Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
We thank Dr. Sidhant Nagrani and Dr. Bisan Salhi for their contributions to previous editions of this chapter.
Miscarriage is defined as the spontaneous termination of pregnancy before 20 weeks of gestational age.
Approximately 25% of women experience some bleeding during pregnancy.
Most miscarriages are due to uterine malformations or chromosomal abnormalities.
Beta hCG discriminatory levels for detecting intrauterine pregnancy are usually considered to be 6500 mIU/mL for transabdominal ultrasonography and 1000 to 2000 mIU/mL for transvaginal ultrasonography.
Treatment of the patient with incomplete miscarriage includes expectant management, medical management, or surgical evacuation.
All patients with pregnancy loss require patient education, return precautions, and support through the grieving process.
Acute complications of pregnancy can appear in all trimesters and pose challenges in diagnosis and management for the emergency clinician. Life-threatening disorders, such as ectopic pregnancy in early pregnancy, pregnancy-induced hypertension in mid to late pregnancy, and abruptio placentae in late pregnancy, are relatively common. Emergency clinicians must consider the signs and symptoms, stage of pregnancy, and hemodynamic stability of the patient in developing diagnostic and treatment strategies.
Miscarriage, the most common serious complication of pregnancy, is defined as the spontaneous termination of pregnancy before 20 weeks of gestation. Fetal demise after 20 weeks of gestation or when the fetus is more than 500 g is considered premature birth. Early pregnancy loss is defined as a nonviable, intrauterine pregnancy with either an empty gestational sac or gestational sac containing an embryo or fetus without fetal cardiac activity within the first 12 weeks and 6 days of gestation.
Early pregnancy loss is common, with 80% occurring in the first trimester, and it is seen in 10% of recognized pregnancies. This estimation is likely low because it is difficult to measure those pregnancies lost before clinically confirmed but still recognized by the patient. Embryonic and fetal loss after implantation occur in up to one-third of detectable pregnancies. The risk of miscarriage rises with increasing maternal age (a fivefold increase in those >40 years compared with those 25 to 29 years), increasing paternal age, alcohol use, increased parity, history of prior miscarriage, poorly controlled diabetes mellitus and thyroid disease, obesity, low pre-pregnancy body mass index, maternal stress, smoking, alcohol, and caffeine consumption, and history of vaginal bleeding.
Approximately 25% of pregnant patients experience some bleeding. It is estimated that up to 50% of all women who have bleeding during early pregnancy miscarry before 20 weeks’ gestation, although the risk is probably higher in the emergency department (ED) population. Patients who have an intrauterine pregnancy with fetal cardiac activity visualized on ultrasound examination have a much lower risk of miscarriage (3% to 6%), although vaginal bleeding is a high-risk indicator, even when a viable fetus is present. Those with a history of bleeding in singleton pregnancies who do not miscarry may have otherwise normal pregnancies, although they have an increased risk for preterm premature rupture of membranes, abruption, previa, stillbirth, and congenital abnormalities.
Most miscarriages are due to uterine malformations or chromosomal abnormalities, which account for the majority that occur within 10 weeks of gestation. In some cases, the ovum never develops (anembryonic gestation). In most early miscarriages, fetal death precedes clinical miscarriage, often by several weeks. Although clinical symptoms of miscarriages are most common between 8 and 12 weeks of gestation, sonographic evidence in most cases demonstrates death before 8 weeks; if fetal viability can be demonstrated by cardiac activity and a normal sonogram, the subsequent risk of fetal loss decreases significantly.
Maternal factors that increase the risk of miscarriage include age greater than 30 years, congenital anatomic defects, uterine scarring, leiomyomas, and cervical incompetence. Other conditions associated with increased miscarriage rates include toxins (e.g., alcohol, tobacco, and cocaine), autoimmune factors, endocrine disorders, a prior history of miscarriage, and maternal infections.
Miscarriage is broadly divided into four categories. The first is a threatened miscarriage , in which the patient presents with vaginal bleeding but is found to have a closed internal cervical os. The risk of miscarriage in this population is estimated at 35% to 50%, depending on the patient’s risk factors and severity of symptoms. The second category is an inevitable miscarriage when the internal os is open. The third is an incomplete miscarriage , where the products of conception are present at the cervical os or in the vaginal canal. The last category is a completed miscarriage , which occurs when the uterus has expelled all fetal and placental material, the cervix is closed, and the uterus is contracted. A gestational sac should be visualized for diagnosis because the cervix may close after an episode of heavy bleeding and clot passage without or after only partial expulsion of the products of conception. Unless an intact gestational sac is passed and recognized, a completed miscarriage is diagnosed only after dilation and curettage (D&C) with pathologic confirmation of gestational products, demonstration by sonography of an empty uterus with a prior known intrauterine pregnancy (IUP), or reversion to a negative pregnancy test result. This may take up to several weeks after the initial presentation.
Missed abortion is a relatively obsolete term referring to the clinical failure of uterine growth over time. Instead, the terms anembryonic gestation (when no fetus is visualized on ultrasound), first- or second-trimester fetal death (failure to see fetal cardiac activity with at least a 5-mm crown-rump length), and delayed miscarriage are more appropriate.
