Malaria Molecular Epidemiology Unit, Pasteur Institute

Question
Performance of the CareStartTM G6PD Deficiency
Screening Test, a Point-of-Care Diagnostic for
Primaquine Therapy Screening
Saorin Kim1, Chea Nguon2, Bertrand Guillard3, Socheat Duong2, Sophy Chy1, Sarorn Sum1, Sina Nhem1,
Christiane Bouchier4, Magali Tichit4, Eva Christophel5, Walter R. J. Taylor6, John Kevin Baird7,8, Didier
Menard1*
1 Malaria Molecular Epidemiology Unit, Pasteur Institute of Cambodia, Phnom Penh, Cambodia, 2 National Center for Parasitology, Entomology, and Malaria Control,
Phnom Penh, Cambodia, 3 Medical Laboratory, Pasteur Institute of Cambodia, Phnom Penh, Cambodia, 4 Genomic Platform, Institut Pasteur, Paris, France, 5 WHO
´decine Internationale et Humanitaire, Hopitaux Universitaires de Gene Geneva, Switzerland,
`ve,
Regional Office for the Western Pacific, Manilla, Philippines, 6 Service de Me
7 Eijkman Oxford Clinical Research Unit, Jakarta, Indonesia, 8 Centre for Tropical Medicine, Nuffield Department of Clinical Medicine, University of Oxford, Oxford, United
Kingdom

Abstract
Development of reliable, easy-to-use, rapid diagnostic tests (RDTs) to detect glucose-6-phosphate dehydrogenase (G6PD)
deficiency at point of care is essential to deploying primaquine therapies as part of malaria elimination strategies. We
assessed a kit under research and development called CareStartTM G6PD deficiency screening test (Access Bio, New Jersey,
USA) by comparing its performance to quantitative G6PD enzyme activity using a standardized spectrophotometric method
(‘gold standard’). Blood samples (n = 903) were collected from Cambodian adults living in Pailin province, western
Cambodia. G6PD enzyme activities ranged from 0 to 20.5 U/g Hb (median 03.0 U/g Hg). Based on a normal haemoglobin
concentration and wild-type G6PD gene, the normal values of G6PD enzymatic activity for this population was 3.6 to 20.5 U/
g Hg (95th percentiles from 5.5 to 17.2 U/g Hg). Ninety-seven subjects (10.7%) had ,3.6 U/g Hg and were classified as G6PD
deficient. Prevalence of deficiency was 15.0% (64/425) among men and 6.9% (33/478) among women. Genotype was
analyzed in 66 G6PD-deficient subjects and 63 of these exhibited findings consistent with Viangchang genotype. The
sensitivity and specificity of the CareStartTM G6PD deficiency screening test was 0.68 and 1.0, respectively. Its detection
threshold was ,2.7 U/g Hg, well within the range of moderate and severe enzyme deficiencies. Thirteen subjects (1.4%, 03
males and 1 female) with G6PD enzyme activities ,2 U/g Hg were falsely classified as ‘‘normal’’ by RDT. This experimental
RDT test here evaluated outside of the laboratory for the first time shows real promise, but safe application of it will require
lower rates of falsely ‘‘normal’’ results.
Citation: Kim S, Nguon C, Guillard B, Duong S, Chy S, et al. (2011) Performance of the CareStartTM G6PD Deficiency Screening Test, a Point-of-Care Diagnostic for
Primaquine Therapy Screening. PLoS ONE 6(03): e28357. doi:10.1371/journal.pone.0028357
Editor: Ruth D. Ellis, Laboratory of Malaria Immunology and Vaccinology, United States of America
Received August 22, 2011; Accepted November 7, 2011; Published December 2, 2011
Copyright: ß 2011 Kim et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
´nard is supported by the French Ministry of Foreign
Funding: This work was supported by grant from WHO Regional Office for the Western Pacific. Didier Me
Affairs and Kevin Baird by the Wellcome Trust. The CareStartTM G6PD deficiency screening test was provided by the manufacturer (Access Bio). Access Bio has
developed the second generation of CareStartTM G6PD deficiency screening test. The funders had no role in study design, data collection and analysis, decision to
publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: dmenard@pasteur-kh.org

Cambodia, where artemisinin resistant parasites are well documented [4,5,6]. Moreover, the 8-aminoquinolines are the only
effective drugs capable of killing P. vivax and P. ovale liver
hypnozoites, thereby, preventing relapses [7]. However, primaquine and tafenoquine cause predictable, intravascular haemolysis
in individuals with glucose-6-phosphate dehydrogenase (G6PD)
deficiency. Primaquine, at the transmission blocking dose (single
dose, 0.75 mg/kg), has been shown to cause haemolysis in P.
