Characterizing human movement patterns using GPS data loggers in an area of persistent malaria in Zimbabwe along the Mozambique border.

BACKGROUND: Human mobility is a driver for the reemergence or resurgence of malaria and has been identified as a source of cross-border transmission. However, movement patterns are difficult to measure in rural areas where malaria risk is high. In countries with malaria elimination goals, it is essential to determine the role of mobility on malaria transmission to implement appropriate interventions. METHODS: A study was conducted in Mutasa District, Zimbabwe, to investigate human movement patterns in an area of persistent transmission along the Mozambique border. Over 1 year, a convenience sample of 20 participants/month was recruited from active malaria surveillance cohorts to carry an IgotU® GT-600 global positioning system (GPS) data logger during all daily activities. Consenting participants were tested for malaria at data logger distribution using rapid antigen diagnostic tests and completed a survey questionnaire. GPS data were analyzed using a trajectory analysis tool, and participant movement patterns were characterized throughout the study area and across the border into Mozambique using movement intensity maps, activity space plots, and statistical analyses. RESULTS: From June 2016-May 2017, 184 participants provided movement tracks encompassing > 350,000 data points and nearly 8000 person-days. Malaria prevalence at logger distribution was 3.7%. Participants traveled a median of 2.8 km/day and spent a median of 4.6 h/day away from home. Movement was widespread within and outside the study area, with participants traveling up to 500 km from their homes. Indices of mobility were higher in the dry season than the rainy season (median km traveled/day = 3.5 vs. 2.2, P = 0.03), among male compared to female participants (median km traveled/day = 3.8 vs. 2.0, P = 0.0008), and among adults compared to adolescents (median total km traveled = 104.6 vs. 59.5, P = 0.05). Half of participants traveled outside the study area, and 30% traveled into Mozambique, including 15 who stayed in Mozambique overnight. CONCLUSIONS: Study participants in Mutasa District, Zimbabwe, were highly mobile throughout the year. Many participants traveled long distances from home, including overnight trips into Mozambique, with clear implications for malaria control. Interventions targeted at mobile populations and cross-border transmission may be effective in preventing malaria introductions in this region.

Steep Rebound of Chloroquine-Sensitive Plasmodium falciparum in Zimbabwe.

Removal of chloroquine from national malaria formularies can lead to the reversion of resistant Plasmodium falciparum to wild-type. We report a steep decline in chloroquine-resistant P falciparum within 10 years of national discontinuation of chloroquine monotherapy in Zimbabwe. Drug resistance surveillance is a vital component of malaria control programs, and the experience with chloroquine in Zimbabwe and elsewhere in sub-Saharan Africa is illustrative of the potentially rapid and dramatic impact of drug policy on antimalarial resistance.

Sulfadoxine-pyrimethamine parasitological efficacy against Plasmodium falciparum among pregnant women and molecular markers of resistance in Zambia: an observational cohort study.

BACKGROUND: The World Health Organization recommends the provision of intermittent preventive treatment during pregnancy (IPTp) with sulfadoxine-pyrimethamine (SP) at 4-week intervals from gestational week 13 to delivery in areas of moderate to high malaria transmission intensity. However, the effect of IPTp-SP has been compromised in some areas due to parasite resistance, raising the importance of parasitological and chemoprophylactic surveillance, and monitoring SP-resistance markers in the Plasmodium falciparum population. METHODS: Between November 2013 and April 2014 in Nchelenge, Zambia, 1086 pregnant women received IPTp-SP at antenatal-care bookings. Blood samples were collected on day 0, and on day 28 post-treatment to test for malaria parasites and to estimate SP parasitological efficacy in the treatment and prevention of parasitaemia. A random sample of 96, day 0 malaria-positive samples were analysed to estimate the prevalence of SP-resistance markers in the P. falciparum population. RESULTS: The overall parasitological and prophylactic failure among women who had paired day 0 and day 28 blood slides was 18.6% (95% CI 15.5, 21.8; 109 of 590). Among pregnant women who had asymptomatic parasitaemia on day 0, the day 28 PCR-uncorrected parasitological failure was 30.0% (95% CI 23.7, 36.2; 62 of 207) and the day 28 PCR-corrected parasitological failure was 15.6% (95% CI: 10.6, 20.6; 32 of 205). Among women who tested negative at day 0, 12.3% (95% CI: 9.0, 15.6; 47 of 383) developed parasitaemia at day 28. Among the 96 malaria-positive samples assayed from day 0, 70.8% (95% CI: 60.8, 79.2) contained the DHPS double (Gly-437 + Glu-540) mutation and 92.7% (95% CI: 85.3, 96.5) had the DHFR triple (Asn-108 + Ile-51 + Arg-59) mutation. The quintuple mutation (DHFR triple + DHPS double) and the sextuple mutant (DHFR triple + DHPS double + Arg-581) were found among 68.8% (95% CI: 58.6, 77.3) and 9.4% (95% CI: 4.2, 16.0) of samples, respectively. CONCLUSION: The parasitological and chemoprophylactic failure of SP, and the prevalence of resistance markers in Nchelenge is alarmingly high. Alternative therapies are urgently needed to safeguard pregnant women against malarial infection.

