This study has shown that LAMP with high PPV and NPV as compared to nested PCR, can be applied for malaria therapeutic follow-up. HRP-2-RDTs have similar sensitivity as microscopy but less specificity, which can either induce to underdiagnosis because of the moderate NPV or over-diagnosis due to the poor PPV.
Methods for the detection of malaria are central to an elimination programme. Current diagnostic methods applied in endemic areas are either time-consuming, or insensitive for use in conditions of low infection
[13, 24]. Optical microscopy is often taken as the gold standard for diagnosis, but its limited sensitivity due to low parasitaemia, common in low endemic areas, makes it unreliable
[25, 26]. A previous study using Bayesian latent class models revealed the danger of statistical analysis based on microscopy as a reference test
. Using nested PCR as a gold standard, this study has shown that good agreement with nested PCR is achieved with LAMP. LAMP for molecular detection of P. falciparum has been compared with other diagnostic tests, but both sensitivity and specificity of LAMP varied in different studies
[23, 27–30]. A recent report on the evaluation of LAMP for malaria diagnosis at a field clinic found that PPV and NPV of LAMP were 100% and 98%, respectively, which are similar to our results, indicating that LAMP is an effective tool for malaria diagnosis in a field setting
The inconsistent results between nested PCR and LAMP were found in two cases (one with PCR negative but LAMP positive while one with PCR positive but LAMP negative). This discrepancy may be expected when both nested PCR and LAMP are performed by two technicians independently. It has been reported that the sensitivity of a given PCR assay varies between laboratories. Although, the variations are relatively minor, they primarily diminish the ability to detect low-level and mixed infections
. These problems may also occur in LAMP based assays. The case with LAMP positive but nested PCR negative was falciparum malaria with extreme low parasitaemia while the case with nested PCR positive but LAMP negative harboured a mixed infection with P. falciparum and P. vivax. Low parasitaemia and mixed infection may limit the detection, even by use of molecular methods, but this cannot be the excuse for the contradiction in results between LAMP and nested PCR.
It has been well documented that blood compositions, such as haemoglobin and IgG/IgM, can interfere with the performance of PCR
[32–34]. A rapid extraction of DNA from filter paper may be favourable for LAMP but a conventional method is needed for the nested PCR. In practice, the risk of variation in results increases as in the case of nested PCR where more steps in the preparation of DNA purification are required. Although errors in the performance of nested PCR and LAMP cannot be excluded, a good participation of well-trained technicians and good controls greatly reduce mistakes. To make sure no amplification in the negative controls, the three- room rule has been applied for the setup; one for DNA extraction, one for reaction preparation, and one for amplification – cycler (PCR) and heater (LAMP). The three- room policy has minimized the false positivity caused by contamination. The cause for discrepancy in results between nested PCR and LAMP is not clear and no available tests can verify the difference unless they were reconciled and repeated. The need of accurate, sensitive, affordable and easy to run for the LAMP warrants further investigation. More sensitive LAMP assay to detect malaria parasites targeted mitochondria DNA has been developed
Malaria can be over-diagnosed if febrile children had positive HRP-2-RDTs but negative microscopy. One third of children (23/68) with positive HRP-2-RDTs were negative by microscopy and this can be attributed to low parasite density or to the delayed clearance of circulating HRP-2 antigen. Further verification by molecular tests, positive results from nested PCR and LAMP (10 out of 23) provided the evidence for the low parasitaemia while negative results from nested PCR and LAMP (13 out of 23) explained the possibility for the residual circulating HRP-2 antigens. Other causes for false-positive RDT results, such as gametocytaemia and the presence of the serum rheumatoid factor cannot be excluded
[24, 35–37]. At follow-up on day 7, these 13 children revealed the same test results as prior to the initial treatment which suggests over-treatment for malaria. Although the causes of fever origin were unknown, these children became afebrile and left without any sequelae after the anti-malaria treatment.
