Large tubo-ovarian abscess following retention of a Copper-T intrauterine device inserted ten years prior in an obese psychiatric patient

Introduction

Pelvic inflammatory disease (PID), also known as upper genital tract infection, including salpingitis and tubo-ovarian abscess, is associated with the infection caused by multiple pathogens, such as Neisseria gonorrhoeae, Chlamydia trachomatis and Trichomonas vaginalis. These microorganisms alter the normal immunological response, predisposing individuals to an inflammatory response due to potential pathogens such as Escherichia coli, Bacteroides spp. and Enterococcus faecalis. The risk factors for PID include positive cervical cultures for N. gonorrhoeae or C. trachomatis, younger age, lower socioeconomic status, substance abuse, other sexually transmitted infections, sexual partners with urethritis of gonorrhea, and the absence of mechanical and/or chemical contraceptive barriers. Previous diagnoses of PID also increase risk. Clinical forms of PID include silent, acute, chronic PID and tubo-ovarian abscess (TOA). Silent PID is an exclusion diagnosis, suspected in cases of tubal-factor infertility without a history of upper tract infection; adhesions and hydrosalpinx may be present, but most commonly, the fallopian tubes appear macroscopically normal. The clinical presentation of acute pelvic inflammatory disease includes uterine, adnexal or cervical tenderness, most commonly after menstruation. Other diagnostic indicators include fever, mucopurulent cervical discharge, cervical secretion rich in leukocytes, elevated inflammatory probes, and positive cervical cultures for N. gonorrhoeae and/or C. trachomatis. In acute salpingitis, sonographic findings may reveal a distended tube with anechoic or echogenic fluid inside, thickened fallopian walls, images of incomplete internal septa and a “cogwheel” appearance in cross section, often with positive color Doppler. Chronic PID is usually diagnosed histologically; hydrosalpinx and chronic pelvic pain may be present. A tubo-ovarian abscess occurs when the suppurated fallopian tube adheres to the ovary, typically presenting unilaterally, with potential involvement of the bladder, bowel and contralateral adnexa. Progression of the suppuration can cause abscess rupture and consecutive peritonitis⁽¹⁾.

Pelvic inflammatory disease guidelines from 2024 recommend as treatment for PID ceftriaxone i.v. or i.m. at 1 g as a single daily dose, combined with doxycycline i.v. at 100 mg twice daily (or orally if tolerated), followed by doxycycline orally at 100 mg, twice daily, plus metronidazole orally at 500 mg, twice daily, for a total of 14 days, or clindamycin intravaginally at 100 mg, once daily, for six days. An alternative parenteral regimen includes clindamycin i.v. at 900 mg, three times daily, plus gentamicin i.m. or i.v. at 3-6 mg/kg, once daily (with renal function monitoring), followed by clindamycin orally at 450 mg, four times daily, or doxycycline orally at 100 mg, twice daily, plus metronidazole orally at 500 mg, twice daily, for 14 days, or clindamycin intravaginally at 100 mg, once daily, for six days⁽²⁾.

Drainage of a tubo-ovarian abscess is indicated in cases with no improvement within 2-3 days of treatment after modifying the antimicrobial regimen, as well as in larger abscesses (more than 8 cm in diameter) as a first-line treatment, concurrently with antibiotics. Drainage can be performed with CT or sonographic guidance via transabdominal, transvaginal, transgluteal, or trans­rectal routes. Exploratory laparoscopy or laparotomy is not usually recommended in the initial management of TOA, although surgical treatment can be performed after achieving the “cold abscess” state⁽¹⁾.

Case report

This case report describes a 40-year-old woman, non-smoker, with a copper T intrauterine device (IUD) inserted ten years prior. She was referred to our hospital for hospitalization and management after being diagnosed in the outpatient setting with a left adnexal tumoral mass. Previously, she was admitted to a psychiatry hospital for a recurrent depressive episode. Her medical history includes bronchial asthma and class I obesity, with a Body Mass Index of 34 kg/m². She had one birth by caesarean section and four elective abortions.