Decreased prevalence of the Plasmodium falciparum Pfcrt K76T and Pfmdr1 and N86Y mutations post-chloroquine treatment withdrawal in Katete District, Eastern Zambia.

BACKGROUND: In 2002, Zambia withdrew chloroquine as first-line treatment for Plasmodium falciparum malaria due to increased treatment failure and worldwide spread of chloroquine resistance. The artemisinin combination regimen, artemether-lumefantrine, replaced chloroquine (CQ) as first choice malaria treatment. The present study determined the prevalence of CQ resistance molecular markers in the Pfcrt and Pfmdr1 genes in Eastern Zambia at 9 and 13 years after the removal of drug pressure. METHODS: Samples collected from Katete District during the drug therapeutic efficacy assessments conducted in 2012 and 2016 were assayed by polymerase chain reaction (PCR) and restriction fragment length polymorphisms (RFLP) to determine the prevalence of genetic mutations, K76T on the Pfcrt gene and N86Y on the Pfmdr1 gene. A total of 204 P. falciparum-positive DBS samples collected at these two time points were further analysed. RESULTS: Among the samples analysed for Pfcrt K76T and Pfmdr1 N86Y in the present study, 112 (82.4%) P. falciparum-infected samples collected in 2012 were successfully amplified for Pfcrt and 94 (69.1%) for Pfmdr1, while 69 (65.7%) and 72 (68.6%) samples from 2016 were successfully amplified for Pfcrt and Pfmdr1, respectively. In 2012, the prevalence of Pfcrt 76K (sensitive) was 97.3%, 76T (resistant) was 1.8%, and 0.8% had both 76K and 76T codons (mixed). Similarly in 2012, the prevalence of Pfmdr1 86N (sensitive) was 97.9% and 86Y (resistant) was 2.1%. In the 2016 samples, the prevalence of the respective samples was 100% Pfcrt 76K and Pfmdr1 86N. CONCLUSION: This study shows that there was a complete recovery of chloroquine-sensitive parasites by 2016 in Katete District, Eastern Zambia, 13 years following the withdrawal of CQ in the country. These findings add to the body of evidence for a fitness cost in CQ-resistant P. falciparum in Zambia and elsewhere. Further studies are recommended to monitor resistance countrywide and explore the feasibility of integration of the former best anti-malarial in combination therapy in the future.

Detecting Plasmodium falciparum in community surveys: a comparison of Paracheck Pf® Test and ICT Malaria Pf® Cassette Test to polymerase chain reaction in Mutasa District, Zimbabwe.

BACKGROUND: Microscopy and rapid diagnostic tests (RDTs) are the main techniques used to diagnose malaria. While microscopy is considered the gold standard, RDTs have established popularity as they allow for rapid diagnosis with minimal technical skills. This study aimed to compare the diagnostic performance of two Plasmodium falciparum histidine-rich protein 2 (PfHRP2)-based RDTs (Paracheck Pf® Test (Paracheck) and Malaria Pf™ ICT (ICT)) to polymerase chain reaction (PCR) in a community survey. METHODS: A cross-sectional study was conducted between October 2012 and December 2014 in Mutasa District, Manicaland Province, eastern Zimbabwe. Households were randomly selected using satellite imagery, and 224 households were visited. Residents present in the household on the date of the visit were recruited for the study. Participants of all age groups from the selected households were screened with Paracheck and ICT RDTs in parallel. Dried blood spots (DBS) and thin and thick smears were collected. Parasite DNA extracted from the DBS was subjected to nested PCR targeting the Plasmodium cytochrome b mitochondrial gene. Data analysis was performed using the Cohen's Kappa test to determine the interrater agreement and the sensitivity and specificity of the diagnostic test were reported. RESULTS: Results from a total of 702 participants were analysed. Most were females, 397 (57%), and the median age of participants was 21 years with an interquartile range of 9-39 years. Of those who were screened, 8 (1.1%), 35 (5.0%), and 21 (2.9%) were malaria parasite positive by microscopy, RDT and PCR, respectively. Paracheck and ICT RDTs had a 100% agreement. Comparing RDT and PCR results, 34 participants (4.8%) had discordant results. Most of the discordant cases were RDT positive but PCR negative (n = 24). Half of those RDT positive, but PCR negative individuals reported anti-malarials to use in the past month, which is significantly higher than reported anti-malarial drug use in the population (p < 0.001). The participant was febrile on the day of the visit, but relying on PfHRP2-based RDT would miss this case. Among the diagnostic methods evaluated, with reference to PCR, the sensitivity was higher with the RDT (52.4%) while specificity was higher with the microscopy (99.9%). The positive predictive value (PPV) was higher with the microscopy (87.5%), while the negative predictive values were similar for both microscopy and RDTs (98%). Overall, a strong correlated agreement with PCR was observed for the microscopy (97.9%) and the RDTs (95.2%). CONCLUSIONS: Paracheck and ICT RDTs showed 100% agreement and can be used interchangeably. As malaria transmission declines and Zimbabwe aims to reach malaria elimination, management of infected individuals with low parasitaemia as well as non-P. falciparum infection can be critical.

