INTRODUCTION — The term retained products of conception (RPOC) refers to placental and/or fetal tissue that remains in the uterus after a spontaneous pregnancy loss (miscarriage), planned pregnancy termination, or preterm/term delivery. The presence of RPOC after a spontaneous pregnancy loss distinguishes an incomplete from a complete miscarriage.
This topic will review the evaluation and management of patients who present with bleeding and/or signs of infection after uterine evacuation in the first half of pregnancy (eg, miscarriage, termination). Management of RPOC after preterm/term delivery and the initial management of an incomplete miscarriage are discussed separately. (See "Overview of postpartum hemorrhage" and "Postpartum endometritis" and "Pregnancy loss (miscarriage): Clinical presentations, diagnosis, and initial evaluation".)
INCIDENCE — The reported incidence of RPOC varies widely and depends on several factors, including initial treatment (higher with nonsurgical versus surgical management), criteria for diagnosis (signs/symptoms versus laboratory/pathology/imaging results), and duration of follow-up (RPOC often resorb or are passed) [1-17].
CLINICAL MANIFESTATIONS — The characteristic clinical manifestations of RPOC include one or more of the following: uterine bleeding, pelvic pain, fever, and/or uterine tenderness. These clinical findings are nonspecific; moreover, it is normal to have some postabortal bleeding and discomfort.
Uterine bleeding — Patients routinely have some uterine bleeding after a miscarriage or pregnancy termination, so it can be difficult to distinguish normal from abnormal bleeding. A reasonable approach is to assume that bleeding is probably abnormal if it is heavy (ie, has the potential to result in anemia [passage of large clots or flow that is significantly greater than menses, or not diminishing over time]) or prolonged (ie, lasting longer than three weeks).
Prospective studies of patients with miscarriage report an average of 8 to 11 bleeding days after identification of the miscarriage; however, the bleeding often lasted two weeks or more [1,3]. Following medical treatment of miscarriage, the average number of bleeding days was 9 to 12, but many patients bled for more than 14 days [1,18]. The duration of vaginal bleeding was significantly shorter after surgical evacuation of the uterus. These studies generally included an intervention if symptoms continued beyond 7 to 21 days; few allowed the bleeding to continue for more than three to six weeks.
Fever and/or pain — Necrotic RPOC are prone to infection by cervicovaginal flora. Clinical manifestations include fever, abnormal uterine bleeding, crampy lower abdominal pain, and uterine tenderness.
Fever is never normal following evacuation of the uterus, but cramping for a few hours or days is common. Most patients will obtain adequate pain relief with a short course of analgesics, such as nonsteroidal anti-inflammatory drugs. Patients with poorly controlled pain or pain that does not resolve over a few days or is increasing may have RPOC. If the patient has been instrumented, the possibility of uterine perforation with visceral injury should be excluded. (See "Uterine perforation during gynecologic procedures".)
Clostridial toxic shock syndrome has been reported following first trimester miscarriage and medical abortion, and almost all of the cases were fatal. These patients generally present with abdominal pain, chills, tachycardia, and profound leukocytosis, but they are often afebrile [19,20]. (See "Clostridial myonecrosis" and "First-trimester pregnancy termination: Medication abortion".)
Amenorrhea — Menses typically resume by six weeks post miscarriage or pregnancy termination. If the patient fails to menstruate by this time, viable trophoblastic tissue may be present.
Bleeding — The differential diagnosis of uterine bleeding after uterine evacuation includes:
●RPOC (including placenta accreta spectrum)
●Ectopic pregnancy (see "Ectopic pregnancy: Clinical manifestations and diagnosis")
●Trauma (cervical or vaginal laceration, uterine arteriovascular malformation, uterine perforation)
●Gestational trophoblastic disease
Infection — Fever and pain after uterine evacuation may be related to:
●Endometritis (see "Postpartum endometritis")
●Pelvic inflammatory disease (see "Pelvic inflammatory disease: Clinical manifestations and diagnosis")
●Other abdominopelvic infections, possibly related to uterine perforation if the uterus has been instrumented
DIAGNOSTIC EVALUATION — Routine evaluation for RPOC after all cases of miscarriage or pregnancy termination is likely to result in false positive diagnoses and unnecessary interventions since RPOC do not always lead to morbidity. On the other hand, evaluation is indicated in patients whose symptoms fall outside the normal range, such as those with bleeding that is heavy or prolonged, and those with fever, uterine tenderness, or significant abdominopelvic pain. Combined clinical, laboratory, and sonographic evaluation of these patients can help direct management .
