Introduction
There is strong epidemiologic and biomolecular evidence supporting the protective effects of hemoglobin S and C on severe malaria, but their protective effects on uncomplicated and asymptomatic malaria are more contested.1–5 In southeast Asia hemoglobin E (HbE; β26Glu→Lys) predominates, with reported prevalence of 30–45% (as high as 74% documented among an ethnic minority group in northeastern Thailand).6,7 Those with homozygous hemoglobin E (HbEE) have mild disease with anemia, splenomegaly, and hemolysis, while the heterozygous state is generally asymptomatic.8 A study suggested a selective population advantage of high HbE allele frequencies,9 consistent with higher rates of hemoglobinopathies in malaria-infected areas, a theory proposed by Haldane over 60 years ago.10
However, there is a general paucity of evidence on the role of HbE in malaria pathogenicity. One study showed the rate of development of severe, acute malaria was only 2.4% for heterozygous hemoglobin E (HbAE) patients compared with 18% for patients without HbE.11 In Myanmar, patients with HbAE and normal hemoglobin (HbAA) showed similar parasitemia rates and clinical severity of malaria. However, the number of HbEE patients was small and none had severe disease.12 Another study with acute falciparum malaria patients being treated with artemisinin derivatives showed that patients with HbAE had significantly faster parasite clearance compared with controls with no hemoglobinopathies, a difference not demonstrated with the use of other antimalarial drugs.13 The authors hypothesize that this result might be related to potentiating effects of the oxidative reactions of artemisinins. One study in Thailand showed that there was no significant difference in allele frequency of HbE between patients with mild malaria and those with cerebral malaria.7 This study only had three HbEE patients, none of which developed cerebral malaria.
During a 3-year malaria epidemiologic project in the Chittagong Hill Districts of Bangladesh, an area hypoendemic for P. falciparum and P. vivax, we first investigated the prevalence of HbE and glucose-6-phosphate dehydrogenase (G6PD) deficiency among the two most populous ethnic minority groups. A separate subsequent malaria case–uninfected control study assessed the association of HbE in malaria cases detected principally by passive means compared with age and location matched uninfected controls, to address the contribution of HbE to malaria hot spots. Additional questions involved the relationship of HbE and initial parasite density as well as the clearance of parasites 2 days after treatment initiation.
Methods
The study area consisted of two unions in the Chittagong Hill Districts of Bangladesh, an area now known as hypoendemic for P. falciparum and P. vivax, with more than a dozen resident ethnic groups. As part of an epidemiology cohort study, we monitored populations year round in all age groups, performing demographic surveys, active and passive surveillance, entomological sampling and mapping. The unions consisted of about 24,000 people, divided into 24 geographic clusters of about 2.5 km2 and similar population sizes of about 1,000 persons.14,15 Informed consent was obtained from all adult participants and guardians of child participants. This study has been approved by the Johns Hopkins Bloomberg School of Public Health and International Center For Diarrheal Disease Research, Bangladesh (icddr,b) Institutional Review Board (IRB) committees.
There were three aspects to the analysis in this study: 1) determining the cross-sectional prevalences of HbE and G6PD deficiency in two of the more populous local ethnic groups in which blood sampling took place in the dry season (December 2010 and April–May 2011), 2) examining the associations between HbE and P. falciparum malaria cases (dates of malaria range from May 29, 2010 to February 16, 2012) compared with uninfected controls with HbE sampling performed from May 2012 to August 2013 (3 months to several years after malaria episode, with all but a few over a year later), and 3) assessing the association of HbE on both initial parasite density and absence of parasites 2 days after initiation of treatment of P. falciparum–infected cases, as a proxy for clearance.
Estimation of prevalences of HbE and G6PD deficiency.
A total of 112 Marma and 90 Khyang individuals, aged 19–31 years, already part of the larger epidemiologic study, were randomly selected using a random number generator with blood analysis done in December 2010 and March–April 2011. In the Marma tribe, there were 73 females and 39 males. In the Khyang tribe, there were 50 females and 40 males. Five women were pregnant at the time of blood draw based on documented pregnancy and birth records from our larger cohort study.
