Mycobaterium fortuitum disseminated infection in an immunocompetent patient without predisposing factors
- 1 Division of Infectious Diseases, Inselspital University Hospital Bern, Bern, Switzerland
- 2 Neuroradiology, Geneva University Hospitals, Geneva, Switzerland
- 3 HIV/AIDS Unit, Division of Infectious Diseases, Geneva University Hospitals, Geneva, Switzerland
- 4 Division of Pulmonology, Department of Medicine, Geneva University Hospitals, Geneva, Switzerland
- Correspondence to Dr Stephanie d'Incau; stephanie.dincau@insel.ch
Abstract
Most Mycobacterium fortuitum infections described involve direct inoculation through skin lesions. We describe the case of a patient without risk factors who presented with an intracranial mass and a pulmonary infection with M. fortuitum. As M. fortuitum are rarely pathogens, there is little knowledge about the optimal treatment and outcome of such infections: what is the best mode of administration, what is the best therapy duration and is surgery always required are some of the unanswered questions. In our patient, surgical removal of the mass associated with a 1-year antimycobacterial therapy led to a full recovery. Even though M. fortuitum was rapidly identified in sputum, it was initially considered non-pathogenic and the definitive diagnosis required almost 6 weeks of investigations. New molecular techniques will probably lead to more identifications of M. fortuitum in the next few years and a better knowledge of their possible pathogenicity and optimal treatment.
Background
Mycobacterium fortuitum is a member of the rapidly growing mycobacteria (RGM). There are six groups of RGM representing about one half of all the mycobacterial species. RGM usually grow within 7 days. They are saprophytes, found in soils, dust, water, animals, hospital settings (tap water) and various solutions and instrumentation (responsible for nosocomial outbreaks). RGM are very difficult to eradicate as they survive in nutriment-poor environment, form biofilms and are resistant to standard disinfectants and high temperatures.1
M. fortuitum infections are rare, usually result from a direct inoculation (following surgery/catheters or trauma) and involve mainly wound infections. Most of the time, the presence of M. fortuitum in a respiratory specimen is considered to be ‘non-pathogenic’ but a few pulmonary infections have also been described, usually associated with a predisposing pulmonary disease. Rarely M. fortuitum is associated with central nervous system (CNS) infections but never without a prior breach.1
Case presentation
In August 2018, a patient from a sub-Saharan country consulted at the emergency department with a 1-month history of left eye ptosis and abduction limitation. He also had left hemicrania headache for the prior 6 months and a weight loss of 6 kg associated with asthenia.
His medical history included a schistosomiasis treated in 2016 and 2018 and a self-resolved hepatitis B infection. Since he arrived in Switzerland in 2015, he only travelled to Tchad once in 2018.
On admission (day 1), the patient was febrile (39°C). Neurological examination revealed an abduction, adduction and vertical gaze palsy of the left eye associated with a left ptosis. We also noted a hyperaesthesia of the left side of the face and a left hypoacusis. These signs combined suggested a cavernous sinus syndrome.
Laboratory findings were non-contributive. Notably, there was no inflammatory syndrome (C-reactive protein and blood cells count normal).
A brain CT scan demonstrated a tissular lytic lesion of the petroclival left synchondrosis, extending to the carotid canal and the cavernous sinus up to the orbital apex (figure 1A). On the upper pulmonary images, multiple bilateral pulmonary micronodules and nodules randomly distributed were visible. A brain MRI was performed and better characterised the perineural extension of the trigeminal nerve (figure 1B–F).
CT scanner illustrates lysis of the petrous apex and wall of the left carotid canal (arrows in A). MRI imaging shows an infiltrative process responsible for enlargement of the left cavernous sinus (arrows in B–D) and associated fluid in the mastoid (E,F).
This patient had no prior immunosuppression. HIV test was negative twice.
These findings were at that time highly suspicious of a cerebral tumour process with pulmonary metastasis.
Investigations
We performed extensive cerebrospinal fluid (CSF) and blood infectious analyses and sputum examinations. Results are presented in table 1.
