Despite advanced diagnostic methods and therapeutic approach, the differential diagnosis of FUO remains to be one of the sophisticated challenges all over the world. The most common causes of FUO can be classified into four main etiology: infections, malignancy, non-infectious inflammatory disorders (NIID, such as autoimmune and rheumatic diseases, vasculitis syndromes and granulomatous disorders), and miscellaneous conditions. However, infections caused by various b pathogens are still the leading cause for classical FUO around the world especially in developing countries [1, 2].
Actinomyces are a genus of gram-positive, rod-shaped ,non-acid-fast, non-spore forming anaerobic or microaerophilic bacteria with a tendency to produce branching filaments and are the normal colonizing flora of the oral cavity, pharynx, gastrointestinal and genitourinary tract. Six of these are considered to be pathogenic in humans, including A. israelii, A. naeslundi, A. odontolyticus, A. viscosus, A. meyeri and A. gerencseriae. They can affect many sites of humans, such as the cervicofacial (50-60%), thoracic actinomycosis (15-20%), abdominopelvic (20%), the central nervous system (2%), and rarely involves cutaneous, ophthalmic, musculoskeletal disease, pericarditis, genitourinary or disseminated disease. As for thoracic actinomycosis, the most common etiology of infection is poor dental hygiene and diabetes, HIV infection, malignancies, steroid, obesity, alcoholism, and age > 65 years are risk factors for developing thoracic actinomycosis. Otherwise, patients with chronic obstructive pulmonary disease, pulmonary sequelae following tuberculosis and with the aspiration of foreign bodies[7, 8], are considered to also be risk factors.
Dyspnea, cough, expectoration, hemoptysis and chest pain are the most common and typical presenting symptoms. In some developing countries, blood-stained sputum is also main one symptom. Interestingly, recurrent high fever was the most primary symptom the patient had in our report.
To separate, culture, and ultimately diagnosis of thoracic actinomycosis are always challenging and delayed. There are some reasons as follows. Radiologic modalities are usually inadequate for diagnosing and only could provide some useful information such as chronic segmental airspace consolidation that irregularly have thickened bronchial walls with central necrotic low attenuation areas and peripheral enhancement, a mass-like shadow, cavitation, pleural thickening, pleural effusions, hilar or mediastinal lymphadenopathy, bronchial and bronchiolar dilatation[12-15]. However, the typical CT feature of thoracic actinomycosis is only seen in less than 50% of patients as a retrospective study reported. The best standard of thoracic actinomycosis was that actinomyces colonies was visualized in histology. However, the rate of Actinomyces isolation using culture tests is only approximately 7% and a retrospective, multicenter study reported that up to 40% of cases were culture negtive. In our case, culture tests of the BALF were negative result. Using conventional microbiological tools of several attempts of samples, we could not diagnose what kind of infection in our patient. Fortunately, with the development of techniques, the Bruker MALDI-TOF BioTyper MS (MBT) and 16S rRNA gene sequencing have emerged as new reliable tools for the identification of Actinomyces[18, 19] in recent years. And some recent studies have shown that next-generation sequence (NGS) may significantly enhance the rare or unusual microbiological diagnostic of the culture-negative specimens in several clinical situations by allowing rapid identification of bacteria in different samples, such as blood samples, sputum, body fluids, or tissues [20-22]. Subsequently, we performed molecular identification by NGS from the BALF sample and finally detected the microbe. Meantime, there were also other bacteria and fungi in this specimen by NGS, indicating that contamination from the oral cavity was a consideration.
We think that the real pathogen of our patient isn’t other microbes that just are contamination than A. gerencseriae because of the following reasons: first the patient was treated with cephalosporin and moxifloxacin to cover common gram-positive and negative bacillias, however, they are not effective. Secondly, there are 6 organisms that are members of oral microflora, constituting the flora of the patient's mouth identified by next-generation sequencing in the patient. V. parvula has been implicated in septicemia, meningitis, endocarditis, peritonitis and osteomyelitis [23, 24] , however, it is generally considered to be contaminant bacteria. V.atypica is one early-colonizing members of the dental plaque biofilm, but to our knowledge have not been recognized as causative agents of disease in respiratory system. There are only occasional reports of endocarditis due to S. parasanguinis. P. melaninogenica is often associated with pyomyositis. Interestingly, P. salivae can be isolated from the mouths of humans, but has not been reported as a pathogen in humans. Moreover, R. dentocariosa and R. mucilaginosa is usually recognized as an opportunistic pathogen with keratitis, endocarditis, peritonitis, arthritis and bacteremia[28-32]. Of course, R. mucilaginosa is reported with pneumonia, it usually affects immunocompromised hosts as the review reports. Candida albicans infect humans only under certain conditions such as immunodeficiency. In contrast, the ability of A. gerencseriae to act as infectious agents is known. However, it’s difficult to isolate this pathogen from the involved samples by normal ways because the specimen must be collected under strictly anaerobic conditions, and mixed infections are inevitable. Finally, we decided to do an experimental treatment for A. gerencseriae diagnosed by NGS and the clinical recovery of our patient also indicated that the infection was caused by A. gerencseriae.
DNA sequencing technologies were invented in the early 1970’s and the NGS revolution appeared due to cut down the cost and time of sequencing in 2005. They are useful in a variety of areas such as cancer, haematology, neurology and epidemiology[35-37]. Because it takes less time and less priori knowledge of the type of microbial to come to an conclusion, there are many promising applications on epidemiology including: infectious disease diagnostics, hospital infection microbial surveillance programs, community outbreak investigations and antimicrobial resistance. However, there are challenges for NGS to widely apply to clinical treatment. For example, detecting many microorganisms directly from a patient sample requires physicians to determine that which is the real pathogen and which is likely (but not absolutely) a contaminant for a patient with non-specific symptoms. Morever, it takes more money than other conventional tools. With the development of the NGS, there is no doubt that it will become more and more useful and reliable to help solve medical diagnostic dilemmas.
Although thoracic actinomycosis is relatively difficult to isolate and identify, prognosis is relatively good if it is properly treated. There are no definitive and standard guidelines for kinds of drugs and duration of antibiotics treatment for thoracic actinomycosis yet, penicillin G has been recommended as the preferred drug of choice for treatment. The optimal treatment duration is 2 to 6 weeks of intravenous penicillin followed oral penicillin or amoxicillin for 6 to 12 months[39, 40]. When the patients are allergic to penicillin, tetracyclines are available for mild presentations. Worth mentioned is that actinomyces are anaerobic, metronidazole should be avoided in the treatment of pulmonary actinomycosis. Moreover, surgical intervention might be necessary if we don’t evaluate the nature of the lesions and alleviate severe symptoms. In our case, this old woman was treated with intravenous sulbenicillin for 11 days and further treated with oral amoxicillin for 1 month.
To the best of our knowledge, this case is meaningful and instructive because it’s the first case of NGS is used for thoracic actinomycosis, that is a cause of FUO. The NGS, particularly in cases of FUO in conventional microbiology, is a promising and advantageous technology for earlier diagnosis and more success. It gives new insights in the management of FUO, especially in infectious diseases.
In obscure causes of FUO, clinicians should consider the possibility of thoracic actinomycosis and actively use next-generation sequencing to help diagnose.