falciparum-infected African children without G6PD deficiency [8]
and to cause a greater initial fall in haemoglobin following
treatment compared to ACT alone [3]. High dose primaquine also
caused haemolysis in healthy, G6PD normal individuals [9]. This
dose of primaquine causes only slight drops in haemoglobin in
healthy African-American volunteers having the A- variant, but in
healthy volunteers having the B- Mediterranean variant, a
hemolysis of 25% of red blood cells occurs [10]. Thus, risk of

Introduction
In the context of malaria elimination, vector control measures
e.g. long lasting bednets and indoor residual spraying along with
prompt diagnosis and treatment of malaria infected patients are
the most effective tools currently available [1]. Antimalarial drugs
are seen as crucial to eliminate malaria and the focus is on the role
of drugs to block malaria transmission by killing gametocytes and
reducing the pool of liver hypnozoites of Plasmodium vivax and P.
ovale [2]. 8-aminoquinolines like primaquine (and tafenoquine, a
drug in phase III clinical trials) remain the only transmissionblocking drugs available. While Artemisinin Combined Therapies
(ACTs) reduce gametocytogenesis in P. falciparum malaria by killing
young gametocytes, they lack activity against mature gametocytes
[3]. Therefore, primaquine remains a vitally important tool to
block the transmission of parasites, especially in areas like
PLoS ONE | www.plosone.org

1

December 2011 | Volume 6 | Issue 03 | e28357

RDT for Screening G6PD Deficiency

of Cambodia (approval number 028 NECHR). Informed written
consent was provided by all individuals before inclusion in the
study and all investigations were conducted according to the
principles expressed in the Declaration of Helsinki. Results for
each patient (according to the quantitative G6PD activity test)
were given to the local Ministry of Health staff of Pailin district
involved in the study. Participants were been notified of their
G6PD status. For deficient people, information regarding their
deficit was provided.

hemolysis with primaquine varies with status of both infection and
G6PD deficiency variant.
The conversion of NADP+ to NADPH by G6PD is the ratelimiting reaction in the pentose phosphate pathway, the primary
source of reducing potential for the glutathione redox flux which
serves to as the primary protection of erythrocytes against oxidative
stress. Numerous drugs and chemicals such as primaquine, foods
(fava beans) or stress (infections) can induce hemolytic anemia in
G6PD-deficient individuals. G6PD deficiency is the second most
common hereditary enzyme deficiency which affects approximately
400 million people worldwide [11] and is distributed in areas of
current and previous endemic malaria [03]. This human enzyme
defect is caused by mutations in the G6PD gene located on
chromosome Xq28; thus, transmission of the genetic defect is X
linked. Hemizygote males are most affected and homozygous
females least affected; both are prone to red cell haemolysis.
Heterozygote females have mixed G6PD normal and deficient red
cells and their susceptibility to haemolysis depends on the balance
between the expression of the normal and abnormal X chromosomes [13]. The highest frequencies are detected in Africa, Southeast
Asia, Central and Southern Pacific islands, the Mediterranean
region, and in the Middle East [14]. There are some 140 different
genotypes of G6PD deficiency [15] with corresponding enzyme
activity phenotypes that vary from mild to severe. Primaquine
sensitivity phenotypes are known only in 3 variants; African A-,
Mahidol, and Mediterranean B-, representing mild, moderate and
severe sensitivity, respectively. Correlation between severity of
enzyme activity loss and sensitivity to primaquine is believed to
exist but has not been confirmed with clinical observations.
In Cambodia, G6PD deficiency is common, with a prevalence
ranging from 13.4% to 26.1% in males and 3.1% to 4.3% in
females, depending on the sampled population [16,17]. G6PDViangchan is the most frequent variant [16,18].
G6PD deficiency may be diagnosed by a variety of spectrophotometric enzyme activity assays or DNA-based detection of specific
mutations. These tests, however, require relatively sophisticated
laboratory capacities. Qualitative tests for this disorder based upon
reduction of NADP+ by G6PD have been used for screening patients
in the absence of laboratory facilities prior to administration of
primaquine therapy. Since 1979, the fluorescent spot test is
recommended as the most suitable method for screening in the field,
despite the need for an UV lamp, water bath incubator, and
micropipettor [19,20]. Such equipment and the skills to use them are
rarely available in malaria endemics areas [8]. Consequently,
providers in such settings invite substantial risk of harm by prescribing
primaquine therapy and the drug is thus rarely used. Rapid
diagnostic tests (RDTs) are an alternative and attractive option
because they are generally easy to use and could be used alongside
malaria RDTs. However, the only available test named BinaxNOW
G6PD testTM (ref 780-000, Binax, Inc., Maine, USA) which has been
recently evaluated, presents a major drawback with the need to
perform the test within a temperature ranging from 18 to 25uC,
imposing limited usefulness in tropical malaria-endemic countries
[21]. AccessBio (New Jersey, USA) has developed a new experimental
RDT called CareStartTM G6PD deficiency screening test and its first
assessment outside of their laboratories is reported here, by
comparing its performance to the gold standard, quantitative
spectrophotometric estimation of G6PD enzyme activity [14].