The goal of the evaluation is to determine whether RPOC are the probable source of the patient's symptoms, or whether another diagnosis is more likely (see 'Differential diagnosis' above). If RPOC are thought to be causing the symptoms, treatment with surgical evacuation or medication is warranted. In patients who have undergone surgical intervention, the possibility of a procedure-related complication needs to be considered (cervical or vaginal laceration, uterine perforation, traumatic arteriovenous malformation). (See 'Surgical intervention' below.)
Since RPOC and other postabortal complications can be life-threatening, prompt assessment for severe hemorrhage or sepsis is crucial.
History — The patient should be asked details of the antecedent pregnancy, including the gestational age of the pregnancy (determined by last menstrual period, timing of fertility treatment, or prior ultrasound examination), timing of tissue passage or uterine evacuation, and whether any tissue had been examined by the patient's clinician or a pathologist. If an intrauterine pregnancy has not been confirmed by prior ultrasound or examination of tissue, then ectopic pregnancy must be excluded as part of the evaluation.
The amount and duration of bleeding should be determined. While it is difficult to quantify vaginal bleeding, the passage of large blood clots (larger than a plum or fist), bleeding that saturates a full-size sanitary pad in an hour, syncope, or orthostasis suggest excessive bleeding.
The intensity, location, and pattern of pain should be determined. Pain that is not controlled with oral analgesia or is becoming worse over time suggests a pathologic process.
Symptoms of uterine infection include fever, lower abdominal or pelvic pain, and possibly chills or purulent vaginal discharge. Fever that is accompanied by palpitations, dyspnea, severe abdominal pain, or mental status changes can signify sepsis. (See "Sepsis syndromes in adults: Epidemiology, definitions, clinical presentation, diagnosis, and prognosis" and "Pathophysiology of sepsis".)
Physical examination — Generalized evaluation of the patient should assess for restlessness, inability to move without pain, mental status changes, diaphoresis, pallor, or dyspnea, any of which can signify serious illness, such as sepsis. Likewise, tachycardia, hypotension, orthostasis, tachypnea, and/or hypoxemia often signify serious illness. (See "Definition, classification, etiology, and pathophysiology of shock in adults".)
The abdomen is evaluated for distention, rigidity, rebound or involuntary guarding, suggestive of an inflammatory intraabdominal process. (See "Evaluation of acute pelvic pain in nonpregnant adult women".)
On speculum examination, the amount and character of uterine bleeding are assessed. Active bleeding that fills the upper vagina during the examination is excessive. Blood clot or tissue may extrude from the cervical os or lie in the vagina; it should be examined for evidence of fetal parts, membranes, or villi (to identify villi, it helps to float the tissue in saline) and then sent for pathologic analysis. If a large cotton swab can be passed into the uterine cavity, the internal cervical os is dilated, which suggests RPOC since the cervix usually closes rapidly (within a day) after a complete uterine evacuation.
The bimanual examination should evaluate for cervical motion tenderness, significant uterine tenderness or enlargement, and adnexal masses or tenderness.
Significant uterine tenderness, heavy vaginal bleeding, cervical dilation, uterine enlargement, or signs of systemic infection should prompt further evaluation for RPOC.
Laboratory — Laboratory studies are often normal and thus of limited value. However, markedly abnormal values help guide treatment recommendations.
●Complete blood count and differential – An elevated white blood cell count supports the diagnosis of uterine infection but can be a normal finding in postpartum patients secondary to the physiologic leukocytosis of pregnancy. However, a rising neutrophil count associated with elevated numbers of bands is suggestive of an infectious process. A low hemoglobin/hematocrit supports a diagnosis of heavy or prolonged bleeding and likely warrants consideration of surgical intervention and assessment for coagulopathy.
●Endometrial cultures – General endometrial cultures are not performed routinely when infection is suspected because of the difficulty in obtaining a clean specimen through the cervix. Furthermore, they yield results too late for clinical use, while rarely changing treatment; thus, they are not cost-effective when performed as a routine procedure in patients with fever. Endometrial biopsy could be useful in diagnosing a more chronic process (chronic RPOC or chronic endometritis associated with RPOC). (See "Postpartum endometritis".)
●Blood cultures – Blood cultures are costly and the initial choice of antibiotic therapy has to be made before the results are available. However, they can be useful in guiding antimicrobial treatment if the patient fails to respond to empiric therapy or appears acutely ill with fever and chills. Interestingly, only a single organism may be identified despite polymicrobial endometrial infection.