Malaria knowledge and bed net use in three transmission settings in southern Africa.

BACKGROUND: Insecticide-treated nets (ITNs) reduce malaria morbidity and mortality in endemic areas. Despite increasing availability, the use of ITNs remains limited in some settings. Poor malaria knowledge is a barrier to the widespread use of ITNs. The goal of this study was to assess the levels of malaria knowledge and evaluate factors associated with bed net use among individuals residing in three regions of southern Africa with different levels of malaria transmission and control. METHODS: A cross-sectional study was conducted on a sample of 7535 residents recruited from 2066 households in Mutasa District, Zimbabwe (seasonal malaria transmission), Choma District, Zambia (low transmission) and Nchelenge District, Zambia (high transmission), between March 2012 and March 2017. A standardized questionnaire was used to collect data on demographics, malaria-related knowledge and use of preventive measures. Multivariate logistic regression analyses were used to assess determinants of bed net use. RESULTS: Most of the 3836 adult participants correctly linked mosquito bites to malaria (85.0%), mentioned at least one malaria symptom (95.5%) and knew of the benefit of sleeping under an ITN. Bed net ownership and use were highest in Choma and Nchelenge Districts and lowest in Mutasa District. In multivariate analyses, knowledge of ITNs was associated with a 30-40% increased likelihood of bed net use after adjusting for potential confounders across all sites. Other factors significantly associated with bed net use were age, household size and socioeconomic status, although the direction, strength and size of association varied by study site. Importantly, participants aged 5-14 years had reduced odds of sleeping under a bed net compared to children younger than 5 years. CONCLUSION: Relevant knowledge of ITNs translated into the expected preventive behaviour of sleeping under a bed net, underscoring the need for continued health messaging on malaria prevention. The implementation and delivery of malaria control and elimination interventions needs to consider socioeconomic equity gaps, and target school-age children to ensure access to and improve utilization of ITNs.

Distinct parasite populations infect individuals identified through passive and active case detection in a region of declining malaria transmission in southern Zambia.

BACKGROUND: Substantial reductions in the burden of malaria have been documented in parts of sub-Saharan Africa, with elimination strategies and goals being formulated in some regions. Within this context, understanding the epidemiology of low-level malaria transmission is crucial to achieving and sustaining elimination. A 24 single-nucleotide-polymorphism Plasmodium falciparum molecular barcode was used to characterize parasite populations from infected individuals identified through passive and active case detection in an area approaching malaria elimination in southern Zambia. METHODS: The study was conducted in the catchment area of Macha Hospital in Choma District, Southern Province, Zambia, where the parasite prevalence declined over the past decade, from 9.2% in 2008 to less than 1% in 2013. Parasite haplotypes from actively detected, P. falciparum-infected participants enrolled in a serial cross-sectional, community-based cohort study from 2008 to 2013 and from passively detected, P. falciparum-infected individuals enrolled at five rural health centres from 2012 to 2015 were compared. Changes in P. falciparum genetic relatedness, diversity and complexity were analysed as malaria transmission declined. RESULTS: Actively detected cases identified in the community were most commonly rapid diagnostic test negative, asymptomatic and had submicroscopic parasitaemia. Phylogenetic reconstruction using concatenated 24 SNP barcode revealed a separation of parasite haplotypes from passively and actively detected infections, consistent with two genetically distinct parasite populations. For passively detected infections identified at health centres, the proportion of detectable polyclonal infections was consistently low in all seasons, in contrast with actively detected infections in which the proportion of polyclonal infections was high. The mean genetic divergence for passively detected infections was 34.5% for the 2012-2013 transmission season, 37.8% for the 2013-2014 season, and 30.8% for the 2014-2015 season. The mean genetic divergence for actively detected infections was 22.3% in the 2008 season and 29.0% in the 2008-2009 season and 9.9% across the 2012-2014 seasons. CONCLUSIONS: Distinct parasite populations were identified among infected individuals identified through active and passive surveillance, suggesting that infected individuals detected through active surveillance may not have contributed substantially to ongoing transmission. As parasite prevalence and diversity within these individuals declined, resource-intensive efforts to identify the chronically infected reservoir may not be necessary to eliminate malaria in this setting.

