Mycobacterium fortuitum

Associated infections

• Skin, soft tissue and wound infection following surgery, trauma or injection
• Post-surgical and catheter-related bloodstream infection
• Surgical site infection following cosmetic, cardiac and breast procedures
• Furunculosis associated with footbaths and cosmetic procedures
• Chronic pulmonary infection, typically in structural lung disease
• Keratitis and other post-procedural ocular infection
• Post-injection abscess following intramuscular or subcutaneous administration
• Pseudo-infection arising from contaminated specimens or instruments

Transmission route

Transmission is environmental rather than person-to-person. Exposure occurs through direct inoculation of contaminated water or instruments into skin, soft tissue, the eye or the bloodstream, or through inhalation of aerosols generated at showers, taps and other outlets. There is no evidence of an animal or human reservoir sustaining the organism, and no requirement for source isolation of colonised or infected patients on the grounds of onward spread.

The practical routes of concern in a healthcare facility all involve water or water-wetted surfaces reaching a vulnerable site. Inoculation events have followed injection through skin cleansed with contaminated solution, surgical procedures where instruments or irrigation contacted non-sterile water, and cosmetic procedures using footbaths or equipment rinsed in tap water. Aerosol exposure is relevant to pulmonary disease, and arises wherever water is atomised — showers, spray taps, humidifiers and, in the reprocessing context, air purge and drying stages that expel residual water as a fine mist.

M. fortuitum is directly relevant to endoscope reprocessing. Because it colonises plumbing biofilm and tolerates chlorine, it can pass through a facility's water supply into an automated endoscope reprocessor and contaminate the final rinse. A bronchoscopy-unit investigation in Spain traced nine patients with M. fortuitum-positive respiratory specimens to tap water feeding a washer-disinfector that lacked a bacterial-retentive filter, with molecular typing matching patient and water isolates in seven cases. Contaminated rinse water recontaminates an otherwise correctly disinfected endoscope at the last step of the cycle, and the organism may subsequently establish biofilm within the reprocessor plumbing and within endoscope channels, where it is protected from routine chemical decontamination. Detection of M. fortuitum in final rinse water therefore indicates a failure of the water treatment train, the terminal filtration, or the reprocessor itself, and warrants immediate investigation rather than repeat sampling alone. AS/NZS 5369 requires NTM to be included in the microbiological monitoring of endoscope rinse water for this reason.

The same reasoning extends beyond flexible endoscopy. In a central sterile department, water used for the final rinse of instruments before packaging and sterilisation is a potential vehicle, and while terminal steam sterilisation will kill the organism, residual biofilm and endotoxin deposited on lumened instruments are not removed by heat. In dental practice, dental unit waterlines present a well-recognised biofilm habitat with narrow-bore tubing, long stagnation periods and low flow, and NTM including M. fortuitum have been recovered from such lines. Water delivered to a handpiece or air-water syringe during a surgical dental procedure contacts exposed tissue directly.

Relevance in endoscopy and reprocessing

M. fortuitum sits at the intersection of the two properties that make an organism dangerous in flexible endoscope reprocessing: it survives the chemistry, and it lives in the water. High-level disinfectants are formulated to achieve mycobactericidal activity, and glutaraldehyde, ortho-phthalaldehyde and peracetic acid will all inactivate M. fortuitum under validated conditions of concentration, temperature and contact time. The documented failures have almost never involved a disinfectant that was intrinsically incapable of killing the organism. They have involved organic soil shielding the cells because manual cleaning was inadequate, disinfectant diluted below minimum effective concentration, channels not perfused because a connector was omitted or a channel was blocked, or — most importantly for water quality — a correctly disinfected instrument being rinsed afterwards with contaminated water. The Spanish bronchoscopy investigation is the clearest illustration: the disinfection step worked, and the rinse step undid it.

Biofilm converts a transient contamination event into a persistent one. Once M. fortuitum attaches to the interior of a rinse water line, a reprocessor tank, a rinse-water storage vessel or an endoscope channel, it produces an extracellular matrix that limits disinfectant penetration and physically protects cells from shear. Cells recovered from mature biofilm are substantially less susceptible than the same strain grown planktonically, and a high-level disinfection cycle validated against suspended organisms cannot be assumed to clear an established channel biofilm. Damage to channel lining — scratches, cracks, delamination at the distal bending section — creates sheltered sites where biofilm is effectively unreachable by any chemical process, which is why instruments with a persistent positive history are eventually withdrawn rather than reprocessed repeatedly. Rinse-water contamination is the mechanism that seeds these sites in the first place.

Drying and storage determine whether a small residual inoculum dies or multiplies. M. fortuitum tolerates desiccation better than most vegetative bacteria, but it cannot replicate without moisture, and residual water in a channel is the single most important determinant of bacterial load at the point of use. Thorough alcohol flush and forced-air drying, followed by hang or cabinet storage that maintains channel dryness, are the controls that prevent overnight amplification. A drying cabinet supplied with air that is inadequately filtered, or a cabinet that dries the exterior but does not actively perfuse the channels, provides false assurance. In CSD and dental settings the analogous concerns are lumened instruments retaining rinse water before packaging, and dental unit waterlines where overnight stagnation in warm, narrow tubing is close to ideal for NTM regrowth; both benefit from the same principle that water left in a lumen is water that will grow something.