●Serum hCG level – A single qualitative or quantitative human chorionic gonadotropin (hCG) concentration is not useful, as hCG can remain positive for weeks after an uncomplicated miscarriage or pregnancy termination. Alternatively, an hCG level that is less than the assay threshold does not exclude the possibility of RPOC as the cause of the patient's symptoms since necrotic RPOC may remain in the uterus without actively secreting hormone.
In stable afebrile patients with persistent abnormal bleeding over three or more weeks and inconclusive findings on ultrasound examination, serially monitoring hCG levels may identify a plateau or rise, which suggests residual, viable trophoblast related to ectopic pregnancy, ongoing intrauterine pregnancy, or gestational trophoblastic disease .
Several studies have attempted to define the normal decline in hCG levels after pregnancy loss or termination. In one study, for example, the authors analyzed data from 710 patients who had a nondiagnostic transvaginal ultrasound examination on presentation (no intrauterine or extrauterine gestational sac) and whose subsequent serial hCG measurements spontaneously declined to less than 5 milli-international units/mL . It was assumed that these patients had undergone a complete miscarriage, although some may have had spontaneous resolution of an ectopic pregnancy (eg, tubal abortion). The rate of hCG decline depended, in part, on the initial level, with higher initial hCG levels showing a more rapid rate of decline. For 95 percent of patients, the slowest rate of decline over 48 hours was 21 to 35 percent; over seven days, the decline was 60 to 84 percent; and, on average, the hCG level became undetectable by 12 to 16 days of follow-up. Based on the curves generated in this study, when the rate of hCG decline is less than 20 percent over 48 hours or 60 percent in one week, then residual trophoblastic tissue (eg, RPOC, ectopic pregnancy) should be suspected. These data have not been validated in prospective studies.
HCG levels have also been studied following medical abortion. In a prospective study, 217 patients undergoing medical abortion had hCG levels drawn at the time of initial treatment and again 6 to 18 days later . An hCG decline of 80 percent during this interval identified 98.5 percent of patients with a complete abortion . Another smaller prospective study found that the hCG level fell by 68 to 97 percent (mean 91 percent) from the day of mifepristone administration to the fifth day of follow-up . However, they did not have a comparison group of patients with RPOC, which would allow estimation of the test's predictive value.
Imaging studies — Sonography has limited value in triaging patients with suspected RPOC because sonographic findings correlate poorly with clinical symptoms and histologic results. The appearance of necrotic decidua and blood clots can mimic RPOC. There is marked overlap in the sonographic uterine appearance of asymptomatic and symptomatic patients; the endometrial cavity is commonly irregular and thickened and may show prominent color Doppler flow in patients with an uneventful course, as well as those with histologically proven RPOC . The use of three-dimensional ultrasound imaging does not appear to improve diagnostic performance .
Furthermore, there is great variability among studies in terms of diagnostic and patient management criteria, no studies have blinded managing clinicians to the ultrasound results, and ultrasound interpretation often influences the choice to perform uterine evacuation. Thus, although histologic findings can often be correlated with imaging results in symptomatic patients with abnormal ultrasound scans, there is less information about RPOC in symptomatic patients with normal ultrasound scans since surgical intervention may be avoided in this group, or in asymptomatic patients since typically they are not scanned. Likewise, the clinical outcome of expectant management of symptomatic patients with abnormal ultrasound findings cannot be determined since these patients often undergo surgical intervention.
Sonographic parameters that may be helpful in supporting a clinical diagnosis of RPOC include identification of an echogenic mass within the endometrium and the presence of low-resistance flow within the myometrium or just beneath the endometrium. Endometrial thickness has also been studied, but there is no good evidence that it is useful. Specifically, one study showed no association between endometrial thickness following a single dose of misoprostol for spontaneous abortion and the need for later surgical treatment .
●Echogenic mass – The combination of a hyperechoic endometrial mass or solid component within the endometrium and abnormal (heavy or prolonged) bleeding or infection is a sensitive indicator of RPOC [29-32]. On the other hand, in the absence of clinical symptoms, surgical intervention for these ultrasound findings is not indicated, as prevention of future morbidity is unproven.
A separate study reported that an ultrasonographic mass measuring ≥3 cm with vascularity extending more than halfway through the myometrium was both sensitive and specific for predicting hemorrhage after a second-trimester abortion .