High prevalence of dhfr and dhps molecular markers in Plasmodium falciparum in pregnant women of Nchelenge district, Northern Zambia.

BACKGROUND: Sulphadoxine-pyrimethamine (SP) is the recommended drug for intermittent preventive treatment in pregnancy (IPTp) in most African countries, including Zambia. However, malaria is still one of the leading causes of morbidity and mortality in pregnant women despite reports of greater than 50% of women taking at least two doses of SP in IPTp. Studies have shown that resistance to SP is associated with mutations in the dhfr and dhps gene of Plasmodium falciparum. This study examined the prevalence of dhfr and dhps polymorphisms in P. falciparum found in pregnant women of Nchelenge district. METHOD: This cross-sectional study was conducted in 2013 in Nchelenge, a holoendemic area with malaria prevalence estimated at 50% throughout the year. Three rural health centres were randomly selected and a census survey carried out at each health centre. A questionnaire was administered and malaria testing done using RDT and microscopy, with collection of a dried blood spot. A chelex extraction was done to extract parasite DNA from dried blood spots followed by nested PCR and enzyme restriction digestion. RESULTS: Of the enrolled participants (n = 375), the median age of the women was 23. The prevalence of malaria by PCR was 22%. The PCR positive samples examined (n = 72) showed a high prevalence of dhfr triple (Asn-108 + Arg-59 + Ile-59) mutant (68%) and dhps double (Gly -437 + Glu-540) mutant (21%). The quintuple haplotype was found in 17% with 2 samples with an additional Gly-581mutation. In addition 6% mutations at Val-16 were found and none found at Thr-108 respectively, these both confer resistance to cycloguanil. Multivariate analysis showed that there was an association between malaria and women aged 30-34 years old p < 0.05(AOR: 0.36) at 95% CI. CONCLUSION: This study showed a high number of mutations in the dhfr and dhps genes. The high malaria endemicity in the general population of this area may have contributed to the high prevalence of resistant parasites in pregnant women, suggesting a need to examine the efficacy of SP given that it is the only approved drug for IPTp in Zambia.

High burden of malaria infection in pregnant women in a rural district of Zambia: a cross-sectional study.

BACKGROUND: Malaria continues to be a major health problem in low-income countries. Consequently, malaria control remains a public health priority in endemic countries such as Zambia. Pregnant women and children under 5 years of age are among groups at high risk of malaria infection. Malaria infection is associated with adverse birth outcomes that affect the mother, foetus, and infant. Infection with HIV has been shown to increase the risk of malaria infection in pregnancy. The prevalence and the predictors of malaria infection among pregnant women resident in the Nchelenge District of northern Zambia were investigated. METHODS: Between November 2013 and April 2014, pregnant women in the catchment areas of two health centres were recruited during their first antenatal care visit. HIV testing was conducted as part of routine care. In addition, blood samples were collected from 1086 participants and tested for malaria infection using standard microscopy and polymerase chain reaction (PCR) techniques specific for Plasmodium falciparum. Multivariate logistic regression were conducted to examine the predictors of malaria infection. RESULTS: The prevalence of malaria identified by microscopy was 31.8 % (95 % confidence intervals [CI], 29.0-34.5; N = 1079) and by PCR was 57.8 % (95 % CI, 54.9-60.8; N = 1074). HIV infection was 13.2 % among women on their first antenatal visit; the prevalence of malaria detected by PCR among HIV-uninfected and HIV-infected women was 56.7 % (531/936) and 65.2 % (90/138), respectively. In the final model, the risk of malaria infection was 81 % higher among pregnant women recruited from Nchelenge health centre compared to those attending the Kashikishi health centre (adjusted odds ratio = 1.81; 95 % CI, 1.38-2.37, P < 0.001), and HIV-infected women across health centres had a 46 % greater risk of malaria infection compared to HIV-uninfected women (adjusted odds ratio = 1.46; 95 %, 1.00-2.13, P = 0.045). CONCLUSION: High burden of malaria detected by PCR in these pregnant women suggests that past prevention efforts have had limited effect. To reduce this burden of malaria sustainably, there is clear need to strengthen existing interventions and, possibly, to change approaches so as to improve targeting of groups most affected by malaria.

Comparison of a PfHRP2-based rapid diagnostic test and PCR for malaria in a low prevalence setting in rural southern Zambia: implications for elimination.

