Endometritis unrelated to pregnancy

INTRODUCTION — Endometritis refers to inflammation of the endometrium, the inner lining of the uterus. Pathologists have traditionally classified endometritis as either acute or chronic. Acute endometritis is characterized by the presence of microabscesses or neutrophils within the endometrial glands, while chronic endometritis is distinguished by variable numbers of plasma cells within the endometrial stroma [1,2].

In the absence of a tissue sample, other factors can help distinguish between acute and chronic endometrial inflammation. Symptoms alone are not useful since the clinical manifestations of both disorders are similar (abnormal vaginal bleeding and pelvic pain). An exception is fever: patients with acute endometritis frequently have fever, while it is less common in patients with the chronic process. Reviewing the clinical scenario is also helpful for determining whether endometritis is acute or chronic (table 1):

●Acute endometritis in the nonobstetric population is usually preceded by pelvic inflammatory disease (PID) either secondary to a sexually transmitted infection or an invasive gynecologic procedure.

●Chronic endometritis in the nonobstetric population can be due to a number of processes, including infections (eg, chlamydia, tuberculosis, other organisms related to cervicitis and PID), intrauterine foreign bodies or growths (eg, intrauterine contraception, submucous leiomyoma, polyp), and radiation therapy. No etiology is identifiable in approximately one-third of patients [3].

Endometritis unrelated to pregnancy will be discussed here. Endometritis in postpartum patients is reviewed separately. (See "Postpartum endometritis".)

ACUTE ENDOMETRITIS

Pelvic inflammatory disease — PID affects the upper genital tract. Most PID results from ascending migration of pathologic lower genital tract flora to the upper genital tract. Infection may be confined to the tubes (salpingitis) or may involve the ovaries (salpingo-oophoritis) and/or uterine cavity (endometritis). In general, the endometrium and ovaries are less susceptible to infection than the fallopian tube. However, either the endometrium (endometritis) or ovary (oophoritis) may occasionally be the sole focus of infection.

Epidemiology and pathogenesis — Concomitant endometritis occurs in 70 to 90 percent of laparoscopically documented cases of salpingitis. There is no correlation between the degree of tubal damage and the histopathologic severity of endometritis. Patients with signs and symptoms of PID, but who have no laparoscopic evidence of the disease, may have an isolated endometritis that is responsible for their symptoms [4,5].

Endometritis is strongly associated with chlamydial colonization or infection of the cervix [6] (see 'Chlamydia' below). Weaker, but positive, correlations exist between endometritis and cervical gonococcal infection and bacterial vaginosis [7-9]. Ascending infection has been suggested as an explanation for both the association among cervicitis, endometritis, and salpingitis, and the overlap in symptomatology between these infections. (See "Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors".)

Diagnosis — The diagnosis of acute endometritis is made clinically based on criteria for the diagnosis of acute PID. An endometrial biopsy for microbiologic and histologic studies may be helpful for the specific diagnosis of acute endometritis but is not necessary. (See "Endometrial sampling procedures".)

The overlap in symptoms between patients with laparoscopically documented PID and those with isolated endometritis is such that differential diagnosis by any specific clinical criterion is not possible. One large series of 152 patients with suspected PID who underwent both laparoscopy and endometrial biopsy found that 43 (28 percent) had neither endometritis nor salpingitis, 26 (17 percent) had isolated endometritis, and 83 (55 percent) had acute salpingitis [10]. Histologic endometritis was present in 85 percent of patients with confirmed acute salpingitis. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis".)

Although vaginal white blood cells are usually present in patients with histologic endometritis (sensitivity: 91 percent), many patients without endometritis will also have white blood cells noted on saline wet mount of vaginal discharge (specificity: 26 percent) [11]. Thus, this test is useful in excluding upper tract infection in the uncommon cases when it is negative (negative predictive value: 95 percent) and is seldom used.

Treatment — The therapy of acute endometritis related to acute PID is the same as the treatment of acute PID. (See "Pelvic inflammatory disease: Treatment in adults and adolescents".)

Outcome — PID is associated with increased risks of infertility, ectopic pregnancy, chronic pelvic pain, and recurrent PID (see "Pelvic inflammatory disease: Treatment in adults and adolescents"). The presence or absence of endometritis does not appear to be a factor for whether patients with a clinical diagnosis of PID who are treated with antibiotics subsequently achieve pregnancy [12].

