Intended for healthcare professionals


Routine screening of children returning home from the tropics: retrospective study

BMJ 1999; 318 doi: (Published 27 February 1999) Cite this as: BMJ 1999;318:568
  1. Marianne L Brouwer, medical student,a,
  2. Jules J M Tolboom,, senior consultant, (j.tolboom{at},
  3. Jaap H J Hardeman, medical adviserb
  1. aDepartment of Paediatrics University Hospital Nijmegen St Radboud, PO Box 9101, 6500 HB Nijmegen, Netherlands
  2. b Haskoning BV Consultants, PO Box 151, 6500 AD Nijmegen
  1. Correspondence to: Dr Tolboom
  • Accepted 30 October 1998

Parents who move to the tropics with their children often worry about tropical diseases. Most of the data available on screening children returning home from living in the tropics are from 1952-63 from American children.1 More recent findings in asymptomatic people returning to the United Kingdom have been reported.2 We aimed to determine whether it is worthwhile to routinely screen children without symptoms returning home from the tropics.

Subjects, methods, and results

During 1987-95all children visiting our hospital for a medical check up after a stay in the tropics were studied. This check up was obligatory under their parents' terms of employment. Our protocol included a questionnaire (completed by the parents), physical examination of the child, and additional investigations. These investigations included a full blood cell count; eosinophil count; liver function tests; urea and creatinine concentrations; hepatitis A and B serology; thick smear; schistosoma and filaria serology (if the child had stayed in endemic areas); urine analysis; stool culture; analysis of three stool samples for ova and cysts; tuberculin skin test; and a chest xray if indicated. The same specialist (JJMT) performed all consultations. Data were analysed with EpiInfo version 5.0.


A total of 216children (103girls) were seen during 282check ups (some children had lived in the tropics more than once) (table). In 175(62%) cases (check ups), children had lived in sub-Saharan Africa. Ninety three children (43%) were born in the tropics. The length of stay overseas ranged from 3months to 13 years; in 150(53%) cases the stay was 1-3years. Age at first check up ranged from 3months to 16years (mean 4.75,median 3.6). In 29(10%) cases children were seen early, mainly because they had one or more complaints: abdominal pain (13children), diarrhoea (9), or fever (8). These cases were classed as symptomatic. In the 253asymptomatic cases of disease—that is, among children seen at their scheduled appointment—parents reported that 40had abdominal pain and 31had diarrhoea. On physical examination, few abnormalities associated with tropical diseases were detected in children with symptoms: hepatosplenomegaly (in three children) and cutaneous larva migrans (one child).

Number (percentage) of diagnoses of tropical disease found during 282 check ups in 216 children in the Netherlands, 1987-1995

View this table:

Most conditions were diagnosed by laboratory investigation. Twelve of 58(21%) asymptomatic cases of giardiasis were associated with abdominal complaints. Results of tuberculin skin tests were available for 187children, of whom 149had been vaccinated with BCG. In 47(25%), induration was less than 10mm, and two children had indurations of 12mm and 15mm with normal physical findings and normal chest xrays. A total of 112(75%) of the children vaccinated with BCG did not respond to the tuberculin test. Tropical diseases were diagnosed in 99/253 (39%) asymptomatic and 15/29 (52%) symptomatic cases. All children with parasitic infections received appropriate treatment as outpatients. Only one child needed admission; this was for treatment of relapsing fever.


These findings can be considered representative for all children returning home, especially those returning from sub-Saharan Africa. Children should be seen early if they have any symptoms but children without apparent symptoms also may have abdominal complaints. We found a higher incidence of tropical diseases in asymptomatic children than Carroll et al2 (39% v 25%), probably because of our higher incidence of giardiasis (23% v 4%). Most cases of giardiasis in our study were asymptomatic. The incidence of unexplained eosinophilia in 24(10%) of 253asymptomatic cases is similar to that found by others.24 A lack of response to tuberculin skin testing among 75% of the children who had been vaccinated with BCG could be explained by inadequate testing technique, failure of the vaccination to produce tuberculin sensitivity, or diminishing sensitivity occurring over time.5

Routine screening of children without symptoms returning home after living in the tropics is worth while. Laboratory testing will identify most cases of tropical disease. Screening can be carried out by well informed general practitioners using a standard protocol.


We thank the nursing and administrative staff of the paediatric clinic for their contribution in carrying out the protocol.

Contributors: MLB extracted and analysed the data and prepared the initial report as part of the requirement to qualify as doctor. She wrote the first draft of this paper. JJMT designed the study, performed all consultations, and supervised MLB. JHJH helped design the protocol. MLB, JJMT, and JHJH all contributed to drafting the paper. JJMT wrote the final version of the paper and is the guarantor.

Competing interests: None declared.


View Abstract