Investigations performed on cerebrospinal fluid (CSF), blood, sputum and stool
Material | Exams |
CSF | Leucocytes: 1 M/L (0–5), proteins: 0.45 g/L (0.15–0.45), glucose: 3.7 mmol/L (2.8–4) PCR neurotropic viruses panel (enterovirus, HSV-1, HSV-2, parechovirus, VZV): NEG PCR M. tuberculosis: NEG Bacterial broad range PCR: NEG Bacterial cultures: NEG (including mycobacteria, 8-week incubation) |
Blood | Cultures: NEG IGRA: NEG Serologies CMV, EBV, HSV1, HSV2, VZV, HHV6: NEG for recent infection Serologies syphilis, Borrelia burgdorferi, toxoplasmosis, Bartonella henselae, Histoplasma capsulatum, Coccidioides immitis: NEG for recent infection CD4 393 cells/µL Flow cytometry: normal |
Sputum | Cultures : NEG 1×; POS 1× Mycobacteria spp (M. fortuitum identified by sequencing) Direct exam AARB : NEG 2× PCR M. tuberculosis : NEG 2× |
Stool | Parasites : SAF +Baermann test : NEG |
-
AARB, Acid-Alcoholic Resistant Bacilli ; CMV, cytomegalovirus; CSF, cerebrospinal fluid; EBV, Epstein-Barr virus; HHV, human herpes virus; HSV, herpes simplex virus; IGRA, interferon-gamma release assays; NEG, negative; PCR, polymerase chain reaction; POS, positive; SAF, sodium acetate-acetic acid-formalin; VZV, varicella zoster virus.
At day 3, we were informed that one out of two sputum cultures for mycobacteria was positive and M. tuberculosis PCR being negative, a non-tuberculous mycobacterium was assumed. After sequencing, it was identified as M. fortuitum. The second sputum remained sterile. The patient did not have any respiratory symptoms, only one sputum was positive for M. fortuitum and we had an alternative explanation for the pulmonary lesions with the suspicion of an oncological disease affecting the brain and lungs. For these reasons, we considered this germ non-pathogenic and decided initially not to treat. The infectious workup being negative, we suspected the mass to be oncological. The tumour board (composed of ear nose throat (ENT) surgeons, neurosurgeons and oncologists) evaluated the case and recommended an elective ENT endoscopy in search of a cavum tumour and, if possible, a transnasal diagnostic biopsy of the intracranial mass.
At day 31, while the patient was waiting for the endoscopic biopsy, he presented an acute neurological deterioration and a new MRI showed a progression of the mass with extension to the right cavernous sinus and a perineural infiltration of the fifth right cranial nerve (figure 2). The neurosurgeons performed a therapeutic and diagnostic craniotomy with removal of the mass in order to release the left optic nerve and obtain specimens. The surgeons described a white milk-like liquid.
MRI shows the enlarged cavernous sinus with perineural extension of the trigeminal nerve (arrows in A, B).
Histopathological examination revealed an inflammatory tissue with a lymphocytic infiltrate and multiple granulomas, some with necrosis (figure 3). The Ziehl, Grocott and Gram stains were negative. There was no element suggesting a tumour. The PCR for M. tuberculosis was negative on the biopsy specimen (amplification sequence IS6110, includes M. tuberculosis, M. bovis, M. africanum, M. bovis var. BCG and M. microti detection). Cultures for bacteria and fungi remained sterile.
Histopathology of the intracranial mass biopsy showing inflammatory tissue with lymphocytes and multiples granulomas, some necrotising, containing numerous macrophages and giant multinucleated cells.
Differential diagnosis
The presence of an intracranial mass with diffuse bilateral pulmonary nodules suggested a differential diagnosis including an oncological, infectious or immunological disease. According to radiologists and oncologists, the images were suggestive of a tumorous process. Because of the origin of the patient, a disseminated mycobacterial disease was also probable. Finally, considering the simultaneous pulmonary and cerebral involvement, the possibility of sarcoidosis was considered. The normal cellularity of the CSF and the absence of detection of any bacterial or viral pathogens despite a thorough microbiological workup did not support the possibility of other infectious aetiologies. The finding of M. fortuitum in only one sputum analysis was considered non-pathogenic because: (1) the patient had no pulmonary symptoms; (2) only one specimen was positive and (3) there was no prior report of an intracranial mass and a pulmonary infection caused by M. fortuitum without prior breach or associated risk factor. However, the biopsy of the mass showed no tumour cells and multiple granulomas. Two main diagnosis were favoured at that time: mycobacterial infection or sarcoidosis.
Treatment
Glucocorticosteroïds and a standard empirical antituberculous therapy (ethambutol 1000 mg once a day intravenous, pyrazinamide 1000 mg once a day per os, rifampicine 600 mg once a day intravenous, isoniazide 250 mg once a day intravenous) were introduced simultaneously. Additional biological testing for a possible sarcoidosis was inconclusive. The brain specimen was kept in culture for fastidious/slow-growing organisms including mycobacteria.