BACKGROUND: Rapid diagnostic tests (RDTs) detecting histidine-rich protein 2 (PfHRP2) antigen are used to identify individuals with Plasmodium falciparum infection even in low transmission settings seeking to achieve elimination. However, these RDTs lack sensitivity to detect low-density infections, produce false negatives for P. falciparum strains lacking pfhrp2 gene and do not detect species other than P. falciparum. METHODS: Results of a PfHRP2-based RDT and Plasmodium nested PCR were compared in a region of declining malaria transmission in southern Zambia using samples from community-based, cross-sectional surveys from 2008 to 2012. Participants were tested with a PfHRP2-based RDT and a finger prick blood sample was spotted onto filter paper for PCR analysis and used to prepare blood smears for microscopy. Species-specific, real-time, quantitative PCR (q-PCR) was performed on samples that tested positive either by microscopy, RDT or nested PCR. RESULTS: Of 3,292 total participants enrolled, 12 (0.4%) tested positive by microscopy and 42 (1.3%) by RDT. Of 3,213 (98%) samples tested by nested PCR, 57 (1.8%) were positive, resulting in 87 participants positive by at least one of the three tests. Of these, 61 tested positive for P. falciparum by q-PCR with copy numbers ≤ 2 x 10(3) copies/µL, 5 were positive for both P. falciparum and Plasmodium malariae and 2 were positive for P. malariae alone. RDT detected 32 (53%) of P. falciparum positives, failing to detect three of the dual infections with P. malariae. Among 2,975 participants enrolled during a low transmission period between 2009 and 2012, sensitivity of the PfHRP2-based RDT compared to nested PCR was only 17%, with specificity of >99%. The pfhrp gene was detected in 80% of P. falciparum positives; however, comparison of copy number between RDT negative and RDT positive samples suggested that RDT negatives resulted from low parasitaemia and not pfhrp2 gene deletion. CONCLUSIONS: Low-density P. falciparum infections not identified by currently used PfHRP2-based RDTs and the inability to detect non-falciparum malaria will hinder progress to further reduce malaria in low transmission settings of Zambia. More sensitive and specific diagnostic tests will likely be necessary to identify parasite reservoirs and achieve malaria elimination.

Pre-amplification methods for tracking low-grade Plasmodium falciparum populations during scaled-up interventions in Southern Zambia.

BACKGROUND: Malaria is receding in many endemic countries with intervention scale -up against the disease. However, this resilient scourge may persist in low-grade submicroscopic infections among semi-immune members of the population, and be poised for possible resurgence, creating challenges for detection and assessment of intervention impact. Parasite genotyping methods, such as the molecular barcode, can identify specific malaria parasite types being transmitted and allow tracking and evaluation of parasite population structure changes as interventions are applied. This current study demonstrates application of pre-amplification methods for successful detection and genotyping of residual Plasmodium falciparum infections during a dramatic malarial decline. METHODS: The study was a prospective cross-sectional design and based on a 2,000 sq km vicinity of Macha Mission Hospital in southern Zambia. Willing and predominantly asymptomatic residents of all ages were screened for malaria by microscopy during the 2005 and 2008 transmission seasons, with simultaneous collection of dried blood spots (DBS) on filter paper, and extraction of Plasmodium falciparum DNA was performed. Plasmodium falciparum infections were genotyped using a 24 SNP-based molecular barcode assay using real-time PCR. Submicroscopic parasitaemia samples were subjected to pre-amplification using TaqMan PreAmp Master Mix following the manufacturer's instructions before SNP barcode analysis. RESULTS: There was a dramatic decline of malaria between 2005 and 2008, and the geometric mean parasite density (95% CI) fell from 704/µL (390-1,271) in 2005 to 39/µL (23-68) in 2008, culminating in a large proportion of submicroscopic infections of which 90% failed to yield ample DNA for standard molecular characterization among 2008 samples. Pre-amplification enabled successful detection and genotyping of 74% of these low-grade reservoir infections, overall, compared to 54% that were detectable before pre-amplification (p <0.0005, n = 84). Furthermore, nine samples negative for parasites by microscopy and standard quantitative PCR amplification were positive after pre-amplification. CONCLUSIONS: Pre-amplification allows analysis of an otherwise undetectable parasite population and may be instrumental for parasites identification, tracking and assessing the impact of interventions on parasite populations during malaria control and elimination programmes when parasitaemia is expected to decline to submicroscopic levels.

Association between malaria control and paediatric blood transfusions in rural Zambia: an interrupted time-series analysis.

