Docstoc

Imported fever a survival guide

Document Sample
Imported fever a survival guide Powered By Docstoc
					JOURNAL

OF

THE ROYAL

SOCIETY OF

MEDICINE

Volume

93

March

2000

Imported fever: a survival guide
Michael Jacobs MRCP(UK) DTM&H
J R Soc Med 2000;93:124±128 SECTION OF ACCIDENT & EMERGENCY, 18 JUNE 1999

The number of travellers worldwide is growing, and at present about 2 million arrive in the UK from tropical and subtropical regions each year. Most escape illness, but an increasingly common dilemma is what to do about fever in such travellers. Fever is a common presentation of tropical diseases and may be the only presenting feature of serious illness. The primary aim of initial management is to prevent morbidity and death from treatable disease. In addition, the doctor must give thought to transmissibility and public health implications. In this short review I discuss the concepts underlying clinical decision-making and offer an algorithm (Figure 1) that highlights the most important issues in immediate careÐhence, a `survival guide'.
THE IMPORTANCE OF MALARIA

Retrospective analyses of the ®nal diagnoses in patients with fever after travel to the tropics (Figure 2) reveal that about half are a cosmopolitan cause of fever and half a tropical disease1,2. In assessing an individual patient, it can therefore be helpful to formulate two differential diagnoses, ®rst including and then excluding the travel history. Here I focus on tropical diagnoses, of which malaria is by far the commonest. About 1500 cases of malaria are noti®ed in England and Wales each year, probably less than half the true incidence3. Over 50% of malaria cases imported into England and Wales are due to Plasmodium falciparum, which can cause severe disease (Box 1), sometimes rapidly fatal. Mortality is associated with delay in diagnosis; thus, early recognition of malaria is a primary consideration. No clinical features accurately predict malaria4. Almost all patients with malaria report fever, but the fever is not continuous and up to 20% are apyrexial at presentation (analysis of fever patterns is generally unhelpful in diagnosis). Malaria commonly presents as fever of abrupt onset without localizing symptoms or signs (undifferentiated fever), accompanied by headache; however, in many cases there are confusing features, such as abdominal discomfort, diarrhoea or jaundice5,6. In one study 36% of patients attending hospital with malaria had been given speci®c treatment, usually antibiotics, for an incorrect
Department of Infection and Tropical Medicine, Imperial College School of Medicine, Lister Unit, Northwick Park Hospital, Watford Road, Harrow HA1 3UJ, UK E-mail: m.jacobs@ic.ac.uk

124

diagnosis6. Since malaria cannot be excluded by the presence of localizing symptoms or signs (especially since it may coexist with other conditions such as respiratory infections), epidemiological rather than clinical factors determine the need to consider the diagnosis of malaria in a febrile traveller. Malaria is transmitted in many tropical and subtropical regions of the world, but there may be local geographic or seasonal variation in transmission and instability from one year to the next; ideally, up-to-date information on the epidemiology of malaria transmission should be set against a detailed travel history. Since delay in diagnosis may have severe consequences and diagnostic tests for malaria are relatively easy and inexpensive, a pragmatic approach is to assume malaria exposure in all febrile travellers from tropical and subtropical regions unless there are speci®c data to the contrary. Patients may be unaware that they have visited a malarious area and may even have been speci®cally reassured that malaria was absent. Even brief exposure may be suf®cient to acquire the disease. Immunity to malaria is only partial and requires repeated exposure to infection by residence in a highly endemic area; after a move to a non-endemic area immunity wanes rapidly. Half of all cases of malaria in the UK occur in individuals with ethnic origins in the tropics who have gone back to visit friends and relations, nearly one-third are in immigrants or visitors, and less than 20% are in returned tourists7. Chemoprophylaxis is used less by visitors to friends and relations (550%) than by tourists (~75%); although chemoprophylaxis offers 70± 90% protection, the fact that it has been practised assiduously does not rule out the diagnosis. Paradoxically, previous malaria chemoprophylaxis in a febrile patient increases the need for vigilance in respect of malaria, because it may delay the onset of clinical symptoms beyond the usual incubation period, attenuate symptoms, and, by reducing the peripheral blood parasitaemia, reduce the sensitivity of diagnostic tests8±10. These effects may occur long after the last dose of me¯oquine and chloroquine, which have half-lives up to 30 days. It is reasonable to stop malaria chemoprophylaxis in individuals who are undergoing investigation for fever. Some medications used for other indications, such as doxycycline and clarithromycin, also have antimalarial activity and can alter the clinical picture in the same way as chemoprophylaxis.

