Treatment of imported malaria in an ambulatory setting: prospective studyCommentary: Should patients with imported malaria routinely be admitted?

BMJ 2002; 324 doi: (Published 13 April 2002) Cite this as: BMJ 2002;324:875

Treatment of imported malaria in an ambulatory setting: prospective study

  1. Valérie D'Acremont, senior house officera,
  2. Pierre Landry, consultanta,
  3. Roger Darioli, professora,
  4. Dieter Stuerchler, professorb,
  5. Alain Pécoud, directora,
  6. Blaise Genton, senior scientist (Blaise.Genton{at}
  1. a Policlinique Médicale Universitaire, 1005 Lausanne, Switzerland
  2. b Swiss Federal Office of Public Health, Bern, Switzerland
  3. bHospital for Tropical Diseases, London WC1E 6AU
  1. Correspondence to: B Genton
  • Accepted 25 July 2001

Many specialists in tropical medicine consider that patients with imported malaria, at least those with Plasmodium falciparum malaria, should be admitted to hospital, as complications can develop quickly.1 In Switzerland, patients with malaria who lack signs of severe disease are treated as outpatients, because empirical observations of patients with imported malaria show that death is usually due to a delay in diagnosis rather than complications during treatment.2 We conducted a prospective study in the outpatient clinic of a university hospital to assess the safety of treating imported malaria in an ambulatory setting.

Participants, methods, and results

We conducted our study from January 1990 to July 2000. At study entry we used predefined clinical and laboratory criteria (table) to determine if patients with malaria required admission to hospital. If no criteria were met, ambulatory treatment was considered appropriate. Patients received the first dose of drugs under supervision and were kept under surveillance for one hour before being sent home with instructions. Follow up was at the attending doctor's discretion: clinical and parasitological assessments were performed daily until symptoms resolved and one blood slide was clear of parasites.

Predefined clinical and laboratory criteria for admission of patients with malaria to hospital and number of patients primarily admitted to hospital with the condition

View this table:

Overall, 165 (17%) of 958 patients with fever were positive for parasites; 113 (69%) had P falciparum. Seventy one (43%) of the 165 were first generation immigrants and none was white; and 135 (82%) had travelled to Africa. Median age was 33.7 (range 16-76) years, and median time from onset of symptoms to consultation was four days. Seventy seven (47%) patients had a parasitaemia of 0.1% or less and 10 (6%) of 2% or more.

We admitted 36 (22%) patients immediately; 33 had at least one of the predefined criteria for admission (23 had more than one), and 31 had P falciparum. Seven patients met one criterion but were not admitted, six because the doctor did not comply with the recommendations; one patient refused (this patient had a parasitaemia of 4.5% and recovered uneventfully).

We followed up 129 (78%) patients as outpatients (82 with P falciparum); 88 (68%) were treated with mefloquine alone or in combination with sulfadoxine and pyrimethamine. Six of the 129 patients were admitted during treatment. Two patients with P falciparum returned before the scheduled follow up because of dizziness, probably related to an adverse drug event. Two patients with P falciparum were observed for 40 and 60 hours, respectively, because their fever had persisted. One patient with P ovale was admitted after 24 hours because of a change in general condition and vomiting. None of these five patients received second line antimalarial treatment. The sixth patient was admitted after 48 hours with slight cough and difficulty breathing and needed assisted ventilation three days after admission. All patients recovered uneventfully.


Ambulatory management of imported malaria seems to be safe, provided that criteria for severity are considered. Previous studies on imported malaria were conducted in specialised tropical centres, where patients tend to present late with disease and thus have a high risk of complications.3 Our study was conducted prospectively in a setting comparable to primary care. Only 5% of patients were admitted during treatment, and none died. The inclusion of 43% of migrants with some immunity did not bias towards a favourable outcome because their clinical presentation and rate of primary admission were similar to those found for naive travellers.3

Our sample size may not have had enough power to detect a small increase in case fatality rate. However, outpatient management of uncomplicated malaria is standard care in Switzerland, and a review of all reported deaths25 in the past 10 years showed that none could be attributed to the acceleration of malaria that had been judged sufficiently mild to be treated outside hospital.