BACKGROUND: Blood transfusions can reduce mortality among children with severe malarial anaemia, but there is limited evidence quantifying the relationship between paediatric malaria and blood transfusions. This study explores the extent to which the use of paediatric blood transfusions is affected by the number of paediatric malaria visits and admissions. It assesses whether the scale-up of malaria control interventions in a facility catchment area explains the use of paediatric blood transfusions. METHODS: The study was conducted at a referral hospital for 13 rural health centres in rural Zambia. Data were used from facility and patient records covering all paediatric malaria admissions from 2000 to 2008. An interrupted time series analysis using an autoregression-moving-average model was conducted to assess the relationship between paediatric malaria outpatient visits and admissions and the use of paediatric blood transfusions. Further investigation explored whether the use of paediatric blood transfusions over time was consistent with the roll out of malaria control interventions in the hospital catchment area. RESULTS: For each additional paediatric malaria outpatient visit, there were 0.07 additional paediatric blood transfusions (95% CI 0.01-0.13; p < 0.05). For each additional paediatric admission for severe malarial anaemia, there were 1.09 additional paediatric blood transfusions (95% CI 0.95-1.23; p < 0.01). There were 19.1 fewer paediatric blood transfusions per month during the 2004-2006 malaria control period (95% CI 12.1-26.0; p < 0.01), a 50% reduction compared to the preceding period when malaria control was relatively limited. During the 2007-2008 malaria control period, there were 27.5 fewer paediatric blood transfusions per month (95% CI 14.6-40.3; p < 0.01), representing a 72% decline compared to the period with limited malaria control. CONCLUSIONS: Paediatric admissions for severe malarial anaemia largely explain total use of paediatric blood transfusions. The reduction in paediatric blood transfusions is consistent with the timing of the malaria control interventions. Malaria control seems to influence the use of paediatric blood transfusions by reducing the number of paediatric admissions for severe malarial anaemia. Reduced use of blood transfusions could benefit other areas of the health system through greater blood availability, particularly where supply is limited.

Malaria antifolate resistance with contrasting Plasmodium falciparum dihydrofolate reductase (DHFR) polymorphisms in humans and Anopheles mosquitoes.

Surveillance for drug-resistant parasites in human blood is a major effort in malaria control. Here we report contrasting antifolate resistance polymorphisms in Plasmodium falciparum when parasites in human blood were compared with parasites in Anopheles vector mosquitoes from sleeping huts in rural Zambia. DNA encoding P. falciparum dihydrofolate reductase (EC 1.5.1.3) was amplified by PCR with allele-specific restriction enzyme digestions. Markedly prevalent pyrimethamine-resistant mutants were evident in human P. falciparum infections--S108N (>90%), with N51I, C59R, and 108N+51I+59R triple mutants (30-80%). This resistance level may be from selection pressure due to decades of sulfadoxine/pyrimethamine use in the region. In contrast, cycloguanil-resistant mutants were detected in very low frequency in parasites from human blood samples-S108T (13%), with A16V and 108T+16V double mutants (∼4%). Surprisingly, pyrimethamine-resistant mutants were of very low prevalence (2-12%) in the midguts of Anopheles arabiensis vector mosquitoes, but cycloguanil-resistant mutants were highly prevalent-S108T (90%), with A16V and the 108T+16V double mutant (49-57%). Structural analysis of the dihydrofolate reductase by in silico modeling revealed a key difference in the enzyme within the NADPH binding pocket, predicting the S108N enzyme to have reduced stability but the S108T enzyme to have increased stability. We conclude that P. falciparum can bear highly host-specific drug-resistant polymorphisms, most likely reflecting different selective pressures found in humans and mosquitoes. Thus, it may be useful to sample both human and mosquito vector infections to accurately ascertain the epidemiological status of drug-resistant alleles.

Malaria Transmission at The Zimbabwe-Mozambique Border: An Observational Study of Parasitemia by Travel History and Household Location.

Cross-border human population movement contributes to malaria transmission in border regions, impeding national elimination. However, its impact in low-to-moderate transmission settings is not well characterized. This community-based study in Mutasa District, Zimbabwe, estimated the association of parasite prevalence with self-reported overnight travel to Mozambique and household distance to the border from 2012-2020. A fully adjusted Poisson regression model with robust variance estimation was fit using active surveillance data. The population attributable fraction of parasite prevalence from overnight travel was also estimated. The relative risk of testing positive for malaria by rapid diagnostic test declined 14% (prevalence ratio [PR] = 0.86, 95% CI = 0.81-0.92) per kilometer from the border up to 12 km away. Travel to Mozambique was associated with a 157% increased risk (PR = 2.57, 95% CI = 1.38-4.78), although only 5.8% of cases were attributable to overnight travel (95% CI = -1.1% to 12.7%), reflecting infrequent overnight trips (1.3% of visits). This study suggests that transmission in eastern Zimbabwe is driven by increasingly conducive social or environmental conditions approaching the border and low levels of importation from overnight travel. Although day trips to Mozambique during peak biting hours were not assessed, the contribution of such trips to ongoing transmission may be significant. Future malaria control efforts should prioritize high coverage of existing interventions and continued support for community health workers and health facilities at the border, which provide free case management.

