Article Text
Abstract
Multiplex PCR can provide rapid diagnosis for patients presenting with an acute undifferentiated febrile illness. Such technology is useful in deployed settings, where access to conventional microbiological diagnosis is limited. It was used in Sierra Leone to guide management of febrile healthcare workers, in whom Ebola virus disease was a possible cause. In particular, it informed appropriate antibiotic treatment while minimising the risk to clinicians of exposure to the causative organism.
- Ebola
- Gastroenteritis
- PCR
- e.coli
- military
- diarrhoea
Statistics from Altmetric.com
Introduction
Febrile illnesses are common in visitors to the tropics and can present a diagnostic challenge. The potential common causes include diarrhoeal disease, respiratory illness and malaria, but many other diseases have a similar presentation.1 While traditionally described as a haemorrhagic disease, Ebola virus disease (EVD) usually begins with an acute, non-specific febrile illness involving headache, myalgia and lethargy, followed by diarrhoea and vomiting. Thus, in febrile patients with potential exposure to the virus, EVD is a potential diagnosis.
EVD is spread by direct contact with an infected person or by contact with their body fluids and is highly contagious; so, healthcare workers are at risk.2 Individuals who are suspected of having contracted the disease should be rapidly isolated in an Ebola treatment unit (ETU) while awaiting diagnosis. Reverse transcriptase PCR (RT-PCR) testing on blood samples taken 48–72 hours after the onset of symptoms is the only effective ‘rule-out’ test.3
In 2014, a British Army field hospital deployed to Sierra Leone as part of the UK's response to the EVD epidemic in West Africa. A 12-bed ETU was set up at Kerry Town, on the Freetown peninsula, tasked with caring for healthcare workers who had, or were suspected of having, contracted EVD. About 50 military personnel worked in direct contact with EVD-positive patients or their body fluids, and were hence called ‘red-zone workers’.
In planning for deployment, there was concern that red-zone workers presenting with febrile illness would have to be treated as a suspected EVD case and admitted to the military ETU as patients.4–8 This would reduce the ETU's operating capacity and expose them to additional risk, should they not have Ebola. An outbreak of febrile illness affecting many personnel might threaten to close the ETU due to bed shortages and clearly would be a cause of anxiety for all concerned.
To decrease the risk of febrile illness, strict hygiene measures were enforced on all military personnel, including setting up handwashing stations at entrances to communal buildings and food collection points. Personnel were ordered to only consume food from approved sources, and the menu was chosen to minimise uncooked food. Bottled water was used for drinking and brushing teeth. Despite these countermeasures, the risk remained significant.
A multiplex PCR machine (Figure 1) was deployed with the field hospital laboratory at Kerry Town in order to provide additional diagnostic support to primary care clinicians.9 Multiplex PCR involves the simultaneous amplification of two or more specific genetic sequences in a single solution. The DNA or RNA sequences are amplified many times until the concentration is high enough to be detected. Faecal samples can be processed using the gastrointestinal panel, which detects 22 different pathogens responsible for illness, including some not routinely cultured in standard laboratory methods such as enteroinvasive Escherichia coli (EIEC). Samples require about 2 min of hands-on preparation, and results are available within an hour. It has a reported sensitivity of 94.5% for 19 of the 22 targets and has a specificity of at least 97.1% compared with conventional stool culture.10
Personnel were instructed to isolate themselves immediately if they had fever, influenza-like symptoms, abdominal pain or diarrhoea. They were then ordered to request assessment from a primary care clinician. The assessment was conducted in such a way as to minimise clinician's exposure to any pathogens, for example, patients were asked to take their own temperature with a tympanic thermometer. If isolation was considered necessary, patients were bedded down in a dedicated area on the camp. If EVD was considered to be a possible cause, the case was discussed with the senior consultant at the ETU who would decide to isolate on the camp, or to refer to the ETU for admission as a suspected Ebola case.
Unwell patients were instructed to collect fresh stool samples themselves. Failing this, patients were asked to take a rectal swab, a previously validated method of EVD detection.11 The samples were then placed in Cary-Blair transport medium, sealed by the patient in a bag, and the primary care clinician wearing gloves and apron ensured samples were triple-bagged before sending them to the laboratory, a 30-min drive away, where they were processed in a portable compliant isolator.
Case reports of a diarrhoeal outbreak
A male, red-zone worker sought help for a 12-hour history of fever, myalgia, lethargy and headache. He denied respiratory or gastrointestinal symptoms, but had been in direct contact with patients confirmed with Ebola for about 1 month. He denied any breaches to his personal protective equipment. His temperature was 38.5°C, and a limited examination was otherwise normal.
Since he had a fever with no focus of infection and had been exposed to patients with EVD, he was admitted to his own unit as a suspected EVD case. On admission, Ebola RT-PCR from blood (FilmArray BioSurveillance System, Biofire) and malaria antigen tests (BinaxNOW) were both negative. Blood cultures were taken. Full blood count, urea and electrolytes, liver function tests, lactate and C reactive protein were normal. Oral rehydration salts and paracetamol were started. On the second day of his illness, he developed diarrhoea and faecal analysis with multiplex PCR detected Shigella/EIEC; so, azithromycin 500 mg once daily was started. Repeat Ebola RT-PCR blood test at 72 hours was also negative. He was discharged back to camp for isolation until his symptoms had resolved.
Over the next 2 weeks, 11 personnel presented feeling unwell (Table 1). A second red-zone worker (case 3) who had direct exposure to Ebola was also admitted to the ETU. Consistent use of the multiplex PCR allowed characterisation of the gastrointestinal pathogen that was causing the presentations, which meant that when the third person with direct EVD exposure (case 9) presented with the same symptoms, they were able to be isolated in camp, as there was now compelling evidence of a diarrhoeal outbreak, as well as microbiological evidence of the organism responsible from use of the multiplex PCR tests. The primary care team completed this with minimal risk of exposure to the causative organism. Stool samples from three patients were later repatriated to the Public Health England national reference laboratory where EIEC was confirmed.
Discussion
Multiplex PCR has been used in deployed secondary care facilities since 2013.9 This is a novel description of its use by primary care clinicians during a febrile illness outbreak, where Ebola was a possible cause.
This multiplex PCR requires about 2 min of hands-on preparation and gives results in about an hour, compared with the days it takes for microbiological culture, though conventional culture is still required to obtain antibiotic sensitivities. Remote advice from a microbiologist or bioscientist is useful to determine the significance of some results. One test costs around £85, which is considerably more than that for standard culture and could discourage widespread use.12 The device takes minimal training to use and is relatively portable, but in the case of a large outbreak, sample processing would still place an extra burden on small, deployed laboratories.
Conclusions
Multiplex PCR is a simple-to-use, easily deployable technology which assisted rapid diagnosis where no microbiological culture was easily available. It improved case management by informing antibiotic use, and once the cause was known, enabled early, empirical antibiotic treatment for individuals as soon as they became symptomatic. By informing clinicians about local disease trends, it helped to prevent admissions to an ETU, and so avoided the risk of contracting EVD from admission as a suspect case. Ultimately, it helped bring about rapid outbreak control, and in so doing helped keep an ETU operational.
Acknowledgments
Biomedical Scientists, Reference Laboratory, Colindale; Dr A Trask, Aberdeen University.
References
Footnotes
Contributors The corresponding author is responsible for the overall content.
Competing interests None declared.
Ethics approval Anonymised, observational report.
Provenance and peer review Not commissioned; externally peer reviewed.