At day 48, we were informed that a mycobacterium was growing in the brain specimen (12th day of incubation). Until definitive identification was available, we continued rifampicin and isoniazid, stopped pyrazinamide and ethambutol and added amikacin, imipenem and ciprofloxacin (to cover a possible non-tuberculous mycobacteria).
About 10 days later, the mycobacterium was identified as M. fortuitum by sequencing and treatment was adapted accordingly: only amikacin, imipenem and ciprofloxacin were continued (figure 4). The resistance profile is shown in table 2.
Susceptibilities of Mycobacterium fortuitum
Antibiotics | M. fortuitum (intracranial mass) |
Cefoxitin | 32 I |
Imipenem | 4–8 I |
Meropenem | 4 S |
Amikacin | 2 S |
Gentamicin | 16 |
Tobramycin | 64 R |
Ciprofloxacin | 0.125 S |
Levofloxacin | 0.25 S |
Moxifloxacin | <0.0625 S |
Minocycline | 1S |
Doxycycline | <0.5S |
Tigecycline | 2 |
Linezolid | 2–4 S |
Ethambutol | 16 |
Clofazimine | <0.125 |
-
Numbers represent minimal inhibitory concentrations in mg/L (EUCAST (European Committee on Antimicrobial Susceptibility Testing) norms).
-
I, intermediate; R, resistant; S, susceptible.
Timeline of exams, findings and treatments. AMK, amikacin; CIP, ciprofloxacin; DOX, doxycycline, ED, emergency department; EMB, ethambutol; ENT, ear nose throat; IMI, imipenem; INH, isoniazid; LZD, linezolid; MXF, moxifloxacin; PZA, pyrazinamide; RIF, rifampin.
The evolution was good under this regimen, both clinically and radiologically (figure 5).
MRI images: note signal anomalies in the left temporal lobe along the passage of the biopsy needle (arrows A, B) and the almost complete regression of collections (asterisks in C) in the left petrous apex.
After 4 weeks of targeted intravenous therapy, the patient was discharged on oral therapy composed of moxifloxacin, linezolid and doxycycline. The gaze palsy was at that time almost resolved and the patient did not have any other symptoms.
A follow-up brain MRI and lung CT scan were performed approximatively 10 weeks after discharge, showing improvement of the brain lesion and pulmonary nodules. At that time, we decided to stop linezolid in order to avoid neurotoxicity and we pursued moxifloxacin and doxycycline for a total duration of at least 1 year (figure 4).
Outcome and follow-up
To understand why this immunocompetent patient presented with a disseminated infection, we performed high throughput sequencing of a panel of genes involved in immune system disorders. The exome sequencing did not reveal any abnormality in the genes studied (panel of 473 genes).
During follow-up, symptoms improved: eye motility came back to normal and the patient only suffered from intermittent headaches. Brain MRI and lung CT scan were performed every 2–3 months, to ensure that evolution was satisfactory.
Tolerance to antibiotics was regularly assessed through clinical examination, ECG and blood tests. We did not notice any prolonged QT, tendinopathy or confusion, the main side effects of moxifloxacin. Regarding linezolid, cell counts and liver tests were assessed every 2 weeks. Due to a worsening thrombopenia, we reduced the dosage of linezolid from 2×600 mg/day to 1×600 mg/day 1 week after introduction. Platelets level improved and no further toxicity was noticed. The patient also had regular ophthalmological check-ups. The patient had intermittent abdominal discomfort, a possible side effect of doxycycline. As mentioned earlier, after 10 weeks of oral therapy and with satisfactory follow-up, we decided for a step down. In order to avoid an eventual linezolid-induced neuropathy, we chose to stop linezolid first.
However, despite what we considered as a positive clinical response, the patient spontaneously interrupted his treatment for approximatively 4 weeks. On the MRI performed 8 months after the beginning of the treatment (including the interruption and the ‘lost to follow-up’ period), the residual collection was easier seen than on the previous MRI. We therefore decided to reintroduce linezolid 600 mg once a day (in addition to moxifloxacin and doxycycline). We continued to monitor tolerance as previously described. No further toxicity was noticed. The treatment was stopped after a total of 13 months with a favourable radiological and clinical evolution (figure 4). A clinical follow-up was scheduled every 6 months with a brain MRI and chest CT scan for a 2-year period.
