Jump to: Page Content, Site Navigation, Site Search,
You are seeing this message because your web browser does not support basic web standards. Find out more about why this message is appearing and what you can do to make your experience on this site better.
Published 19 August 2009, doi:10.1136/bmj.b2985
Cite this as: BMJ 2009;339:b2985
James Powell, ST1 core medical trainee in infectious diseases
1 Flat 6, 233 Acton Lane, Chiswick, London W4 5DD
jbpowell{at}doctors.org.uk
A 64 year old woman presented to hospital with fevers, chills, lethargy, and poor appetite. The symptoms had begun six days previously, when returning to the United Kingdom after a three week holiday visiting friends and family in Gujarat, India. She also felt nauseous, had a moderately severe bilateral frontal headache, and was constipated. Her only medical history was a thyroidectomy in 1982. Before travelling she received all relevant vaccinations, including hepatitis A and B and typhoid fever. She had not taken malarial prophylaxis. Her only regular medication was thyroxine 150 µg once a day.
On examination the patient seemed well. Her temperature was 37.3°C, pulse 80 beats/min regular, blood pressure 101/67 mm Hg, and respiratory rate 15 breaths/min. Her abdomen was soft but tender in the right upper quadrant, with no demonstrable organomegally; bowel sounds were normal.
Routine investigations showed a haemoglobin of 113 g/l, mean corpuscular volume 85 fl (78-100), white cell count 3.4x109/l (4.5-11.0), neutrophils 2.1x109/l (1.80-7.70), lymphocytes 0.99x109/l (1.10-4.80), platelets 194x109/l (140-450), C reactive protein 126 mg/l (<5), erythrocyte sedimentation rate 40 mm/h, sodium 136 mmol/l (135-145), potassium 3.6 mmol/l (3.5-5.1), creatinine 61 µmol/l (53-97), urea 3.1 mmol/l (2.5-6.7), alanine aminotransferase 68 U/l (0-55), alkaline phosphatase 242 U/l (40-150), total bilirubin 11 µmol/l (3-21), albumin 35 g/l (35-50), and corrected calcium 2.45 mmol/l (2.15-2.58).
Malaria parasites were not seen on three thick and thin blood films over three days. Chest radiography showed no abnormalities.
Short answers
Long answer 1. Diagnosis
Having excluded malaria, the syndrome of fever, chills, malaise, anorexia, headache, and constipation combined with leucopenia and moderately raised liver enzymes in a traveller returning from the Indian subcontinent suggests enteric fever, which has an incubation period of 10-20 days. Most cases seen in the UK are acquired in India by people visiting friends and relatives. Other classic features are a sustained high fever with relative bradycardia and a "rose spots" rash on the central torso. Because the patient is relatively well, is a returning traveller, and received the typhoid vaccine, paratyphoid fever (rather than typhoid fever, the other cause of enteric fever) may be the more likely diagnosis.
Enteric fevers are caused by the Gram negative bacteria Salmonella typhi and Salmonella paratyphi A, B, or C, which cause typhoid and paratyphoid fever, respectively. Enteric fevers are potentially fatal systemic infections and are now mostly travel associated diseases in developed countries.2 Case fatality has been reported to be as high as 20-30%, but appropriate antibiotic treatment reduces this to under 1%.2 3 The disease is usually transmitted by ingestion of food or water contaminated with the faeces of infected people. No non-human vectors exist. Complications that occur in 10-15% of cases include intestinal perforation and haemorrhage, shock, toxic myocarditis, encephalopathy, hepatitis, myocarditis, pneumonia, disseminated intravascular coagulation, thrombocytopenia, and haemolytic uraemic syndrome.4
Epidemiology
Paratyphoid fever has become increasingly common in the UK in returning travellers vaccinated against typhoid.2 In England and Wales in the years 2000-8, laboratory identified cases of S paratyphi A outnumbered those of S typhi in 2001, 2002, 2004, and 2006.5 Several factors are probably responsible for this change. Firstly, the global epidemiology of enteric fever is changing. In India increasing numbers of cases of enteric fever are caused by S paratyphi A,6 and most cases of enteric fever in the UK are contracted from the Indian subcontinent. Secondly, vaccines may protect travellers against typhoid but not paratyphoid fever.