Patient history should include the estimated length of the gestation, time since the last menstrual period, symptoms of pregnancy, including evolution or loss of pregnancy symptoms, degree and duration of bleeding, presence of cramps, pain, or fever, and attempts by the patient to induce miscarriage. Although the history is important, it is not helpful in the classification of the type of miscarriage. In addition, the severity of symptoms does not correlate well with the risk of miscarriage, although cramping and passage of clots are thought more likely to occur as the miscarriage becomes inevitable.
The assessment of the patient who experiences first-trimester vaginal bleeding includes a careful abdominal examination to evaluate for tenderness or peritoneal irritation from a potential ectopic pregnancy and to determine the size of the uterus, which should not be palpable abdominally. A pelvic examination is performed to evaluate whether the cervix is closed or open, look for clots or the products of conception, determine the degree of vaginal bleeding, and assess uterine size and tenderness.
In the patient with second- and third-trimester bleeding, cervical probing should not be done because the uterus is more vascular, and the organized placenta may overlie the cervical os. Parous women normally have an open or lax external os, which is a finding of no significance. The adnexa may be enlarged, often unilaterally, because the corpus luteum is cystic or because the pregnancy is ectopic. Adnexal or uterine tenderness should always raise the possibility of an ectopic pregnancy. Much less commonly, pelvic inflammatory disease can cause uterine and adnexal tenderness during early pregnancy.
Ectopic pregnancy can masquerade as a miscarriage in the early stages of pregnancy and should always be considered in the differential diagnosis. Even in the patient with painless vaginal bleeding, the diagnosis of ectopic pregnancy must be considered. Early ultrasonography is indicated to locate the pregnancy in the patient who has bleeding or pain.
A small amount of bleeding occurs at the time of implantation of the blastocyst into the endometrium and, occasionally, at the time of the first missed menses. Molar pregnancy is also characterized by vaginal bleeding, usually during the late first trimester or second trimester. This condition can be identified by ultrasonography. Cervical and vaginal lesions can also cause local bleeding and can usually be seen on vaginal inspection.
A hemoglobin level is useful to provide a baseline measurement and evaluate the degree of blood loss in women whose bleeding persists. In addition, the Rh type should be determined. Ultrasonography is the primary means of evaluating the health of the fetus as well as its location ( Table 173.1 ). Because historical and clinical estimations of gestational age are often inaccurate, ultrasonography is useful to provide an accurate measure of fetal age and viability ( Box 173.1 ).
Finding | Weeks From LMP | β-hCG (mIU/mL) |
---|---|---|
Gestational sac (25 mm) | 5 | 1000 |
Discriminatory zone | 5–6 | 1000–2000 |
Yolk sac | 6 | 2000 |
Upper discriminatory zone | 6–7 | 3000 |
Fetal pole | 7 | 5700 |
Fetal heart motion | 6–7 | 7000 |
No gestational sac at β-hCG level of 3000 mIU/mL
No yolk sac with gestational sac of 13 mm (or at 32 days since last menstrual period)
5-mm crown-rump length, with no fetal heart tones
No fetus, with gestational sac of 25 mm mean diameter
No fetal heart tones after gestational age of 10–12 wk
β - hCG, Beta subunit of human chorionic gonadotropin.
Serial quantitative hCG levels are used to assess the health of the fetus if sonographic findings are indeterminate or if the gestational age is less than 6 to 7 weeks. The sonographic discriminatory zone is defined as the quantitative hCG level at which a normally developing IUP should reliably be seen. Discriminatory levels are operator- and equipment-dependent and vary by individual patient characteristics, but are usually considered to be 6500 mIU/mL for transabdominal ultrasonography and 1000 to 2000 mIU/mL for transvaginal ultrasonography. Ultrasonography can be performed or repeated when hCG levels rise to 1500 to 3000 mIU/mL. If hCG levels are level or decline, or if sonographic criteria for fetal demise are demonstrated ( Box 173.1 ), the patient should be referred to an obstetrician for follow-up to ensure miscarriage completion and to assess for subsequent complications. Expectant management may be sufficient in the stable patient with threatened miscarriage, as long as ectopic pregnancy has been excluded.
After assessment of hemodynamic status and management of blood loss, a patient with a threatened miscarriage requires very little specific medical treatment. Though expert opinions vary as to whether anti-D immune globulin should be administered to Rh-negative patients after a threatened or spontaneous miscarriage, we recommend administering anti-D immune globulin in the ED to all pregnant patients with bleeding. If anti-D immune globulin is administered, a 50-μg to 120-μg dose is used during the first trimester and a full 300-μg dose after the first trimester. Once evaluated for ectopic pregnancy, the need for a follow-up routine ultrasonogram should be discussed with the patient; the patient should be made aware that the potential for ectopic pregnancy exists until it is excluded by identification of an IUP. In the patient who is planning pregnancy termination, prompt referral should be encouraged and chorionic villi confirmed at the time of uterine evacuation.
