Pelvic Actinomycosis

Introduction Actinomycosis is a chronic bacterial infection caused by Actinomyces, Gram-positive anaerobic bacteria. Its symptomatology imitates some malignant pelvic tumours, tuberculosis, or nocardiosis, causing abscesses and fistulas. Actinomycoses are opportunistic infections and require normal mucous barriers to be altered. No epidemiological studies have been conducted to determine prevalence or incidence of such infections. Objective To analyse the clinical cases of pelvic actinomycosis reported worldwide, to update the information about the disease. Methods A systematic review of worldwide pelvic actinomycosis cases between 1980 and 2014 was performed, utilising the PubMed, Scopus, and Google Scholar databases. The following information was analysed: year, country, type of study, number of cases, use of intrauterine device (IUD), final and initial diagnosis, and method of diagnosis. Results 63 articles met the search criteria, of which 55 reported clinical cases and 8 reported cross-sectional studies. Conclusions Pelvic actinomycosis is confusing to diagnose and should be considered in the differential diagnosis of pelvic chronic inflammatory lesions. It is commonly diagnosed through a histological report, obtained after a surgery subsequent to an erroneous initial diagnosis. A bacterial culture in anaerobic medium could be useful for the diagnosis but requires a controlled technique and should be performed using specialised equipment.


Introduction
Actinomycosis is a chronic bacterial infection, suppurative and granulomatous in nature, caused by bacteria of the genus Actinomyces [1], a group of Gram-positive anaerobic bacteria that form filamentous microcolonies [2], do not form spores, measure up to 1 m diameter, and are slow-growing [1]. Actinomycosis is an uncommon condition whose symptomatology imitates some malignant pelvic tumours, tuberculosis, or nocardiosis because it spreads progressively and continuously [3]. This pathology invades tissue layers, causing the formation of abscesses and fistulae. Its diagnosis is difficult, and it results in increased morbimortality. Actinomyces belong to the phylum Actinobacteria and to the order Actinomycetales. Hundreds of Actinomyces species exist, most of which inhabit the soil. Others are associated with plants, which participate in nitrogen fixation, and a few species live in human beings 2 Canadian Journal of Infectious Diseases and Medical Microbiology respiratory tract, urogenital tract, digestive tract, central nervous system, skin, and soft tissue structures). The most frequent clinical form of the disease is cervicofacial actinomycosis, representing approximately 60% of all reported cases, and is associated with odontogenic infection. Other clinical types include thoracic actinomycosis, the third most common type of actinomycosis, which includes pulmonary, bronchial, and laryngeal actinomycosis [3], and abdominal actinomycosis, where the appendix, caecum, and colon are the most common sites of infection. Actinomycosis of the central nervous system is located chiefly in the cerebral abscess. Actinomycosis of the urogenital tract is the second most common clinical form of actinomycosis, and the principal clinical presentation is pelvic actinomycosis [3,5,8].
Pelvic actinomycosis can affect any age group, with no preference for occupation or season and is secondary to perforation or fistulation [4]. Other possible causes include bacterial vaginosis, which fosters an anaerobic environment and is associated with other microorganisms [51]; the presence of tumours [66]; and the use of IUDs [3][4][5]. The possibility of a contagion through oral sex has been considered because these bacteria are part of the oral cavity microbiota [72]. One possible route of dissemination is through IUDs, which fosters the growth of microorganisms through wires that are left in the exocervix. In addition, the IUD changes the carbohydrate metabolism in endometrial cells, fostering still more inflammation. Another probable route is the perineum, where the microorganisms could extend from the anus up through the cervicovaginal zone [4].
The symptoms of pelvic actinomycosis associated with the use of an IUD can imitate symptoms of gynaecological malignant tumours, uterine myoma, or adenomyosis when presenting as a genital mass without fever [3]. The infection can disseminate to the uterine tubes and can cause salpingitis and the subsequent destruction of the ovarian parenchyma [4]. Organs such as the bladder, ileocaecal (iliac fossa) and rectosigmoid region, colon, urethra, and extension to the skin have been reportedly affected in various published cases.
The diagnosis of pelvic actinomycosis is obtained using various techniques because culturing Actinomyces spp. presents difficulties and also depends on the skill and access to equipment necessary to perform it.
(a) First, the signs and symptoms of the patients are considered and can point to a possible abdominal infection, vaginitis, abscess, or possible tumourforming process. The most common symptoms are weight loss, nonspecific abdominal or pelvic pain, breakthrough bleeding or abundant vaginal flow, and, on rare occasions, fever [3,4,51].
(b) Upon medical exploration, the affected zone is palpated to detect hard masses, and a gynaecological exam is performed to check for inflammation of the vaginal mucous membrane, yellowish secretion with a bad smell, or some visible damage to the mucous membrane [4,51].
(c) In laboratory studies, it is possible to identify leucocytosis, erythropaenia, and high sedimentation rate; high values of C-reactive protein; and tumour marker values within the reference ranges or slightly elevated like Ca 125 (Alpha-fetoprotein), and cancer antigen 15-3 [3,4,51].
(d) Diagnostic images, such as computed tomography, magnetic resonance, ultrasound, X-rays, and laparoscopy are helpful, as they can be used to observe the affected zone, such as a tumour-forming mass that can induce either actinomycosis or a carcinogenic process [4,51,73].
(e) In most cases, histological visualisation of biopsy or aspirated samples is employed, where bacilli in the tissue with their typical ramifications, such as in interconnected breasts, are observed. Cervicovaginal cells are collected for Papanicolaou (Pap) staining. In both cases, they are reported as Microorganisms Similar to Actinomyces (MSA) [4,51]. In many cases, the diagnosis is made a posteriori through a histological examination of samples obtained surgically during laparotomy or laparoscopy, but rarely in a preoperative manner. Histological studies of tissues show inflammatory changes of suppurative and granulomatous nature, connective proliferation, and sulphur granules, which have also been identified in infections caused by Nocardia brasiliensis, Actinomadura madurae, and Staphylococcus aureus. These granules are particles of yellowish colour, which, when viewed by the naked eye, are formed by groups of filamentous Actinomyces surrounded by neutrophils [73].
(f) Two methods exist for completely identifying the causal agent: culture and identification through biochemical tests and identification through sequencing of the 16S rRNA segment, which offers greater precision. Although these methods are very efficient, they are not well reported in the literature due to the conditions under which they must be performed, requiring an anaerobic culture environment and the necessary equipment, which is costly.
The usual treatment for actinomycosis consists of high and prolonged doses of penicillin G (20 million units per day) or amoxicillin for 4 to 6 weeks, followed by penicillin V (4 g per day) orally for 6 to 12 months. Clindamycin, tetracycline, and erythromycin are an alternative in cases of allergy to penicillin [4,5]. In addition to these medicines, it has been observed that Actinomyces is also sensitive to thirdgeneration cephalosporins, ciprofloxacin, trimethoprim-sulfamethoxazole, and rifampicin [4]. However, the elimination of the injured tissue and surgical drainage are necessary measures in some cases [5], and, in these patients, the duration of antimicrobial therapy could be reduced (3 months) [3].
In the review performed by Martínez et al. [74], it is mentioned that there are reports of the presence of Actinomyces in secretions starting from 1877, recorded by Harz. Beedham et al. found that the first reports of intrauterine actinomycosis related to IUDs appeared in the 1920s [76]. Clinical cases of pelvic actinomycosis have been reported in Africa, Oceania, Asia, Europe, and America. However, as pelvic actinomycosis is an uncommon infection, no epidemiological studies have been conducted to determine its prevalence or incidence.

Materials and Methods
We performed this analysis according to the guidelines of the Meta-Analysis of Observational Studies in Epidemiology Group (MOOSE). A systematic review of worldwide cases of pelvic actinomycosis between the years 1980 and 2014 was performed. Studies that described clinical cases of pelvic actinomycosis with a detailed diagnostic method and cross-sectional studies of cases of actinomycosis available in the PubMed, Scopus, and Google Scholar databases were included using the following keywords: Pelvic actinomycosis, Actinomycosis pélvica, "Actinomycosis prevalence", "Prevalencia de actinomycosis", "Actinomyces" AND "female genital tract" and combinations of these terms. The use of quotes was avoided when searching for the terms Pelvic actinomycosis and Actinomycosis pélvica in order to increase the search results.
Abstracts of articles identified to be relevant for the objective of this paper were read; studies whose abstract or full text was unavailable were automatically excluded. When an abstract complied with inclusion criteria, the full text was analysed. Case reports that lacked a diagnostic method and a final diagnosis of pelvic actinomycosis were excluded. Studies published in a language that was not English, Spanish, French, or Portuguese were not included. The following information was extracted and analysed from the compiled studies: year, country, type of study, number of cases, prior use of IUD and duration, initial diagnosis, treatment, definitive diagnosis, and method of definitive diagnosis ( Figure 1).

Results
The search yielded a total of 3852 studies; 3693 were excluded from the title, abstract, and language screening; 96 more were excluded for not being available in full text format and for not meeting the selection criteria when reading the full article.
A total of 63 studies including 86 case reports of pelvic actinomycosis, along with 8 cross-sectional studies of reports examining populations for cases of Microorganisms Similar to Actinomyces (MSA), were included for this review (Figure 1).

Cases of Pelvic Actinomycosis Reported in Africa.
From the African continent, 3 articles of clinical cases were found, totalling 8 clinical cases. The majority of patients were IUD users; however in most cases the type of IUD used was not disclosed. The pathology that was first diagnosed in these cases was an ovarian tumour. The method of diagnosis that was utilised to definitively diagnose patients with actinomycosis was histopathological reporting ( Table 1). The most common treatments were hysterectomy, laparotomy, and antibiotic therapy. No follow-up data was presented.

Cases of Pelvic Actinomycosis Reported in Oceania. From
Oceania, 1 article was published that included 3 clinical cases with the following ages: 56, 70, and 37 years; two of them were copper IUD users. In the three cases, malignant lesions were initially diagnosed; the final diagnosis was performed postoperatively. Salpingo-oophorectomy along with antibiotic therapy was used in all the cases; patients fully recovered after treatment. (Table 2).

Cases of Pelvic Actinomycosis Reported in Asia.
Fourteen articles of 16 clinical cases came from Asia, the age of the patients ranged between 25 and 86 years, and the average age was 45.6 years (SD 15.5). The majority of patients were IUD users, with a usage time of 1 year to more than 20 years; most of the studies did not specify the type of IUD used. However, cases in nonusers were also reported, despite the well-known relationship between IUD use and pelvic actinomycosis. The most common presumptive diagnostic was malignant lesions, while, in other cases, Crohn's disease and acute peritonitis were also suspected. The most utilised diagnostic method was histological reporting after surgical interventions, which were invasive in most cases, such as hysterectomy and salpingo-oophorectomy along with antibiotic therapy. Most of the patients had a full recovery or at least a significant improvement after follow-up; only a case of renal sequelae was reported. (Table 3).

Cases of Pelvic Actinomycosis Reported in Europe.
Twenty clinical case report articles including 39 cases of pelvic actinomycosis originated from Europe, in which ages ranged from 18 to 65 years; average age was 40 years (SD 10.4). The cases principally included female IUD users, with a usage time ranging from 1.5 to 20 years; copper and multiload IUD were the most reported; however in most studies the type of IUD is not specified. The predominant presumptive diagnosis was malignant lesion; other suspected diagnoses included Crohn's disease, acute appendicitis, endometrial infection, pelvic inflammatory disease, and abscesses. Postoperative histopathological reports were the most common definitive diagnostic methods. Other methods of final diagnosis have also been reported, such as the Pap smear, culture, API 20A biochemical assays, and 16S rRNA sequencing techniques. The most common treatments used were damaged tissue excision, laparotomy, and salpingo-oophorectomy together with antibiotic therapy. The majority of the articles do not have follow-up information, nonetheless studies reporting patient follow-up stated that they fully recovered after treatment, and there is one report of death (Table 4).

Cases of Pelvic Actinomycosis Reported in America.
With regard to America, 16 articles with reports of 20 clinical cases exist. The ages of patients ranged between 18 and 58 years with an average age of 39.6 years (SD 9.9). All patients were IUD users except one case, and the time of device use ranged from 22 months to 33 years. Reported types of IUD include Dalkon Shield and Lippes loop. Tubo-ovarian and pelvic abscesses along with malignant lesions were the conditions with the greatest diagnostic confusion. Similar to the other summaries, the postsurgical histological reports were the most reported definitive diagnostic methods. Other methods were also utilised, such as culture, the 16S rRNA sequencing technique, haematoxylin-eosin staining microscopy (HE), and the IUD smear. Salpingo-oophorectomy and laparotomy along with prolonged antibiotic therapy were the most used therapeutic measures. After treatment most of the patients had a full or significant recovery. (Table 5).

Cross-Sectional Studies.
Eight cross-sectional studies of reports worldwide that examine populations for cases of actinomycosis or MSA were analysed. The prevalence of pelvic actinomycosis was low. Likewise, there is a strong relationship between the use of an IUD and the presence of MSA. In this type of report, the diagnosis methods reviewed were the Pap reports. However, it is important to emphasise that what is reported in these analyses are MSA. Only 3 articles reported actinomycosis as such, and only one report completely identified the causal agent through culture and biochemical assays (Table 6).

Discussion
According to the analysis of the articles presented, Europe was the continent on which the greatest number of cases of pelvic actinomycosis was reported, followed by Asia and America. However, it is important to emphasise that this summary of information only gives us an approximation of the real epidemiology of this disease, as the cases presented in this article are only those reported. The youngest cases (18 years) are found in the European and American continents, and the oldest case (86 years) is found in the Asian continent. Actinomycosis is an invasive infection that frequently imitates malignant processes in various anatomical zones. Pelvic actinomycosis involves one of the regions that is most often a source of diagnostic confusion. In this review, it is evident that, in many of the cases presented, an erroneous clinical diagnosis was made, confusing pelvic actinomycosis mainly with malignant lesions. Other common suspects were tubo-ovarian and pelvic abscesses and Crohn's disease. As such, as described by Kayikcioglu et al. [35] and Moniruddin et al. [77], pelvic actinomycosis should be considered in the differential diagnosis in any chronic inflammatory lesion of the viscera located in the pelvic zone to prevent a diagnostic error that could lead to unnecessary invasive treatment.
The diagnosis of pelvic actinomycosis is difficult because it does not produce characteristic disease signs or symptoms. According to what was observed in this analysis, the most utilised diagnostic method in all continents was the histopathological report, which is commonly obtained after a surgical intervention due to an initial diagnostic error. This observation was also made by Purola and Paavonen [78] and more recently by Pérez-López et al. [73]. Other highly reported methods are the Pap test, which is generally reliable, but not unequivocal, as Actinomyces could be confused with similar organisms. Cases in which the causal microorganism is completely identified are few, as are cases identified by culture and biochemical assays. Valour et al. [3] mentioned that culturing bacteria in anaerobic medium is the cornerstone for diagnosing actinomycosis. However, this method requires a very controlled and precise technique and specific equipment. Identification by sequencing of the 16S rRNA segment is another technique that offers greater precision. Currently other authors such as Demirezen et al. [79] report the effectiveness of using specific primers to identify the most common Actinomyces species from patients' swabs samples; this technique is more accurate and faster than all the previous ones; the disadvantage is the high cost of        reagents and the use of special equipment. However, because of the nature of the pathology, there is no early diagnosis, because, as has already been mentioned, the presence of symptoms occurs in advanced stages of the disease. According to the observed reports, we conclude that the presentation of symptoms in pelvic actinomycosis is observed in an advanced period of the pathology, which does not include attack to the general state of health or fever, which is oriented towards an infectious pathology. The manifestations found are occupational masses in the pelvic-abdominal cavity that force the surgical procedures to be performed, and the diagnosis is made up to the time of the histopathological study.
Pelvic actinomycosis is considered to be a rare and unusual disease, although the use of IUDs can promote its appearance. In the articles analysed, the greatest number of patients diagnosed with pelvic actinomycosis on all continents was IUD users, and the periods of use varied widely, from one year to long periods, such as 33 years. However, cases occurred mainly in users that wore IUDs for prolonged periods. Thus, based on experience and observation, it is recommended that IUDs be changed periodically to limit the occurrence of this condition. Some authors, such as Valour et al. [3], recommend changing the IUD every 5 years at a minimum, and others, such as Hernández et al. [4], recommend changes every 3 years. It should be emphasised that those cases in which patients were not IUD users were identified more recently. This observation could suggest that, despite the information that is available regarding the relationship between this condition and IUD use, the aetiology of pelvic actinomycosis could be due to other factors.
Like all review studies, the main limitation of the study was the lack of data reported, another limitation was that not all articles were open access, and there were not enough subscriptions to the respective journals to access them.
During the first reports of this disease, greater numbers of cases were observed in developed countries, but, presently, reports of cases in developing countries and regions such as the Middle East, Southwest Asia, or Latin America are more common. This change could be because, in the first decades of its observation, this condition and its aetiology were unknown and prevention was difficult. However, with the advance of technology, preventative measures directed at high-risk populations in developed countries began to be applied, as opposed to developing countries, where no such actions were taken.