Discussion
In a recent review involving 115 cases of RGM infections over 5 years, only 14 involved a M. fortuitum infection and none of them was a CNS infection.2
In our case, M. fortuitum was first identified in a sputum sample but the patient had no pulmonary symptom. M. fortuitum isolated from pulmonary specimens is often a colonisation, especially in association with bronchiectasis and it does not require a specific treatment.1
Another review of microbiological aetiology of brain abcesses3 reported only one infection due to M. fortuitum out of 50 brain abscesses studied. Only a few case reports describe CNS infection with M. fortuitum. They were all associated with either a previous traumatic or surgical breech or followed insertion of neurosurgical material.4–12 An overview of recently published case reports is presented in table 3.
Overview of recent case reports of M. fortuitum central nervous system (CNS) infections
Reference | Infection | Age and sex |
Predisposing condition | Surgery | Therapy | Treatment duration | Evolution |
Marie et al 13 | Lung infiltrate and multiple cerebral lesions | 63-year-old F | Lipoid pneumonia | No | 1st line: RIF+CIP+CLR 2nd line : AMK+IMI+ CLR followed by IMI+CLR+ CIP |
1st line: 6 months Relapse: 10 months |
Relapse |
Midani et al 7 | VP shunt infection | VP shunt (spina bifida) |
VP shunt removal | TMP/SMX+AMK followed by TMP/SMX |
6 weeks intravenous 6 months po |
Recovery | |
O'Brien and Rawluk8 | Epidural mass | 39-year-old M | Epidural injection | Yes | INH+EMB+RIF+ PZA |
6 weeks | Recovery |
Smith et al 11 | Meningitis | 28-year-old M | AIDS, CD4 24/μL |
No | 1st line: SULF+TET 2nd line: PEN+CRO then INH+EMB+RIF+ PZA 3rd line: AMK+DOX |
Unknown | Relapsed three times then Death (postmortem diagnosis) |
Madaras-Kelly et al 6 | Meningitis | 45-year-old F | Epidural catheter insertion | Catheter removal | RIF+INH+ CIP followed by CLR+IMI+DOX+ CIP |
12 months | After initial poor response and catheter removal, full recovery |
Santamaria-Jaúregui et al 10 | Meningitis +traumatic wound sacral region | 16-year-old M | Accident with wound and foreign body in the sacral region | Drainage +foreign body removal | INH+TMP/SMX | Not specified, at least 12 months (last follow-up) | Recovery at 1-year follow-up |
Fujikawa et al 19 | Meningitis | 57-year-old F | Meningioma surgery | No | CLR+LVX | Unknown | Recovery |
Cadena et al 20 | VP shunt infection +abdominal pseudocyst infection | 14-year-old M | Elective distal shunt revision (congenital hydrocephalus) |
Shunt removal | MXF+MEM+AMK+LZD followed by TMP/SMX+MXF+ MEM |
9 months | Stable at 2-year follow-up |
Tankhiwale and Katkar21 | Subdural empyema | 14-year-old M | None | Drainage | CLR+CIP | 6 weeks | Recovery |
Kell et al 12 | Chronic meningoencephalomyelitis | 33-year-old F | None | VP shunt insertion | TGC+AMK+ MEM followed by TGC+MEM |
Not specified, at least 18 months (last follow-up) | Multiple relapses leading to VP shunt. No relapse afterwards at 18-month follow-up |
-
AMK, amikacin; CIP, ciprofloxacin; CLR, clarithromycin; CRO, ceftriaxone; DOX, doxycycline; EMB, ethambutol; F, female; IMI, imipenem; INH, isoniazid; LVX, levofloxacin; LZD, linezolid; M, male; MEM, meropenem; MXF, moxifloxacin; PEN, penicilline; PZA, pyrazinamide; RIF, rifampin; SULF, sulfadiazine; TET, tetracycline; TGC, tigecycline; TMP/SMX, trimethoprim/sulfamethoxazole; VP, ventriculoperitoneal.;
Only one previous case report describes a disseminated M. fortuitum infection involving lungs and CNS.13 Contrary to our patient, in this case M. fortuitum was never recovered from a CNS specimen, but the evolution of the brain lesions was favourable under specific treatment for M. fortuitum. The patient had a lipoid pneumonia as suspected predisposing factor.