It has been well documented that individuals with low parasite density act as reservoirs for transmission
[4, 6, 8] and are also more prone to having malaria attack
[5, 7] In this study, all febrile children with positive RDTs were treated as malaria at the first instance because patients with positive RDTs in a suspected epidemic can be highly suggestive of falciparum malaria
. In addition, it is not wise to delay the treatment for falciparum malaria until the nested PCR result is available, which takes several hours. LAMP is simpler and faster and can be a potential tool to replace nested PCR, but it still takes time to obtain results as the laboratory is usually distant from collection points. While the costs for the LAMP assay are only about a tenth of that for the conventional PCR
, the reagents and enzymes are still expensive and may restrict its use in malaria-endemic areas. Future development of RDT-based LAMP for the malaria diagnosis should always take into account the affordability by the users in resource-limited countries
. With improvement of infra-structures of health system in the future, LAMP can be useful as a point-of-care test.
In tropical Africa, the differentiation of malaria from other causes of fever in the absence of microscopy is notoriously difficult. Even febrile children were malaria positive by microscopy it is a challenge to distinguish children who really do have severe malaria from those who have severe febrile illness but coincidental parasitaemia, who may have another infection. Although recent studies reported a strong correlation between plasma PfHRP2, disease severity, and outcome, and suggest that plasma PfHRP2 is a prognostic indicator in African children with severe falciparum malaria
[38, 39], measurement of plasma PfHRP2 was not available in this study. On the other hand, current HRP2-based RDTs cannot distinguish between asexual parasitaemia and gametocytaemia, which also contribute to the production of HRP2. Evidence showed that the RDT line intensity did not correlate with parasite density, days of parasitaemia, or disease severity
. Therefore, RDTs cannot be considered quantitative.
False-negative malaria RDT results can occur due to a prozone-like effect in high-density infections
. In this study, two children with high parasite density (~100,000/μl) were negative for HRP-2-RDTs but positive for microscopy, nested PCR, and LAMP. Their initial negative HRP-2-RDT became positive after a 10-dilution of blood was made. However, low parasitaemia below the detection limit of microscopy is the major cause for false-negative HRP-2-RDT results found in this study. False negativity due to the genetic heterogeneity of PfHRP-2 expression is also possible. A recent report showed that falciparum malaria parasites in Africa fail to produce HRP-2 can cause patent bloodstream infections and false-negative RDT results, which were more frequently seen in persons with asymptomatic infections
Artesunate plus amodiaquine combination is one ACT recommended by the WHO for use in malaria control programmes and a first line treatment for African children with uncomplicated malaria
. This recommendation is now a national policy in STP. Clinical trials have shown that artesunate-amodiaquine is a highly efficacious and safe anti-malarial drug
. However, poor adherence cannot achieve its desired therapeutic outcome. Too many tablets is inconvenient for children and patients/guardians may discontinue the treatment when they think their children feel better after fever subsides. A new formulation of artesunate-amodiaquine (Winthrop) based on one tablet per day administration, may improve compliance.
The efficacy of the six-dose in three-day regimen of artemether/lumefantrine has been confirmed in many different patient populations around the world, cure rates on 7,14 and 28 days exceeding 95% in the evaluable population
. The 14-day follow-up revealed that the cure rate for second-line treatment was 100% by microscopy but 92% corrected by nested PCR and LAMP. The 28-day follow-up showed the cure rate of 100% by microscopy, but 83% corrected by nested PCR and LAMP. The lower efficacy of second-line treatment observed in this study can be attributed to the presence of drug-resistant parasites which may contribute to the spread of drug-resistant mutant strains of the malaria parasite, thus complicating the treatment scenario.
The weakness of this study was that it was not randomized and the sample size was relatively small. In addition, both nested PCR and LAMP detect the amplification of target DNA while RDTs is antigen capture and microscopy is morphological identification. With the different targets detected, comparisons among these four different tests may be limited. Latent class models have been widely used to estimate diagnostic tests performance measures, such as sensitivity and specificity, as well as the diseases prevalence, in the absence of a gold standard or perfect reference test. In spite of being frequently used as a reference technique, microscopy is in fact an imperfect gold standard, especially in low parasitaemia as in this malaria control island
Although this study was carried out during the rainy season with a short period time and only older children were enrolled, malaria positive rate in febrile children was high (1/3 by microscopy or 1/2 by the nested PCR), which strongly indicates a high risk of malaria epidemics in a low transmission area, like São Tomé. After an intensive treatment and follow-up, there was no mortality observed among these febrile children.