The speculum examination revealed normal external genital organs and a cervix with intact epithelium, without macroscopic lesions; the IUD wire could not be identified, and there was no vaginal bleeding. Bimanual palpation indicated a normal-sized, mobile, and non-tender uterus, with supple adnexal zones; in the posterior cul-de-sac, a spherical tender tumoral mass measuring about 8 cm with reduced mobility and soft consistency could be palpated. Transvaginal ultrasound showed the uterus in anteversion flexion (AVF), measuring 72/68/42 mm, with a homogeneous myometrium and an 8-mm endometrium. In the Douglas pouch, an acoustically heterogeneous formation measuring 8.9/7.5 cm was noted, suggestive of a left tubo-ovarian abscess (Figure 1). The right adnexa appeared normal on ultrasound, with no anechoic fluid behind the uterus.
 

Blood tests showed leukocytosis (WBC) with neutrophilia (18,530/mm³ with 16,180/mm³), moderate normochromic normocytic anemia (hemoglobin 9.5 g/dL, secondary to the inflammatory process), normal platelet count (343,000/mm³), altered coagulation, and significantly elevated inflammatory markers (fibrinogen 1086 mg/dL, C-reactive protein 463 mg/L, procalcitonin 0.47 ng/mL). CA-125 was slightly elevated (37.6 compared to 35). The cervical cultures tested negative for Enterobacteriaceae, Staphylococcus aureus, Streptococcus spp., Enterococcus spp., and Pseudomonas spp. The patient was transferred to our hospital with the diagnosis of left adnexal tumoral mass and retained IUD.

During hospitalization, she received triple antibiotic therapy (doxycycline, cefuroxime, metronidazole) and anti-inflammatory treatment. On the sixth day, the IUD was removed under sedation. The dynamic of the laboratory test under medical treatment is presented in Table 1.
 

Abdomen and pelvis computed tomography with contrast revealed a left parauterine lesion measuring 86/71 mm, with fluid densities, a vascularized wall, a few septa, and fine densifications of the adjacent adipose tissue (possible left tubo-ovarian abscess), causing a compressive effect on the left ureter (resulting in second-grade ureterohydronephrosis and post-obstructive perfusion disorders of the left kidney). Scanned sections noted bilateral pleurisy (maximum 12 mm on the left side) and small lymph nodes bilaterally at the pulmonary hila, as well as infracarinal. On the eighth day of treatment, following IUD removal, the inflammatory markers showed little improvement (fibrinogen 799 mg/dL, CRP 212 mg/L), along with a rise in hemoglobin (9.7 g/dL), but with thrombocytosis (548,000/mm³) and a stationary leukocyte formula. The patient was discharged with recommendations to continue the antibiotic treatment for three more weeks at home and to return for reevaluation.

After almost one month of antibiotic therapy and IUD removal, the patient was readmitted to our hospital for surgical excision. Blood tests showed a normal leukocyte formula (WBC 8220/mm³, neutrophils 5670/mm³), mild normochromic normocytic anemia (Hb 11 g/dL), and normal platelet count (404,000/mm³). C-reactive protein levels were normal (4.62 mg/L), fibrinogen slightly above the upper limit (467 mg/dL), and ESR remained high at 113 mm/hour. The examination of vaginal secretions indicated frequent yeast cells, rare leukocytes, relatively frequent lactobacilli, and the absence of Neisseria gonorrhoeae and Trichomonas vaginalis. On transvaginal ultrasound, the uterus was in an intermediate position, measuring 82/44/53 mm, with a homogeneous myometrium and a 5-mm endometrium. In the projection zone of the left adnexa, there was an acoustically heterogeneous formation measuring 7/6.2 cm, suggestive of a left tubo-ovarian abscess. The right adnexa appeared normal on ultrasound, with no anechoic fluid behind the uterus (Figure 2).
 