The blood samples were analyzed for malaria by rapid diagnostic test (RDT, FalciVax, Zephyr Biomedical Systems, Verna, Goa, India) and microscopy. Analysis of hemoglobin amounts and types was done using an automated hemoglobin testing system (VARIANT™ II Hemoglobin Testing System, Biorad Incorporated, Hercules, CA). Pregnant woman were excluded from the hemoglobin analysis. Laboratory analysis for G6PD deficiency was performed on the Hitachi 902 instrument (Roche Diagnositcs, Basel, Switzerland) using Randox reagents (Randox Laboratories, Crumlin, United Kingdom). The G6PD level was categorized by the World Health Organization (WHO) classification using a population-defined mean of 11.8 U/g Hb, from Cambodia,16 since we lack a local population-defined mean for Bangladesh. Using 11.8 U/g Hb, we established five classes: Class I: very severely deficient with < 1% residual activity (≤ 0.12 U/g Hb); Class II: severely deficient, > 1–10% residual activity (> 0.12–1.2 U/g Hb); Class III: mildly deficient, > 10–60% residual activity (> 1.2–7.1 U/g Hb); Class IV: normal activity, > 60–150% residual activity (> 7.1–17.7 U/g Hb); and Class V: increased activity, > 150% residual activity (> 17.7 U/g Hb).
Examining the associations between HbE and P. falciparum malaria.
A second and separate aspect of the study was a nested P. falciparum malaria case–uninfected control analysis for HbE status and RDT- or microscopy-positive P. falciparum malaria among the most populous Marma ethnic group. Malaria was detected by both passive and active means from our cohort study data. The malaria case definition required a positive RDT or microscopy and was divided between 1) passive case detection when individuals had sought treatment of fever, history of fever, or other symptoms, and were found to have a positive RDT and/or microscopy and 2) active case detection in which individuals tested positive after being randomly selected for RDT/microscopy testing within a predetermined sampling structure according to cluster and age, as used for the larger cohort study.14,15 From the larger epidemiologic cohort study, 170 passive malaria cases were randomly selected from 472 using a random number generator.14 All cases detected by active surveillance meeting this criteria from our cohort study were also included in this analysis. Uninfected controls were matched by age (6 months–4 years; 5 years–14 years; 15+ years), cluster (by 24 geographic areas), and ethnic group (all Marma). Uninfected controls were chosen initially on the basis of never having had documented positive malaria by RDT/microscopy during the cohort study that began in October 2009. They also had to have been selected for random active surveillance and to have specifically tested negative for malaria within 12 weeks of the time the case tested positive, with preference to those tested closer in time to the case. Participants with HbEE and HbAE were compared with those with HbAA, including those with low A2 (excluding those with β-thalassemia trait and the two cases of P. vivax malaria). The two cases that had malaria twice were only included one time for this analysis. Analysis was done using McNemar's test on matched pairs with a sensitivity analysis using conditional logistic regression, including those without matched pairs.
Using McNemar's test for the passive case–control analysis, 142 matched pairs were available after eliminating those with β-thalassemia. For the 142 eligible pairs, when comparing any HbE versus normal, if we assume a 50% probability of exposure, a power of 80%, a probability of an exposure discordant pair of 50%, and an alpha of 5%, we would at a minimum be able to detect an odds ratio (OR) of 2. When comparing HbEE versus normal, the minimum OR detectable (after reducing sample size by 36%) is 2.8. For the active sample (21 matched pairs), when similar comparisons are made using the same assumptions, we do not have the power to detect anything but the most extreme relationships. For any HbE versus normal, there is a minimum detectable OR of 7.8 and for HbEE versus normal, the minimum detectable OR would be 12.5 or greater.
Assessment of initial parasite density and absence of parasites 2 days after initiation of treatment.
Some of the malaria cases had only an initial RDT available, rather than a blood film for quantification. The geometric means and 95% confidence intervals (CIs) were calculated on the log-normalized parasite density using Student t tests. The means were compared through analysis of variance (ANOVA) tests. We also conducted a case–control analysis comparing the HbE status of subjects from any ethnic group who had persistent parasitemia 2 days after the initiation of treatment with artemether/lumefantrine (N = 48), with those P. falciparum malaria patients with absent parasites 2 days later (N = 288). The controls were a convenience sample selected from those P. falciparum malaria patients who were not positive on Day 2 in our active and passive studies described above.
All analyses were conducted using R Statistical software (Vienna, Austria).17
Results
Population hemoglobin analysis.
