Shigella spp.

Associated infections

• Bacillary dysentery and acute gastroenteritis
• Bacteraemia, most often in malnourished or immunocompromised patients
• Haemolytic uraemic syndrome (S. dysenteriae type 1, Shiga toxin)
• Reactive arthritis
• Seizures and encephalopathy in children
• Toxic megacolon and intestinal perforation
• Rectal prolapse in young children
• Protein-losing enteropathy and malnutrition following prolonged illness

Transmission route

Transmission is faecal-oral. The very low infectious dose makes direct person-to-person contact the dominant route, supplemented by contaminated food and water and, in some populations, sexual transmission. Waterborne outbreaks are documented, generally associated with untreated or inadequately treated supplies, recreational water, or contamination of a distribution system following a loss of pressure or an intrusion event. Shigella is comparatively fragile outside the host and does not establish itself as a persistent environmental coloniser.

In endoscope reprocessing, Shigella is not an expected organism and is not a routine target of final rinse water testing. It does not participate in the oligotrophic biofilm communities that colonise reverse osmosis plant, storage vessels and distribution pipework, it does not survive well in treated water, and it is readily inactivated by validated high-level disinfection. Any confirmed isolation from final rinse water or from a patient-ready endoscope therefore represents a significant abnormal finding indicating faecal contamination of the water path or of the reprocessing process. Appropriate response is immediate quarantine of the reprocessor and affected endoscopes, escalation to infection prevention and the treating service, investigation for cross-connection, backflow or ingress of untreated water, verification of manual cleaning practice and of filter integrity, and assessment of whether a patient look-back exercise is warranted. Given the low infectious dose, the finding should not be attributed to sampling artefact without a documented investigation.

The low infectious dose deserves particular emphasis because it changes how a low count should be read. For most organisms encountered in reprocessing water surveillance, a count near the limit of detection prompts a proportionate response scaled to the number recovered. For Shigella, the number of organisms required to establish infection in a healthy adult is within the range that a single low-count sample could represent, and in a patient with reduced gastric acidity, which describes a large proportion of the gastroenterology population, that threshold is lower still. A low count is therefore not reassuring, and the response should be driven by the identification rather than by the enumeration.

Relevance in endoscopy and reprocessing

There is no established literature attributing Shigella transmission to flexible endoscope reprocessing, and the organism should not be presented as a recognised endoscopy-associated pathogen. Unlike Salmonella, which has documented historical endoscopic transmission, and unlike Pseudomonas and the non-tuberculous mycobacteria, which have well-described roles in reprocessing failures, Shigella appears in reprocessing reference material because of what its detection would indicate rather than because of any body of case reports. The absence of documented transmission almost certainly reflects the organism's fragility outside the host and the effectiveness of routine reprocessing rather than any protective feature of the endoscopy pathway, and biological plausibility for carry-over exists wherever manual cleaning fails.

What a detection does indicate is unambiguous: human faecal material has reached a place it should not be. In a water sample, that means faecal contamination of the treated water path, which in practice means a cross-connection, a backflow event, an intrusion into the distribution system, or contamination introduced at the point of sampling. In an endoscope channel sample, it means gastrointestinal contents from a patient have survived the cleaning and disinfection process, or have been introduced after it. Neither finding is explicable by water system ecology, because the organism has none of the characteristics that would allow it to colonise reprocessing plant. It does not form oligotrophic biofilm, it does not tolerate the low-nutrient conditions of a reverse osmosis loop, it is removed by reverse osmosis and by bacterial-retentive filtration, and it does not persist in a terminal connecting tube or a reprocessor manifold. Searching for a Shigella reservoir within the treated water system is therefore not a productive line of investigation.

Against validated high-level disinfection Shigella has no relevant tolerance whatsoever. It is inactivated readily by peracetic acid, glutaraldehyde and ortho-phthalaldehyde at in-use concentrations, by thermal disinfection at washer-disinfector parameters and by free chlorine at drinking water residuals. Where it survives in a reprocessing setting, the explanation is physical shielding within organic soil, an unbrushed or unperfused channel, a damaged lumen, or an incomplete cycle. Drying and storage matter less for this organism than for the waterborne non-fermenters, since it is not expected to be introduced from rinse water and will not proliferate in a clean channel, though retained moisture will extend the survival of anything already present. For CSD and dental water services the interpretation is identical: a Shigella detection is evidence of faecal ingress or gross process failure, never of waterline colonisation.

