Epidemiology, clinical manifestations, and diagnosis of tubo-ovarian abscess

INTRODUCTION — The female genital tract may be affected by ascending infection, either from sexually transmitted infections or vaginal flora, or occasionally from secondary infections from gastrointestinal sources. Pelvic inflammatory disease (PID) refers to acute and subclinical infection of the upper genital tract, involving any or all of the uterus, fallopian tubes, and ovaries; this is often accompanied by involvement of the neighboring pelvic organs. It results in endometritis, salpingitis, oophoritis, peritonitis, perihepatitis, and/or tubo-ovarian abscess (TOA).

A complication of PID may be a TOA, which is an inflammatory mass involving the fallopian tube, ovary, and, occasionally, other adjacent pelvic organs (eg, bowel, bladder) [1]. This may manifest as a tubo-ovarian complex (an agglutination of those structures) or a collection of pus (TOA). These abscesses are found most commonly in reproductive-age patients and typically result from upper genital tract infection. TOA typically occurs as a complication of PID.

A TOA or complex is a serious and potentially life-threatening condition. Aggressive medical and/or surgical therapy is required, and rupture of an abscess may result in sepsis. The mortality rate associated with TOA was approximately 50 percent or higher prior to the advent of broad-spectrum antibiotics and modern surgical practice [2,3]. In current practice, the mortality rate approaches zero for abscesses that have not ruptured. Current mortality rates for patients with ruptured abscesses are not reported in the literature; data from the 1960s suggested a mortality rate ranging from 1.7 to 3.7 percent [2,4,5].

The epidemiology, clinical manifestations, and diagnosis of TOA are reviewed here. The management options for TOA are reviewed elsewhere. (See "Management and complications of tubo-ovarian abscess".)

Other manifestations of PID are discussed separately.

●(See "Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors".)

●(See "Pelvic inflammatory disease: Clinical manifestations and diagnosis".)

●(See "Pelvic inflammatory disease: Long-term complications".)

●(See "Pelvic inflammatory disease: Treatment in adults and adolescents".)

EPIDEMIOLOGY AND RISK FACTORS — There are few data regarding the incidence of TOA. In the United States from 1983 to 2000, 200,000 females were hospitalized annually for pelvic inflammatory disease (PID) [6]. It is important to note that not all cases of TOA are associated with PID, so some cases may be missed in studies based on PID diagnostic codes. Nevertheless, since approximately one-third of patients who are hospitalized with a diagnosis of PID are found to have a TOA, it can be extrapolated that approximately 66,000 cases of TOA occur annually in the United States [7-9].

The incidence of TOA may be increasing. A study from Norway reported that the number of patients admitted with a diagnosis of PID decreased between the time periods of 1990 to 1992 and 2000 to 2002, but a higher proportion of those patients had TOAs (increased from 26 to 43 percent) [9]. The cause of this increase is not clear, but it may be related to changing practice patterns in the treatment of PID, with a shift from inpatient to outpatient. In current practice, patients admitted to the hospital generally have more severe disease, which may include TOA.

Patients with a TOA are most likely to be between the ages of 15 and 40 years, but age should not exclude the diagnosis.

The risk factors for TOA are the same as for PID and include the following: multiple sexual partners, age between 15 and 25 years, and a prior history of PID. Several cases of TOA following oocyte retrieval for in vitro fertilization have been reported [10-12]. It is important to note, however, that a minority of patients lacking traditional risk factors for PID can and will be diagnosed with a TOA. A clear delineation of the prevalence of this finding or exact demographic profile that predicts that finding is, however, not available. (See "Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors", section on 'Risk factors'.)

Modern intrauterine devices (IUD) cause little, if any, increased risk for PID. The use of an IUD has been suggested as a risk factor for the development of a TOA (specifically unilateral TOA), but this association has not been substantiated by well-designed studies [13-16]. The slight elevated risk of developing PID among users of modern IUDs appears to be primarily limited to the first three weeks after IUD insertion. (See "Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors", section on 'Intrauterine device and tubal ligation'.)

Some data suggest that patients with HIV infection are more likely to develop a TOA than non-infected patients [17,18].

PATHOGENESIS

Progression of infection — The mechanism of the formation of a TOA has not been fully elucidated. TOAs most frequently (but not exclusively) result from upper genital tract infection, usually pelvic inflammatory disease (PID (figure 1)).