Interpreting a detection

A positive M. fortuitum result from a final rinse or reprocessing water sample means that mycobacteria are getting past the barriers intended to exclude them. The first checks are physical rather than microbiological. The terminal bacterial-retentive filter should be examined for correct rating, correct orientation, intact seating and in-service date, since a filter fitted beyond its validated life, wetted and allowed to dry, or bypassed by a poorly sealed housing will pass organisms without any visible sign. Where the filter is intact and current, attention moves upstream to the water treatment train — reverse osmosis membranes, softeners, carbon beds, ultraviolet units and storage vessels are all capable of harbouring biofilm, and carbon filtration in particular removes the disinfectant residual and provides surface area for growth. Reprocessor internals follow: rinse water lines, tanks, dosing lines, non-return valves and the self-disinfection cycle record. Sampling technique is checked last, not first — a mycobacterial result is far more often real than a laboratory artefact, and reflexively attributing it to sampling delays remediation.

A single isolate and a trend carry different weight, but neither is a reason for inaction. One positive from an otherwise clean history is treated as a genuine breach until disproved, with the instruments processed since the last clean result identified and quarantined, and with resampling performed only after a defined remediation step so that the repeat result tests the fix rather than the luck of the draw. Repeat positives across successive rounds, positives from more than one sample point, or positives that recur after remediation indicate established biofilm somewhere in the system rather than an intermittent ingress, and that finding should change the response from cleaning to replacement — of filter housings, of flexible hose, of storage vessels, or of the reprocessor itself where its internal plumbing cannot be verified. Rising counts of any organism on routine total viable count, even when below action level, deserve review alongside NTM results, because both track the same underlying loss of control.

Routine total viable counts do not detect this organism reliably, and services that rely on them are effectively unmonitored for mycobacteria. Standard heterotrophic plate counts are incubated for a short period on general-purpose media at temperatures optimised for common waterborne bacteria; mycobacteria require decontamination of the sample to suppress faster-growing competitors, selective media, and extended incubation, and NTM culture is a separate request that must be specified. M. fortuitum is a rapid grower and is the NTM most likely to appear incidentally, which makes its recovery a useful sentinel: it indicates conditions that will also support the slow-growing species that routine culture will never show. Escalation is warranted immediately on a confirmed result, and should involve the infection prevention team, the water safety group and the facility engineer together, because the fix is almost always an engineering one. Where a bronchoscope is implicated, patient look-back and comparison of patient and water isolates by molecular typing should be considered, since specimens positive for M. fortuitum after a contaminated procedure may otherwise be treated as genuine infection and lead to prolonged, toxic multi-drug therapy.

Antimicrobial resistance

M. fortuitum is intrinsically resistant to the standard anti-tuberculosis agents and to many first-line antibacterials, and treatment relies on susceptibility-directed multi-drug therapy. Isolates are frequently susceptible to amikacin, fluoroquinolones, sulfonamides, doxycycline and imipenem, but the M. fortuitum group characteristically carries an inducible erythromycin ribosomal methylase gene, so apparent macrolide susceptibility on short incubation may not be maintained during therapy; extended incubation or genotypic testing is needed to detect this. The species also possesses a beta-lactamase that limits the utility of most beta-lactams, and monotherapy of any kind risks selection of resistance during the prolonged courses these infections require. Species-level identification within the complex matters, since susceptibility patterns are not uniform across its members.

Beyond antibiotic resistance, the species shows practical tolerance to chlorine-based disinfection at concentrations used in potable supply, and biofilm-associated cells are less susceptible again than planktonic cells. The mechanism is largely structural: the mycolic-acid outer layer is thick, highly hydrophobic and poorly permeable to hydrophilic biocides, so the oxidant reaches its intracellular targets slowly. The same barrier underlies tolerance to chloramine, to low-level disinfectants generally, and to drying. Internalisation within free-living amoebae adds a further layer of protection, since the amoebal cyst wall shields ingested mycobacteria from concentrations of disinfectant that would be lethal to free cells.

The practical consequence for a reprocessing service is that the usual reassurances do not apply. A compliant municipal supply, a satisfactory free chlorine residual, an absence of coliforms or E. coli, and an acceptable heterotrophic plate count are all consistent with the presence of M. fortuitum in the water reaching a reprocessor. Control depends on physical exclusion — validated bacterial-retentive filtration at the point of use, maintained in accordance with its service life — supported by system design that avoids stagnation and dead legs, and by verification through mycobacteria-specific culture. Neither antibiotic susceptibility data nor a chlorinated supply should be relied upon as evidence that a water system is controlled.