●Low resistance Doppler flow – Pulsed Doppler can be applied during ultrasound examination to calculate a resistance index; RPOC are suspected when there is abundant low-resistance flow within the myometrium or just beneath the endometrium [32,34]. The value of this test is unclear since the placental implantation site may appear vascular even after trophoblast has separated, and residual tissue causing subinvolution may be present even in the absence of significant blood flow to that tissue .
In a prospective study, 62 patients underwent ultrasound and Doppler evaluation following miscarriage, and those with a positive Doppler study then underwent uterine evacuation . Twenty-nine percent of the patients had a positive study, and 89 percent of these patients had RPOC on pathologic analysis. Of 44 subjects with a negative Doppler study, 86 percent had an uneventful clinical course, 14 percent had uterine evacuation performed for heavy bleeding, and 33 percent of these patients had RPOC confirmed by pathologic analysis. In a similar prospective study, 20 percent of 65 patients had increased Doppler flow on ultrasound 10 days after surgical abortion. All underwent dilation and curettage, and 77 percent had trophoblast on pathology . As in other studies of ultrasound performance, it is not possible to know how many of the patients with positive Doppler findings would have done well without surgical intervention, and how many of those expectantly managed actually had RPOC within the uterus.
●Endometrial thickness – A thickened endometrium has been used as a sign of RPOC. Endometrial thickness is determined by measuring the width of the endometrium in the anterior-posterior dimension. Relatively arbitrary cut-offs of 8 to 25 mm have been defined in the literature . However, there is no good evidence that any cut-off is clinically useful [31,44-47]. Measurement of endometrial thickness following miscarriage or a pregnancy termination should not be used as a test for RPOC or as a predictor of need for surgical intervention.
Sonohysterography — An advantage of sonohysterography over conventional ultrasound examination is that it can better define focal irregularities in the uterine cavity and distinguish a free-floating echogenic mass (which is likely to pass through the cervix naturally) from a mass adherent to the uterine wall (which may require surgical intervention). However, this test is invasive and costly compared with standard ultrasound examination; there is inadequate evidence to support its use in management of suspected RPOC.
In one small study, 43 percent of the patients with late postabortion and postpartum bleeding and an echogenic mass on conventional ultrasound had either an empty cavity or free floating mass on sonohysterogram . These patients were managed conservatively and did well. Those with an adherent mass underwent surgical intervention, and all had RPOC confirmed by pathology. As with the other ultrasound studies, it is not possible to know if these patients also would have done well if managed conservatively.
Hysteroscopy — Like sonohysterography, hysteroscopy better defines focal irregularities in the uterine cavity than conventional ultrasound examination and can distinguish a free-floating echogenic mass from a mass adherent to the uterine wall; unlike sonohysterography, hysteroscopy potentially allows diagnosis and therapy during a single procedure. If bleeding is not heavy, hysteroscopy can be performed in the office with a flexible hysteroscope and no anesthesia, but requires specialized equipment and expertise.
We prefer hysteroscopy to sonohysterography in stable patients with persistent (three or more weeks) mild symptoms and nondiagnostic ultrasound examination. If RPOC are identified, the tissue can often be removed with a hysteroscopic grasper or with a manual vacuum aspirator.
We do not use either sonohysterography or hysteroscopy in the setting of clinical evidence of infection (fever, significant uterine tenderness, or cervical motion tenderness), as the instillation of fluid may theoretically promote an ascending pelvic infection. Patients with suspected infection and clinical concern for RPOC should be treated with antibiotics and uterine evacuation (see below).
MANAGEMENT — The following discussion applies to patients in whom the suspicion of RPOC is high based on history and physical examination, with or without support from imaging studies.
Patients who are hemodynamically unstable — Patients who are hemodynamically unstable due to excessive uterine bleeding after miscarriage or pregnancy termination should be stabilized with fluid and blood products, as needed, and undergo urgent surgical intervention. Uterotonic drugs (misoprostol 800 mcg per rectum or methylergonovine 0.2 mg orally three or four times per day) may be useful, but have not been studied in this setting. Methylergonovine should be withheld in patients with hypertension or cardiovascular disease.
If an alternate source of significant bleeding is found before surgical evacuation, such as a cervical laceration, it should be repaired first. Uterine evacuation may not be necessary if repair of the laceration controls the hemorrhage.
After uterine evacuation, if removal of clots and RPOC fail to slow the bleeding, uterine tamponade with an intrauterine balloon catheter (eg, Foley, Bakri, BT-Cath) is the next step. Uterine artery embolization may be an option for stabilized patients who do not respond to this therapy, while exploratory laparoscopy or laparotomy and hysterectomy is the last resort.