Selection for chloroquine-sensitive Plasmodium falciparum by wild Anopheles arabiensis in Southern Zambia.

BACKGROUND: The emergence of parasite drug resistance, especially Plasmodium falciparum, persists as a major obstacle for malaria control and elimination. To develop effective public health containment strategies, a clear understanding of factors that govern the emergence and spread of resistant parasites in the field is important. The current study documents selection for chloroquine-sensitive malaria parasites by wild Anopheles arabiensis in southern Zambia. METHODS: In a 2,000-sq km region, mosquitoes were collected from human sleeping rooms using pyrethrum spray catches during the 2006 malaria transmission season. After morphological examination and molecular confirmation, vector mosquitoes were dissected to separate head and thorax from the abdominal section, followed by PCR screening for P. falciparum infection. Human residents of all ages were tested for P. falciparum parasitaemia by microscopy and PCR. Plasmodium falciparum infections were genotyped at the chloroquine resistance-conferring amino acid codon 76 of the PfCRT gene, using PCR and restriction enzyme digestion. RESULTS: In the human population there was nearly 90% prevalence of the chloroquine-resistant PfCRT K76T mutant, with no significant differences in polymorphism among smear-positive and smear-negative (submicroscopic) infections (p = 0.323, n = 128). However, infections in both abdominal and salivary gland phases of the An. arabiensis vector exhibited wild type K76-bearing parasites with up to 9X higher odds (OR (95% CI): 9 (3.7-20.2), p < 0.0005, n = 125), despite having been acquired from humans within a few weeks. CONCLUSIONS: Anopheles arabiensis selects for wild-type K76-bearing P. falciparum during both abdominal and salivary gland phases of parasite development. The rapid vectorial selection, also recently seen with antifolate resistance, is evidence for parasite fitness cost in the mosquito, and may underpin regional heterogeneity in the emergence, spread and waning of drug resistance. Understanding the nature and direction of vector selection could be instrumental for rational curtailment of the spread of drug resistance in integrated malaria control and elimination programmes.

Changes in the burden of malaria following scale up of malaria control interventions in Mutasa District, Zimbabwe.

BACKGROUND: To better understand trends in the burden of malaria and their temporal relationship to control activities, a survey was conducted to assess reported cases of malaria and malaria control activities in Mutasa District, Zimbabwe. METHODS: Data on reported malaria cases were abstracted from available records at all three district hospitals, three rural hospitals and 25 rural health clinics in Mutasa District from 2003 to 2011. RESULTS: Malaria control interventions were scaled up through the support of the Roll Back Malaria Partnership, the Global Fund to Fight AIDS, Tuberculosis and Malaria, and The President's Malaria Initiative. The recommended first-line treatment regimen changed from chloroquine or a combination of chloroquine plus sulphadoxine/pyrimethamine to artemisinin-based combination therapy, the latter adopted by 70%, 95% and 100% of health clinics by 2008, 2009 and 2010, respectively. Diagnostic capacity improved, with rapid diagnostic tests (RDTs) available in all health clinics by 2008. Vector control consisted of indoor residual spraying and distribution of long-lasting insecticidal nets. The number of reported malaria cases initially increased from levels in 2003 to a peak in 2008 but then declined 39% from 2008 to 2010. The proportion of suspected cases of malaria in older children and adults remained high, ranging from 75% to 80%. From 2008 to 2010, the number of RDT positive cases of malaria decreased 35% but the decrease was greater for children younger than five years of age (60%) compared to older children and adults (26%). CONCLUSIONS: The burden of malaria in Mutasa District decreased following the scale up of malaria control interventions. However, the persistent high number of cases in older children and adults highlights the need for strategies to identify locally effective control measures that target all age groups.

Improved assay to detect Plasmodium falciparum using an uninterrupted, semi-nested PCR and quantitative lateral flow analysis.

BACKGROUND: A rapid, non-invasive, and inexpensive point-of-care (POC) diagnostic for malaria followed by therapeutic intervention would improve the ability to control infection in endemic areas. METHODS: A semi-nested PCR amplification protocol is described for quantitative detection of Plasmodium falciparum and is compared to a traditional nested PCR. The approach uses primers that target the P. falciparum dihydrofolate reductase gene. RESULTS: This study demonstrates that it is possible to perform an uninterrupted, asymmetric, semi-nested PCR assay with reduced assay time to detect P. falciparum without compromising the sensitivity and specificity of the assay using saliva as a testing matrix. CONCLUSIONS: The development of this PCR allows nucleic acid amplification without the need to transfer amplicon from the first PCR step to a second reaction tube with nested primers, thus reducing both the chance of contamination and the time for analysis to < two hours. Analysis of the PCR amplicon yield was adapted to lateral flow detection using the quantitative up-converting phosphor (UCP) reporter technology. This approach provides a basis for migration of the assay to a POC microfluidic format. In addition the assay was successfully evaluated with oral samples. Oral fluid collection provides a simple non-invasive method to collect clinical samples.