Unless an IUP is diagnosed, the patient with threatened miscarriage should be given careful instructions on discharge to return if she has signs of hemodynamic instability, pain, or other symptoms that might indicate ectopic pregnancy. In conjunction with gynecologic colleagues, an ED protocol is useful to determine when follow-up sonographic evaluation and serial hCG measurements should be obtained since ultrasonography can be an inaccurate diagnostic tool if the hCG level is below 1500 mIU/mL, vaginal bleeding is significant, or sonographic findings do not include a fetal pole or yolk sac. The patient must be given explicit return precautions and close obstetrical follow-up is essential. Typically, serial hCG measurements are obtained between 48 to 72 hours after the initial ED visit and follow-up sonographic evaluation is obtained within 3 to 7 days of ED presentation.
Fifty percent or more of women with threatened miscarriage who are seen in the ED ultimately miscarry, and there is no proven treatment to prevent miscarriage. In most cases, spontaneous miscarriage is the body’s natural method of expelling an abnormal or undeveloped (blighted) pregnancy. Thus, a major goal of early management should be patient education and support. Patients should be advised that moderate daily activities do not affect the pregnancy. Tampons, intercourse, and other activities that might induce uterine infection should be avoided as long as the patient is bleeding, and she should return immediately for fever, abdominal pain, or an increase in bleeding. Cramping from a known IUP can be safely treated with acetaminophen or oral synthetic narcotics, if needed. If the patient passes tissue, it should be brought to a provider to be examined for products of conception because differentiation of fetal parts or villi from decidual slough or casts is difficult.
Patient counseling is paramount with threatened miscarriage and education of the ED staff on this topic is critical. Determination of fetal viability can be helpful in reassuring the mother or preparing her for probable fetal loss. Miscarriages are associated with a grieving process, which is frequently more difficult because early pregnancy is unannounced, and early fetal death is not publicly recognized. Because many women consider that minor falls, injuries, or stress during the first trimester can precipitate miscarriage, patients should be reassured that they have done nothing to cause miscarriage. Patients should be made aware that miscarriage is common, grieving is normal, and counseling may be beneficial. A follow-up appointment should be scheduled after miscarriage to support the patient in resolving such issues.
Treatment of the patient with incomplete miscarriage includes expectant management, medical management, or surgical evacuation. When the miscarriage is incomplete, the uterus may be unable to contract adequately to limit bleeding from the implantation site. Bleeding may be brisk, and gentle removal of fetal tissue from the cervical os with ring forceps during the pelvic examination often slows bleeding considerably. Manual uterine aspiration performed in the ED may also be appropriate in cases of brisk uterine hemorrhage as a result of early pregnancy loss or retained products of conception up to 12 weeks gestational age.
Management of patients with presumed completed miscarriage is more complicated. If the patient brings passed tissue with her, this should be sent to the pathology department for evaluation. Unless an intact gestational sac or fetus is visualized, it is rarely clear clinically whether miscarriage is complete. In women with a history consistent with miscarriage who have minimal remaining intrauterine tissue as determined by ultrasonography, expectant management is safe, but only if ectopic pregnancy can be excluded. If endometrial tissue is not seen with ultrasonography, bleeding is mild, and gestational age is less than 8 weeks, curettage is frequently unnecessary, and the patient can be safely observed by a gynecologist for serial hormonal assays. It is estimated that 65% of women with first-trimester miscarriage complete the miscarriage without intervention. However, the need for later visits and procedures may be decreased by uterine curettage, particularly if the fetal pole or a gestational sac is visible on the sonogram at the time of evaluation. Medical management with misoprostol (800 μg intravaginal for one dose) instead of dilation and curettage is also an option and has a success rate of 80% to 91%. The patient should be instructed to return if uncontrolled bleeding, severe pain or cramping, fever, or tissue passage occurs. Follow-up is recommended in 1 or 2 weeks to ensure that the miscarriage is complete.
After miscarriage, the patient should be advised that fetal loss can cause psychological stress. Follow-up in 1 or 2 weeks with a gynecologist should be provided. There is no conclusive evidence to support the use of antibioitics after D&C or miscarriage, and some evidence has suggested that the side effects of treatment may outweigh any potential benefit. For that reason, we do not recommend the routine use of antibiotics after a miscarriage. Ergonovine or methylergonovine (0.2 mg orally bid) can be used to stimulate uterine involution. The patient should be advised to return if signs of infection (e.g., fever, uterine tenderness) occur, bleeding resumes, or further tissue is passed.
An ectopic pregnancy can masquerade as a threatened miscarriage in the early stages of pregnancy and should always be considered in the differential diagnosis.
Because the history and physical examination of the patient with ectopic pregnancy are insensitive and nonspecific, pelvic ultrasonography and determination of serum hCG levels are essential to locate the pregnancy in any patient who has abdominal pain or vaginal bleeding and a positive pregnancy test result.
Ultrasonographic detection of an IUP is likely at hCG levels higher than 1500 to 2000 IU/L.
Ectopic pregnancy is a pregnancy implanted outside the uterus, most commonly in the fallopian tube. It is increasing in frequency and the third leading cause of maternal death, responsible for 4% to 10% of cases. Ectopic pregnancy is estimated to account for approximately 1% to 2% of all pregnancies, although national estimates of incidence are difficult to determine. Although the incidence of ectopic pregnancy is highest in women aged 25 to 34 years, the rate is highest among older women and women belonging to minority groups. Simultaneous intrauterine and extrauterine gestations (heterotopic pregnancy) have historically been rare, occurring in approximately 1 in 4000 pregnancies; women who have undergone assisted reproduction techniques with embryo transfer are at high risk of one of the pregnancies being ectopic. The incidence of ectopic pregnancy among women presenting to the ED with vaginal bleeding or pain in the first trimester is approximately 10%, but may be as high as 16%.