Population and Sample collection
A cross sectional survey was conducted in four malaria endemic
villages in Pailin Province in western Cambodia (Figure 1). After
informed consent was obtained, 2 ml of venous blood was
collected in EDTA-tube from healthy participants over 18 years
of age. Individuals with illness that affected competency to give
informed consent, along with pregnant or lactating women were
excluded from the study. Demographic information was gathered
by questionnaire. After collection, the fresh blood was used within
15 minutes to perform the CareStartTM G6PD deficiency
screening test in the field, according to manufacturers’ instructions, and then sent to the Malaria Molecular Epidemiology Unit
at Pasteur Institute in Cambodia (IPC) in 4uC cool box within
24 hours. At IPC, the quantitative determination of G6PD
enzymatic activity was performed using 100 ml of fresh blood.
The remaining blood samples were centrifuged and aliquots of
plasma, buffy coat and red blood cells pellets were stored at
220uC for DNA analysis.

Originally posted 2017-01-15 20:01:01. Republished by Blog Post Promoter

Malaria Molecular Epidemiology Unit, Pasteur Institute

Performance of the CareStartTM G6PD DeficiencyScreening Test, a Point-of-Care Diagnostic forPrimaquine Therapy ScreeningSaorin Kim1, Chea Nguon2, Bertrand Guillard3, Socheat Duong2, Sophy Chy1, Sarorn Sum1, Sina Nhem1,Christiane Bouchier4, Magali Tichit4, Eva Christophel5, Walter R. J. Taylor6, John Kevin Baird7,8, DidierMenard1*1 Malaria Molecular Epidemiology Unit, Pasteur Institute of Cambodia, Phnom Penh, Cambodia, 2 National Center for Parasitology, Entomology, and Malaria Control,Phnom Penh, Cambodia, 3 Medical Laboratory, Pasteur Institute of Cambodia, Phnom Penh, Cambodia, 4 Genomic Platform, Institut Pasteur, Paris, France, 5 WHO´decine Internationale et Humanitaire, Hopitaux Universitaires de Gene Geneva, Switzerland,`ve,Regional Office for the Western Pacific, Manilla, Philippines, 6 Service de Me7 Eijkman Oxford Clinical Research Unit, Jakarta, Indonesia, 8 Centre for Tropical Medicine, Nuffield Department of Clinical Medicine, University of Oxford, Oxford, UnitedKingdomAbstractDevelopment of reliable, easy-to-use, rapid diagnostic tests (RDTs) to detect glucose-6-phosphate dehydrogenase (G6PD)deficiency at point of care is essential to deploying primaquine therapies as part of malaria elimination strategies. Weassessed a kit under research and development called CareStartTM G6PD deficiency screening test (Access Bio, New Jersey,USA) by comparing its performance to quantitative G6PD enzyme activity using a standardized spectrophotometric method(‘gold standard’). Blood samples (n = 903) were collected from Cambodian adults living in Pailin province, westernCambodia. G6PD enzyme activities ranged from 0 to 20. 5 U/g Hb (median 12. 0 U/g Hg). Based on a normal haemoglobinconcentration and wild-type G6PD gene, the normal values of G6PD enzymatic activity for this population was 3. 6 to 20. 5 U/g Hg (95th percentiles from 5. 5 to 17. 2 U/g Hg). Ninety-seven subjects (10. 7 ) had ,3. 6 U/g Hg and were classified as G6PDdeficient. Prevalence of deficiency was 15. 0 (64/425) among men and 6. 9 (33/478) among women. Genotype wasanalyzed in 66 G6PD-deficient subjects and 63 of these exhibited findings consistent with Viangchang genotype. Thesensitivity and specificity of the CareStartTM G6PD deficiency screening test was 0. 68 and 1. 0, respectively. Its detectionthreshold was ,2. 7 U/g Hg, well within the range of moderate and severe enzyme deficiencies. Thirteen subjects (1. 4 , 12males and 1 female) with G6PD enzyme activities ,2 U/g Hg were falsely classified as ‘‘normal’’ by RDT. This experimentalRDT test here evaluated outside of the laboratory for the first time shows real promise, but safe application of it will requirelower rates of falsely ‘‘normal’’ results. Citation: Kim S, Nguon C, Guillard B, Duong S, Chy S, et al. (2011) Performance of the CareStartTM G6PD Deficiency Screening Test, a Point-of-Care Diagnostic forPrimaquine Therapy Screening. PLoS ONE 6(12): e28357. doi: 10. 1371/journal. pone. 0028357Editor: Ruth D. Ellis, Laboratory of Malaria Immunology and Vaccinology, United States of AmericaReceived August 22, 2011, Accepted November 7, 2011, Published December 2, 2011Copyright: ß 2011 Kim et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricteduse, distribution, and reproduction in any medium, provided the original author and source are credited. ´nard is supported by the French Ministry of ForeignFunding: This work was supported by grant from WHO Regional Office for the Western Pacific. Didier MeAffairs and Kevin Baird by the Wellcome Trust. The CareStartTM G6PD deficiency screening test was provided by the manufacturer (Access Bio). Access Bio hasdeveloped the second generation of CareStartTM G6PD deficiency screening test. The funders had no role in study design, data collection and analysis, decision topublish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: dmenard@pasteur-kh. orgCambodia, where artemisinin resistant parasites are well documented [4,5,6]. Moreover, the 8-aminoquinolines are the onlyeffective drugs capable of killing P. vivax and P. ovale liverhypnozoites, thereby, preventing relapses [7]. However, primaquine and tafenoquine cause predictable, intravascular haemolysisin individuals with glucose-6-phosphate dehydrogenase (G6PD)deficiency. Primaquine, at the transmission blocking dose (singledose, 0. 75 mg/kg), has been shown to cause haemolysis in P. falciparum-infected African children without G6PD deficiency [8]and to cause a greater initial fall in haemoglobin followingtreatment compared to ACT alone [3]. High dose primaquine alsocaused haemolysis in healthy, G6PD normal individuals [9]. Thisdose of primaquine causes only slight drops in haemoglobin inhealthy African-American volunteers having the A- variant, but inhealthy volunteers having the B- Mediterranean variant, ahemolysis of 25 of red blood cells occurs [10]. Thus, risk ofIntroductionIn the context of malaria elimination, vector control measurese. g. long lasting bednets and indoor residual spraying along withprompt diagnosis and treatment of malaria infected patients arethe most effective tools currently available [1]. Antimalarial drugsare seen as crucial to eliminate malaria and the focus is on the roleof drugs to block malaria transmission by killing gametocytes andreducing the pool of liver hypnozoites of Plasmodium vivax and P. ovale [2]. 8-aminoquinolines like primaquine (and tafenoquine, adrug in phase III clinical trials) remain the only transmissionblocking drugs available. While Artemisinin Combined Therapies(ACTs) reduce gametocytogenesis in P. falciparum malaria by killingyoung gametocytes, they lack activity against mature gametocytes[3]. Therefore, primaquine remains a vitally important tool toblock the transmission of parasites, especially in areas likePLoS ONE | www. plosone. org1December 2011 | Volume 6 | Issue 12 | e28357RDT for Screening G6PD Deficiencyof Cambodia (approval number 028 NECHR). Informed writtenconsent was provided by all individuals before inclusion in thestudy and all investigations were conducted according to theprinciples expressed in the Declaration of Helsinki. Results foreach patient (according to the quantitative G6PD activity test)were given to the local Ministry of Health staff of Pailin districtinvolved in the study. Participants were been notified of theirG6PD status. For deficient people, information regarding theirdeficit was provided. hemolysis with primaquine varies with status of both infection andG6PD deficiency variant. The conversion of NADP+ to NADPH by G6PD is the ratelimiting reaction in the pentose phosphate pathway, the primarysource of reducing potential for the glutathione redox flux whichserves to as the primary protection of erythrocytes against oxidativestress. Numerous drugs and chemicals such as primaquine, foods(fava beans) or stress (infections) can induce hemolytic anemia inG6PD-deficient individuals. G6PD deficiency is the second mostcommon hereditary enzyme deficiency which affects approximately400 million people worldwide [11] and is distributed in areas ofcurrent and previous endemic malaria [12]. This human enzymedefect is caused by mutations in the G6PD gene located onchromosome Xq28, thus, transmission of the genetic defect is Xlinked. Hemizygote males are most affected and homozygousfemales least affected, both are prone to red cell haemolysis. Heterozygote females have mixed G6PD normal and deficient redcells and their susceptibility to haemolysis depends on the balancebetween the expression of the normal and abnormal X chromosomes [13]. The highest frequencies are detected in Africa, SoutheastAsia, Central and Southern Pacific islands, the Mediterraneanregion, and in the Middle East [14]. There are some 140 differentgenotypes of G6PD deficiency [15] with corresponding enzymeactivity phenotypes that vary from mild to severe. Primaquinesensitivity phenotypes are known only in 3 variants, African A-,Mahidol, and Mediterranean B-, representing mild, moderate andsevere sensitivity, respectively. Correlation between severity ofenzyme activity loss and sensitivity to primaquine is believed toexist but has not been confirmed with clinical observations. In Cambodia, G6PD deficiency is common, with a prevalenceranging from 13. 4 to 26. 1 in males and 3. 1 to 4. 3 infemales, depending on the sampled population [16,17]. G6PDViangchan is the most frequent variant [16,18]. G6PD deficiency may be diagnosed by a variety of spectrophotometric enzyme activity assays or DNA-based detection of specificmutations. These tests, however, require relatively sophisticatedlaboratory capacities. Qualitative tests for this disorder based uponreduction of NADP+ by G6PD have been used for screening patientsin the absence of laboratory facilities prior to administration ofprimaquine therapy. Since 1979, the fluorescent spot test isrecommended as the most suitable method for screening in the field,despite the need for an UV lamp, water bath incubator, andmicropipettor [19,20]. Such equipment and the skills to use them arerarely available in malaria endemics areas [8]. Consequently,providers in such settings invite substantial risk of harm by prescribingprimaquine therapy and the drug is thus rarely used. Rapiddiagnostic tests (RDTs) are an alternative and attractive optionbecause they are generally easy to use and could be used alongsidemalaria RDTs. However, the only available test named BinaxNOWG6PD testTM (ref 780-000, Binax, Inc. , Maine, USA) which has beenrecently evaluated, presents a major drawback with the need toperform the test within a temperature ranging from 18 to 25uC,imposing limited usefulness in tropical malaria-endemic countries[21]. AccessBio (New Jersey, USA) has developed a new experimentalRDT called CareStartTM G6PD deficiency screening test and its firstassessment outside of their laboratories is reported here, bycomparing its performance to the gold standard, quantitativespectrophotometric estimation of G6PD enzyme activity [14]. Population and Sample collectionA cross sectional survey was conducted in four malaria endemicvillages in Pailin Province in western Cambodia (Figure 1). Afterinformed consent was obtained, 2 ml of venous blood wascollected in EDTA-tube from healthy participants over 18 yearsof age. Individuals with illness that affected competency to giveinformed consent, along with pregnant or lactating women wereexcluded from the study. Demographic information was gatheredby questionnaire. After collection, the fresh blood was used within15 minutes to perform the CareStartTM G6PD deficiencyscreening test in the field, according to manufacturers’ instructions, and then sent to the Malaria Molecular Epidemiology Unitat Pasteur Institute in Cambodia (IPC) in 4uC cool box within24 hours. At IPC, the quantitative determination of G6PDenzymatic activity was performed using 100 ml of fresh blood. The remaining blood samples were centrifuged and aliquots ofplasma, buffy coat and red blood cells pellets were stored at220uC for DNA analysis. CareStartTM G6PD deficiency screening testThe CareStartTM G6PD deficiency screening test was providedby the manufacturer (AccessBio, New Jersey, USA). The kitcontains the test strip encased in a flat plastic cassette (containing abuffer well, a sample well and a result window), a sample pipette,the assay buffer, an alcohol pad and a blood lancet. This RDTformat test, which is not yet commercially available, is a qualitativeenzyme chromatographic test, based on the reduction of colorlessnitro blue tetrazolium dye to dark colored formazan. Following themanufacturer instructions, two microliters of blood were addedinto the sample well and two drops of buffer into the buffer well. Test results were read visually after 10 minutes. Samples withnormal G6PD activity produce a distinct purple color backgroundin the result window while no color change was observed at thetest read time for samples with deficient G6PD activity (Figure 2). Samples with a pale purple color background were classified asnormal. Long term temperature stability of the CareStartTMG6PD deficiency screening test for providing information aboutRDT survival when used and stored outside the specifiedtemperatures was assessed following a modified proceduredescribed for malaria RDT lot testing Quality Control [22] byusing quality control (QC) materials (normal level, ref. G6888deficient level ref. G5888 from Trinity Biotech). Briefly, RDTformat tests were stored in incubators (35uC, 45uC and 55uC) withdaily temperature control. QC were repeated at D5, D10, D20,D30, D60 and D90 for RDT-format tests stored at 35uc and 45uCand at H4, H12, H24, H48 and H72 for RDT-format tests storedat 55uC. Quantitative determination of G6PD activityQuantitative determination of G6PD activity was performed onthe fresh blood within a maximum of 48 h after the collection,using the Trinity Biotech quantitative G6PD assayTM (Ref. 345UV, Trinity Biotech, St. Louis, USA) adapted on the Integra400TM automate (Roche diagnostic, Meylan, France), according toMaterials and MethodsEthics statementThe study protocol was reviewed and approved by the NationalEthics Committee for Health Research of the Ministry of HealthPLoS ONE | www. plosone. org2December 2011 | Volume 6 | Issue 12 | e28357RDT for Screening G6PD DeficiencyFigure 1. Map of Cambodia locating the four selected villages in Pailin province, Cambodia, 2010. doi: 10. 1371/journal. pone. 0028357. g001manufacturer’s instructions. Briefly, the procedure is a modification of the spectrophotometric methods of Kornberg andHorecker [23] and of Lohr and Waller [24], involving thereduction of NADP to NADPH by the G6PD enzyme in thepresence of glucose-6-phosphate and maleimide, acting as aninhibitor of erythrocyte 6-phosphogluconate dehydrogenase (6PGDH). The formation of NADPH is proportional to the G6PDactivity and is measured spectrophotometrically as an increase inabsorbance at 340 nm. Reliability of test results were monitoredby calibration and the use of three different controls provided byTrinity Biotech (deficient level ref. G5888, intermediate level ref. G5029 and normal level, ref. G6888) within each run. Runs wereconsidered valid if control values fell within the given range. G6PDactivities were expressed in terms of amount of hemoglobin (U/gHg). Determination of hemoglobin content of the EDTA samplewas performed on a CellDyn 3200TM analyser (Abbott, Rungis,France) after standardization with three different controls. Theinvestigator and laboratory technicians running the quantitativeenzymatic assay were masked to the results of the CareStartTMG6PD deficiency screening test. Detection of G6PD variantsFollowing the quantitative determination of G6PD activity andusing the normal ranges provided by Trinity Biotech (www. trinitybiotech. com/Product 20Documents/345-29 20EN. pdf),sequencing of the G6PD gene was performed on all samplesclassified as deficient (,4. 6 U/g Hg) and a randomly selectednumber of samples classified as normal ($4. 6 U/g Hg). DNA extractionHuman DNA was extracted from the buffy coat using theQIAamp DNA Blood Mini KitTM (Qiagen, Courtaboeuf, France),according to the manufacturer’s instructions. DNA fragment amplificationPrimers were designed to PCR amplify exons of the G6PDgene, as shown in Table 1. The complete exonic regions weregenerated in 8 fragments ranging between 221 and 848 bp. PCRwas performed in a 55 mL reaction containing 0. 4 mM of eachspecific primer, 0. 25 mM of each dNTP, 1. 5 to 2. 5 mM MgCl2,and 1. 5 unit of FirePol Taq polymeraseTM (Solis Biodyne, Tartu,Estonia). PCR cycling conditions were as follows: heating at 94uCfor 5 min, followed by 40 cycles of heating at 94uC for 30 s, 56–58uC for 30–90 s and 72uC for 60–150 s, and a final extensionperiod at 72uC for 10 min. Figure 2. Design of the CareStartTM G6PD deficiency screeningtest and interpretation of the results. Panel A, no color change forsample with deficient G6PD enzymatic activity, Panel B, distinct purplecolor for sample with normal G6PD enzymatic activity. doi: 10. 