Usually there is not enough time for formal imaging studies prior to surgery. If an ultrasound examination is possible intraoperatively or at the bedside, evidence of free fluid in the abdomen, which suggests uterine perforation or ruptured ectopic pregnancy, can alter the surgical plan to laparotomy or laparoscopy.
Surgical intervention is discussed below. (See 'Surgical intervention' below.)
Patients with sepsis — Patients with suspected sepsis due to infected RPOC should be managed by prompt evacuation of the uterus after stabilization and initiation of intravenous broad spectrum antibiotic therapy; delay in evacuation may be fatal since antibiotics may not eradicate infection in poorly perfused necrotic tissue. Sepsis can be defined as clinical diagnosis of infection with two or more of the following: temperature >38.5°C or <35°C; heart rate >90 beats/min; respiratory rate >20 breaths/min or PaCO2 <32 mmHg; and WBC >12,000 cells/mm3, <4000 cells/mm3, or >10 percent immature (band) forms.
In patients who have been instrumented, generalized abdominal tenderness, guarding, tachycardia, high fever, and prostration suggest advanced sepsis. These patients require aggressive therapy with broad spectrum intravenous antibiotics, uterine re-evacuation, assessment for uterine perforation and visceral injury, and monitoring and support in an intensive care unit. (See "Pregnancy loss (miscarriage): Clinical presentations, diagnosis, and initial evaluation", section on 'Septic abortion'.)
Severely ill patients who do not respond to adequate medical therapy and surgical evacuation may have bowel injury, pelvic abscess, or clostridial myometritis. These patients require exploration of the abdomen and pelvis and may need hysterectomy.
Surgical intervention is discussed below. (See 'Surgical intervention' below.)
Medically stable patients with endometritis — Patients with uncomplicated endometritis (low grade fever, mild uterine tenderness, empty uterus on ultrasound examination) are managed with a trial of broad spectrum antibiotics, with coverage of anaerobes (eg, cefotetan [2 g intravenously] plus doxycycline [100 mg intravenously or orally] every 12 hours). This regimen can be completed as an outpatient oral regimen for a 14-day course. An alternative outpatient regimen is ceftriaxone 250 mg intramuscularly in a single dose plus doxycycline 100 mg orally twice a day for 14 days with or without metronidazole 500 mg orally twice a day for 14 days.
The presence of a focal echogenic abnormality in the endometrium, particularly with Doppler evidence of blood flow, suggests RPOC complicating endometritis and the possible need for surgical intervention. In these cases, antibiotics alone may not reach the poorly perfused nidus of infection. The positive predictive value of ultrasound findings (echogenic mass, hematometra, endometrium >3 cm thick, Doppler flow) is low in stable, minimally symptomatic patients, but higher in a symptomatic, infected patient. Surgical intervention is discussed below. (See 'Surgical intervention' below.)
If the ultrasound is negative or borderline (possible small amount of material in the uterus), then an otherwise stable patient can be given antibiotics and monitored. An evacuation procedure can be avoided if there is an excellent response to medical therapy within 24 hours. Failure to respond adequately to antibiotics should prompt evacuation. (See "Postpartum endometritis".)
Medically stable patients with prolonged bleeding — Surgical intervention may be initiated to manage bothersome bleeding lasting more than three weeks, or sooner if there is strong suspicion for RPOC and the bleeding is persistently heavy. Bleeding will eventually resolve as the RPOC are resorbed, but surgical management has the potential for immediate removal of the retained tissue, which is preferable for many patients.
Surgical intervention — Uterine evacuation may be accomplished with sharp or suction curettage, or with directed hysteroscopic removal. Some patients with RPOC will present with an open cervix, and will not require further dilation. If dilation is needed, the cervix is dilated just enough to allow the curette or suction catheter to enter the uterine cavity. (See "Dilation and curettage".)
Hysteroscopic removal has the advantage of providing simultaneous diagnosis and treatment; it is most appropriate for use in patients without hemorrhage or clinical signs of infection and may also be used for patients with prolonged irregular bleeding or persistent signs of RPOC on imaging. If abnormal tissue is visualized, it can be removed with a resectoscope, hysteroscopic morcellator, or hysteroscopic grasper [49,50]. The procedure has the proposed advantages of visualizing complete tissue removal and avoiding damage to unaffected areas of endometrium. In one meta-analysis of observational studies assessing outcomes after hysteroscopic resection of RPOC, success rates were high (91 percent; 95% CI 0.83-0.96) and there was a low rate of postsurgical intrauterine adhesions (0.8 percent) .