Implantation of the fertilized ovum occurs approximately 8 or 9 days after ovulation. Risk factors for an abnormal site of implantation include prior tubal infection (50% of cases), anatomic abnormalities of the fallopian tubes, assisted reproduction (especially multiple embryo transfers), and abnormal endometrium (host factors). This results in failure of the embryo to implant in the endometrium. The risk of ectopic pregnancy increases approximately threefold after a patient has had pelvic inflammatory disease (PID). If the patient is currently using an intrauterine device (IUD), increased risk can occur from complicating PID or from failure of the IUD to prevent pregnancy while preventing endometrial implantation. All forms of contraception, except the IUD and tubal sterilization, decrease the incidence of ectopic pregnancy. After an ectopic pregnancy, the risk of a subsequent ectopic pregnancy can be as high as 22%, depending on the characteristics and treatment of the ectopic pregnancy (e.g., location of implantation, surgical vs. medical management; Box 173.2 ).
Tubal surgery (for tubal sterilization or ectopic pregnancy)
Pelvic inflammatory disease
Smoking
Advanced age
Prior spontaneous abortion
Medically induced abortion
History of infertility
Intrauterine device
When abnormal implantation occurs in the fallopian tubes, on the ovaries, or in the cervix, the pregnancy usually grows at a less than normal rate, which can result in abnormally low or declining hCG production. Even if exceedingly low, a single hCG measurement cannot exclude the diagnosis of ectopic pregnancy. Blood leaks intermittently through the tubal wall or out the fimbrial ends, with spillage into the peritoneal cavity. Bleeding and other symptoms are usually intermittent. Three outcomes are possible: spontaneous involution of the pregnancy, tubal abortion into the peritoneal cavity or vagina, or rupture of the pregnancy with internal or vaginal bleeding. Implantation in the uterine horn (cornual pregnancy) is particularly dangerous because the growing embryo can use the myometrial blood supply to grow larger (10–14 weeks of gestation) before rupture occurs. Cornual pregnancy accounts for 2% to 4% of all ectopic pregnancies and can be difficult to identify by ultrasonography.
The classic clinical picture of ectopic pregnancy is a history of delayed menses, followed by abdominal pain and vaginal bleeding in a patient with known risk factors. Unfortunately, this history is neither sensitive nor specific. Risk factors for ectopic pregnancy are absent in almost half of patients. Of patients with symptomatic ectopic pregnancy, 15% to 20% have not missed a menstrual period, and occasionally the patient has no history of vaginal bleeding. Abdominal pain varies and can be described as crampy, intermittent, severe, or even absent.
The physical findings in ectopic pregnancy are likewise variable. Vaginal bleeding, uterine or adnexal tenderness, or both in the patient with a positive pregnancy test result should trigger consideration of ectopic pregnancy. Tachycardia is not always present, even with significant hemoperitoneum; the hemoglobin level is usually normal, and hypotension may be seen. The presence of peritoneal signs, cervical motion tenderness, or lateral or bilateral abdominal or pelvic tenderness indicates an increased likelihood of ectopic pregnancy. If peritoneal irritation is present, pain can preclude an accurate bimanual examination. Adnexal masses are palpated in only 10% to 20% of patients with ectopic pregnancy.
Vaginal bleeding is often mild. Heavy bleeding with clots or tissue usually suggests a threatened or incomplete miscarriage, although the patient with an ectopic pregnancy who has decreasing hormonal levels may experience endometrial sloughing, which can be mistaken for passage of fetal tissue. Passed tissue should be examined, as with cases of miscarriage, in tap water or saline (or under low-power microscopy). Unless fetal parts or chorionic villi are seen, ectopic pregnancy should not be excluded in the patient with bleeding or passage of tissue.
The spectrum of clinical presentations in ectopic pregnancy is wide, so the differential diagnosis includes essentially all first-trimester complications. Threatened miscarriage, the most common alternative diagnosis, can be recognized by sonographic evidence of an IUP, healthy or failed. Hypovolemia may be seen, particularly in incomplete miscarriage, but hypotension without significant vaginal hemorrhage is highly suggestive of ectopic pregnancy. Identification of fetal parts or chorionic villi in tissue expelled or obtained during D&C is useful to confirm a complication of IUP, although this is not sufficient to exclude ectopic pregnancy in a patient with an increased risk of heterotopic gestation, such as the patient undergoing assisted reproduction treatment.
A ruptured corpus luteum cyst should also be considered in the first trimester when bleeding is associated with peritoneal pain or irritation. The corpus luteum normally supports the pregnancy during the first 7 or 8 weeks. Rupture causes pelvic pain and peritoneal irritation. Ultrasonography is helpful if it reveals an IUP (except in patients with in vitro fertilization). During early gestation, when ultrasonography is nondiagnostic, free fluid is usually visible by ultrasonography, and serial observation may be required. If the patient is unstable, especially if an IUP cannot be identified by ultrasonography, laparoscopy or in rare cases laparotomy, may be required to differentiate between the two conditions.