Initially, blood samples were taken for hemoglobin analysis on 202 study participants, 19–31 years of age, from the two most populous ethnic groups, in the dry season (December 2010 and late April to early May 2011). None of the participants tested positive for malaria by RDT or microscopy at this time. In total, 36% were heterozygote for HbE and 8% were homozygous, as shown in Table 1. The Khyang ethnic group had 11% (N = 10) homozygous for HbE and 38% (N = 34) heterozygous, while the Marma ethnic group had 5% (N = 6) homozygous and 34% (N = 38) heterozygous. These differences were not significant by Fisher's exact test (two-sided P value of 0.12). Hemoglobin Hope, a beta chain Gly136(H14)Asp mutation, was found in 1.5% of those sampled (N = 3). Low hemoglobin A2 was found in 5% (N = 10). Hemoglobin A2, a normal variant consisting of two alpha and two delta chains, may be elevated in people with β-thalassemias and has a lowered value in people with iron-deficient anemia or α-thalassemias, persistence of fetal hemoglobin, or hemoglobin H (thalassemia intermedia).18
Hemoglobin type and quantity analysis for random sampling of the Marma and Khyang ethnic groups excluding six pregnant women (N = 196)
| Frequency distribution | Number | Percentage | Hemoglobin mean (g/dL) | Hemoglobin quartiles 25%; 75% | Hemoglobin minimum; maximum |
|---|---|---|---|---|---|
| HbEE | 16 | 8.2 | 12.2 | 10.9; 13.1 | 9.8; 14.8 |
| HbAE | 69 | 35.2 | 12.5 | 11.4; 13.7 | 7.1; 16.0 |
| Normal study | 98 | 50.0 | 12.9 | 11.7; 14.1 | 6.9; 18.1 |
| Hb Hope | 3 | 1.5 | 12.5 | 11.4; 14.3 | 8.9; 14.6 |
| Low Hb A2 level | 10 | 5.1 | 11.45 | 10.8; 12.8 | 7.9; 13.2 |
| Total | 196 | 100.0 | 12.6 | 11.5; 13.8 | 6.9; 18.1 |
HbAE = heterozygous hemoglobin E; HbEE = homozygous hemoglobin E; Hb Hope = a beta chain Gly136(H14)Asp mutation.
The overall mean hemoglobin among the 196 (6 removed because of pregnancy at the time of blood draw) malaria-negative Marma and Khyang people (aged 19–31 years) tested in the initial prevalence survey was 12.6 g/dL (13.6 g/dL male, 12.0 g/dL female). We did not find a significant difference in hemoglobin levels by race or by HbE type in this population when controlling for age, sex, and pregnancy status (P value = 0.43).
G6PD deficiency.
Of the 202 participants tested for G6PD deficiency, the mean was 7.42 U/g Hb, with 58.9% having normal G6PD activity, 34.7% having mild deficiency, and 6.4% having severe deficiency (Figure 1). Unlike with hemoglobin levels, where no significant difference was found by ethnic group, we found that there was a significantly higher proportion of cases that were G6PD deficient among the Marma, with a mean 6.68 U/g Hb, compared with that of Khyang who had a mean G6PD activity of 8.35 U/g Hb, a difference of 1.68 (95% CI = −2.57 to −0.78), and a P value of 0.0003 using Welch two-sample t test. Thus, half the Marma population had at least a mild deficiency compared with 30% of the Khyang population.
Frequency of glucose-6-phosphate dehydrogenase (G6PD) deficiency and World Health Organization (WHO) classification categories. The frequency of 202 G6PD enzyme levels from the random sampling of Marma and Khyang tribal members. Class I: very severely deficient with < 1% residual activity (≤ 0.12 U/g Hb); Class II: severely deficient, > 1–10% residual activity (> 0.12–1.2 U/g Hb); Class III: mildly deficient, > 10–60% residual activity (> 1.2–7.1 U/g Hb); Class IV: normal activity, 60–150% residual activity (7.1–17.7 U/g Hb); Class V: increased activity, > 150% residual activity (> 17.7 U/g Hb).
Citation: The American Society of Tropical Medicine and Hygiene 93, 2; 10.4269/ajtmh.14-0623
Frequency of glucose-6-phosphate dehydrogenase (G6PD) deficiency and World Health Organization (WHO) classification categories. The frequency of 202 G6PD enzyme levels from the random sampling of Marma and Khyang tribal members. Class I: very severely deficient with < 1% residual activity (≤ 0.12 U/g Hb); Class II: severely deficient, > 1–10% residual activity (> 0.12–1.2 U/g Hb); Class III: mildly deficient, > 10–60% residual activity (> 1.2–7.1 U/g Hb); Class IV: normal activity, 60–150% residual activity (7.1–17.7 U/g Hb); Class V: increased activity, > 150% residual activity (> 17.7 U/g Hb).