Three concerns of instrumenting the postabortal uterus are (1) infection from introduction of cervicovaginal flora; (2) uterine perforation, which may injure blood vessels or internal organs (bowel or bladder) and require arterial embolization or surgical repair; and (3) development of synechiae.
Three randomized trials have evaluated perioperative antibiotic use for surgical evacuation of an incomplete miscarriage. One trial assigned 240 patients to intravenous doxycycline or placebo and found the same 6 percent rate of infection in both groups . Notably, this study had a high loss-to-follow-up rate of 30 percent. A second trial including 140 patients prescribed oral tetracycline or placebo for one week following the procedure and also found no difference in infection rate but did note a high rate of noncompliance with the medication . A third trial randomly assigned over 3400 patients in low-resource countries to either antibiotic or placebo prior to surgical evacuation for RPOC following spontaneous abortion . Compared with placebo, pretreatment with a combination of doxycycline and metronidazole did not significantly decrease the rate of subsequent infection as defined by clinician judgment, though there was a significant reduction from 2.6 to 1.5 percent when stricter infection criteria were applied.
There are, however, high-quality data that prophylactic antibiotics reduce periabortal endometritis in all subgroups of patients undergoing pregnancy termination, even those at low risk of infection (ie, no history of pelvic inflammatory disease and a negative preoperative chlamydia culture). Although this evidence is indirect, we suggest antibiotic prophylaxis prior to surgical evacuation of RPOC. We use a single dose of a broad spectrum antibiotic at the time of the procedure, but give a full course of endometritis treatment for patients who are clinically infected.
If uterine perforation occurs, abdominal exploration is warranted if electrosurgical energy or suction curettage was used at the time of perforation, if the patient is hemodynamically unstable, or if there are signs of severe uterine bleeding or vascular or visceral injury. The original procedure can be completed with use of sharp curettage under direct vision of the exterior surface of the uterus. If uterine perforation occurred during cervical dilation or with a blunt instrument, with no suction or electrosurgical energy source, the risk of vascular or visceral injury is low and close observation may be sufficient. The procedure can be completed under ultrasound guidance (with use of sharp curettage rather than use of suction or an electrosurgical device) and the patient can be observed for signs of complications. Another option is to complete the procedure under laparoscopic guidance: laparoscopy allows for direct visualization of the perforation site and retraction of surrounding structures. (See "Uterine perforation during gynecologic procedures".)
Patients with suspected perforation should be treated with antibiotics, such as a single dose of ceftriaxone 250 mg intramuscularly or doxycycline 100 mg orally twice per day plus metronidazole 500 mg orally three times per day for one week.
A series of 105 patients undergoing hysteroscopic morcellation of placental remnants showed a 5 percent rate of uterine perforation, which appears relatively high. However, there is no directly comparative study to dilation and curettage, during which perforations may go unidentified.
Uterine synechiae, also known as Asherman syndrome when accompanied by symptoms (eg, infertility, recurrent pregnancy loss, amenorrhea), are the result of scarring within the uterine cavity. The synechiae may range from inconsequential filmy adhesions to very thick scars that obliterate the endometrium, leading to amenorrhea/hypomenorrhea and infertility . The major risk factor for developing uterine synechiae is curettage of a pregnant or recently pregnant uterus, particularly in the setting of an infection, and the risk appears to increase with the number of surgical procedures [55-58]. As an example, a hysteroscopic study of 147 patients who had had prior curettage for miscarriage reported 15 percent had synechiae after one or two curettage procedures and 32 percent had synechiae after three procedures . The scar tissue appeared denser after the second or third curettage. The benefits of a second procedure in a stable patient should be weighed against this risk.
It is possible that intrauterine adhesions are secondary to the inflamed retained tissue, rather than the curettage procedure itself. This has been suggested by the findings of diminished fertility and more hypomenorrhea when curettage with positive placental pathology is compared with curettage with negative pathologic findings . From this perspective, removal of the retained tissue is desirable, and directed hysteroscopy may be the best approach for patients with persistent RPOC. A direct comparison of hysteroscopic resection to curettage for RPOC showed a significantly lower rate of adhesions with hysteroscopy, presumably because global endometrial damage is avoided . (See "Intrauterine adhesions: Clinical manifestation and diagnosis".)
Medical intervention — The benefit of misoprostol over expectant management for incomplete abortion has yet to be established , though it is often prescribed. Misoprostol (a prostaglandin E1 analog) is the most commonly used drug for medical treatment of incomplete miscarriage . Its safety has been established by multiple randomized and controlled trials. Notably, these trials have excluded patients with suspected or confirmed infection or active hemorrhage; these patients have been treated with antibiotics (if infected) and/or surgical (suction) evacuation.