Conditions favouring a good outcome in our setting included a clinic that was open all hours, specialist supervision, and easy accessibility because of short distances to travel. Mefloquine is not standard treatment internationally, although alternative drugs such as atovaquone with proguanil or artemether with lumefantrine may help to manage uncomplicated malaria in an ambulatory setting.4


Contributors: VD'A was responsible for data entry, management and analysis, interpretation of results, and writing the manuscript. PL was responsible for clinical supervision, data collection, and interpretation of results. RD developed and designed the protocol. DS collected and analysed the mortality data. AP had overall clinical responsibility. BG was responsible for the development and design of the protocol, clinical supervision, data analysis, interpretation of results, and writing the manuscript. All authors reviewed the manuscript and agreed on the content. AP and BG will act as guarantors for the paper.


  • Funding None.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.

Commentary: Should patients with imported malaria routinely be admitted?

  1. Christopher J M Whitty, consultant physician (c.whitty{at},
  2. Diana N J Lockwood, consultant physician
  1. a Policlinique Médicale Universitaire, 1005 Lausanne, Switzerland
  2. b Swiss Federal Office of Public Health, Bern, Switzerland
  3. bHospital for Tropical Diseases, London WC1E 6AU
  1. Correspondence to: C J M Whitty

    Malaria remains an important infection in Europe, with several thousand cases imported each year. D'Acremont et al pose an important question—should these cases routinely be admitted for treatment? Most people agree that non-falciparum malaria can be treated without admission. Falciparum malaria has more serious implications. It claims over a million lives a year, including some in Europe. Several factors combine to make it difficult to assess severity at the time of diagnosis. If the parasites are mature almost all will be sequestrated, giving a misleadingly low peripheral parasitaemia. Conversely, if the parasites are young the patient may seem well but deteriorate rapidly over 24 hours as the parasites mature and begin to sequester in vital organs. This can occur despite adequate treatment; drugs such as quinine have a limited effect on early stages. Patients may therefore deteriorate rapidly despite adequate antimalarial treatment.

    D'Acremont et al present data that at first sight support treating patients with malaria as outpatients, provided that strictly applied criteria identify those needing admission. The finding—that only 6% treated as outpatients needed subsequent admission—is reassuring. This is, however, potentially misleading. Many of those treated as outpatients had non-falciparum malaria, and such patients almost never need admission. Overall, 34% of all patients with falciparum malaria met the study criteria for moderate to severe disease requiring admission. Of the 82 patients with falciparum malaria treated as outpatients five were readmitted, one requiring ventilation. It is also unclear from the paper in which patients malaria was diagnosed by a positive slide result and in which by immunological methods alone (a group with a different prognosis).

    Additionally, study criteria used to identify moderate to severe disease may be difficult to generalise. Subjective criteria such as “poor general condition” are difficult to assess and standardise in patients with malaria, even for specialist centres. Busy casualty departments in general hospitals will find it no easier. Even the harder criteria have pitfalls; in particular the admission of patients with a parasitaemia of 2% seems reassuring, but in the last 100 consecutive patients with falciparum malaria seen at our hospital, 23% had an increase in parasitaemia over the first 24 hours of treatment, including eight increasing above 2%, one with increase from 1.3% to 32%, and one from 0.2% to 8.4%. As a minor point, mefloquine, the main drug used in this study, is not used as first line treatment for malaria in most centres and may well be better adhered to by patients than quinine—which, although safe and effective, has major short term side effects and has to be taken for longer.

    Conventional practice is to admit all patients with falciparum malaria because initial assessment can be misleading—even for specialist centres—and otherwise fit patients can deteriorate markedly, despite appropriate treatment. This study opens this practice up for debate, but it does not provide adequate justification for changing practice—yet.

    View Abstract