JOURNAL

OF

THE

ROYAL SOCIETY

OF

MEDICINE

Volume 93

March

2000

Figure 1 Algorithm for initial management of the febrile traveller

In 75% of cases, P. falciparum malaria presents within one month of exposure, in 90% by two months. Presentation beyond 6 months is uncommon. By contrast, disease due to P. vivax, P. ovale or P. malariae can develop even years after exposure; fortunately, these species do not cause rapidly progressive life-threatening illness so delay in recognition is unlikely to be disastrous. Occasionally malaria can be excluded because of its minimum incubation period, 8 days. If a febrile patient has visited a known or suspected malarious area within the past six months, then P. falciparum

malaria is an urgent diagnostic consideration. Microscopic examination of thick and thin ®lms is the most accurate method of diagnosis. The timing of ®lms in relation to the fever is unimportant. Thick ®lms are most sensitive but are not routinely performed in non-specialist laboratories. Thin ®lms permit reliable species identi®cation and evaluation of the parasitaemiaÐimportant for decisions on immediate treatment. A basic scheme of initial malaria treatment is outlined in the algorithm (Figure 1). For practical purposes, all P. falciparum should be assumed to be chloroquine-

125

JOURNAL

OF

THE ROYAL

SOCIETY OF

MEDICINE

Volume

93

March

2000

Figure 2 Aetiology of fever after travel to the tropics. Figures are percentages of patients. UTI=Urinary tract infection; TB=tuberculosis. & Maclean et al.; & Doherty et al. [Adapted from Maclean et al. (Ref 1) and Doherty et al. (Ref 2)]

Box 1 Severe complications of Plasmodium falciparum malaria
Neurological Impaired higher cerebral function Impaired conciousness Seizures Haematological Severe anaemia Macroscopic haemoglobinuria Disseminated intravascular coagulation Renal Acute renal failure Pulmonary Pulmonary oedema Adult respiratory distress syndrome Cardiovascular Shock Metabolic Metabolic acidosis Hypoglycaemia

126

resistant and the usual choice for treatment is quinine. If there is doubt regarding the species of malaria, the patient should be treated for P. falciparum until the ®lm has been reviewed (e.g. by the Malaria Reference Laboratory). Because P. falciparum malaria has serious complications and can progress after the start of treatment, patients with this infection are best admitted to hospital.

In expert laboratories, 98% or more of individuals with a ®nal diagnosis of malaria are positive on the ®rst set of ®lms (a thick and thin ®lm made from a single blood sample)2,6, but this ®gure can be much lower in laboratories with less experience. Therefore, a single negative set of ®lms weighs against but does not exclude malaria. A falsenegative result from a malaria ®lm is usually a consequence of a low level of parasitaemia dif®cult to detect by microscopy. Since the parasite count correlates strongly with disease severity, a negative ®lm provides reassurance that a well patient is unlikely to become seriously ill within the next 24 hours. In individuals at risk of malaria with a negative ®rst blood ®lm, repeat malaria ®lms every 12±24 hours are usually performed. After three negative ®lms, malaria is highly unlikely, but it remains a possibility until the illness resolves or an alternative diagnosis is con®rmed. Rapid diagnostic tests for malaria are likely to revolutionize the approach to diagnosis. Simple bedside tests for the detection of P. falciparum, based on detection of histidine-rich protein 2 (HRP-2), are now readily available (Parasight F, Becton Dickinson and ICT Malaria P.f., ICT Diagnostics) and in travellers they have detected P. falciparum with a sensitivity similar to that of microscopy and are highly speci®c10,11. While blood ®lms remain the primary means for diagnosing malaria, these tests can be a useful adjunct. Rapid testing for P. falciparum is likely to be most contributory in patients who have a serious illness compatible with malaria but a negative blood ®lm; the hazards of a false negative are thus reduced. Laboratory ®ndings suggestive of malaria are a normal or low white cell

JOURNAL

OF

THE

ROYAL SOCIETY

OF

MEDICINE

Volume 93

March

2000

count (95±99% of cases) and low platelets (50±80% of cases)5,6. The rapid tests based on HRP-2 detect only P. falciparum and a positive test is suf®cient evidence to start treatment. Rapid tests for other species of malaria are being developed. Empiric therapy with quinine is rarely indicated if blood ®lms and rapid testing are negative. There is a remote possibility that an individual with severe P. falciparum malaria will have a negative blood ®lm together with a falsenegative rapid antigen test. Empiric quinine could be used as a component of treatment for a patient with a lifethreatening illness of uncertain cause in whom malaria is possible, but in these circumstances other empiric therapy guided by the clinical picture (e.g. antibiotics for sepsis, acyclovir for encephalitis) will also be essential.