In 2006 around 500 laboratory confirmed cases of enteric fever were reported in England, Wales, and Northern Ireland,7 most of them (94%) in travellers to the Indian subcontinent.7 More than 76% of people who contracted an enteric fever had travelled to visit friends or relatives.7 This is reflected by figures showing that between May 2006 and April 2007, 85% of enteric fevers occurred in those of Indian, Pakistani, or Bangladeshi ethnicity.7
The risk of contracting enteric fever is greatest when visiting areas with high incidence rates. The Indian subcontinent, South East Asia, and Far East Asia have the highest incidence worldwide of >100 per 100 000. Areas such as Africa, South America, Russia, and China have incidence rates of 1-100 per 100 000, and in North America and the UK rates are <1 per 100 000.2
Other risk factors for travellers include visiting rural areas, increased length of stay in an endemic area, and visiting relatives or friends. Such people are thought to be less likely to seek advice before travel, less likely to receive the typhoid vaccine, and less likely to take precautions with food and water during their stay.2
Vaccination
Two vaccines for typhoid fever are currently licensed in the UK: an intramuscular polysaccharide and oral live attenuated vaccine. Neither protects against paratyphoid fever, and immunity to typhoid can be "overcome" by a high inoculating dose of the bacterium.2 The maximum efficacy achieved by either of these vaccines is 75%, with cumulative three year efficacies of 50-60%.3 Protection wanes with time, and boosters are needed yearly with the oral vaccine and every three years with the polysaccharide vaccine.3
Long answer 2. Confirmatory tests
Cultures should be 100% specific. However the use of antibiotics in endemic areas makes accurate sensitivities and specificities hard to define.8 Blood cultures form the cornerstone for diagnosis and are 40-80% sensitive.8 Sensitivities are higher if multiple blood samples of adequate volume are obtained (10-15 ml for adults and school children).4 Sensitivities are lower when stool and urine are used, and these samples are not positive until one week after infection.8
If the patient is already taking antibiotics blood cultures may be negative, and bone marrow culture—which is the gold standard for diagnosis—will provide the best chance of culturing the organism.4 Bone marrow culture is more sensitive than blood culture, but the invasive nature of the procedure means that it is rarely performed.8
Long answer 3. Treatment
General advice
Antibiotics are the only effective treatment for enteric fevers. Most patients without complications can be treated with oral drugs on an outpatient basis.4 Paratyphoid fever needs to be treated in the same way as typhoid fever, because clinical disease and complications associated with S paratyphi A infection can be indistinguishable from those of S typhi.2
Antibiotic resistance is a growing problem that makes the first choice of treatment difficult, so expert advice should be sought. Two types of resistance are recognised—firstly, multidrug resistant (MDR) strains that are resistant to drugs such as chloramphenicol, ampicillin, and co-trimoxazole; and secondly, those resistant to fluoroquinolones such as ciprofloxacin.4 Of all the laboratory confirmed cases of enteric fever in the UK from May 2006 to April 2007, 92% of typhoid and 93% of paratyphoid cases acquired in India were ciprofloxacin resistant.7
WHO guidelines
Guidelines from the World Health Organization for the optimal treatment of uncomplicated enteric fever recommend that fluoroquinolones such as ciprofloxacin can be used for five to seven days for fully sensitive strains or MDR strains.4 Oral cefixime can also be used for seven to 14 days to treat MDR strains. Ceftriaxone for 10-14 days or oral azithromycin for seven days are recommended for the treatment of fluoroquinolone resistant strains.4 WHO guidelines for severe or complicated disease recommend intravenous treatment, which should be extended to 10-14 days in all cases.4
Conclusion
With susceptible organisms, fluoroquinolones may be superior to chloramphenicol and cefixime for reducing clinical relapse and better than ceftriaxone and cefixime for reducing clinical failure.9 Therefore, once sensitivities are known ciprofloxacin is a good choice of antibiotic, but caution is advised because sensitivity only to nalidixic acid confirms true quinolone sensitivity.10 It is worth noting that even if resistant to fluoroquinolones the bacteria may be sensitive to amoxicillin or chloramphenicol, which are recommended by WHO as "alternative" effective drugs for treating enteric fever.4
Initial empirical treatment must be based on the likely resistance pattern of the organism according to region of the world in which the disease was acquired. Cases acquired on the Indian subcontinent must be assumed to be resistant until proved otherwise, because MDR and fluoroquinolone resistance are common.6 7 Subsequent treatment can be guided by the resistance pattern of any cultured organisms and clinical response. If no organism is cultured and a presumptive diagnosis of enteric fever is made, resistance should be assumed. In such cases, intravenous therapy with ceftriaxone or oral therapy with cefixime or azithromycin would cover MDR and fluoroquinolone resistant organisms.
Long answer 4. Advice
Up to 2-5% of people with enteric fever become chronic carriers (they excrete the bacteria in their stools for more than one year).3 The risk increases with increasing age and presence of a biliary tract abnormality.3
Microbiological clearance is defined by negative stool cultures. The number of negative cultures needed depends on whether the person has confirmed enteric fever, is a contact of a case, or is a carrier. Antibiotics should be used to achieve microbiological clearance.1
With regard to exclusion from work, school, or other institutional settings, each case must be considered individually. Factors such as type of employment, provision of sanitation facilities, and standards of personal hygiene should be taken into account.1 Decisions should be made by people experienced in such evaluations.
Relapses can occur in as many as 20% of apparently successfully treated cases, although rates vary according to the antibiotic used.4 Patients should therefore be followed up in clinic and be aware of recurring symptoms in the following weeks.
Long answer 5. Who to inform
Clinical suspicion is all that is needed to notify the Health Protection Unit, and the notification can be amended if the diagnosis is proved wrong. Any undue delay in notifying may mean that the legal requirement has not been discharged.1
To be valid, each notification should be signed by the notifying medical practitioner. However, in practice, telephone or telefax notifications are allowed.1
Patient outcome
Blood cultures from this patient grew S paratyphi A, which was resistant to ciprofloxacin and naladixic acid but sensitive to ceftriaxone, amoxicillin, and trimethoprim. Stool cultures yielded no growth.
This patient was initially treated empirically with ceftriaxone 2 g intravenously daily. Culture and sensitivity results became available after two days of intravenous therapy, and the patient was started on amoxicillin 500 mg orally three times a day. The patient was discharged to complete 14 days of antibiotic treatment in total.
The patient was seen one month later in clinic. She had remained well, with no fevers, and she had a good appetite. Her liver function tests were returning to normal.
Cite this as: BMJ 2009;339:b2985
Provenance and peer review: Not commissioned; externally peer reviewed.