during which expectant management was selected. However, conservative care was also associated with excessive mortality rates. Patton (1906) reported a 26-percent maternal death rate with expectant care. Another early study of adnexal masses in pregnancy summarized the many complications with conservative management. These included torsion (33 percent), cyst rupture (5 percent), suppuration (14 percent), and significant dystocia leading to cesarean delivery (16 percent) (Spencer, 1920). Caverly (1931) reported a 30-percent spontaneous abortion rate in cases expectantly managed.

From these experiences, treatment evolved to that of expectant observation until the second trimester, after which time, any mass that persisted was excised surgically. This regimen gained popular acceptance not only because of the high maternal complication rate if the mass remained, but also because of the 2- to 8-percent risk of ovarian malignancy. Following the introduction of modern anesthetic techniques, antibiotics, and blood banking, maternal and fetal morbidity and mortality rates declined. Yet, despite these improvements, management of adnexal masses during pregnancy is still challenging.

The reported incidence of adnexal masses that are discovered during pregnancy varies considerably. Prior to the introduction of sonography, Grimes and colleagues (1954) reported an incidence of 1.2 percent of pregnancies from their private practice. As expected, with the advent of sonographic examination of nearly all pregnancies, the reported incidence is appreciably higher and ranges from 5 to 25 percent in the first trimester (Condous, 2004; Yazbek, 2007). In the second trimester, in a study of more than 24,000 women, an incidence of 4.9 percent was reported (Goh, 2013). The proportion of adnexal masses in pregnancy that are malignant is also variably reported and ranges from 2 to 8 percent. Most of these are tumors of low malignant potential (Leiserowitz, 2006; Ngu, 2014).

The differential diagnosis of an adnexal mass in pregnancy depends on the stage of pregnancy at the time of diagnosis and the sonographic appearance of the mass (Table 14-1). Many women during a routine, first-trimester sonographic examination will have an incidental finding of a small, cystic adnexal mass measuring less than 3 cm. These typically represent functional corpus luteum cysts. In approximately 5 percent of scans, such simple sonographic masses will measure 3 cm or greater (Glanc, 2007; Yazbek, 2007). Most of these larger cysts are also functional, and only 0.7 to 1.7 percent persist beyond the first trimester (Condous, 2004; Yazbek, 2007). Simple cysts that persist into the second trimester are less likely to be functional cysts, and etiologies include ovarian cystadenomas, hydrosalpinx, paraovarian or paratubal cysts, endometriomas, or thecalutein cysts.

Of complex adnexal masses, the most common etiology in pregnancy is benign mature cystic teratoma, colloquially called a dermoid cyst (Yacobozzi, 2012). Other complex masses may be endometriomas; benign cystadenomas, especially mucinous cystadenomas; or malignant tumors. Solid masses are less frequent but may represent ectopic pregnancy, pedunculated leiomyomas, or ovarian fibromas. Malignant adnexal masses can be simple but more often appear complex or solid.

There are also nonneoplastic adnexal masses to consider. For example, a chronic tuboovarian complex or an ectopic pregnancy may appear as a complex mass, even if an intrauterine pregnancy is also seen. Although simultaneous intrauterine and ectopic pregnancies are rare, a clinician should consider the possibility, particularly in a woman who has undergone assisted reproductive technology (ART) treatments. Of 553,577 pregnancies reported to the National ART Surveillance System for the decade 2001 to 2011, Perkins and associates (2015) found the heterotopic pregnancy rate approximated 0.09 percent in pregnancies resulting from ART. Bello and colleagues (1986) reported the incidence of combined ectopic and intrauterine pregnancies as 1 in 3899 to 15,000 pregnancies.

Torsion

Incidence and Etiology

A major problem with adnexal masses in pregnancy is torsion of adnexal components. Most often, the ovary and fallopian tube rotate as a single entity (Figs. 14-1 and 14-2). Infrequently, an ovary may alone turn about its mesovarium, and rarely a fallopian tube twists alone about the mesosalpinx. The reported incidence of adnexal torsion during pregnancy is 1 to 5 cases per 10,000 spontaneous pregnancies. It may be as high as 16 percent in pregnancies conceived using ART and complicated by ovarian hyperstimulation (Mashiach, 1990). This increased rate may be explained by the greater tendency during ART treatment to develop large, multicystic ovaries that result from the artificial hormonal stimulation.

In general, factors associated with a higher rate of torsion include masses that are 6 to 10 cm in diameter and a gestational age of 10 to 17 weeks. Indeed, 37 percent of torsion cases present between 15 and 16 weeks’ gestation. Mechanistically, displacement of the adnexa by the rapidly growing uterus is one explanation for this timing (Koo, 2011; Yen, 2009). Of 33 pregnant women with adnexal torsion in one study, three presented with symptoms 3 weeks postpartum, suggesting that torsion may also be associated with uterine involution (Yen, 2009).