Interpreting a detection

A confirmed Shigella isolate from a water, final rinse or endoscope sample is an abnormal result with no benign routine explanation, and should be handled as a potential serious incident from the point of notification. The organism is not treated water flora, is not an environmental coloniser of reprocessing plant, and is not a plausible incidental laboratory contaminant. The first step is to confirm with the laboratory that the identification is genuine and to establish the species, since the identification is technically demanding given the organism's close relationship to E. coli and since S. dysenteriae type 1 carries additional implications through Shiga toxin production.

Sampling and handling artefact must be actively considered, but because the reservoir is exclusively human, the mechanisms are specific and identifiable rather than generic. The realistic possibilities are contamination of the sample or the sampler's hands or gloves in the decontamination area, contact between the sample and a drain, sink or soiled item, contamination by a sampler who is themselves a case or a convalescent excreter, or laboratory cross-contamination from a concurrently processed faecal specimen. Each can be checked directly: identify who took the sample and what they had handled, confirm outlet disinfection and pre-flush, confirm use of a sterile neutralising container, check the transport interval, and ask the laboratory what else was processed in the same batch. Establishing artefact requires a positive finding of one of these mechanisms; the absence of an obvious water-system explanation is not itself evidence of artefact, and given the low infectious dose the default assumption must be that the result is real.

Where artefact is not demonstrated, the investigation should proceed along both a water track and a process track. On the water side, review backflow prevention devices and any potential cross-connection between treated water and mains, waste, drain or non-potable services; review recent plumbing work, temporary hoses and bypasses; check the terminal filter's integrity and change record and whether its housing has been disturbed; verify reverse osmosis performance and whether product water has been blended; and consider whether drain or waste water could enter the reprocessor chamber. On the process side, audit manual cleaning by direct observation, covering brush selection and condition, channel access and perfusion, detergent dilution and contact time, leak testing, the delay between procedure and cleaning, and whether any cycle has been interrupted or aborted. Where the isolate came from an endoscope channel rather than from water, the process track is the priority and the water review is secondary.

The usual distinction between a single isolate and a trend does not apply. One confirmed isolate is sufficient to justify immediate quarantine of the affected reprocessor and of every endoscope processed on it since the last satisfactory result, and to justify a full investigation. Waiting for confirmation on a repeat sample before acting is not appropriate for an organism with this infectious dose. Escalation to infection prevention should be immediate, and should extend to the facility executive and the relevant state or territory health authority where the result is confirmed and a patient exposure pathway cannot be excluded, shigellosis being notifiable in all Australian jurisdictions. Isolates must be retained for typing so that any subsequent clinical isolate can be compared. The decision on patient look-back belongs to infection prevention and should take account of the interval since the last satisfactory result, the number of patients affected, and whether any has presented with a compatible illness.

Antimicrobial resistance

Antimicrobial resistance in Shigella is advancing rapidly and the organism is designated a priority antibiotic-resistant pathogen by the World Health Organization. Multidrug resistance to ampicillin and trimethoprim-sulfamethoxazole is long established, and resistance to fluoroquinolones and azithromycin has become common in many regions. Ciprofloxacin resistance is now widespread in isolates acquired in South and Southeast Asia, and azithromycin resistance has spread through sexually transmitted lineages internationally.

Extensively drug-resistant strains, defined as resistant to ampicillin, azithromycin, ceftriaxone, ciprofloxacin and trimethoprim-sulfamethoxazole, have increased in prevalence in recent years and leave very limited oral treatment options. These strains have been reported in Australia and elsewhere, predominantly as S. sonnei, and the combination of a very low infectious dose with near-complete oral treatment failure makes them a substantial public health concern. Where such an isolate is identified, susceptibility testing against the full available panel and infectious diseases consultation are essential, and treatment may require intravenous therapy for an illness that would otherwise be managed orally or supportively.

Antimicrobial therapy is indicated more often in shigellosis than in most other causes of bacterial gastroenteritis, because treatment shortens illness and reduces the duration of shedding, which matters for an organism transmitted so efficiently person to person. This makes the loss of oral options clinically significant rather than merely academic, and it also means that a resistant isolate cannot be relied upon to be cleared quickly from a case or carrier.

As with the other Enterobacterales, antimicrobial resistance has no bearing on susceptibility to physical or chemical decontamination. Shigella is inactivated by chlorine at drinking water residuals, by peracetic acid, glutaraldehyde and ortho-phthalaldehyde at high-level disinfection concentrations, by thermal disinfection and by steam sterilisation, regardless of its antibiogram. What resistance changes is the consequence of a failure to remove it, and an extensively drug-resistant isolate recovered from any point in a reprocessing pathway should lower the threshold for escalation, external notification and patient look-back.