Gynecologic procedures — Endometritis is an uncommon complication of transcervical diagnostic gynecologic procedures. As an example, there were only two cases of postoperative infection in a series of 927 diagnostic and operative hysteroscopies; both infections occurred in patients undergoing hysteroscopic myomectomy [13]. The most common pathogens are chlamydia and gonorrhea, but other cervicovaginal organisms can be involved.

Screening for chlamydia and gonorrhea before minimally invasive procedures confined to the endometrium, such as hysteroscopy, intrauterine contraception insertion, and endometrial biopsy, is not necessary. If cultures are done, appropriate treatment should be administered to patients with positive results, but prophylactic treatment in the absence of culture results, or in patients with negative cultures, is not indicated [14,15].

The American College of Obstetricians and Gynecologists does not recommend routine use of antibiotic prophylaxis prior to cervical tissue excision procedures (eg, cervical biopsy, endocervical curettage), endometrial biopsy, laparoscopic procedures without entry into bowel or vagina, hysteroscopy, intrauterine device insertion, and dilation and curettage for nonpregnancy indications because of the low risk of procedure-associated infection [15]. However, prophylactic antibiotics are recommended in the following situations:

●Prior to hysterosalpingography (HSG) and sonohysterography in patients with a history of pelvic infection – If dilated tubes are noted during the procedure, doxycycline 100 mg orally twice daily can be continued for five days to reduce the incidence of postprocedure PID. In patients with no history of pelvic infection, HSG and sonohysterography can be performed without prophylactic antibiotics. However, if the procedure shows dilated fallopian tubes, doxycycline 100 mg twice daily for five days is suggested. However, data supporting use of prophylactic antibiotics in patients with dilated tubes undergoing HSG are sparse [16].

●Before surgical termination of pregnancy – Randomized trials have consistently shown that antibiotic prophylaxis decreases the risk of postabortal infection [17]. This is discussed in detail separately. (See "First-trimester pregnancy termination: Uterine aspiration", section on 'Antibiotic prophylaxis' and "Second-trimester pregnancy termination: Dilation and evacuation", section on 'Prophylactic antibiotics'.)

CHRONIC ENDOMETRITIS

Unknown etiology — Chronic endometritis is detected histologically in approximately 8 percent of endometrial specimens [18]. There is no apparent etiology in as many as one-third of these patients, thus making idiopathic endometritis the most common cause of chronic endometritis in patients who are not pregnant/postpartum [3]. Chronic endometritis may be associated with adverse reproductive outcomes, such as implantation failure and recurrent miscarriage, and thus may be more prevalent in patients with infertility [19].

Clinical manifestations and diagnosis — Patients with symptomatic chronic endometritis usually present with abnormal uterine bleeding, which may consist of intermenstrual bleeding, spotting, postcoital bleeding, menorrhagia, or amenorrhea. Vague, crampy lower abdominal pain accompanies the bleeding or may occur alone. The most common finding on physical examination is uterine tenderness or cervical motion tenderness. However, in contrast to acute endometritis, many patients have no symptoms and a completely normal examination.

Patients with abnormal vaginal bleeding often undergo an endometrial biopsy as part of their diagnostic evaluation. In these patients, histology that reveals plasma cells in the endometrial stroma is diagnostic of chronic endometritis, but it cannot be assumed that this is the cause of the abnormal vaginal bleeding. An endometrial culture should be performed and may be positive for an infectious organism, despite negative endocervical cultures. The diagnosis of chronic endometritis of unknown etiology is made if culture results are negative, ultrasound does not reveal a submucous fibroid, and there is no history of radiation therapy or recent intrauterine contraception (see below). Although culture results are negative, these patients are more likely to have a history of genital tract infection than patients without chronic endometritis [20].

Hysteroscopy using fluid for distending the uterine cavity may be a useful and reliable technique for detecting chronic endometritis [21,22]. When hyperemia, mucosal edema, and micropolyps were used as diagnostic parameters, hysteroscopy showed a diagnostic accuracy of 93.4 percent. These authors found over 70 percent of chronic endometritis resulted from common bacteria and mycoplasma [23].