Citation: The American Society of Tropical Medicine and Hygiene 93, 2; 10.4269/ajtmh.14-0623
Frequency of glucose-6-phosphate dehydrogenase (G6PD) deficiency and World Health Organization (WHO) classification categories. The frequency of 202 G6PD enzyme levels from the random sampling of Marma and Khyang tribal members. Class I: very severely deficient with < 1% residual activity (≤ 0.12 U/g Hb); Class II: severely deficient, > 1–10% residual activity (> 0.12–1.2 U/g Hb); Class III: mildly deficient, > 10–60% residual activity (> 1.2–7.1 U/g Hb); Class IV: normal activity, 60–150% residual activity (7.1–17.7 U/g Hb); Class V: increased activity, > 150% residual activity (> 17.7 U/g Hb).
Citation: The American Society of Tropical Medicine and Hygiene 93, 2; 10.4269/ajtmh.14-0623
Case–control analysis of HbE and malaria.
From the larger epidemiologic cohort, 170 passive malaria cases were randomly selected from a total of 472 cases appearing over the 2-year period. A total of 146 uninfected controls matched by age, cluster, and ethnic group were identified, located, and tested for hemoglobin amount and type. Of these 146 matched pairs, three were eliminated with β-thalassemia trait, one was eliminated as the case had P. vivax malaria, leaving 142 P. falciparum cases and matched uninfected controls (Table 2). We found significantly higher odds of malaria among patients with HbEE compared with those with HbAA (including those with low A2), based on our sample of age, location and ethnic group matched pairs using McNemar's exact test (OR = 5.0, 95% CI = 1.07–46.93; P value = 0.039). If the matched design is ignored, the unpaired analysis has similar prevalence of HbEE among cases and controls (10 and 12 out of the 142, respectively), while the paired analysis reached significance in the HbEE matched comparison of 10 to 2 normal Hb. Although not statistically significant, we found the opposite trend when comparing those with HbAE to HbAA (OR = 0.71, 95% CI = 0.42–1.19; P value = 0.22). A sensitivity analysis using conditional logistic regression, including the cases or controls that were collected without a match for both the active and passive analysis, resulted in similar ORs and no change in the conclusions of the study.
Hemoglobin type and malaria in matched pairs for passive Plasmodium falciparum malaria case—uninfected control analysis
| Uninfected controls | Cases | Total | ||
|---|---|---|---|---|
| HbEE | HbAE | Normal | ||
| HbEE | 1 | 7 | 2*† | 10 |
| HbAE | 1 | 26 | 38†‡ | 65 |
| Normal | 10*† | 27†‡ | 30 | 67 |
| Total | 12 | 60 | 70 | 142 |
HbAE = heterozygous hemoglobin E; HbEE = homozygous hemoglobin E.
McNemar's test results:
For HbEE vs. HbAA analysis: Odds ratio (OR) = 5.0, 95% confidence interval (CI) = 1.07–46.93; McNemar's exact two-tailed P value = 0.039.
For any HbE vs. HbAA analysis: OR = 0.93, 95% CI = 0.58–1.48; McNemar's exact two-tailed P value = 0.82.
For HbAE vs. HbAA analysis: OR = 0.71, 95% CI = 0.42–1.19; McNemar's exact two-tailed P value = 0.22.
Conditional logistic regression results:
HbEE vs. HbAA: OR = 5.0, 95% CI = 1.1–22.8; P value = 0.038.
Any HbE vs. HbAA: OR = 0.92, 95% CI = 0.59–1.45; P value = 0.73.
HbAE vs. HbAA: OR = 0.71, 95% CI = 0.43–1.16; P value = 0.18.
To examine malaria cases identified from the active surveillance, 24 randomly selected subjects who tested positive for malaria during the active surveillance were compared with age, cluster, and ethnic group matched uninfected controls. Of the 24 infected subjects, 23 had participating matched pairs. Of these, one case and one control had β-thalassemia trait, leaving 21 matched pairs for analysis. With this small sample size, we did not see any significant difference in proportion having HbEE and HbAE among matched malaria cases and controls. The conditional logistic regression analysis did not change this finding.