Due to the localisation of the lesion, we also looked for skull floor M. fortuitum infection reports. None could be found. The first CT scan of our patient described also an otomastoiditis: therefore, we looked for similar cases. The most extensive paper we could find reviewed 16 cases of non-tuberculous otomastoiditis over a 12-year period in two Swedish hospitals.14 Only two cases were due to M. fortuitum, of which only one showed a CNS invasion. Both patients were children (aged 6): one suffering from a serous otitis media and the other from recurrent otitis media. They presented with secretion and polyps evolving for months and both required a mastoidectomy. This is therefore a very different presentation from our patient.
We also searched for reports of sinus M. fortuitum infection but found none.
In our case, we postulated that the route of entry was probably respiratory with a pulmonary subclinical involvement accompanied by an initial local invasion through the sinuses, even if the CT scan at the time of the presentation did not reveal any sign of sinus infection.
This is to our knowledge the first case describing a disseminated M. fortuitum infection with intracranial invasion and lung involvement without any underlying immunodeficiency or traumatic trigger. Of note, our patient did not have any lung disease predisposing to mycobacterial infection.
The treatment was chosen based on the literature available at that time and the susceptibility profile. There are no controlled trials for treatment of RGM infections. To prevent emergence of resistance during therapy, a multidrug regimen is recommended for treatment of M. fortuitum infections. Serious or disseminated infections require an initial intravenous therapy.15 RGM do not respond to antituberculous agents but are sensitive to various standard antibiotics. M. fortuitum is considered the ‘most antibiotic sensitive’ RGM. It is usually sensitive to quinolones, imipenem and amikacin,2 the empirical regimen chosen for this patient while awaiting for the resistance profile.
The oral therapy was later decided taking into account CNS drug penetration, toxicity profile and susceptibility pattern. Moxifloxacin, doxycycline and linezolid have been used to treat various CNS infections and their CSF concentrations have been previously documented. These molecules are often cornerstones of treatment of RGM infections; therefore, we chose this combination for oral therapy.15 16
It is important to note that an inducible macrolide resistance (erm) gene was identified in M. fortuitum, which may lead to in vivo resistance despite in vitro susceptibility. Macrolides therefore are not an appropriate therapy.
The final diagnosis required almost 2 months and a craniotomy. This surgery was performed because of neurological deterioration. The goal was primarily therapeutic; the initial plan was to perform a transnasal diagnostic biopsy. The presence of M. fortuitum in a specimen obtained through sinuses might have been considered non-pathogenic and even expected in a patient ‘colonised’ by M. fortuitum in the lungs. Indeed, we initially found M. fortuitum in one sputum sample and considered it as non-pathogenic. M. fortuitum is known to colonise the respiratory tract especially in patients with underlying pulmonary disease; its pathogenicity is still unclear.17
American Thoracic Society/Infectious Diseases Society of America (ATS/IDSA) criteria for diagnosis of non-tuberculous mycobacteria disease include pulmonary symptoms, nodules or cavities on chest radiograph or high-resolution CT scan showing multifocal bronchiectasis with small nodules, and tree in bud appearance and appropriate exclusion of other diagnoses. Microbiological criteria require at least two positive sputum samples or one bronchial wash or lavage.18 Initially, our patient did not fulfil these criteria since he had no underlying pulmonary disorder such as bronchiectasis; only one sputum was positive and we had at that time another plausible diagnosis which was a tumorous process. This suspicion led us to investigate primarily the CNS mass. A different approach could have been to perform a bronchial lavage first. The presence of M. fortuitum in a bronchial fluid culture would probably not have changed the decision to perform a biopsy of the mass: M. fortuitum CNS infections without predisposing factors have never been reported and a definitive pathological diagnosis of the mass was warranted.
The craniotomy surgery was therefore a key element for the final diagnosis and the successful therapy.
Learning points
-
M. fortuitum infections are rare. Their diagnostic is difficult (requiring sequencing a good-quality specimen), suggesting that their frequency and their pathogenicity are probably underestimated.
-
It is to our knowledge the first case report of a disseminated infection in a patient without predisposing factors.
-
The number of M. fortuitum infections will probably increase in the next years with the increased performance of diagnostic tools (such as identification by sequencing).
-
In order to gain expertise in the diagnosis and treatment of such infections, it is of crucial importance to communicate about them.
Footnotes
-
Contributors Sd: design and conceptualisation of the case report, involved in clinical care, acquisition of data, analysis and interpretation of data, drafting of manuscript. M-IV: involved in interpretation of CT scan and MRI, revision of manuscript for intellectual content. AC and J-PJ: involved in clinical care, acquisition of data, analysis and interpretation of data, revision of manuscript for intellectual content.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Patient consent for publication Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2020. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer