The epidemiology, clinical presentation and treatment outcomes in CNS actinomycosis: a systematic review of reported cases

Actinomycosis is a rare, subacute to chronic granulomatous infection caused by gram-positive anaerobic Actinomyces species [1]. The Actinomyces genus of the family Actinomycetaceae currently comprises around 42 species [1‐3]. Most infections are caused by Actinomyces israelii and A. gerencseriae [4]. Actinomyces are non-spore-forming, non-acid-fast, filamentous microorganisms which are commensals in the human oropharynx (found in gingival crevices, dental plaques, tonsillar crypts), gastrointestinal and genitourinary flora [5, 6]. These organisms usually manifest as a slowly progressive, mass-like lesion which ultimately forms draining sinus tracts [6]. Actinomycosis is a great masquerader that can mimic malignancy and other indolent infections like fungal, nocardiosis and mycobacterial diseases. Of note, later two genera belong to the same order of Actinomycetales [7]. The Cervicofacial disease is the most common presentation of actinomycosis (around 50% of the cases), followed by abdominal (20%) and thoracic disease (15–20%) [5, 8‐10].

Of note, actinomycosis is usually described as an indolent infection with protean manifestations; however, CNS involvement is the most severe form of actinomycosis [11]. CNS actinomycosis is a rare disease with an exact incidence that is difficult to ascertain due to the rarity of the disease. The majority of data is available as anecdotal evidence in the form of case reports and expert opinions. Friedman et al. performed the first systematic review of cases with CNS actinomycosis in 1937 [12]. However, there were a lot of cases where no differentiation was made between nocardiosis and actinomycosis. In 1964, Bolton et al. reviewed 17 cases with CNS actinomycosis [13]. The first comprehensive systematic review was published by Smego et al. in 1987 [14], which included 70 cases of CNS actinomycosis. To better understand the clinical spectrum and various predictors of outcome, we performed an updated systematic review of CNS actinomycosis and analyzed the cases reported from 1988 to 2022.

The initial database search revealed a total of 1291 case records that were analyzed for the final inclusion. After removing duplicate records, 870 records were assessed for further inclusion. After the removal of ineligible articles, a total of 118 patients with CNS actinomycosis (95 articles) were analyzed in this systematic review (Fig. 1). Out of 118 patients, 29 were from a case series, and the remaining 89 were individual data. The reference list of all reported cases in this review is given in supplementary material (Additional file 1: s4).

Actinomycosis is a rare, invasive bacterial disease characterized by chronic granulomatous infection and pus discharging sinuses. CNS involvement is uncommon in actinomycosis, but can cause significant morbidity and mortality. Moreover, data are scarce due to the disease's rarity and the low index of suspicion. Most literature is available in the form of anecdotal data with a lack of more extensive studies and treatment guidelines. The treatment decision for CNS disease depends on the literature predominantly focused on cervicofacial and pulmonary actinomycosis, which have a different clinical spectrum from CNS disease. This review of published cases of CNS actinomycosis was conducted to provide insight into the aforementioned factors.

Actinomycosis is usually considered a disease with low virulence. CNS spread primarily depends on mucosal breach and contiguous spread from cervicofacial areas, skull osteomyelitis and less commonly from hematogenous spread from distant foci (lungs, abdomen, pelvis) [25]. Trauma, surgery, dental extraction, and ear infections are the usual trigger for invasive CNS actinomycosis. The incidence of CNS disease is 1–2% in all disseminated cases of actinomycosis [26, 27]. In this review, disseminated infection was found in 20% of the patients, which was probably due to hematogenous spread. The primary source of actinomycosis was the lungs in 10% of the cases. A previous review by Smego et al. described the relatively high proportion of pulmonary actinomycosis (27%) in disseminated cases [14]. In their report, cyanotic heart disease (CHD) was also described as a risk factor for actinomycosis brain abscess. However, we did not find any patients with CHD. The common imitators of actinomycosis are Nocardia and tuberculosis, which typically develop in immunodeficient patients, CNS actinomycosis is not usually associated with immunodeficiency. Only 12% of the cases were found immunodeficient. A large multicenter study reported the low prevalence of actinomycosis in renal transplant recipients (0.02%), and none of these patients had CNS disease [28]. Smego et al. also reported that only 7% of patients were immunocompromised [14]. This also emphasizes the need for a high index of suspicion of CNS actinomycosis in immunocompetent patients. The clinical presentation of CNS actinomycosis depends on the site and extent of the lesion. Focal neurological deficits and headache are the common presentations of CNS disease, which makes it indistinguishable from other chronic CNS infections (e.g., nocardiosis and neuroaspergillosis). The indolent nature and lack of fever at presentation prompt an erroneous diagnosis of malignancy in some cases of actinomycosis. Meningitis is an uncommon presentation in CNS actinomycosis. In this review, the incidence of meningitis was 21% which was significantly lower compared to the previous review (46%) [14]. Moreover, isolated meningitis was seen in only 11% of the cases. A high proportion of meningitis in previous studies might be related to delayed presentation, leading to the rupture of foci into the subarachnoid space, producing meningeal signs and CSF abnormalities.