Adnexa receive blood supply from adnexal branches of both the uterine and ovarian vessels. Of these, torsion usually involves the uterine branches, which travel with and then twist with the uteroovarian ligament and proximal fallopian tube. During torsion, low-pressure veins draining an adnexum are compressed early by the twisting pedicle, but high-pressure arteries initially resist compression. As a result of this continued inflow but arrested egress of blood, the adnexum becomes congested and edematous but does not infarct. Because of this, cases of early torsion can often be conservatively managed at the time of surgery. With continued stromal swelling, however, arteries can become compressed, leading to infarction and necrosis that necessitate adnexectomy. Grossly, twisted adnexa are enlarged and often appear hemorrhagic.

Diagnosis

Symptoms may include acute-onset pain that is constant or intermittent, nausea and vomiting, and signs of peritoneal irritation (Hasson, 2010). The pain usually is localized to the involved side, with radiation to the flank, groin, or thigh. Low-grade fever suggests adnexal necrosis.

Sonography plays an essential role, and specific findings have been described. First, multiple follicles rimming an enlarged ovary reflect ovarian congestion and edema described in the last section. The twisted pedicle may appear as a bull’s-eye target, whirlpool, or snail shell, that is, a rounded hyperechoic structure with multiple, inner, concentric hypoechoic rings. In affected women, transvaginal color Doppler sonography (TV-CDS) may show disruption of normal adnexal blood flow. In some cases, however, incomplete or intermittent torsion may variably display both venous and arterial flow using TV-CDS. Thus, disruption of vascular flow, when present, is highly suggestive of torsion. But torsion cannot be excluded on the basis of a normal Doppler study alone, especially with clinically suggestive signs and symptoms. Computed tomography (CT) or magnetic resonance (MR) imaging is usually not required. These may be helpful in complicated cases or in those with ambiguous clinical presentation, such as that seen with incomplete or chronic torsion.

Treatment

Salvage of the involved adnexa, resection of any associated cyst or tumor, and possible oophoropexy are treatment goals. However, adnexal necrosis or rupture with hemorrhage may mandate adnexal removal.

Torsion may be evaluated by laparoscopy or laparotomy. Previously, to avoid possible thrombus release and subsequent embolism, adnexal untwisting was eschewed, and adnexectomy was the standard. However, evidence does not support this. McGovern and coworkers (1999) reviewed nearly 1000 cases of torsion and found that pulmonary embolism was rare and occurred in only 0.2 percent. Notably, these cases of embolism were associated with adnexal excision, and none were linked to untwisting of the pedicle. In a study of 94 women with adnexal torsion, Zweizig and associates (1993) reported no increased morbidity in women undergoing untwisting of the adnexum compared with those undergoing adnexectomy.

For these reasons, detorsion of the adnexum is generally recommended. Within minutes following untwisting, congestion is relieved, and ovarian volume and cyanosis typically diminish. For many, absence of these changes may prompt adnexal removal. A persistently engorged, black-blue ovary with foci of hemorrhage, however, is not pathognomonic for necrosis, and the ovary may still recover. In one study of 102 cases of ovarian torsion in nonpregnant women, patients underwent detorsion without oophorectomy, regardless of ischemic appearance of the adnexa. Surgery included detorsion alone (35 percent), ovarian cyst aspiration (33 percent), cystectomy (30 percent), and oophoropexy for repeat torsion (1 percent). No cases were complicated by thromboembolism postoperatively. Postoperative sonography done 8 to 10 weeks later showed ovaries with normal size and normal follicular development in more than 90 percent of patients. Almost 14 percent of the entire cohort underwent subsequent surgery for unrelated issues, and all but one patient had normal-appearing adnexa. Six women later underwent in vitro fertilization (IVF) with oocytes retrieved from the previously ischemic ovaries, and all were fertilized (Oelsner, 2003). Despite these encouraging findings, if conservative management of a dusky adnexum is elected, vigilance for fever, leukocytosis, and peritoneal signs is required to exclude later postoperative adnexal necrosis. For the gravida, this potential complication may make conservative management of the necrotic-appearing adnexum less advantageous.

Following detorsion, there is no consensus as to the management of the reperfused adnexum. Ovarian cysts or masses are ideally excised to prevent repeat torsion and to exclude cancer in suspect masses. Cystectomy in a hemorrhagic, edematous ovary, however, may technically be difficult. Surgical steps of cystectomy and adnexectomy are illustrated on page 232. Importantly, a preserved adnexum can twist again at a later time, and this is especially true for adnexa with long uteroovarian ligaments. To minimize this risk, one intraoperative technique suitable for pregnancy can be considered. With it, a running stitch is placed through the length of the ligament and then tied to create an accordion-like shortening of the ligament.

Malignancy

Incidence and Etiology

Ovarian malignancy is certainly one of the worst complications of an adnexal mass. The incidence of ovarian malignancy during pregnancy ranges from 1 in 20,000 to 1 in 50,000 births (Palmer, 2009; Smith, 2003). Among women with adnexal masses in pregnancy, the reported rate of cancer varies between 2 and 8 percent (Leiserowitz, 2006; Ngu, 2014). Masses greater than 10 to 15 cm in diameter are associated with a higher rate of malignancy (Koo, 2011; Yen, 2009). In one study, an increase in mass diameter greater than 0.35 cm/wk was associated with a higher malignancy rate (8.3 percent) compared with the cancer rate (0.88 percent) in slower growing tumors (Yen, 2009).