Treatment — Empiric antimicrobial therapy appears to improve symptoms and histology in some patients with chronic endometritis [24]. We suggest a course of doxycycline (100 mg orally twice daily for 10 to 14 days) for patients with chronic endometritis of unknown etiology [25]. The rationale is that negative cultures may represent false negative tests for detecting chlamydia [6] or the presence of other difficult to cultivate microbes, and doxycycline is active against some common bacteria and mycoplasma.

If the patient remains symptomatic after treatment with doxycycline and cultures are negative, the provider should investigate other etiologies of abnormal uterine bleeding and/or abdominal pain.

If the patient is allergic to doxycycline, we suggest a five-day course of azithromycin: 500 mg orally on the first day and then 250 mg orally on days 2 through 5. This provides coverage for chlamydia, common bacteria, and mycoplasma.

Chlamydia — Chlamydia trachomatis is an infectious cause of chronic endometritis in the setting of PID or cervicitis. In one study, as an example, C. trachomatis was detected using polymerase chain reaction (PCR) or immunohistochemistry in 24 percent of endometrial tissue samples with plasma cell endometritis compared with 4 percent of those with no evidence of plasma cell endometritis [6]. By comparison, another study which used PCR to detect C. trachomatis in 43 specimens of histopathologically diagnosed chronic endometritis detected C. trachomatis in only one case in which dense plasma cell infiltrates were present; concurrent C. trachomatis infection of the cervix was documented [26].

Symptoms of chronic chlamydial endometritis (intermenstrual bleeding, dyspareunia, and pelvic pain) are similar to those of chronic PID or chronic endometritis from other etiologies. However, the Centers for Disease Control and Prevention (CDC) minimal diagnostic criteria for PID appear less sensitive for diagnosis of chronic endometritis (sensitivity 33 percent) than for PID [27].

The recommended treatment is 10 to 14 days administration of broad-spectrum antibiotics effective against C. trachomatis and other probable pathogens. (See "Treatment of Chlamydia trachomatis infection".)

Tuberculous endometritis — Approximately one-third of the world's population is infected with Mycobacterium tuberculosis, and it is a leading cause of infertility in endemic countries. The global incidence of tuberculosis (TB) is growing at approximately 0.4 percent per year, and much faster in sub-Saharan Africa and in countries of the former Soviet Union [28]. TB is a relatively common cause of upper genital tract infection in patients from Nepal and Northern India [29]. However, it is a rare cause of upper genital tract infection in developed countries.

Epidemiology — Tuberculous endometritis most commonly occurs in young patients from countries where TB is endemic and among older patients exposed to TB before the advent of effective chemoprophylaxis. (See "Epidemiology of tuberculosis".)

Pathogenesis — Genital tract TB usually arises from hematogenous spread from a pulmonary or other nongenital tract focus. Inoculation of the genital tract from other intraperitoneal foci or from male sexual partners with tuberculous epididymitis is unusual. (See "Clinical manifestations, diagnosis, and treatment of miliary tuberculosis".)

The fallopian tube and endometrium are the most frequently affected sites in the female genital tract, although the cervix and ovary can also be involved [30]. The vagina and vulva are rarely infected. (See "Tuberculosis: Natural history, microbiology, and pathogenesis".)

Clinical manifestations — Genital tract TB usually results from reactivation of a focus of infection which originated at the time of initial infection; thus, clinical manifestations may not appear until 10 years after the initial seeding of the genital tract. Infertility is the chief presenting complaint among young patients with genital tract TB [30,31]. Genital involvement without renal TB is more common in females than in males [32]. Unusual patterns of vaginal bleeding, such as changes in menses, amenorrhea, and postmenopausal bleeding are other common presentations of disease. Approximately 25 to 35 percent of patients with pelvic TB complain of vague, chronic lower abdominal or pelvic pain. Tuberculous peritonitis and ascites, which are usually secondary to direct hematogenous seeding of the peritoneum, are infrequently observed.

Diagnosis — The diagnosis should be suspected in patients with symptoms suggestive of TB, residence in or travel to endemic areas, and a chest roentgenogram with evidence of healed pulmonary TB. However, a significant number of patients with pelvic TB may have a normal chest x-ray and no prior history of TB [33]. A skin test with a purified protein derivative of tuberculin is the best way to screen for asymptomatic infection. (See "Pulmonary tuberculosis: Clinical manifestations and complications" and "Tuberculosis infection (latent tuberculosis) in adults: Approach to diagnosis (screening)".)

A hysterosalpingogram performed as part of an infertility evaluation may show some or all of the characteristic changes of tuberculous infection: beading; sacculation; sinus formation; and rigid pipestem patterning.

The diagnosis should be confirmed by histologic examination of a surgical specimen (showing typical granuloma) or by an acid-fast stain or culture of menstrual blood or tissue from an endometrial biopsy. Menstrual fluid, which may be collected in a vaginal cup or cervical cap, is more sensitive than biopsy specimens for the diagnosis of tuberculous endometritis [31]. Culture and histology are also more sensitive than acid-fast stains. (See "Diagnosis of pulmonary tuberculosis in adults".)

Treatment — The treatment of genital tract TB is primarily medical and consists of combination drug therapy for 9 to 12 months. Surgery is indicated if the symptoms or the physical examination suggest persistence or increase in disease despite adequate therapy or if culture or biopsy results demonstrate resistant organisms. Total abdominal hysterectomy and bilateral salpingo-oophorectomy is the definitive surgical procedure for pelvic TB. However, early aggressive medical therapy may allow conservative procedures in patients who wish to attempt a future pregnancy. (See "Treatment of drug-susceptible pulmonary tuberculosis in nonpregnant adults without HIV infection".)

Intrauterine foreign objects, intrauterine growths, and radiation therapy — The copper intrauterine contraception induces a foreign body reaction in the endometrium; the resulting inflammatory response prevents viable sperm from reaching the fallopian tubes [34]. This reaction subsides after removal of the device. In addition, patients with noncopper intrauterine contraception who have discomforting symptoms (eg, irregular bleeding, crampy pain) are more likely to have histological evidence of endometritis compared with those who are asymptomatic [35].

Intrauterine growths (eg, submucous leiomyoma or polyps) in the uterine cavity and radiation therapy can also induce chronic endometrial inflammation [34,36]. Excision of the fibroid or polyp may lead to resolution.

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: Gynecologic infectious diseases (non-sexually transmitted)".)

SUMMARY AND RECOMMENDATIONS

●Classification – Pathologists have traditionally classified endometritis as either acute or chronic (table 1), which is based on histological findings.

Symptoms alone are not useful for distinguishing these entities since the clinical manifestations of both disorders are similar (abnormal uterine bleeding, pelvic pain, uterine tenderness). An exception is fever: patients with acute endometritis frequently have fever, while it is less common in patients with the chronic process. (See 'Introduction' above.)

●Acute endometritis – Acute endometritis in the nonobstetric population is a common component of acute pelvic inflammatory disease (PID); it is usually present along with acute salpingitis. (See 'Pelvic inflammatory disease' above.)

•Transcervical gynecologic procedures are an uncommon cause of acute endometritis. For this reason, we suggest not performing tests for chlamydia and gonorrhea before minimally invasive procedures confined to the endometrium (Grade 2C). (See 'Gynecologic procedures' above.)

•To prevent postprocedure endometritis, we recommend antibiotic prophylaxis before surgical termination of pregnancy (Grade 1A). We suggest its use before hysterosalpingography or sonohysterography in patients with a history of pelvic infection and after these procedures in those found to have hydrosalpinges (Grade 2C). (See 'Gynecologic procedures' above.)

●Chronic endometritis – As many as one-third of nonobstetric patients with chronic endometritis have no identifiable etiology; infection, intrauterine growths/foreign bodies, and radiation therapy account for the remainder. (See 'Chronic endometritis' above.)

•In contrast to acute endometritis, many patients with chronic endometritis have no symptoms and a completely normal examination. (See 'Clinical manifestations' above.)

•We suggest a course of doxycycline (100 mg orally twice daily for 10 days) for patients with chronic endometritis of unknown etiology (Grade 2C). (See 'Unknown etiology' above.)

•Targeted therapy is appropriate when the cause of endometritis is known. (See 'Chronic endometritis' above.)