Next we examined if the initial parasite density of malaria cases might be different by hemoglobin status. Of the 24 cases from active surveillance (some were RDT only), we found an initial parasite density on 20, with a geometric mean of 1,320 (95% CI = 694–2,511). Of the 170 passive P. falciparum malaria cases, 122 had available parasite density, with a geometric mean of 4,254 (95% CI = 332–5,455). The mean parasite density did not significantly differ by hemoglobin type in either the active (F = 0.25, P value = 0.78) or passive studies (F = 1.13, P value = 0.34) (Table 3).
Hemoglobin type and initial parasite density
| HbEE | HbAE | β-Thalassemia trait | HbAA | |
|---|---|---|---|---|
| Active cases (N = 23)* | ||||
| Absent parasitemia N (%) | – | 2 (18) | 0 (0) | 1 (9) |
| Positive parasitemia N (%) | – | 9 (82) | 1 (100) | 10 (91) |
| Range (parasites/μL)† | – | 200–17,320 | – | 200–11,200 |
| Geometric mean parasite density (parasites/μL)†‡ (95% CI) | – | 1,694 (561–5,113) | 1,000 | 1,084 (391–3,009) |
| Passive cases (N = 147)§ | ||||
| Absent parasitemia N (%) | 1 (8) | 9 (14) | 0 (0) | 15 (22) |
| Positive parasitemia N (%) | 12 (92) | 54 (86) | 3 (100) | 53 (78) |
| Range (parasites/μL)† | 480–31,400 | 240–112,000 | 3,280–14,240 | 480–60,000 |
| Geometric mean parasite density (parasites/μL)†∥ (95% CI) | 2,232 (845–5,900) | 4,175 (2,815–6,193) | 6,175 (947–40,271) | 4,912 (3,425–7,044) |
CI = confidence interval; HbAA = normal hemoglobin; HbAE = heterozygous hemoglobin E; HbEE = homozygous hemoglobin E.
Blood samples not collected for one of 24 active cases (1 HbEE).
Range and geometric mean parasite density are calculated including all positive values.
ANOVA comparison of geomeans F = 0.25, P value = 0.78.
Blood films not available for 22 of 169 passive cases (1 β–thalassemia trait, 8 HbAE, 1 HbEE, 12 HbAA).
ANOVA comparison of geomeans F = 1.13, P value = 0.34.
The last part of the analysis related to an examination of the hemoglobin status of those patients who had positive P. falciparum parasitemia (possible delayed clearance) 2 days after initiating treatment with artemether/lumefantrine (N = 47). This group of 47 patients, parasite positive on day 2, was made up of seven malaria case found positive through our active surveillance and 40 patients with malaria detected by the passive surveillance, representing a range of ethnicities including 31 Marma, four Chakma, one Tripura, two Khyang, seven Bengali, and two from other ethnic groups. The matched infected controls in this analysis were all the P. falciparum cases for which we had conducted hemoglobin type analysis and had cleared parasites by day 2 (20 active and 165 passive). We also compared the Marma cases that had persistent parasitemia to controls, as all controls were Marma. Although the point estimates of the ORs were below 1 (suggestive of a possible faster clearance with HbE), we did not see a significant difference of absence of parasites by day 2 stratified by HbE status (Table 4). This result did not change if we separated active and passive malaria cases.
Hemoglobin type in parasite clearance analysis
| HbEE N (%) | HbAE N (%) | Any HbE N (%) | β-Thalassemia trait N (%) | HbAA N (%) | N | |
|---|---|---|---|---|---|---|
| All day 2 positive cases | 3* (6) | 16 (34) | 19‡ (40) | 3 (6) | 25 (53)*‡ | 47 |
| Day 2 positive Marma cases | 2† (6) | 11 (35) | 13§ (42) | 1 (3) | 17 (55)†§ | 31 |
| Day 2 negative Marma cases | 13*† (7) | 80 (43) | 93‡§ (50) | 5 (3) | 87 (47)*†‡§ | 185 |
HbAA = normal hemoglobin; HbAE = heterozygous hemoglobin E; HbE = hemoglobin E; HbEE = homozygous hemoglobin E.
For HbEE vs. HbAA analysis using all day 2 + cases: OR = 0.80, 95% CI = 0.14–3.26; Fisher's exact two-tailed P value = 1.
For HbEE vs. HbAA analysis using only Marma day 2 + cases: OR = 0.79, 95% CI = 0.08–4.01; Fisher's exact two-tailed P value = 1.