Because the history and physical examination of the patient with ectopic pregnancy are insensitive and nonspecific, ancillary studies are essential to locate the pregnancy in any patient who has abdominal pain or vaginal bleeding and a positive pregnancy test result. Ultrasonography and hormonal assays are the most commonly used ancillary tests and advances in these technologies have allowed for more accurate detection and exclusion of ectopic pregnancy in patients with first trimester bleeding or pelvic pain. Laparoscopy may be the most efficient diagnostic tool in the hemodynamically unstable patient.
Ultrasonography is the primary method used to locate early gestation, establish gestational age, and assess fetal viability. Transabdominal ultrasonography is most useful for identification of IUPs with fetal cardiac activity and exclusion of ectopic pregnancy, except in patients at high risk for heterotopic pregnancy because of infertility procedures. Transvaginal ultrasonography is more sensitive, recognizes IUP earlier than transabdominal ultrasonography, and is diagnostic in up to 80% of stable patients presenting in the first trimester. Transvaginal ultrasound requires operator training and can be limited by device availability and quality.
As home pregnancy tests and access to early ultrasound becomes more prominent, the risk of having a “pregnancy of unknown location” by ultrasound rises. Indeterminate sonograms, which demonstrate neither an IUP nor extrauterine findings suggestive of ectopic pregnancy, occur in approximately 20% of ED evaluations of women with first-trimester bleeding or pain. Ectopic pregnancy is more likely among this subgroup with indeterminate sonograms if the hCG level is less than 1000 mIU/mL and the uterus is empty. Endometrial debris and fluid in the uterus do not exclude ectopic pregnancy.
An indeterminate ultrasound study usually does not result in a diagnosis of normal pregnancy. In one series of more than 1000 pelvic ultrasound examinations, 53% of indeterminate ultrasound studies resulted in a diagnosis of embryonic demise, 15% were ectopic pregnancies, and only 29% had an IUP. However, correlation of sonographic results with quantitative hCG measurements can add to the predictive value. With no intrauterine pregnancy on transvaginal ultrasound and hCG greater than 1500 mIU/mL, ectopic pregnancy should be suspected keeping in mind that ectopic pregnancies can be discovered at any level of hCG. Normal pregnancy is unlikely if no gestational sac is seen by transvaginal ultrasonography with an hCG level higher than 1000 to 2000 mIU/mL, depending on the institution’s discriminatory zone. Additional ultrasound findings that may predict early spontaneous abortion include crown rump length or fetal heart rate below the fifth percentile and fetal heart rate below 130 beats per minute. The differential diagnosis includes miscarriage and ectopic pregnancy in these patients. Unfortunately, levels of approximately 1500 mIU/mL develop in only approximately 50% of patients with ectopic pregnancies (see Table 173.1 ). Sonographic findings in a patient with suspected ectopic pregnancy are listed in Box 173.3 and illustrated in Figures 173.1 to 173.5 .
“Double” gestational sac
Intrauterine fetal pole or yolk sac
Intrauterine fetal heart activity
Pregnancy in fallopian tube (see Fig. 173.1 )
Ectopic fetal heart activity (see Fig. 173.2 )
Ectopic fetal pole
Moderate or large cul-de-sac fluid without intrauterine pregnancy
Adnexal mass without intrauterine pregnancy a
A complex mass is the most suggestive of ectopic pregnancy, but a cyst can also be seen with ectopic pregnancy.
Empty uterus (see Fig. 173.3 )
Nonspecific fluid collections (see Fig. 173.4 )
Echogenic material
Abnormal sac (see Fig. 173.5 )
Single gestational sac
Quantitative hCG levels serve two primary functions—serial levels can be used in the stable patient who can be observed as an outpatient, and a single level can be correlated with sonographic results for improved interpretation. Beginning 8 or 9 days after ovulation, serum hCG levels normally double every 1.8 to 3 days for the first 6 or 7 weeks of pregnancy. An initial quantitative level can be measured at the time of the ED visit, particularly if the sonogram is indeterminate or gestational age is estimated as less than 6 weeks. A repeated level should be measured 48 to 72 hours later. A doubling or rise of hCG by 66% generally indicates a viable intrauterine pregnancy, however approximately 15% of normal IUPs have a minimal rise in hCG, requiring a third serial test. A rapid decline in hCG tends to indicate miscarriage whereas a slow decline can indicate an ectopic pregnancy.
Single quantitative hCG levels can also be useful in conjunction with ultrasonography; normal IUPs should be visible transvaginally at 1000 to 2000 mIU/mL hCG or higher (see Table 173.1 ). A benign course for ectopic pregnancy cannot be assumed with low hCG levels. Ruptured ectopic pregnancies requiring surgery have been reported with very low or absent levels of hCG.
Serum progesterone levels have been studied as an additional or alternative marker to determine which patients need further evaluation and follow-up for possible ectopic pregnancy though it is not a standard tool in the ED. The progesterone level rises earlier than the hCG level in normal pregnancy and plateaus with levels higher than 20 ng/mL, so measurement of serial levels over time is not necessary. Levels below 5 ng/mL exclude a viable IUP with rare exceptions and can be used in combination with serum hCG ratios at presentation and 48 hours. Patients determined to have low risk of ectopic pregnancy by this algorithm can avoid additional testing and be managed conservatively. A progesterone level should be sent when the hCG levels are low, ultrasonography is indeterminate, and the emergency clinician is considering consultation for D&C or laparoscopy.