A TOA may also arise from local spread of infection associated with uncontrolled inflammatory diseases of the bowel [19,20], appendicitis, adnexal surgery, or, occasionally, from hematologic spread. Pelvic abscesses that do not involve the tubo-ovarian complex often differ in etiology and management from TOA and are not discussed here. Rarely, patients who have undergone previous hysterectomy (without adnexectomy) may still be at risk for the formation of a TOA, likely from local and/or hematogenous spread from nongynecologic organ systems [21]. (See "Posthysterectomy pelvic abscess".)

It is unclear why some patients with upper genital tract infection develop a TOA, while most do not [8,13,22]. In the majority of cases, the infection appears to originate within the lower genital tract from a sexually transmitted pathogen or from a patient's endogenous flora. Infection then ascends to the fallopian tubes. Ascending inflammation can progress to cause not only damage to the endothelium of the fallopian tube (destroying both the secretory and ciliated cells) but also edema of the infundibulum of the tube. This results in tubal blockage due to clubbing, ischemia, and necrosis, which contributes to the development of pyosalpinx. Prior infection may also result in abnormal tubal architecture that predisposes to TOA.

Delay in diagnosis because of a failure to recognize or report symptoms of PID does not appear to offer a complete explanation for progression to TOA. This is supported by the fact that a sizable percentage of patients with TOA do not report specific antecedent acute symptoms directly attributable to PID [8].

TOAs have traditionally been classified into either tubo-ovarian complexes (an agglutination of pelvic organs with or without bowel) or a collection of pus that can be approached for drainage. Different treatment approaches are required depending upon these two forms (eg, drainage is typically feasible only for collections).

If TOA remains untreated, uncontrolled tubal infection with associated tissue invasion and destruction typically produce a copious purulent exudate with gradually increasing tissue edema. This edema in turn may compromise local blood supply and result in tissue necrosis. Tubal structures may adhere to and coalesce with the adjacent ovarian tissue (and often adjacent nongenital tract tissues) to form a complex mass. Necrosis inside this complex mass may result in one or more abscess cavities with an anaerobic environment that allows or even promotes the growth of numerous anaerobic bacteria that were previously present in the endogenous pelvic flora [8,13]. In addition, worsening infection may result in sepsis.

Microbiology — TOAs are typically polymicrobial. The bacteria found in TOA are often similar to those found in patients with uncomplicated PID. There is also some overlap between the organisms commonly isolated in patients with bacterial vaginosis (BV) and those isolated from patients with TOA. While ascension of organisms associated with BV into the upper genital tract may theoretically contribute to the development of a TOA [23], this link has not been confirmed. (See "Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors", section on 'Microbiology' and "Bacterial vaginosis: Clinical manifestations and diagnosis", section on 'Pathogenesis and microbiology'.)

Mixtures of aerobic, facultative, and anaerobic organisms have been isolated from TOA [7,8]. Common organisms include Escherichia coli, aerobic streptococci, Bacteroides fragilis, Prevotella, and other anaerobes, such as Peptostreptococcus [7]. Scattered case reports highlight the potential for an occasional unusual pathogen to be isolated from a TOA, including Candida species, Pasteurella multocida, Salmonellae species, and Streptococcus pneumoniae [24-26]. Rarely, particularly in an immunocompromised patient and patients with HIV, Mycobacterium tuberculosis may also be causative of a TOA [27,28].

TOA that occurs in patients with an intrauterine device (IUD), typically long-term users, are often associated with a specific anaerobic pathogen, Actinomyces israelii [15,16]. The reason for the relationship between this potential pathogen and IUD users remains controversial. Management of patients with an IUD who develop a PID is discussed separately. (See "Intrauterine contraception: Management of side effects and complications".)

Neisseria gonorrhoeae and Chlamydia trachomatis are rarely isolated from the abscess cavity of a TOA. The role of these organisms appears to be limited to antecedent infections, such as cervicitis or PID. As an example, in a series of 232 patients with a TOA, N. gonorrhoeae was found in the endocervix in approximately 33 percent of cases but was present in less than 4 percent of the associated TOAs [7]. Some data suggest that N. gonorrhoeae facilitates invasion of the upper genital tract by lower genital tract flora, thus indirectly increasing the risk of progressive infection [29].

CLINICAL PRESENTATION — TOA is typically (but not exclusively) considered a complication of pelvic inflammatory disease (PID), and the classic presentation is the same as for PID, including acute lower abdominal pain, fever, chills, and vaginal discharge [7,8,22]. Most patients will not be overtly septic-appearing if their TOA is intact. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis", section on 'Clinical features'.)

However, the presentation of some patients with TOA differs from the classic scenario. Fever is not present in all patients, and some patients report only low-grade nocturnal fevers or chills. Also, not all patients present in an acute fashion. These variations in clinical presentation were illustrated in one of the largest series of patients with TOA (n = 175) [7]. As many as 40 percent of patients were afebrile upon presentation, 25 percent complained of chronic rather than acute abdominal pain, and 23 percent had normal white blood cell counts. Infrequently, patients with TOA (often not associated with PID) present with seemingly unrelated symptoms, such as diffuse persistent upper abdominal pain or a change in bowel habits.

Ruptured abscess — A "ruptured" TOA refers to an abscess that has begun to leak inflammatory contents into the abdominal cavity and thus requires immediate surgical exploration. Rupture is probably a misnomer, as the process is more like a leak, slow or fast; it is not typical for the abscess to burst open. Patients with a ruptured TOA in the large majority of cases present with an acute abdomen and signs of sepsis, although sepsis is not consistently present on retrospective evaluation at the time of surgery. Approximately 15 percent of patients with TOA present with signs and symptoms suggestive of a ruptured TOA. If this is suspected, surgical evaluation is recommended. (See 'Excluding sepsis' below and "Management and complications of tubo-ovarian abscess".)

EVALUATION OF PATIENTS WITH SUSPECTED TOA — The evaluation of patients with suspected TOA includes making the diagnosis of pelvic inflammatory disease (PID) and then pursuing further testing to evaluate for TOA. In addition, other etiologies of the clinical presentation should be excluded. (See 'Differential diagnosis' below.)

Establishing a diagnosis of PID — Patients with a presentation consistent with PID or TOA should be evaluated for PID with a complete history and pelvic examination. Laboratory testing includes a complete blood count and cervical testing for gonococcus and chlamydia. A pregnancy test should be performed since this may affect choice of antimicrobial therapy and clinical course.

The diagnosis and evaluation of PID are discussed in detail separately. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis".)

Further evaluation for TOA — TOA is a potentially life-threatening condition. Once patients have been diagnosed with PID, it is a clinical priority to decide whether they require further evaluation for TOA.

Further evaluation for TOA involves pelvic imaging. A very low threshold is warranted for performing imaging among all patients admitted with PID, given the serious nature of a TOA and the frequent inability to adequately assess for masses on pelvic examination due to pelvic tenderness. We routinely order imaging studies for patients with a diagnosis of PID who have one or more of the following characteristics:

●Acutely ill.

●Significant abdominopelvic tenderness precluding a complete pelvic examination.

●Adnexal mass noted on examination, particularly a tender mass [7,8,22].

●Lack of, or poor response to, antibiotic therapy – For patients who had earlier negative imaging studies for TOA (at least 48 to 72 hours previously) and continue to worsen on broad-spectrum antibiotic therapy, imaging may be repeated to evaluate again for a TOA.

Laboratory evaluation — Leukocytosis is found in most, but not all, patients with TOA [7,8,22]. Of all patients admitted with a diagnosis of PID, multiple studies have suggested that patients older than age 40 and those with higher laboratory markers of inflammation (white blood cell [WBC] count, erythrocyte sedimentation rate [ESR], and/or C-reactive protein [CRP]) are more likely to have a TOA versus PID alone [30-33]. As an example, in a prospective cohort study of 94 patients with PID, CRP levels were found to be a strong laboratory predictor of TOA, followed by WBC count and ESR, with CRP levels >49.3 mg/L suggesting the presence of TOA (85 percent sensitivity and 93 percent specificity) [33]. In addition, increasing CRP values during hospitalization were an indicator of failed medical therapy and the need for surgical intervention.

Imaging studies — The most useful and commonly performed studies in the evaluation for a TOA are pelvic ultrasonography or pelvic computed tomography (CT).

Choice of imaging modality is often institution specific, with radiologic potential for imaging-guided drainage procedures and local expertise being relevant considerations. When the clinical scenario suggests benefit from serial imaging, it is recommended to perform the same modality in succession, allowing for more objective comparisons over time.

Ultrasound is typically the first-line imaging study since it produces high-quality images of the upper genital tract, is less expensive than CT, and does not expose the patient to radiation. Ultrasound is also preferred when other genital tract pathology associated with pelvic pain must be excluded (eg, ovarian cyst, ectopic pregnancy, degenerating uterine fibroid).

CT is preferred in patients in whom pathology associated with the gastrointestinal tract must be excluded (eg, appendicitis, phlegmon, or abscess associated with inflammatory bowel disease). A few studies suggest use of CT may pose a slight advantage in terms of increased sensitivity to detect a TOA when compared with ultrasound (78 to 100 versus 75 to 82 percent, respectively). However, CT imaging is more costly than ultrasonography. In addition, the reported higher sensitivity of CT scan also assumes the patient had both intravenous and oral contrast, which may not be possible in ill patients for various reasons (eg, inability to tolerate oral intake of contrast) [22,34,35].

Ultrasonographic images of TOAs typically demonstrate complex multilocular masses that often obliterate the normal adnexal architecture and/or cul-de-sac anatomy (image 1). These masses also often appear to contain speckled fluid and internal echoes consistent with inflammatory debris [7,8,35]. This would be compared with a typical noninflammatory mass with clear fluid and a thin rim.

Specific findings noted on CT that are consistent with a TOA include thick-walled, rim-enhancing adnexal masses. Inflammatory masses are also often multilocular and may have an increased fluid density suggestive of purulent exudate [36]. CT imaging in some patients with TOA may show evidence of thickened fluid-filled tubes consistent with pyosalpinx, adjacent regional bowel thickening, and mesenteric stranding.

In addition, CT (and occasionally ultrasound) may show changes suggestive of abscess rupture, with free abdominopelvic thickened fluid (often pus) in the pelvis.

There are few data about other imaging modalities (eg, magnetic resonance imaging, radionuclide scanning) for the diagnosis of TOA. Such imaging modalities have limited use in routine clinical practice and are not routinely used.

Surgical evaluation — Surgical evaluation is frequently indicated for TOA (picture 1) in several clinical contexts, including:

●Abscess rupture – Abscess rupture is suspected due to a finding of an acute abdomen or of signs of sepsis. (See 'Ruptured abscess' above and "Evaluation and management of ruptured ovarian cyst", section on 'Infection or malignancy'.)

●Postmenopausal patients – A finding of a TOA in a postmenopausal patient often warrants surgical intervention since this may be associated with gynecologic malignancy [37-39]. In a systematic review including nine studies (199 patients) evaluating TOA in postmenopausal patients, the risk of malignancy ranged from 2.5 to 47 percent [40]; this risk may be lower than previously appreciated and suggests that, in some cases, conservative management may be appropriate [41].

A scoring system has been developed that can be used at the time of admission and can predict the likelihood of requiring surgical intervention, thus helping guide decision-making for these patients. This builds on previous data suggesting larger abscesses have a much higher chance of requiring surgical management and adds a few additional data points (age, WBC count) to the previously identified risks of larger abscess size and bilaterality. In this investigation, an abscess diameter of 7 cm or larger was most predictive of necessitating surgical intervention, with age, WBC count, and bilaterality also being important, but less so than abscess diameter [42]. Additionally, one study suggested that increased body mass index appeared to be a relevant predictor of failed medical management [43].

Surgical treatment of TOA is discussed in detail separately. (See "Management and complications of tubo-ovarian abscess", section on 'Drainage and surgery'.)

DIAGNOSIS — TOA is most often a clinical diagnosis based upon the finding of an inflammatory adnexal mass on pelvic imaging in a patient who meets diagnostic criteria for pelvic inflammatory disease (PID); treatment may be initiated based upon this clinical diagnosis. As noted above, there are other clinical scenarios that do not follow this typical picture of PID combined with an adnexal mass. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis", section on 'Diagnosis'.)

Purulent material obtained from a pelvic mass using imaging-guided drainage adds further support for the diagnosis of TOA by confirming the presence of an abscess, but the anatomic source is not always correctly identified (ie, TOA versus diverticular abscess).

A definitive diagnosis of TOA can be made only with direct visualization of the abscess during an invasive surgical procedure, such as laparoscopy or laparotomy.

DIFFERENTIAL DIAGNOSIS — The differential diagnosis for TOA includes a wide array of other conditions that may also cause lower abdominal or pelvic pain, similar to the differential diagnosis for pelvic inflammatory disease (PID (table 1)).

The differential diagnosis of the cause of lower abdominal pain in a patient includes diseases of the following organ systems:

●Reproductive tract – PID, ovarian mass, ruptured ovarian cyst, ovarian torsion, degenerating uterine fibroid, dysmenorrhea, ectopic pregnancy (including infected ectopic pregnancy), septic abortion, miscarriage.

●Gastrointestinal – Appendicitis, gastroenteritis, inflammatory bowel disease, irritable bowel syndrome, diverticulitis/diverticular abscess, constipation, cholecystitis, bowel perforation, colorectal cancer.

●Urinary tract – Cystitis, pyelonephritis, nephrolithiasis, urethritis.

The absence of fever (and/or the presence of other atypical presentations, as noted above) does not entirely exclude TOA since some patients with TOA are afebrile. (See 'Clinical presentation' above.)

Additional features that may help to exclude other causes include gastrointestinal symptomatology and related gastrointestinal imaging findings, as well as laboratory results from urinary tract evaluations. Occasionally, diagnostic laparoscopy is indicated to further clarify the diagnosis and offer therapy.

FURTHER EVALUATION TO DETERMINE MANAGEMENT — After making the diagnosis of TOA, evaluation of the severity of illness and determination of the etiologic organism help to guide management. Management is discussed in detail separately. (See "Management and complications of tubo-ovarian abscess".)

Excluding sepsis — Patients with TOA may present with or develop signs of sepsis, such as hemodynamic and/or respiratory instability as well as significant laboratory aberrations, etc. A complete evaluation should include vital signs, physical examination, and laboratory testing to exclude sepsis, particularly in ill-appearing patients. The diagnosis and evaluation of sepsis are discussed in detail separately. (See "Definition, classification, etiology, and pathophysiology of shock in adults".)

Determining choice of therapy — Blood cultures may aid choice of antimicrobial agents. For patients who undergo percutaneous drainage or surgery, fluid from the TOA cavity may be cultured. (See "Management and complications of tubo-ovarian abscess", section on 'Antibiotic therapy'.)

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

●A tubo-ovarian abscess (TOA) is an inflammatory mass involving the fallopian tube, ovary, and, occasionally, other adjacent pelvic organs (eg, bowel, bladder). This may manifest as a tubo-ovarian complex (an agglutination of those structures) or a collection of pus (TOA). (See 'Introduction' above.)

●TOAs are found most commonly in reproductive-age patients and typically result from upper genital tract infection. TOAs are typically polymicrobial. Mixtures of aerobic, facultative, and anaerobic organisms have been isolated from TOAs using modern culture techniques with attention to specific anaerobic methodology. Common organisms include Escherichia coli, aerobic streptococci, Bacteroides fragilis, Prevotella, and other anaerobes, such as Peptostreptococcus. (See 'Pathogenesis' above and 'Microbiology' above.)

●The classic presentation of TOA includes acute lower abdominal pain, fever, chills, and vaginal discharge. However, the presentation of many patients with TOA differs from this classic scenario. (See 'Clinical presentation' above.)

●Patients with a ruptured TOA classically present with an acute abdomen and signs of sepsis, although this is not invariably noted. A ruptured abscess may be life-threatening and requires immediate surgical exploration. (See 'Ruptured abscess' above.)

●TOA is most often a clinical diagnosis based upon the finding of an inflammatory adnexal mass on pelvic imaging in a patient who meets diagnostic criteria for pelvic inflammatory disease (PID). If a drainage procedure or surgery is performed, these may provide additional information to support or confirm the diagnosis. (See 'Diagnosis' above.)

●Imaging studies are a crucial component of the diagnosis of TOA. Ultrasound is typically the first-line imaging study. It is also useful for excluding other genital tract pathology (eg, ovarian cyst, ectopic pregnancy, degenerating uterine fibroid). Computed tomography may also be used, particularly when pathology associated with the gastrointestinal tract must be excluded (eg, appendicitis, phlegmon, or abscess associated with inflammatory bowel disease). (See 'Imaging studies' above.)

●Detecting TOA in patients with PID is a clinical priority. TOA is most commonly detected based on findings from pelvic imaging. We order imaging studies for patients with a diagnosis of PID who have the following characteristics: acutely ill; lack of, or poor response to, antibiotic therapy; adnexal mass noted on examination; or significant abdominopelvic tenderness precluding a complete pelvic examination. (See 'Further evaluation for TOA' above.)