For any HBE vs. HbAA analysis: OR = 0.71, 95% CI = 0.34–1.45; Fisher's exact two-tailed P value = 0.40.
For any HBE vs. HbAA analysis using only Marma day 2 + cases: OR = 0.72, 95% CI = 0.30–1.67; Fisher's exact two-tailed P value = 0.44.
Discussion
In southeast Asia, an area that was hyperendemic for malaria, HbE has high prevalences.
In the Chittagong Hill Districts, malaria has persisted in this forested, hilly environment with seasonal heavy rains. In this study, a little under half of the randomly sampled Marma and Khyang populations had some variant of HbE. Among mild clinical malaria cases collected through passive sampling, there is some evidence of an increased proportion of homozygous HbEE compared with HbAA among those with malaria compared with matched uninfected controls. There was no significant difference in HbAE levels compared with normal levels among cases and controls. With our asymptomatic malaria case–control study collected through active sampling, we had too small sample size to define the relationship between malaria and HbE.
Differences have been noted in hospitalized, severe malaria cases in regard to HbE status with some evidence of a protective effect of HbEE12 and HbAE11 from severe complications of malaria in a hospital setting. In this study, we did not have enough severe malaria cases to analyze the association between severe disease and HbE status, but rather we compared those without infection to those who presented with mild clinical infection in a community setting. It is possible that the protective effects may only apply to reducing severe symptoms, a result found by Hutagalung and others,11 rather than preventing infection or onset of mild symptoms of infection. As we lacked sufficient power to compare asymptomatic cases found through active surveillance, it is not possible to parse out this difference with this analysis. However, as Billo and others19 have suggested, it is possible that the cross-sectional analyses used were not robust enough to identify these differences and that a longitudinal approach was needed. The small sample size and lack of significance when comparing the marginal frequencies of hemoglobin status by malaria suggest that further research on this question is needed to fully understand the relationship of HbE and malaria. With increased power, factors such as age and geographic clusters could be further explored. Furthermore, a study including populations with asymptomatic/submicroscopic infections, clinical mild malaria infections, as well as severe malaria may help to parse out whether the associations we found relate to HbEE or HbAE being associated with the initial malaria infection, becoming clinically symptomatic, and/or having severe symptoms and complications.
We did not find any evidence that hemoglobin type is associated with either geometric mean parasite density on the day of diagnosis or clearance of parasitemia by day 2 after beginning treatment. The day 2 estimates were approximate, as the day 2 sample was collected at any time during day 2 and could have ranged by a number of hours, possibly impacting our results. Bangladesh, despite repeated measurements, has yet to show delayed clearance with artemisinin treatment as seen in Cambodia.20–23
With respect to G6PD deficiency, we found 6% with severe deficiency in this population. This suggests one should screen for deficiency with a point-of-care test in this population before use of a 14-day course of primaquine used to eliminate the liver stage in P. vivax infections or long half-life tafenoquine.24,25 Single dose primaquine at 0.25 mg base/kg is well tolerated even in those severely deficient for G6PD, and is recommended by the WHO in addition to artemisinin combination therapy, for all but pregnant woman and infants under 1 year as a gametocytocidal drug for those with P. falciparum infections as part of an elimination strategy.26 The high levels of G6PD deficiency in this area should thus not be a reason to stop the implementation of these recommendations.
Limitations of this study include inability to locate all of the controls that had been selected for our cases. The population that did not have controls was slightly (although not statistically significantly) younger. As age and location may be related to our outcome, this was a limiting factor of our analysis. The analysis, particularly for the malaria cases detected through active surveillance, was limited by sample size. For the elimination studies, the number of hours defined as ‘2 days’ varied depending on when during the day the patients were originally tested/treated and when field workers were able to return to their villages. Thus the timing is not as controlled nor measured as often as some hospital-based parasite elimination studies. With respect to G6PD measurements, reticulocyte count was not taken at the time of blood draw. As patients were not ill at the time of blood draw, this is unlikely to substantially impact results. However, we would underestimate true rates of deficiency in the rare cases of high reticulocytosis elevating levels of G6PD.
In summary, we found high levels of G6PD deficiency, HbE, and anemia in this population. Uncomplicated malaria was more common among patients with homozygous HbE compared with controls, while no difference was found in the prevalence of malaria in patients with HbAE compared with normal patients. Focal populations of certain ethnic groups with remnant high prevalence of HbEE may provide an increased risk of uncomplicated malaria secondary to either longer duration of infection or increased risk of infection, and thus malaria control in these populations is necessary for elimination programs to be successful.