Malaria research challenges in low prevalence settings.

The prevalence of malaria has reduced significantly in some areas over the past decade. These reductions have made local elimination possible and the research agenda has shifted to this new priority. However, there are critical issues that arise when studying malaria in low transmission settings, particularly identifying asymptomatic infections, accurate detection of individuals with microparasitaemic infections, and achieving a sufficient sample size to have an adequately powered study. These challenges could adversely impact the study of malaria elimination if they remain unanswered.

Temporal and spatial patterns of serologic responses to Plasmodium falciparum antigens in a region of declining malaria transmission in southern Zambia.

BACKGROUND: Critical to sustaining progress in malaria control is comprehensive surveillance to identify outbreaks and prevent resurgence. Serologic responses to Plasmodium falciparum antigens can serve as a marker of recent transmission and serosurveillance may be feasible on a large scale. METHODS: Satellite images were used to construct a sampling frame for the random selection of households enrolled in prospective longitudinal and cross-sectional surveys in two study areas in Southern Province, Zambia, one in 2007 and the other in 2008 and 2009. Blood was collected and stored as dried spots from participating household members. A malaria rapid diagnostic test (RDT) was used to diagnose malaria. An enzyme immunoassay (EIA) was used to detect IgG antibodies to asexual stage P. falciparum whole parasite lysate using serum eluted from dried blood spots. The expected mean annual increase in optical density (OD) value for individuals with a documented prior history of recent malaria was determined using mixed models. SatScan was used to determine the spatial clustering of households with individuals with serological evidence of recent malaria, and these households were plotted on a malaria risk map. RESULTS: RDT positivity differed markedly between the study areas and years: 28% of participants for whom serologic data were available were RDT positive in the 2007 study area, compared to 8.1% and 1.4% in the 2008 and 2009 study area, respectively. Baseline antibody levels were measured in 234 participants between April and July 2007, 435 participants between February and December 2008, and 855 participants between January and December 2009. As expected, the proportion of seropositive individuals increased with age in each year. In a subset of participants followed longitudinally, RDT positivity at the prior visit was positively correlated with an increase in EIA OD values after adjusting for age in 2007 (0.261, p = 0.003) and in 2008 (0.116, p = 0.03). RDT positivity at the concurrent visit also was associated with an increase in EIA OD value in 2007 (mean increase 0.177, p = 0.002) but not in 2008 (-0.063, p =0.50). Households comprised of individuals with serologic evidence of recent malaria overlapped areas of high malaria risk for serologic data from 2009, when parasite prevalence was lowest. CONCLUSIONS: Serological surveys to whole asexual P. falciparum antigens using blood collected as dried blood spots can be used to detect temporal and spatial patterns of malaria transmission in a region of declining malaria burden, and have the potential to identify focal areas of recent transmission.

What Heterogeneities in Individual-level Mobility Are Lost During Aggregation? Leveraging GPS Logger Data to Understand Fine-scale and Aggregated Patterns of Mobility.

Human movement drives spatial transmission patterns of infectious diseases. Population-level mobility patterns are often quantified using aggregated data sets, such as census migration surveys or mobile phone data. These data are often unable to quantify individual-level travel patterns and lack the information needed to discern how mobility varies by demographic groups. Individual-level datasets can capture additional, more precise, aspects of mobility that may impact disease risk or transmission patterns and determine how mobility differs across cohorts; however, these data are rare, particularly in locations such as sub-Saharan Africa. Using detailed GPS logger data collected from three sites in southern Africa, we explore metrics of mobility such as percent time spent outside home, number of locations visited, distance of locations, and time spent at locations to determine whether they vary by demographic, geographic, or temporal factors. We further create a composite mobility score to identify how well aggregated summary measures would capture the full extent of mobility patterns. Although sites had significant differences in all mobility metrics, no site had the highest mobility for every metric, a distinction that was not captured by the composite mobility score. Further, the effects of sex, age, and season on mobility were all dependent on site. No factor significantly influenced the number of trips to locations, a common way to aggregate datasets. When collecting and analyzing human mobility data, it is difficult to account for all the nuances; however, these analyses can help determine which metrics are most helpful and what underlying differences may be present.