The advantages of misoprostol over other drugs (including prostaglandin E2) are its low cost, low incidence of side effects when given intravaginally, stability at room temperature, and ready availability. The risk of a major complication is rare.
It is important to note that trials of medical management of RPOC have generally involved patients with incomplete, inevitable, and missed abortion, rather than those with tissue left behind after a medical or surgical pregnancy termination. It is unclear whether results from these trials can be extrapolated to the latter group. One center offered a choice of dilation and curettage or a two-day misoprostol course to all patients with cramping and/or bleeding after first-trimester surgical abortion. Ninety-three percent of those receiving misoprostol had complete resolution of their symptoms, and the authors concluded that this strategy may decrease the use of repeat surgical management by 79 percent. However, it is not clear how many of these patients actually had RPOC and what the success rate would have been with expectant management alone .
It is reasonable to consider medical management in any stable, consenting patient with RPOC who wishes to avoid a repeat surgical procedure, though patients should be advised that its superiority to expectant management is unproven. Nonsurgical management may be preferable for patients who wish to avoid the risks of a surgical procedure, particularly if they have had prior uterine evacuation and are concerned about uterine injury or synechiae. The medical management of incomplete abortion is discussed in detail separately. (See "Pregnancy loss (miscarriage): Description of management techniques", section on 'Medication management'.)
No trials have evaluated the use of prophylactic antibiotics in uninfected patients undergoing medically induced evacuation of RPOC with misoprostol. These patients should be counseled about signs and symptoms of infection, and treated promptly with antibiotics and surgical evacuation if endometritis is suspected.
Expectant management — Expectant management is an option for patients with RPOC who have stable vital signs and no evidence of infection. Two meta-analyses have reviewed outcomes when treating incomplete miscarriage expectantly. Although these studies are limited by heterogeneity in diagnosis and management protocols, they report success rates of 50 to 85 percent at one to two weeks of follow-up, and up to 90 percent when subjects are followed for six weeks [11,63]. Notably, these studies were also limited to patients who had undergone spontaneous miscarriage, rather than those who had prior uterine evacuation. (See "Pregnancy loss (miscarriage): Description of management techniques", section on 'Expectant management'.)
In a retrospective study including 44 patients with suspected RPOC after spontaneous or induced abortion and managed expectantly, more patients were observed without intervention than required a later intervention (77 versus 23 percent, respectively) . Initial heavy bleeding and increased vascularity of the retained tissue by ultrasound were the best predictors of a later clinically indicated procedure.
No trials have evaluated the use of prophylactic antibiotics in uninfected patients undergoing expectant management; we do not use antibiotic prophylaxis. These patients should be counseled about signs and symptoms of infection, and treated promptly with antibiotics and surgical evacuation if endometritis is suspected.
Asymptomatic patients with suspected RPOC — For asymptomatic patients in whom ultrasound suggests RPOC, we suggest expectant management rather than intervention, especially in the first one to two weeks following initial pregnancy loss or evacuation. While evacuation procedures may hasten resolution, RPOC can resolve spontaneously and complications (eg, uterine perforation, intrauterine synechiae) from evacuation procedures (though rare) can occur; we have cared for patients with clinical Asherman syndrome whose only prior procedure was a uterine aspiration procedure. It is unclear, however, whether the procedure itself, the retained tissue, or both confer such risk.
In our practice, we typically reassess asymptomatic patients after six weeks if the patient has not resumed menses or if they develop symptoms of RPOC. For patients uncomfortable with waiting six weeks, intervention can be performed sooner after discussion of the potential risks and benefits, including the possibility of a false-positive ultrasound diagnosis for RPOC. (See 'Clinical manifestations' above.)
In one randomized trial including 59 patients with evidence of RPOC on ultrasound performed one to two weeks after misoprostol treatment for spontaneous abortion, those managed expectantly compared with immediate curettage had lower success rates (defined as no POCs by later hysteroscopy or uneventful follow-up course; 76 versus 97 percent, respectively) . Two of 29 patients in the expectant management group underwent delayed emergency curettage for hemorrhage. Notably, this study had a very low threshold to define RPOC (endometrial thickness >10 mm), and the inclusion of patients with a false-positive diagnosis likely increased their reported success rates; no immediate histology results for the curettage group were reported. A follow-up study of this group combined with a larger, nonrandomized cohort showed similar rates of future pregnancy for expectant management versus curettage .
In another randomized trial including 40 patients with a missed abortion treated with misoprostol and a postprocedure ultrasound demonstrating an endometrial thickness ≤20 mm, those treated with expectant managed compared with immediate manual vacuum aspiration (MVA) had higher rates of persistent signs of RPOC (50 versus 10 percent, relative risk [RR] 0.11, 95% CI 0.02-0.61) . Other outcomes (eg, duration of bleeding, duration of pain) were similar between groups.
Patients with abnormal placentation — Abnormally adherent placenta is generally described later in pregnancy but has been reported as early as the first trimester. While uncommon, it can lead to major hemorrhage and hysterectomy at the time of uterine evacuation, even in the first trimester . A number of cases have presented as RPOC, with prolonged or heavy bleeding following prior uterine evacuation [69,70]. The condition may be considered when a patient has had prior cesarean births or other uterine procedures (eg, myomectomy), particularly with a lower uterine segment pregnancy implantation, or when there are multiple failed attempts at uterine evacuation . Such patients should be referred to experienced specialists for evaluation and treatment, which may include laparoscopic excision , chemotherapy , uterine artery embolization , or hysterectomy . (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences" and "Cesarean scar pregnancy".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pregnancy termination".)
SUMMARY AND RECOMMENDATIONS
●Clinical findings – Retained products of conception (RPOC) refers to placental and/or fetal tissue that remains in the uterus after spontaneous pregnancy loss (miscarriage), planned pregnancy termination, or preterm/term delivery. RPOC present with uterine bleeding, pelvic pain, fever, and/or uterine tenderness. These clinical findings are nonspecific; moreover, some postabortal bleeding and discomfort are normal. If viable trophoblastic tissue is present, the patient may not resume normal menstrual cycles after six weeks of follow-up. (See 'Introduction' above and 'Clinical manifestations' above.)
•Routine evaluation for RPOC after pregnancy loss or termination is likely to result in false positive diagnoses and unnecessary interventions since RPOC do not always lead to morbidity. On the other hand, evaluation is indicated in patients whose symptoms fall outside the normal range, such as those with bleeding that is heavy (greater than menses) or prolonged (over three weeks), and those with fever, uterine tenderness, or abdominopelvic pain that is worsening or cannot be controlled with oral analgesics. (See 'Diagnostic evaluation' above.)
•The goal of the evaluation is to determine whether RPOC are the probable source of the patient's symptoms or whether another diagnosis is more likely. (See 'Differential diagnosis' above.)
•The presence of a focal hyperechoic mass in the endometrium, particularly with evidence of blood flow by Doppler imaging, is the best test for prediction of RPOC. However, the decision to intervene should be based on clinical need rather than on an isolated ultrasound finding. (See 'Imaging studies' above.)
•Hemodynamically unstable patients – Patients who are hemodynamically unstable from hemorrhage or sepsis should undergo prompt surgical evacuation. Sepsis should also be treated with broad spectrum intravenous antibiotics. (See 'Patients who are hemodynamically unstable' above and 'Patients with sepsis' above.)
•Patients with endometritis – If diagnostic ultrasound of medically stable patients with uterine infection does not show RPOC, we suggest treatment with antibiotics alone rather than uterine evacuation (Grade 2C). These patients should be monitored closely for clinical improvement, and evacuation should be performed if there is no improvement within 24 hours. (See 'Medically stable patients with endometritis' above.)
•Patients with prolonged bleeding – For patients who are bothered by prolonged bleeding (lasting more than three weeks), we suggest cavity evaluation with potential uterine evacuation rather than medical intervention or expectant management (Grade 2C). Hysteroscopy is our preferred modality in this situation, both for diagnostic confirmation and visually confirmed, targeted removal of the tissue. Bleeding will eventually resolve as the RPOC are resorbed, but surgical management has the potential for immediate removal of the retained tissue, which is preferable for many patients. (See 'Medically stable patients with prolonged bleeding' above.)
•Asymptomatic patients – For asymptomatic patients in whom ultrasound suggests RPOC, we suggest expectant management rather than surgical intervention (Grade 2C). While evacuation procedures may hasten resolution, RPOC can resolve spontaneously and complications (eg, uterine perforation, intrauterine synechiae) from evacuation procedures (though rare) can occur. We typically reassess such patients after six weeks if the patient has not resumed menses or if they develop symptoms of RPOC. (See 'Asymptomatic patients with suspected RPOC' above.)