ACKNOWLEDGMENTS
We acknowledge the surveillance staff at the icddr,b field office in Bandarban for helping to arrange this study. We express our gratitude to the late Ashish Chowdhury who helped with the G6PD analysis. We also thank Ciprian M. Crainiceanu for his statistical support and express gratitude to the local population for their participation in the study and their enthusiasm.
- 1.↑
Agarwal A, Guindo A, Cissoko Y, Taylor JG, Coulibaly D, Kone A, Kayentao K, Djimde A, Plowe CV, Doumbo O, Wellems TE, Diallo D, 2000. Hemoglobin C associated with protection from severe malaria in the Dogon of Mali, a west African population with a low prevalence of hemoglobin S. Blood 96: 2358–2363.
- 2.
Modiano D, Luoni G, Sirima BS, Simpore J, Verra F, Konate A, Rastrelli E, Olivieri A, Calissano C, Paganotti GM, D'Urbano L, Sanou I, Sawadogo A, Modiano G, Coluzzi M, 2001. Haemoglobin C protects against clinical Plasmodium falciparum malaria. Nature 414: 305–308.
- 3.
Mockenhaupt FP, Ehrhardt S, Cramer JP, Otchwemah RN, Anemana SD, Goltz K, Mylius F, Dietz E, Eggelte TA, Bienzle U, 2004. Hemoglobin C and resistance to severe malaria in Ghanaian children. J Infect Dis 190: 1006–1009.
- 4.
May J, Evans JA, Timmann C, Ehmen C, Busch W, Thye T, Agbenyega T, Horstmann RD, 2007. Hemoglobin variants and disease manifestations in severe falciparum malaria. JAMA 297: 2220–2226.
- 5.↑
Williams TN, Mwangi TW, Wambua S, Alexander ND, Kortok M, Snow RW, Marsh K, 2005. Sickle cell trait and the risk of Plasmodium falciparum malaria and other childhood diseases. J Infect Dis 192: 178–186.
- 7.↑
Naka I, Ohashi J, Nuchnoi P, Hananantachai H, Looareesuwan S, Tokunaga K, Patarapotikul J, 2008. Lack of association of the HbE variant with protection from cerebral malaria in Thailand. Biochem Genet 46: 708–711.
- 8.↑
Williams TN, 2006. Human red blood cell polymorphisms and malaria. Curr Opin Microbiol 9: 388–394.
- 9.↑
Ohashi J, Naka I, Patarapotikul J, Hananantachai H, Brittenham G, Looareesuwan S, Clark AG, Tokunaga K, 2004. Extended linkage disequilibrium surrounding the hemoglobin E variant due to malarial selection. Am J Hum Genet 74: 1198–1208.
- 11.↑
Hutagalung R, Wilairatana P, Looareesuwan S, Brittenham GM, Aikawa M, Gordeuk VR, 1999. Influence of hemoglobin E trait on the severity of falciparum malaria. J Infect Dis 179: 283–286.
- 12.↑
Oo M, Tin S, Marlar T, O'Sullivan WJ, 1995. Genetic red cell disorders and severity of falciparum malaria in Myanmar. Bull World Health Organ 73: 659–665.
- 13.↑
Hutagalung R, Wilairatana P, Looareesuwan S, Brittenham GM, Gordeuk VR, 2000. Influence of hemoglobin E trait on the antimalarial effect of artemisinin derivatives. J Infect Dis 181: 1513–1516.
- 14.↑
Ahmed S, Galagan S, Scobie H, Khyang J, Prue CS, Khan WA, Ram M, Alam MS, Haq MZ, Akter J, Glass G, Norris DE, Nyunt MM, Shields T, Sullivan DJ, Sack DA, 2013. Malaria hotspots drive hypoendemic transmission in the Chittagong Hill Districts of Bangladesh. PLoS One 8: e69713.
- 15.↑
Khan WA, Sack DA, Ahmed S, Prue CS, Alam MS, Haque R, Khyang J, Ram M, Akter J, Nyunt MM, Norris D, Glass G, Shields T, Haq MZ, Cravioto A, Sullivan DJ Jr, 2011. Mapping hypoendemic, seasonal malaria in rural Bandarban, Bangladesh: a prospective surveillance. Malar J 10: 124.
- 16.↑
Khim N, Benedet C, Kim S, Kheng S, Siv S, Leang R, Lek S, Muth S, Chea N, Chuor CM, Duong S, Kerleguer A, Tor P, Chim P, Canier L, Witkowski B, Taylor WR, Menard D, 2013. G6PD deficiency in Plasmodium falciparum and Plasmodium vivax malaria-infected Cambodian patients. Malar J 12: 171.
- 17.↑
Team RR, 2008. A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
- 18.↑
Alperin JB, Dow PA, Petteway MB, 1977. Hemoglobin-A2 levels in health and various hematologic disorders. Am J Clin Pathol 67: 219–226.
- 19.↑
Billo MA, Johnson ES, Doumbia SO, Poudiougou B, Sagara I, Diawara SI, Diakite M, Diallo M, Doumbo OK, Tounkara A, Rice J, James MA, Krogstad DJ, 2012. Sickle cell trait protects against Plasmodium falciparum infection. Am J Epidemiol 176 (Suppl 7): S175–S185.
- 20.↑
Ashley EA, Dhorda M, Fairhurst RM, Amaratunga C, Lim P, Suon S, Sreng S, Anderson JM, Mao S, Sam B, Sopha C, Chuor CM, Nguon C, Sovannaroth S, Pukrittayakamee S, Jittamala P, Chotivanich K, Chutasmit K, Suchatsoonthorn C, Runcharoen R, Hien TT, Thuy-Nhien NT, Thanh NV, Phu NH, Htut Y, Han KT, Aye KH, Mokuolu OA, Olaosebikan RR, Folaranmi OO, Mayxay M, Khanthavong M, Hongvanthong B, Newton PN, Onyamboko MA, Fanello CI, Tshefu AK, Mishra N, Valecha N, Phyo AP, Nosten F, Yi P, Tripura R, Borrmann S, Bashraheil M, Peshu J, Faiz MA, Ghose A, Hossain MA, Samad R, Rahman MR, Hasan MM, Islam A, Miotto O, Amato R, MacInnis B, Stalker J, Kwiatkowski DP, Bozdech Z, Jeeyapant A, Cheah PY, Sakulthaew T, Chalk J, Intharabut B, Silamut K, Lee SJ, Vihokhern B, Kunasol C, Imwong M, Tarning J, Taylor WJ, Yeung S, Woodrow CJ, Flegg JA, Das D, Smith J, Venkatesan M, Plowe CV, Stepniewska K, Guerin PJ, Dondorp AM, Day NP, White NJ, 2014. Spread of artemisinin resistance in Plasmodium falciparum malaria. N Engl J Med 371: 411–423.
- 21.
Starzengruber P, Swoboda P, Fuehrer HP, Khan WA, Hofecker V, Siedl A, Fally M, Graf O, Teja-Isavadharm P, Haque R, Ringwald P, Noedl H, 2012. Current status of artemisinin-resistant falciparum malaria in south Asia: a randomized controlled artesunate monotherapy trial in Bangladesh. PLoS One 7: e52236.
- 22.
Noedl H, Se Y, Schaecher K, Smith BL, Socheat D, Fukuda MM, Consortium AS, 2008. Evidence of artemisinin-resistant malaria in western Cambodia. N Engl J Med 359: 2619–2620.
- 23.↑
Noedl H, Se Y, Sriwichai S, Schaecher K, Teja-Isavadharm P, Smith B, Rutvisuttinunt W, Bethell D, Surasri S, Fukuda MM, Socheat D, Thap LC, 2010. Artemisinin resistance in Cambodia: a clinical trial designed to address an emerging problem in southeast Asia. Clin Infect Dis 51: E82–E89.
- 24.↑
Li QG, O'Neil M, Xie LS, Caridha D, Zeng Q, Zhang J, Pybus B, Hickman M, Melendez V, 2014. Assessment of the prophylactic activity and pharmacokinetic profile of oral tafenoquine compared to primaquine for inhibition of liver stage malaria infections. Malar J 13: 141.
- 25.↑
Fernando D, Rodrigo C, Rajapakse S, 2011. Primaquine in vivax malaria: an update and review on management issues. Malar J 10: 351.
- 26.↑
Global Malaria Programme, 2012. Single Dose Primaquine as a Gametocytocide in Plasmodium falciparum Malaria. Available at: http://www.who.int/malaria/mpac/sep2012/en/.