Dilation and evacuation can be used in patients without a viable IUP or ectopic pregnancy on ultrasonography to differentiate intrauterine miscarriage from ectopic pregnancy. Identification of chorionic villi in endometrial samples is seen in approximately 70% of patients and excludes ectopic pregnancy, except in patients undergoing assisted reproduction. Identification of chorionic villi can be made, even in 50% of women with an empty uterus on ultrasonography, and limits the need for laparoscopy to exclude ectopic pregnancy in this population.
Although it is invasive, laparoscopy is extremely accurate as a diagnostic (and therapeutic) procedure for possible ectopic pregnancy. It is the diagnostic treatment of choice in unstable first-trimester patients with peritoneal signs and is also indicated in patients with peritoneal fluid or an ectopic gestation in the pelvic cavity. Medical alternatives for the management of ectopic pregnancy have resulted in decreased indications for laparoscopy in stable patients.
Approximately 20% of women with ectopic pregnancies manifest signs and symptoms warranting immediate intervention. This includes patients with hypovolemia, large amounts of peritoneal fluid, or an open cervical os. For patients with signs of hypovolemia, rapid volume resuscitation should be instituted with intravenous (IV) fluids and blood products as necessary, and a baseline hemoglobin level and type and crossmatch should be obtained. If the patient remains unstable, immediate surgery is warranted. Laparoscopy may be indicated for patients who stabilize with treatment or those who are hemodynamically stable but exhibit peritoneal signs on abdominal examination. One study has reported that identification of free fluid in the Morison pouch on bedside ultrasonography predicts the need for operative intervention in most cases in patients with suspected ectopic pregnancies. A D&C or evacuation procedure with examination of the endometrial contents for products of conception can be performed urgently in the unstable patient with an open cervical os. All patients with ectopic pregnancy who are Rh-negative should be given Rh immune globulin, 50 μg intramuscularly.
In stable patients with first-trimester bleeding, the goal is to exclude ectopic pregnancy in a timely manner. In the patient with pain by history or examination or risk factors for ectopic pregnancy, ultrasonography should be performed before discharge.
In low-risk patients with minor symptoms or bleeding, ectopic pregnancy is still a possibility. In most cases, ultrasonography is the initial screening tool because it provides the most accurate and rapid information ( Fig. 173.6 ). If an IUP is not seen, quantitative hCG levels help risk stratify these patients. In all cases, if the patient is discharged, careful instructions are given for symptoms that would require her earlier return. An alternative strategy uses hCG levels first. However, waiting times for the serum assay can increase ED length of stay. In addition, ultrasonography is usually diagnostic of IUP or ectopic pregnancy, even if the hCG level is less than 1000 mIU/mL.
A minority of patients have indeterminate sonographic results and hCG levels below 1000 mIU/mL. When the hCG levels never rise to the discriminatory zone, the differential diagnosis includes intrauterine fetal demise and ectopic pregnancy. Early D&C with identification of the products of conception can be useful in the patient with nonrising hCG levels to detect chorionic villi and confirm a failed IUP or strongly suggest ectopic pregnancy. Alternatively, hCG levels can be followed until they reach zero, particularly if initial levels are low.
Although laparotomy may be required for patients who have an ectopic pregnancy, an increasing number of surgeries are being performed through the laparoscope. Salpingostomy is preferred to salpingectomy if the patient is stable and the procedure is technically feasible. Overall, the advent of transvaginal ultrasonography has resulted in earlier diagnosis and a trend toward nonoperative management.
Medical management is a safe and cost-effective treatment for the stable patient with minimal symptoms, especially when future fertility is desired. Methotrexate (50 mg/m2 IM or 1 mg/kg IM, alternating with folinic acid) is the drug most commonly used to treat early ectopic pregnancy. It interferes with fetal DNA synthesis and causes destruction of rapidly dividing fetal cells and involution of the pregnancy. Medical treatment is used most often for patients who are hemodynamically stable, with a tubal mass smaller than 3.5 cm in diameter, no fetal cardiac activity, and no sonographic evidence of rupture. Although there is no agreed on hCG cutoff for single-dose methotrexate, studies have suggested that increasing hCG levels are significantly correlated with methotrexate failure. Medical therapies are associated with an 85% to 93% success rate, with no significant difference between single- and multiple-dose protocols. Pelvic pain is common in patients receiving methotrexate (60%), even when it is used successfully. Indications of methotrexate failure and need for rescue surgery include decreasing hemoglobin levels, significant pelvic fluid, and unstable vital signs. All patients receiving methotrexate require close follow-up until the hCG level reaches 0, which may take 2 or 3 months.
Molar pregnancy, also known as a hydatidiform mole, comprises a spectrum of diseases characterized by disordered proliferation of chorionic villi. In the absence of fetal tissue, the pregnancy is termed a complete hydatidiform mole . Complete moles are caused by the fertilization of an ovum without maternal DNA and the subsequent duplication of the haploid genome. The term incomplete mole refers to a mole that is caused by the fertilization of a normal ovum by two sperm. The duplication of the triploid karyotype causes some fetal tissue to be present, along with focal trophoblastic hyperplasia. In approximately 19% of molar pregnancies, neoplastic gestational disease develops, with persistence of molar tissue after the pregnancy has been evacuated. Metastatic diseases can develop, requiring chemotherapy and intensive oncologic management.