After obtaining the patient’s informed consent, she was prepared for the surgical intervention. Under general anesthesia, exploratory laparotomy was performed. At the examination of the peritoneal cavity, the following were noted: the right adnexa appeared normal macroscopically, while the left adnexa was enlarged, approximately 8/6 cm, intensely adherent to the left uterine border, small intestine, rectum and visceral peritoneum, with the appearance of false membranes – meticulous viscerolysis was performed. Upon mobilizing the left adnexa, gray purulent fluid was spontaneously evacuated and collected for bacteriological examination (negative for Enterobacteriaceae, Staphylococcus aureus, Streptococcus spp., Enterococcus spp., Pseudomonas spp. and anaerobes). Left adnexectomy was performed following meticulous digital debridement up to the Douglas pouch, where the abscess had been extended (Figure 3). Lavage in abundance with betadine solution and pelvic drainage were conducted. The extracted specimens were sent for histopathological examination. The postoperative evolution was favorable under anticoagulant, antibiotic (ceftriaxone), anti-inflammatory and hydroelectrolytic rebalancing therapy. The dynamic of the laboratory test after surgical excision is presented in Table 2. The patient was discharged surgically cured on the fifth day.
 

The patient’s follow-up revealed a favorable evolution with a fully recovery, without complications or recurrence.

The histopathological examination of the surgical specimen revealed a salpingeal wall with marked diffuse polymorphous inflammatory infiltration, edema, and focal transwall hyperemia with areas of substance loss in the wall; fibrino-leukocytic deposits on the tubal serosa. The histopathological findings are compatible with the clinical diagnosis of a tubal abscess.

Discussion

The factors associated with the failure of antibiotic therapy in patients with tubo-ovarian abscesses include advanced age, larger abscess size, history of intrauterine device use, postmenopausal status, underlying medical conditions such as diabetes mellitus or pelvic inflammatory disease, the presence of fever, elevated white blood cell count, and increased levels of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP)⁽³⁾.

In a 2020 report analyzing 146 patients diagnosed with TOA, divided into two groups – those successfully treated with antibiotics alone and those requiring surgical treatment after antibiotic therapy failure –, the predictive risk factors for antibiotic treatment failure were examined. According to the study, the antibiotic treatment is likely to be ineffective in women over 41 years and 6 months of age, with a BMI over 26.72 kg/m², a CRP level above 143.5 mg/L at admission, and an abscess diameter greater than 6.25 cm⁽⁴⁾. In our case, three out of four criteria were above the cutoff level (BMI 34 kg/m², CRP 463 mg/L, abscess diameter 8.9 cm), thus predicting the failure of antibiotic treatment alone.

A 2022 retrospective study tracking the evolution of inflammatory markers and leukocyte formula after the surgical management of TOA highlighted the following findings. All inflammatory markers increased within the first 48 hours postoperatively and subsequently decreased over time. On average, WBC levels were the first to normalize postoperatively (2.5 days; 95% CI; 1-4.3), followed by the neutrophil-to-lymphocyte ratio (7.1 days; 95% CI; 4.7-10.8), and CRP (after more than 14 days). The study concluded that the CRP levels are not useful for assessing the postoperative success, as this inflammatory marker normalizes later, while WBC levels and the neutrophil-to-lymphocyte ratio are valuable in evaluating the operated TOA patient⁽⁵⁾. In our case, in the first day postoperatively the WBC count was 15,550/mm³ and the neutrophil-to-lymphocyte ratio was 11,440/2980 (3.83). In the third day, the WBC count was 7740/mm³, the neutrophil-to-lymphocyte ratio was 5350/1850 (2.89) and the CRP remained high at 40 mg/L.

A retrospective study from 2020 included 50 patients diagnosed with TOA between 2013 and 2017, divided into two groups: initial medical treatment (further subdivided into a successful medical group and late surgical treatment) and early surgical treatment. The medical treatment included triple antibiotic therapy (ceftriaxone 2 g/day, azithromycin 500 mg/day, and metronidazole 1000 mg/day). The surgical treatment was preferably performed via laparoscopic drainage, with other methods, including ultrasound-guided drainage, CT-guided drainage and laparotomy. The study showed that patients initially treated medically required late surgical intervention more frequently when the abscesses were larger (73.9 mm versus 53.5 mm in the successful medical group; p=0.19) and when the WBC and CRP values were higher (WBC 15,900/mm³ versus 13,500/mm³ in the successful medical group; p=1.00, and CRP 263.5 mg/L versus 171 mg/L in the successful medical group; p=0.43). Despite the promising results, these findings are not statistically significant (p>0.05). The duration of hospitalization was also analyzed between the two groups, being significantly shorter in the successful medical group compared to the late surgical group (p=0.01), but without statistically significant differences from the early surgical group⁽⁶⁾.