TROPICAL INFECTIONS OTHER THAN MALARIA

Once malaria has been considered as the cause of fever in a traveller, the remaining differential diagnosis will often be extensive. Simple laboratory tests, especially the white cell count and platelets, can give useful clues. Enteric fever (Salmonella typhi or paratyphi infection) is typically associated with a normal or low white cell count and a normal platelet count. Enteric fever is transmitted (by faecal±oral spread) throughout the developing world, particularly the Indian subcontinent. Existing vaccines have an ef®cacy of about 70% against typhoid and less against paratyphoid, so vaccination does not exclude the diagnosis. Enteric fever has an incubation period of up to two months but usually develops within three weeks. Unlike malaria, the illness is usually of slow onset; and, although almost all patients are pyrexial at presentation, up to 20% do not volunteer fever as a symptom. A typical patient with enteric fever has headache, cough and gastrointestinal disturbance. Both diarrhoea and constipation are common. Relative bradycardia (in relation to body temperature) has poor discriminative value, and rose spots are present in less than half of cases and easily missed in dark-skinned individuals. Up to 40% of cases have crackles or wheezes on chest examination. Complications in enteric fever occur after the ®rst week of illness, and in most cases the results of blood, stool and urine culture can safely be awaited before therapy (the Widal test is not reliable and is not used where culture facilities are available). Cultures may be negative, particularly in individuals who have received antibiotics, and empiric treatment may be indicated. When malaria has been ruled out, enteric fever is the most common cause of fever lasting 10 days or more12. In patients who are seriously ill with undifferentiated fever, cipro¯oxacin for enteric fever is a reasonable component of empiric treatment. S. typhi is widely resistant to antibiotics such

as chloramphenicol and ampicillin, which were previously recommended for treatment. The early clinical stages of viral hepatitis (A, B, or E) can be dif®cult to diagnose since they commonly manifest as undifferentiated fever with normal white cell count and platelets. Raised transaminases are often the ®rst evidence of viral hepatitis, which can then be diagnosed by speci®c serology. One must remember that a mild rise in transaminases is non-speci®c and can occur in malaria, typhoid, dengue and bacterial sepsis. In viral hepatitis, the fever and systemic symptoms typically subside at the onset of jaundice. The coincidence of high fever and jaundice suggests other diagnoses such as malaria or ascending cholangitis, especially if the patient is severely ill. Neutrophil leucocytosis is a predictor of bacterial infection. In addition to common community-acquired syndromes (pneumonia, urinary tract infection, streptococcal pharyngitis, etc.), leptospirosis is an important cause of fever in adventure travellers who have freshwater contact (such as rafting) in South-east Asia13. In most cases leptospirosis is anicteric and there may be nothing to suggest the diagnosis other than the exposure history. Neutrophilia is also a feature of amoebic liver abscess. Presentation is usually subacute and weight loss may be the predominant symptom. Jaundice is unusual and most patients do not report diarrhoea. Amoebic abscess deserves consideration in any patient with appropriate exposure, even if months earlier, and clinical or radiological evidence of changes at the right lung base (in which case imaging below the diaphragm, with ultrasound or computed tomography, is indicated). Infections with unicellular organisms such as amoebae and plasmodia do not give rise to eosinophilia. Eosinophilia with fever, though an uncommon combination, suggests an invasive worm infection such as acute schistosomiasis (Katayama fever). Specialist advice is needed. In dengue, as in malaria, the white cell count is usually normal or low with thrombocytopenia. Dengue is the most common arboviral disease of man, and is transmitted by day-biting mosquitoes through tropical and subtropical regions of the world. Transmission is most intense in Southeast Asia, the Caribbean and South America; cases from Africa are seldom imported into the UK. Dengue has a short incubation period (up to 10 days) and typically presents with abrupt onset of fever, headache, retro-orbital pain and severe myalgias. Rash, if present, is an important clue. In the acute phase, diagnosis can be made by a polymerase chain reaction test; serology is not usually positive until after defervescence but paired sera can be used to make a retrospective diagnosis. There is no speci®c treatment and the illness usually resolves spontaneously within 5±7 days. Uncommonly, dengue in travellers causes haemorrhagic fever, which is treated with supportive

