The present study provides, for the first time, evidence that a malaria diagnostic strategy based on a RDT followed by immediate or delayed microscopy reading at first attendance is safe and does not expose travellers or migrants to an increased risk of severe malaria or death. These findings can probably be generalized to most setting in non-endemic countries since they derive from data collected in predominantly non-immune patient population under routine clinical and laboratory conditions of an ordinary outpatient clinical and emergency hospital ward. No patient with a negative RDT developed severe malaria, despite a planned delay before getting the blood smear results out of hours. This was true even when established timelines between RDT and microscopy were not complied to. The use of RDTs was not associated with the development of complications since all 6 cases who developed severe malaria after admission had a positive RDT at first testing. The availability nowadays of artemisinin-based combination therapy (ACT) for the treatment of uncomplicated malaria renders the strategy with RDT even safer since ACT is very effective and well-tolerated [17, 22]. ACT should be given in the outpatient department or emergency ward immediately after a positive RDT result while waiting for the BS result, even in uncomplicated cases. Similarly, quinine should be administered straight after a positive RDT result for severe cases.
Even if the study was not designed to validate the accuracy and performance of RDTs, which has already been extensively demonstrated, RDTs were as good as microscopy to diagnose malaria. Indeed 5 malaria diagnoses were based on positive RDT results only (negative blood slide) and 4 on blood smear results only (negative RDT result). Recently, Gillet et al.[23, 24] and Luchavez et al. demonstrated that the prozone effect (false-negative or false-low results, due to an excess of either antigen or antibody) exists, but has so far only been described with histidine-rich protein 2 tests [23, 24]. Negative results were rare compared to an increase in test line intensity after dilution. In our study, none of the negative RDT results was explained by the prozone effect. This strategy of performing an immediate blood smear in the presence of danger signs or a thrombocytes count < 100 G/l (higher pre-test probability) should prevent delay in the diagnosis of malaria in case of false negative RDT result in the presence of hyperparasitaemia.
Diagnostic strategies based on RDT have been adopted in other centres managing non-immune patients [10, 26] but, this is the first study assessing the safety of such strategy. Because of the retrospective design of the study, we were able to assess the strategy under routine clinical and laboratory practice. As imported malaria is a rare disease and severe malaria even rarer, it was not possible to perform a non-inferiority trial comparing the strategy with and without RDT. Also there are concerns not to use RDTs in a setting where not all laboratory technicians are familiar with malaria parasites, especially out of hours, which might have resulted in missed malaria, and hence higher rate of complications.
The less rigorous follow-up of the malaria negative patients is a limitation in the overall assessment. However, it is highly unlikely that secondary malaria cases were missed after having attended the outpatient clinic since feverish patients are advised to come back daily to repeat malaria tests, especially so if symptoms persist or worsen. In addition, the University Hospital in Lausanne has a long tradition of reference centre for travel related diseases in the area and is easily accessible. At least, secondary malaria deaths that would have occurred outside the hospital have been virtually excluded by our investigation of malaria death records in the region. This study was undertaken in one hospital only and should be repeated in different settings to accumulate more evidence. Since non-falciparum malaria patients and children were excluded, the safety of this strategy should be confirmed for malaria due to other species (especially so since RDT of last generation do detect vivax with excellent sensitivity) and in a paediatric population .
In conclusion, this study - conducted in a routine clinical and laboratory non-endemic setting without 24-hour expert microscopy available – provides some evidence that a malaria diagnostic strategy based on RDTs followed by immediate or delayed microscopy reading is safe. Indeed no patients with a negative RDT developed severe malaria or died. This study adds information about the safety of a malaria diagnostic strategy based on RDTs, of which accuracy and performance have been extensively demonstrated. There was also a clear benefit of using RDT, as it allowed decreasing significantly the delay before getting a test result (and thus onset of appropriate treatment), even during laboratory working hours and increasing overall sensitivity when combined with microscopy. The results of this analysis provide evidence and lessons for considering large-scale implementation of malaria diagnostic strategies that include RDTs in non-endemic settings.