In a study of more than 4.8 million obstetric patients in California, 9375 were diagnosed with an adnexal mass. Of these, 87 women had ovarian cancer, and another 115 gravidas had low-malignant-potential tumors. More than 80 percent of cases were International Federation of Gynecology and Obstetrics (FIGO) stage I, and the overall mortality rate due to ovarian cancer was 4.7 percent. The authors attributed the low mortality rate to a higher rate of germ cell tumors—39 percent, younger patients—30 years, and early-stage disease compared with nonpregnant women in the California cancer registry database (Leiserowitz, 2006).

Diagnosis

Commonly, pelvic pain, constipation, and back pain may be earlier signs of ovarian cancer. Unfortunately, these often mirror pregnancy symptoms and can delay diagnosis. Moreover, many ovarian masses are asymptomatic and only detected during routine prenatal sonographic examination.

Sonography is the preferred initial evaluation if an adnexal mass is suspected. The typical imaged appearance of ovarian cancer varies but is usually complex. Peritoneal implants with more advanced-stage tumors are rarely detected sonographically. In cases with more complicated anatomy, MR imaging may add supplemental information. With suspected cancer, tumor markers are obtained as outlined on page 230.

Management

If malignancy is strongly suspected, preoperative consultation with a gynecologic oncologist is prudent. During surgery for a highly suspicious adnexal mass, pelvic washings are obtained for cytologic analysis once the abdomen is entered. If frozen-section histopathologic analysis of the mass verifies malignancy, cancer staging begins with careful inspection of all accessible peritoneal and visceral surfaces (Giuntoli, 2006; Yazigi, 1988). Biopsies are taken from the diaphragmatic surface and peritoneum; omentectomy is done; and pelvic and infrarenal paraaortic lymph nodes are sampled. Depending on uterine size, some of these components, especially lymphadenectomy, may not be technically feasible. If there is advanced disease, bilateral adnexectomy and omentectomy will decrease most tumor burden. In early pregnancy, hysterectomy and aggressive surgical debulking procedures may be elected.

Cyst Rupture or Obstructed Labor

Rupture of an adnexal cyst is uncommon and is reported to occur in <1 percent of cases (Naqvi, 2015). Symptoms are similar to those with torsion and may include acute abdominal pain associated with peritoneal signs. If the cyst is hemorrhagic, there may be associated hypotension due to hemoperitoneum. Again, sonography can aid diagnosis. Hemoperitoneum is seen as anechoic or complex fluid in the cul-de-sac. With advancing volumes, fluid extends up the paracolic gutters to Morison pouch (Fig. 14-3).

Large adnexal masses located in the anterior or posterior cul-de-sacs can potentially obstruct labor. In these cases, they act similarly to a leiomyoma in the lower uterine segment (Goh, 2013).

Sonography

Several diagnostic modalities have been evaluated to help differentiate between benign and malignant adnexal masses and to predict the likelihood of emergent surgical intervention during pregnancy. These include sonography and Doppler interrogation, CT, and MR imaging, as well as the use of serum tumor markers.

With sonography, several investigators have assessed characteristics of adnexal masses during pregnancy to predict malignancy (Bromley, 1997; Chiang, 2004; Yacobozzi, 2012; Zanetta, 2003). Simple cysts usually have thin walls and are anechoic. If there is acute or chronic hemorrhage, then the cyst may be echogenic or show a reticular pattern (Fig. 14-4). A large fluid-filled or multilocular cyst that persists in the second trimester is more likely to be a serous or mucinous cystadenoma than a functional cyst (Fig. 14-5). These tumors sometimes have thin septations.

Complex adnexal masses that are encountered during pregnancy vary sonographically. Those most commonly seen are benign mature cystic teratomas, which are bilateral in approximately 10 percent of cases. Their characteristic sonographic appearance includes hyperechoic mural nodules, acoustic shadowing, and a fat-fluid interface (Fig. 14-6). Additionally, hyperechoic lines or dots represent hair oriented lengthwise or on end, respectively. Endometriomas are seen in 3 to 10 percent of pregnant women (Fig. 14-7). These tumors are characterized by diffuse homogenous low-level internal echoes, and septations may be present. In some cases, their appearance mimics a hemorrhagic cyst.

Solid tumors are less common. Leiomyomas are the most common solid adnexal masses in pregnancy. As shown in Figure 14-8, they appear similar to an ovarian fibroma, that is, hypoechoic, heterogeneous, and solid with acoustic shadowing.

Theca-lutein cysts are typically bilateral and associated with increased human chorionic gonadotropin (hCG) production, which is often seen with multifetal gestation or gestational trophoblastic disease. Their characteristic appearance includes enlarged ovaries with multiple anechoic cysts (Fig. 14-9). These cysts usually resolve spontaneously after delivery as hCG levels dissipate.