127

JOURNAL

OF

THE ROYAL

SOCIETY OF

MEDICINE

Volume

93

March

2000

measures14. Dengue is not transmitted directly from person to person, so special isolation procedures are not required. In contrast to dengue, nosocomial spread of Lassa, Ebola, Marburg and Crimean±Congo haemorrhagic fever (CCHF) viruses has occurred. These diseases do not represent a major public health threat, despite a high casefatality rate. Although they are extremely rare in travellers, vigilance must be maintained to protect nursing, medical and particularly laboratory staff. Doctors in the UK involved in immediate care of febrile travellers must be familiar with the guidelines for management of potential viral haemorrhagic fevers15. These fevers usually lack identifying features in the early stages and case identi®cation then depends upon recognition of the major epidemiological risksÐnamely, travel to rural sub-Saharan West Africa (except CCHF, which is distributed sporadically in Africa, the eastern Mediterranean, Middle East and parts of southern Asia) and contact with known or suspected cases. The upper limit of the incubation period is 21 days, beyond which these diseases are effectively excluded.
CONCLUSION

REFERENCES

The immediate management of a febrile traveller demands consideration of a large and sometimes unfamiliar differential diagnosis, and a stepwise approach will assist safe management. In this review I have emphasized the diagnosis of malaria, and the strategy outlined should prove useful in conjunction with careful clinical assessment of the individual patient for both tropical and cosmopolitan diseases. discussions.
Acknowledgment

I thank Dr R Davidson for helpful

1 MacLean JD, Lalonde RG, Ward R. Fever from the tropics. Travel Med Advisor 1994;5:27.1±27.14 2 Doherty JF, Grant AD, Bryceson ADM. Fever as the presenting complaint of travellers returning from the tropics. Quart J Med 1995;88:227±81 3 Davidson RN, Scott JA, Behrens RH, Warhurst D. Under-reporting of malaria, a noti®able disease in Britain. J Infection 1993;26:348±9 4 Svenson JE, Gyorkos TW, MacLean JD. Diagnosis of malaria in the febrile traveller. Am J Trop Med Hyg 1995;53:518±21 5 Froude JRL, Weiss LM, Tanowitz HB, Wittner M. Imported malaria in the Bronx: Review of 51 cases recorded from 1986±1991. Clin Infect Dis 1992;15:774±80 6 Svenson JE, MacLean JD, Gyorkos TW, Keystone J. Imported malaria. Clinical presentation and examination of symptomatic travellers. Arch Intern Med 1995;155:861±8 7 Behrens RH. Travel morbidity in ethnic minority travellers. In: Cook GC, ed. Travel-associated Disease. London: Royal College of Physicians of London, 1995:93±100 8 Lewis S, Davidson RN, Ross E, Hall AP. Severity of imported falciparum malaria: effect of taking antimalarial prophylaxis. BMJ 1992;305:741±3 9 Day J, Behrens RH. Delay in onset of malaria with me¯oquine prophylaxis. Lancet 1995;345:398 10 Humar A, Ohrt C, Harrington MA, Pillai D, Kain KC. Parasight F test compared with the polymerase chain reaction and microscopy for the diagnosis of Plasmodium falciparum malaria in travelers. Am J Trop Med Hyg 1997;56:44±8 11 Jelinek T, Grobusch MP, Schwenke S, et al. Sensitivity and speci®city of dipstick tests for rapid diagnosis of malaria in nonimmune travelers. J Clin Microbiol 1999;37:721±3 12 Humar A, Keystone J. Evaluating fever in travellers returning from tropical countries. BMJ 1996;312:953±6 13 van Crevel R, Speelman P, Gravekamp C, Terpstra WJ. Leptospirosis in travellers. Clin Infect Dis 1994;19:132±4 14 Jacobs MG, Brook MG, Weit WR, Bannister BA. Dengue haemorrhagic fever: a risk of returning home. BMJ 1991;302:828±9 15 Advisory Committee on Dangerous Pathogens. Management and Control of Viral Haemorrhagic Fevers. London: Stationery Of®ce, 1996

128


				
DOCUMENT INFO
Shared By:
Stats:
views:15
posted:1/19/2010
language:English
pages:5
Description: Imported fever a survival guide