Early molar pregnancy is usually not clinically apparent. The most well-described risk factor for the development of a molar pregnancy is extreme maternal age. Many patients present with abdominal pain, nausea and vomiting, or vaginal bleeding, and it may be difficult to differentiate these patients from those with threatened miscarriage or ectopic pregnancy by historical features alone. Patients sometimes seek treatment for apparent persistent hyperemesis gravidarum from high circulating levels of hCG, bleeding or intermittent bloody discharge, or respiratory distress; failure to hear fetal heart tones during the second trimester is the usual initial clue to diagnosis. If molar pregnancy spontaneously aborts, it is usually in the second trimester (before 20 weeks), and the patient or physician may note the passage of grapelike hydatid vesicles. Uterine size is larger than expected by date (by >4 weeks) in approximately 30% to 40% of patients. Theca lutein cysts may be present on the ovaries as a result of excessive hormonal stimulation, and torsion of affected ovaries can be seen.
The characteristic sonographic appearance of hydropic vesicles within the uterus, described as a snowstorm appearance, is highly suggestive of a diagnosis of molar pregnancy ( Fig. 173.7 ). Alternatively, cystic changes are seen in partial molar pregnancies. In some cases, a partial molar pregnancy is detected only on pathologic examination of abortion specimens. Complications of molar pregnancy include preeclampsia or eclampsia, which can develop before 24 weeks of gestation, respiratory failure or distress from pulmonary embolization of trophoblastic cells, hyperemesis gravidarum, and uterine bleeding. Ultrasonography usually provides the diagnosis of a complete molar pregnancy in the second-trimester patient who has “threatened miscarriage” or during sonographic assessment for fetal well-being and size. However, ultrasonography is only 58% sensitive, and diagnosis of a partial mole is made in only 17% of cases. Up to two-thirds of molar pregnancies are diagnosed by pathologic specimens after miscarriage.
Molar pregnancies are managed with uterine dilation and curettage (D&C). Following evacuation of a molar pregnancy, patients must be monitored in the outpatient setting for trophoblastic sequelae. Patients are at increased risk of an invasive mole, a benign tumor that invades the uterine wall and metastasizes to the lungs or vagina, or choriocarcinoma, a malignant tumor that invades the uterine wall and metastasizes to the lungs, brain, and liver via the patient’s vasculature. Patients who present to the ED with complications of bleeding metastases are managed with a combination of chemotherapy, radiation, and surgery.
Bleeding during the second trimester (14 to 24 weeks) is not benign and is associated with a 33% risk of fetal loss. Management is supportive and expectant because fetal rescue is impossible at this level of fetal immaturity.
The major conditions associated with vaginal bleeding in the second half of pregnancy include abruptio placentae and placenta previa. Patient history, physical examination, and results of ultrasonography can be used to distinguish them.
All patients with painless, second-trimester vaginal bleeding should be assumed to have placenta previa until proven otherwise. Digital or instrumental probing of the cervix should be avoided until the diagnosis has been excluded via ultrasound.
Abruptio placentae consists of a wide spectrum of severity of symptoms and risk. Up to 20% of women will have no pain or vaginal bleeding. Assessment is generally based on clinical features, coagulation parameters, and signs of fetal distress.
Bleeding during the second half of pregnancy occurs in approximately 4% of pregnancies. Only 20% of miscarriages occur after the first trimester, and the most important differential diagnoses after 12 to 14 weeks of gestation are abruptio placentae and placenta previa. The cause is often not determined, although occult marginal placental separations, which can be recognized only by placental inspection at delivery, are believed to come from a common source of bleeding above the cervix. Other causes of late vaginal bleeding include early labor, various cervical and vaginal lesions, lower genital tract infections, and hemorrhoids.
Bleeding during the second trimester before the fetus is potentially viable (14 to 24 weeks) is not benign. One-third of fetuses are ultimately lost when maternal bleeding occurs. Management is supportive and expectant because fetal rescue is not possible at this level of fetal immaturity. In the third trimester, vaginal bleeding is still associated with significant morbidity in approximately one-third of women; treatment includes consideration of urgent delivery.
Abruptio placentae is a separation of the placenta from the uterine wall and complicates roughly 1% of pregnancies. Small subclinical or marginal separations may go undetected until the placenta is examined at delivery and probably account for many of the other self-limited episodes of bleeding for which no diagnosis is made. In cases of nontraumatic abruptio placentae, spontaneous hemorrhage into the decidua basalis occurs, causing separation and compression of the adjacent placenta. Small amounts of bleeding may be asymptomatic and remain undetected until delivery. In other cases, the hematoma expands and extends the dissection. Bleeding may be concealed or may be clinically apparent if dissection occurs along the uterine wall and through the cervix. Placental separation may be acute or may be an indolent problem throughout late pregnancy.
