| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Misdiagnosis of malaria is a frequent mistake that can have deadly consequences. In resource-rich populations, malaria is uncommon and therefore may not be recognized by clinicians who have insufficient training in tropical diseases, or by laboratory technicians who have insufficient experience interpreting blood smears. In these cases, the non-immune victim, typically a recently returned traveler, is at great risk of severe disease and death if infected with Plasmodium falciparum.1 In resource-poor countries where malaria is a major scourge, patients suffer the opposite problem: clinicians tend to diagnose many if not most febrile illnesses as malaria and often microscopy is not available for confirmation. Recent studies have highlighted the increased mortality by individuals diagnosed with blood smear–negative malaria, presumably because they were not properly treated for whatever disorder other than malaria was causing the fever.2 In this issue of the journal, Nyunt and others describe their experience using laser desorption mass spectrometry (LDMS) as a new tool for malaria diagnosis in pregnant women.3
Microscopy has been the standard for malaria diagnosis since not long after Laveran discovered the parasite in 1880, and Romanowsky improved the methods for staining the parasite in 1891. Microscopy has several advantages, including the ability to speciate and quantify the parasites, thereby providing information that is essential for deciding on the proper treatment of the patient. However, microscopy also has several disadvantages that limit its application in different settings. Reading a blood smear is subjective and requires experienced personnel, microscopes are expensive and often reliant on electricity that may not be available, and in the case of pregnancy malaria parasites that are sequestered in the placenta may not be detected by smears prepared using peripheral blood samples.
Immunochromatographic assays, also called rapid diagnostic tests or dipsticks, have emerged recently as an alternative to microscopy (reviewed by Moody4). These assays typically use a capture antibody and conjugated detection antibody to detect malarial antigen in blood samples. The major benefits of this approach include the rapid turnaround time and the ease of use, which allows inexperienced laboratory or clinical staff to make on-the-spot diagnoses. In cases in which the antigen being detected is secreted or released by the parasite, peripheral blood samples may detect tissue-sequestered parasites even when no parasites are seen on peripheral blood smear.5,6 Many dipsticks also allow the user to speciate the offending parasites.
However, dipsticks have not yet been developed that allow the user to quantify the parasite burden, which clinicians use to guide and assess therapy. Furthermore, the cost of dipsticks has limited their use. In the past, when commonly used anti-malarials such as chloroquine or pyrimethamine-sulfadoxine cost 10–20 U.S. cents for an entire treatment course, many ministries of health decided that the $1 cost (or more) for each dipstick could not be justified. This equation will change in the future, as more expensive combination therapies including artemisinin-containing combination therapy are introduced as first-line regimens to replace treatments that have lost efficacy to drug-resistant parasites.7 An increased awareness of the cost in human lives of misdiagnosing malaria may also change this equation.
Laser desorption mass spectrometry is able to detect heme from hemozoin, the breakdown product of hemoglobin that is commonly referred to as parasite pigment. Importantly, heme from hemozoin yields a spectral signature that is distinct from that of heme from normal hemoglobin, allowing LDMS to discriminate infected from uninfected red blood cells.8,9 Although the sensitivity of LDMS to detect heme is theoretically equivalent to that in a single infected red blood cell, the sensitivity in this pioneering study by Nyunt and others was approximately 100–1,000 parasites/µL of blood, or roughly similar to routine microscopy. Optimization of LDMS for malaria diagnosis should improve this level of sensitivity. Nevertheless, in this study LDMS detected parasites in 15 pregnant women (of 45 study volunteers) whose peripheral blood smear yielded negative results on microscopy. Compared with the polymerase chain reaction (PCR), LDMS had a sensitivity of 52% and a specificity of 92%.
Notably, an improvement in diagnostic sensitivity may not translate into improved patient care. For example, pregnancy malaria is associated with low birth weight and maternal anemia, outcomes that increase mortality risk for infant and mother. The most sensitive technique for detecting parasites is the PCR. However, whereas maternal parasitemia detected by microscopy is associated with maternal anemia and low birth weight, there was no difference in these outcomes between uninfected women versus women with submicroscopic parasitemia detected by a PCR.10–12 Dipsticks are less sensitive than the PCR, and malaria diagnoses obtained by dip-stick are associated with pregnancy outcomes, similar to diagnoses obtained by microscopy.10
Nyunt and others tout LDMS as a tool that is more sensitive than microscopy, is potentially rapid, and requires nothing more than a lancet, blood collection container, and water, after the initial investment in the instrument. The results presented by Nyunt and others suggest that LDMS may be able to provide some quantification of the parasite burden. However, using LDMS to quantify parasitemia will be complicated by mixed species infections; the hemozoin content in ring-stage parasites is lower than in mature-stage parasites, and P. falciparum infections yield exclusively ring-stage parasites in the peripheral blood circulation, while the non–P. falciparum species yield all parasite stages.
In the study by Nyunt and others,3 two samples that were negative by PCR were positive by LDMS. The investigators speculated that the hemozoin in these samples may have originated from pigmented macrophages that remained in the circulation after infection had resolved. Because pigment can persist in the placenta for significant periods after resolution of parasitemia, it will be important to determine how frequently this may confound LDMS diagnosis of malaria in pregnant women. In future studies, LDMS on peripheral samples should be compared with histologic examination of the placenta for the presence of pigment in the absence of parasitemia.
Laser desorption mass spectrometry does not speciate parasite forms. However, proper treatment of malaria requires parasite speciation, and proper treatment is one element in the difficult effort to stem the spread of drug-resistant parasites. Owing to its simplicity, speed, and cheapness (after the initial investment in the apparatus as well as the infrastructure required to sustain it), LDMS may be useful for rapidly screening many samples for infection, and subsequently using microscopy or dipsticks to speciate those samples that are positive.
Diagnostics remains a problem for malaria because microscopy is difficult to deploy in many areas despite its time-proven value. Finding cheap, rapid, robust methods for malaria diagnosis that require minimal training for the technician would have a major impact on malaria management. Laser desorption mass spectrometry is a new and interesting approach, and deserves careful consideration along with immunochromatography, PCR, and other novel formats that can be implemented in either resource-rich or resource-poor environments. Complex technologies such as LDMS and PCR are unlikely to be suitable in most resource-poor settings for the foreseeable future, although miniaturized mass spectrometers that are being developed (http://www.jhuapl.edu/programs/rtdc/Pathogens/Malaria.html) may allow the deployment of LDMS at remote sites. The best solution may be a combination of different techniques, and these should be assessed in careful comparative studies that examine the impact on patient outcomes as well as the sensitivity and specificity of each approach.
Received May 16, 2005. Accepted for publication May 22, 2005.
* Address correspondence to Patrick Duffy, Malaria Antigen Discovery Program, Seattle Biomedical Research Institute, Seattle, WA 98109. E-mail: p.duffy{at}sbri.org 
Authors’ address: Patrick Duffy and Michal Fried, Malaria Antigen Discovery Program, Seattle Biomedical Research Institute, Seattle, WA 98109, E-mails: p.duffy{at}sbri.org and michal.fried{at}sbri.org.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