A prospective study conducted between 2015 and 2017 in Israel, which included 94 patients diagnosed with TOA, evaluated the utility of daily CRP measurement in predicting the need for early surgical intervention. In the requiring surgery group, the CRP values followed an upward trend from the day of admission to the second day and from the second day to the third day (128.26 mg/L, 173.75 mg/L and 214.66 mg/L, respectively; p<0.05 for both). In the group that did not require surgery, the same values were analyzed, showing a downward trend from the second, third and fourth days, while a plateau was observed between the day of admission and the second day (110 mg/L, 120.49 mg/L, 97.52 mg/L and 78.45 mg/L, respectively)⁽⁷⁾. In our case, the C-reactive protein was not measured daily, but every other day. Thus, the CRP levels decreased from the day of the first admission compared to day 3, day 3 compared to day 5, and day 5 compared to day 8 (463 mg/L versus 345 mg/L versus 265 mg/L versus 212 mg/L). Still, it should be added that, beside the triple antibiotic therapy that was initialized on the first day, the IUD was removed on the sixth day.

A case study from September 2024 followed a 61-year-old patient with an IUD and pseudoactinomycotic radiate granules (PAMRAGs) diagnosed on endometrial biopsy using Periodic acid-Schiff-diastase, Gram stain, and modified acid-fast bacilli stains, all of which were negative. PAMRAGs are nonpathogenic microorganisms, presenting a diagnostic challenge due to their resemblance to actinomycotic granules (AMGs)⁽⁸⁾. Actinomyces are Gram-positive, filamentous, anaerobic bacteria commonly found in the oral cavity, reproductive tract and gastrointestinal tract. Actinomyces spp. are associated with PID and TOA in patients with IUDs (the foreign body acting as a risk factor for microbial aggregation)⁽⁹⁾. This case report is valuable as it highlights an alternative pathogenesis in the TOA development. In our case, no etiological agent was found, but the tested bacteria did not include this microorganism.

A 2020 study from Türkiye included 124 patients diagnosed with TOA with a present IUD, all initially receiving medical treatment and later divided into a successful group (29.8%) if surgery was not required, and a failed group (70.2%) if surgery was necessary. The IUD was removed just before discharge in the medically treated group and during surgery in the other group. The study showed that the duration of IUD presence was significantly longer in the surgically treated group compared to the medically treated group (7.61±3.34 years versus 4.41±1.95 years; p<0.001). Other factors were also evaluated, among which statistically significant ones included: patients’ age (36.03±9.04 in the successful group versus 39.46±7.49 in the failed group; p=0.03), parity (1.92±0.89 in the successful group versus 2.47±1.12 in the failed group; p=0.01), maximum TOA size in cm (3.73±1.33 in the successful group versus 6.51±2.19 in the failed group; p<0.001), and WBC count/mm³ (11.150±4300 in the successful group versus 14.430±6720 in the failed group; p=0.007). The best calculated cutoff value for TOA size predictive of medical treatment success was 4.5 cm (sensitivity of 85.1% and specificity of 78.4%), and for IUD duration, it was 5.5 years (sensitivity of 67.8% and specificity of 78.4%)⁽¹⁰⁾. In our case, the large abscess size (8.9 cm) and the longer presence of the IUD (10 years) indicated the failure of the medical treatment alone and the requirement of later surgical intervention. Also, in our case, the patient presented multiple risk factors for failure of medical treatment alone, such as age (40 years old), maximum TOA size (8.9 cm) and WBC count (18,530/mm³).

Conclusions

The peculiarities of the case presented above include the inability to identify a causative agent despite the final histopathological diagnosis of a tubal abscess, the success of the combined treatment – initially medical with triple empirical antibiotic therapy followed by “cold” excision of the abscess after infection source eradication. Another notable aspect is the patient’s medical history, which likely contributed to a delayed diagnosis, allowing the abscess to reach a large size with minimal symptoms, as well as the presence of an intrauterine device for approximately ten years.

Autori pentru corespondenţă: Romina-Marina Sima E-mail: romina.sima@umfcd.ro

CONFLICT OF INTEREST: none declared.

FINANCIAL SUPPORT: none declared.

This work is permanently accessible online free of charge and published under the CC-BY.