Abruptio placentae is most clearly associated with maternal hypertension and preeclampsia. It is also more common with maternal age younger than 20 or older than 35 years of age, parity of three or more, unexplained infertility, history of smoking, thrombophilia, prior miscarriage, prior abruptio placentae, and cocaine use. Placental separation can also be associated with blunt trauma to the abdomen. In such cases, the cause appears to be shearing of a nonelastic placenta from the easily distorted elastic uterine wall at the time of traumatic impact. Intimate partner violence affects 4% to 8% of pregnancy and torso injuries are reported in 21.5% of cases placing the patient at risk for premature labor, abruption, uterine rupture, or fetal death. A significant etiology of trauma in pregnancy is motor vehicle accidents in which placental abruption may complication up to 40% of patients severely injured in a motor vehicle accident.
Vaginal bleeding occurs in 70% of patients with abruptio placentae. Blood is characteristically dark and the amount is often insignificant, although the mother may have hemodynamic evidence of blood loss. Uterine tenderness or pain is seen in approximately two-thirds of women; uterine irritability or contractions are seen in one-third. With significant placental separation, fetal distress occurs and the maternal coagulation cascade may be triggered, causing disseminated intravascular coagulation (DIC).
There is a wide spectrum of severity of symptoms and risk in placental separation. About 10% of women will present only with occult bleeding. Assessment is generally based on clinical features, coagulation parameters, and signs of fetal distress. Mild abruption is characterized by slight vaginal bleeding, little or no uterine irritability, absence of signs of fetal distress, and normal coagulation. As the separation becomes more extensive, it is associated with increased vaginal bleeding (or hidden maternal blood loss), increased uterine irritability with or without tetanic contractions, declining fibrinogen levels, evidence of fetal distress, and maternal tachycardia. In severe abruptio placentae (15% of cases), the uterus is tetanically contracted and very painful, maternal hypotension results from visible or concealed uterine blood loss, fibrinogen levels are less than 150 mg/dL, and fetal death can occur. Ultrasonography is insensitive in the diagnosis of abruptio placentae, often because the echogenicity of fresh blood is similar to that of the placenta. Symptomatic or even fetus-threatening abruption can occur in the presence of a normal sonogram.
Fetal distress and death occur in approximately 15% of patients with abruptio placentae by interruption of placental blood and oxygen flow. Risk of fetal death increases in proportion to the percentage of the placental surface involved and rapidity of separation. Fetal distress may result from the loss of placental blood flow, associated maternal hemorrhage (into the uterine cavity or externally), increased uterine tone, or resultant DIC. Maternal death can result, usually from coagulopathy or exsanguination. Fetomaternal transfusion can occur. Placental separation also predisposes the mother to amniotic fluid embolism.
The main alternative diagnosis in the woman with late-pregnancy bleeding is placenta previa, which is usually associated with painless, bright red bleeding and is excluded with ultrasonography. Lower genital tract or rectal lesions and blood-tinged cervical mucous plug are also considerations.
In the patient with abdominal pain but no vaginal bleeding, abruptio placentae with concealed hemorrhage must be distinguished from other causes of abdominal pain in later pregnancy—complications of preeclampsia, pyelonephritis, various liver diseases, gallbladder disease, appendicitis, and ovarian torsion. Uterine irritability caused by abruptio placentae can also be confused with early labor. If the patient has acute catastrophic hypotension, amniotic fluid embolus, with or without abruptio placentae, and uterine rupture must be considered.
Placenta previa, or implantation of the placenta over the cervical os, is the other major cause of bleeding episodes during the second half of pregnancy. The risk of placenta previa is increased with maternal age, smoking, multiparity, cesarean section, prior miscarriage or induced abortions, and preterm labor. Bleeding occurs when marginal placental vessels implanted in the lower uterine segment are torn, either as the lower uterine wall elongates or with cervical dilation near the time of delivery. Early bleeding episodes tend to be self-limited unless separation of the placental margin is aggravated by iatrogenic cervical probing or the onset of labor.
Painless, fresh vaginal bleeding is the most common symptom of placenta previa. In approximately 20% of cases, some degree of uterine irritability is present, but this is generally minor. Vaginal examination usually reveals bright red blood from the cervical os. All patients with painless, second-trimester vaginal bleeding should be assumed to have placenta previa until proven otherwise. Digital or instrumental probing of the cervix should be avoided until the diagnosis is excluded via ultrasound because this can precipitate severe hemorrhage in a patient with asymptomatic or minimally symptomatic placenta previa. Speculum examination of the vagina and cervix should be limited to an atraumatic partial speculum insertion to identify whether the bleeding is coming from the cervical os (and a presumed placenta previa), hemorrhoids, or a vaginal lesion that might not require urgent management.
Most cases of placenta previa identified during the mid-trimester resolve by the time of delivery as the lower uterine segment elongates and the placenta no longer overlaps the cervical os. Central or total previa, which occurs in approximately 20% of cases, can, however, cause severe hemorrhage, with the risk of exsanguination for the fetus and mother.
Ultrasonography is the diagnostic procedure of choice for localization of the placenta and diagnosis of placenta previa. Accuracy is excellent, but visualization of the placenta and of the internal cervical os is required. The bladder should be emptied before examination for suspected placenta previa to avoid overdiagnosis of placenta previa. Transvaginal ultrasonography is safe and even more accurate for visualization of the relationships between the placenta and internal os.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here