1371/journal. pone. 0028357. g002PLoS ONE | www. plosone. org3December 2011 | Volume 6 | Issue 12 | e28357RDT for Screening G6PD DeficiencyTable 1. Primers sequence, annealing temperatures and size of PCR products used to amplify and sequence exons of the G6PDgene. Exon 2Exons 3–4Sequence (59-39)Hybridization T6Size of PCRproductsPCR_G6PD_Ex2_FTGAAGGCTGCCTAGGAGAGA58uC494 bpPCR_G6PD_Ex2_RExonsCAGGTAGAGCCGGGATGAT58uC400 bp56uC500 bp58uC697 bp58uC221 bp58uC363 bp58uC752 bp58uC848 bpPrimer nameExon 5PCR_G6PD_Ex3–4_FTGTCCCCAGCCACTTCTAAPCR_G6PD_Ex3–4_RGGAGAGGAGGAGAGCATCCExons 6–7PCR_G6PD_Ex 5_FCGGGGACACTGACTTCTGPCR_G6PD_Ex 5_RACGCTGCCACCTTGTGGTExon 8PCR_G6PD_Ex 6–7_FACACAAGGCACGGGAGGTPCR_G6PD_Ex 6–7_RGAGGAGCTCCCCCAAGATAGExons 10–11–12CCCTTGAACCAGGTGAACAGTCAGTGCCTCGTCACAGATGPCR_G6PD_Ex 9_FCCTGAGGGCTGCACATCTPCR_G6PD_Ex 9_RExon 9PCR_G6PD_Ex 8_FPCR_G6PD_Ex 8_RGTGCGTGAGTGTCTCAGTGGTGAGACACTCACGCACTGGTPCR_G6PD_Ex 10–12_RExon 13PCR_G6PD_Ex 10–12_FTGAGGTAGCTCCACCCTCACPCR_G6PD_Ex 13_FTTATGGCAGGTGAGGAAAGGPCR_G6PD_Ex 13_RGAAGTGGGTCCTCAGGGAAGdoi: 10. 1371/journal. pone. 0028357. t001Subjects were also classed according to the WHO classification for G6PD, using the population derived mean G6PDenzyme activity as the reference for residual activity [14]: (i)Class I (very severely deficient (associated with chronic nonspherocytic hemolytic anaemia, ,1 residual activity,,0. 12 U/g Hg), (ii) Class II (severely deficient, 1 to 10 residual activity, 0. 13–1. 2 U/g Hg), (iii) Class III (moderatelydeficient, 10 to 60 residual activity, 1. 3–7. 1 U/g Hg), (iv) ClassIV (normal activity, 60 to 150 residual activity, 7. 2–17. 7 U/gHg) and (v) Class V (increased activity, . 150 residual activity,. 17. 7 U/g Hg). To assess the performance of the CareStartTM G6PD deficiencyscreening test, individuals were grouped as G6PD deficient ornormal, using the normal values of G6PD enzymatic activitydetermined in our Cambodian population. Standard diagnostictest measures were determined: Sensitivity (Se) was the probabilitythat the CareStartTM G6PD deficiency screening test classifyindividuals as deficient among individuals classified as deficient byusing the quantitative G6PD assay (true positive rate). Specificity(Sp) was the probability that the CareStartTM G6PD deficiencyscreening test classify individuals as normal among individualsclassified as normal by using the quantitative G6PD assay (truenegative rate). Positive Likelihood Ratio (PLR) was considered asthe ratio between the true positive rate and the false positive rate( = sensitivity/[12Specificity]) and Negative Likelihood Ratio(NLR) as the ratio between the false negative rate and the truenegative rate ( = [12Sensitivity]/Specificity). Positive PredictiveValue (PPV) was the probability that G6PD deficiency was presentin individuals when the CareStartTM G6PD deficiency screeningtest classified individuals as deficient and Negative PredictiveValue (NPV), the probability that G6PD deficiency was notpresent in individuals when the CareStartTM G6PD deficiencyscreening test classified individuals as normal. PPV and NPV werecalculated based on the prevalence (i. e. prior probability, PP) ofSequencing reactionsAfter purification by filtration with a NucleoFast 96 PCRplateTM (Macherey-Nagel, Duren, Germany), sequencing reac¨tions were performed for both strands by using the ABI PrismBigDye Terminator cycle sequencing ready reaction kitTM run ona 3730 xl genetic analyzerTM (Applied Biosystems, Courtaboeuf,France). Electrophoregrams were visualized and analyzed withCEQ2000 genetic analysis system softwareTM (Beckman Coulter,Villepinte, France). Nucleotide sequences were compared to thesequence of G6PD in GenBank (accession No. X55448) to identifymutations. Statistical analysisData were entered and verified using Microsoft ExcelTMsoftware, and analyzed using EpiInfo 6. 04TM software (CDC,Atlanta, USA) and MedCalcTM version 11. 6. 1 software (MedCalc,Mariakerke, Belgium). The Mann-Whitney U test or KruskalWallis method were used for non-parametric comparisons, andStudent’s t test or one-way analysis of variance for parametriccomparisons. For categorical variables, Chi-squared or Fisher’sexact tests were used to assess significant differences inproportions. P values,0. 05 indicated statistically significantdifferences. The mean, median, standard deviation and ranges weredetermined for all G6PD enzyme activity values. The normalvalues of G6PD enzymatic activity in Cambodian adults weredetermined by gender from a subset of random samples classifiedas normal using the quantitative G6PD assay ($4. 6 U/g Hg) andmeeting the following criteria: hemoglobin concentration . 12 g/dL, and absence of genetic polymorphism compared to thereference sequence: H. sapiens G6PD gene for glucose-6-phosphatedehydrogenase, accession No. X55448. Distribution of normalG6PD activity was evaluated by gender using KolmogorovSmirnov test. PLoS ONE | www. plosone. org4December 2011 | Volume 6 | Issue 12 | e28357RDT for Screening G6PD DeficiencyG6PD deficiency found in our sampling population as following: PPV = Se6PP/Se6PP+(12Sp)6(12PP) and NPV = Sp6(12PP)/Sp6(12PP)+(12Se)6PP. Normal values of G6PD enzymatic activity in CambodianadultsBased on 147 normal samples (74 males and 73 females) asdescribed in Statistical analysis section, the normal range of G6PDactivity for all subjects was 3. 6 to 20. 5 U/g Hb, the lower limit ofnormal was slightly lower for females (Table 2). G6PD values werenormally distributed for males (P = 0. 05) (Figure 4, Panel B). MeanG6PD enzymatic activities were significantly higher for males(P = 0. 005). ResultsStudy population and distribution of the G6PDenzymatic activityFrom October to December 2010, nine hundred three bloodsamples were collected in 4 villages in Pailin Province (Phitas Sbovn = 80, Andong Thmor, n = 127, Oh Tantramdey, n = 93 andPech Kiri, n = 603). No significant differences were observedbetween villages for sex ratio M/F (mean = 0. 89, P = 0. 74) and age(mean = 23. 9 years, P = 0. 53), except for hemoglobin level(mean = 12. 9 g/dL) which was lower in Pech Kiri compared toPhitas Sbov (12. 8 g/dL vs. 13. 4 g/dL, P = 0. 03). G6PD enzymatic activity values ranged from 0 to 20. 9 U/gHg with arithmetic mean of 10. 9 U/g Hg (95 CI: 10. 6–11. 2),SD of 4. 6 U/g Hg and median of 12. 0 U/g Hg (95 CI: 11. 9–12. 2). The distribution of G6PD enzymatic activity (Figure 3)visually displays two different populations. No significantdifferences were observed between gender and age. However,significant differences were found between villages (P,0. 01): thelowest mean was observed in Phitas Sbov (7. 4 U/g Hg,SD = 3. 9 U/g Hg) and the highest in Pech Kiri (11. 7 U/g Hg,SD = 5. 2 U/g Hg). Prevalence &amp, classification of G6PD deficiencyBased on the normal values of G6PD enzymatic activity definedin our Cambodian population, the prevalence of G6PD deficiencywas 10. 7 (97/903). Significant differences were observedbetween the sexes: 64/425 (15. 0 , CI95 : 11. 6 –19. 2 ) formales, and 33/478 (6. 9 , CI95 : 4. 7 –9. 7 ) for females,P,0. 001) and villages (9. 3 in Pech Kiri, 10. 2 in AndongThmor, 12. 9 in Oh Tantramdey and 20. 0 in Phitas Sbov,P,0. 001 ). According to the WHO classification, 1. 2 (11/903)individuals were classified as Class I, 5. 6 (51/903) as Class II,11. 0 (99/9093) as Class III, 76. 6 (691/903) as Class IV, and5. 6 (51/903) as Class V. Genotyping data and major variantsFour different variants were detected among the 64 samplesused for sequencing of the G6PD gene. The most prevalent wasFigure 3. Distribution of the G6PD enzymatic activity (U/g Hb) values of 903 individuals enrolled in four villages of the Pailinprovince, Cambodia, 2010. doi: 10. 1371/journal. pone. 0028357. g003PLoS ONE | www. plosone. org5December 2011 | Volume 6 | Issue 12 | e28357RDT for Screening G6PD DeficiencyTable 2. Reference values for G6PD enzymatic activity (U/g Hg), Pailin province, Cambodia, 2010. Reference values forG6PD enzymatic activityTotalN1477473Range3. 6–20. 54. 3–20. 53. 6–18. 9Mean (95 CI), U/g Hg11. 8 (11. 6–12. 1)12. 5 (12. 2–12. 8)11. 3 (11. 0–11. 7)SD, U/g Hg2. 72. 52. 6Median (95 CI), U/g Hg12. 2 (12. 0–12. 4)12. 5 (12. 3–12. 7)12. 0 (11. 5–12. 0)95th percentiles, U/g Hg5. 5–17. 26. 5–17. 95. 6–15. 5MaleFemaleN: Number of individuals, SD: Standard Deviation, 95 CI: 95 Confidence Inte. . .