The diagnosis of actinomycosis is based on a constellation of histopathological features and microbiological isolation of the organism [6, 29]. However, failure to isolate the organism from culture does not exclude the diagnosis. Many of these patients had suppurative disease and received multiple antibiotics that could preclude culture positivity. In this review, nearly half of the patients were culture-negative. Actinomycosis is a fastidious organism which requires a prolonged incubation period (14–21 days) in an anaerobic environment [6]. Suppression by concomitant pathogens and inadequate culture media are the other reasons for culture negativity. Actinomycosis israelii is the most common causative agent for CNS disease. We found A. meyeri and A. viscosus as emerging species in the last two decades causing CNS infections, which were not identified in a previous review published in 1987 [14]. Both these species are known to cause invasive actinomycosis and bacteraemia (3 out of 4 cases in this review). We found a low CSF culture positivity, which may be explained due to the low prevalence of meningitis and the inadequate samples sent for culture. CSF cytology showed predominantly neutrophilic leucocytosis, in contrast to the lymphocytic predominance seen in other mimickers like neuroaspergillosis [30]. Another characteristic of actinomycosis infection is the concomitant microorganism which grew in the culture. This polymicrobial infection acts in synergy, and the other pathogens decrease the oxygen tension, thus enhancing the growth of actinomycosis. In addition, the host defence mechanism is also suppressed in this setting [31, 32].

The demonstration of characteristic sulfur granules also supports the diagnosis of actinomycosis though it can also be seen in nocardiosis and botryomycosis [33]. Sulfur granules have a tiny, yellowish tinge composed of filamentous bacteria, and at the periphery, there are eosinophilic filaments with club ends [5, 33]. CNS cases usually have delayed presentation; by that time, the lesion may have extensive fibrosis, which can impair the identification of sulfur granules. Recent reports highlight the role of targeted histological evaluation in increasing the detection of actinomycosis [34]. The advent of molecular techniques (PCR, 16S rRNA gene sequencing) allows the rapid identification of actinomycosis species. Recent reports have also shown promising results of MALDI-TOF comparable to 16S rRNA sequencing [35]. Neuroimaging in CNS actinomycosis shows brain abscess in more than 50% of the cases [14], similar to nocardiosis and neuroaspergillosis [30, 36]. The high predilection of the temporal lobe was reported in the previous systematic review [14]. However, in this review, the frontal and frontal-parietal lobes were the most common area affected.

In this study, the mortality rate was 11% in all treated cases, which is lower compared to previous review (28%), [14]. In last few decades, prompt diagnosis, availability of antibiotics and early aggressive surgery could be the potential factors contributing to low mortality. Medical management of CNS actinomycosis includes a prolonged high dose of antimicrobials. Determining the optimal duration and use of combination antibiotics remains elusive in CNS disease. The decision should be individualized based on the severity of the disease, clinical response and immune status of the patients [1, 6]. We recommend at least 3–6 months of antibiotics in CNS actinomycosis, which can be extended if an adequate clinical response is not achieved or source control (surgery) is not possible. Antibiotic susceptibility is usually not essential, with most isolates showing a good sensitivity to beta-lactam antibiotics (penicillin, amoxicillin, ceftriaxone, meropenem, piperacillin-tazobactam), [1]. High-dose penicillin is the most widely used therapy for CNS disease [14]. However, we found the emerging use of ceftriaxone, ampicillin and meropenem. Amoxicillin, clindamycin, doxycycline and cotrimoxazole are good options for maintenance therapy. Of note, the clinician should be aware of a few actinomycosis species (A. turicensis and A. europaeus) that showed resistance to clindamycin, macrolides and quinolones [37, 38]. Metronidazole and aminoglycosides usually do not have any activity against actinomycosis and should be avoided [39]. Though combination therapy was the preferred approach based on this review, it did not improve the outcome of CNS actinomycosis. Further studies should focus on this elusive question. Surgical intervention (debridement of necrotic tissue, abscess drainage) becomes vital in CNS disease. In this review, early surgery was found to be an independent factor in improving the outcomes. Thus, it should not be reserved only for cases with a poor response to antibiotics. Whether it will shorten the hospital stay and antibiotic duration is a matter of further studies.

This systematic review has a few limitations; firstly, case reports are inherently biased; a heterogeneous population makes the outcome and risk factor analysis difficult. Data regarding the antimicrobial susceptibility reports are not available. Many cases were excluded due to a lack of information regarding treatment and follow-up.

Despite the indolent nature of the disease, CNS actinomycosis carries significant mortality. Early suspicion and differentiation from tuberculosis, nocardiosis and malignancy is vital to improve outcomes. The diagnostic utility of molecular methods could be helpful in such settings. We emphasize early aggressive surgery and prolonged antibiotics in CNS actinomycosis to improve clinical outcomes. The use of combination antibiotics is debatable and requires further evidence in terms of prospective studies.