Abstract
Objectives:
Until late 2015, Botswana recommended preventive treatment for pregnant women in malarial regions with chloroquine and proguanil (CP). The guideline change provided an opportunity to evaluate CP and adverse birth outcomes.
Methods:
The Tsepamo Study performed birth outcomes surveillance at large delivery centers throughout Botswana. We evaluated adverse birth outcomes from 2015 to 2017 at three hospitals where 93% of CP use was recorded. Outcomes included neonatal death (NND), small for gestational age (SGA), very SGA, stillbirth (SB), preterm delivery (PTD), and very PTD. Logistic regression analysis (unadjusted and adjusted) was conducted for each adverse birth outcome.
Results:
During the study period, 5,883 (26%) of 23,033 deliveries were exposed to CP, with the majority (65%) in the most malaria-endemic region. At this site, there was a trend or an association between CP use and reduction of three adverse birth outcomes: PTD (aOR 0.85, 95%CI 0.76–0.96), vPTD (aOR 0.83, 95%CI 0.68–1.01), and NND (aOR 0.65, 95% CI 0.42–1.00). However, at the least malaria-endemic site, the association was in the opposite direction for SB (aOR 1.54, 95%CI 1.08–2.22), SGA (aOR 1.24, 95% CI 1.06–1.44), and vSGA (aOR 1.42, 95%CI 1.14–1.77). The association between CP and reduced PTD was present among women without HIV (aOR 0.77, 95%CI 0.67–0.89) but not among women with HIV (aOR 1.09, 95%CI 0.78–1.35).
Conclusions:
Antimalarial prophylaxis was associated with improved birth outcomes in the most malaria-endemic region of Botswana, but not elsewhere. This finding supports current WHO guidance to use prophylaxis strategies among pregnant women in highly malaria-endemic regions. Further studies of the risks and benefits of specific antimalarial regimens in pregnancy are warranted, particularly in areas with lower incidence of malaria.
Keywords: Antimalarial prophylaxis, HIV, pregnancy, birth outcomes, Botswana
INTRODUCTION
To decrease maternal and infant consequences of placental malaria infection, WHO recommends intermittent preventative treatment in pregnancy (IPTp) in regions of moderate to high malaria transmission and in vulnerable populations in Africa.1–3 Current guidance supports sulfadoxine-pyrimethamine (SP), but due to the emergence of SP-resistant malaria, drug-drug interactions, and adverse drug effects, many African countries have not implemented WHO’s IPTp policy.1–3 Alternatives to IPTp-SP are needed, and the effect of specific antimalarial strategies on birth outcomes requires further research to guide policy decisions.
Decisions regarding antimalarial prophylaxis in pregnancy are more complicated in the setting of HIV infection. There are limited and inconsistent data on whether adverse birth outcomes are associated with use of certain antimalarials among women living with HIV that are on ART. While the role of HIV-malaria co-infection in mother-to-child transmission has been studied in detail,4–7 the deleterious effects of co-infection on birth outcomes are less well studied, and the few studies performed are conflicting.4,7–9 In addition, recent studies have demonstrated associations between antiretroviral treatment (ART) during pregnancy and adverse birth outcomes.10,11 However, it is unknown how concomitant use of ART and antimalarials may affect birth outcomes.
Botswana is an African country with endemic HIV throughout the country and malaria in the northern region. From 2007 to early 2015, the Botswana government recommended the use of a combined regimen of preventive treatment for pregnant women with chloroquine and proguanil (CP) throughout pregnancy (instead of the WHO-recommended IPTp-SP).12 Because the effectiveness of CP was not known, in 2015 the Ministry of Health in Botswana revised its guidelines to end prophylactic CP in pregnancy and recommended the use of vector control measures (bed nets), and intermittent testing for malaria infection.13
The Tsepamo Study performs ongoing birth outcomes surveillance throughout Botswana, including during the period of the CP guideline change in 2015. This surveillance offered an ideal opportunity to evaluate the effect of CP, by HIV status, on adverse birth outcomes.
METHODS
Study design and population
We performed a retrospective cohort study utilizing data from the Tsepamo Study, which extracted obstetric information and birth outcomes data at eight large delivery centers throughout Botswana. This sub-analysis used obstetrical records obtained from the three government facilities in Botswana (Nyangabgwe, Ghanzi and Maun) where 93.4% of antimalarial prophylaxis was used between January 1, 2015, and September 30, 2017. After excluding births among women on monotherapy (n=118) and those on non-prophylactic antimalarial treatment (n=3), there were 23,033 births exposed to either prophylaxis or no prophylaxis with chloroquine and proguanil (CP). The obstetric records consist of maternal medical history, demographics, pregnancy laboratory biomarkers/values, birth outcomes, maternal HIV status, other maternal diagnosis, and history of medication use (antimalarials and ART). We started capturing antimalarial use for pregnant women who had a delivery date from September 26, 2015. To avoid immortal time bias, we restricted our study to women who received CP administration from their first ANC visit. Ethical approval was granted by the Office of Human Research Administration at Harvard T.H. Chan School of Public Health, and the Health Research and Development Committee in Botswana.
Exposures and Outcomes
The primary exposure of this study was antimalarial exposure in pregnancy consisting of continuous preventive treatment with CP. Per the Guidelines for the Diagnosis and Treatment of Malaria in Botswana, the dose for CP administration was proguanil 200mg/daily and chloroquine 5mg/kg of body weight weekly given throughout pregnancy. CP administration was recommended to start in the first trimester at the initial ANC visits. We considered births as exposed to CP when there was documented CP use in the obstetrical record during pregnancy, and as unexposed to CP when no documentation of antimalarial use was recorded throughout pregnancy. HIV exposure was documented by a positive maternal HIV diagnostic test result during or before pregnancy. Births to mothers who stopped antimalarials during pregnancy were excluded from the analyses.
The primary outcomes of this study were neonatal death (NND), defined as deaths before 28 days of age among neonates who had never left the hospital; small for gestational age (SGA), defined as birth weight below the 10th percentile for gestational age; very small for gestational age (vSGA), defined as birth weight below the 3rd percentile for gestational age; stillbirth (SB), defined as fetal death with Apgar scores of 0, 0, and 0; preterm delivery (PTD), defined as birth at gestational age less than 37 weeks; and very preterm delivery (vPTD), defined as birth at gestational age less than 34 weeks.
Statistical analysis
For the primary outcome, logistic regression analysis (unadjusted and adjusted) was conducted for each adverse birth outcome (NND, SGA, vSGA, SB, PTD, and vPTD). Antimalarial use in pregnancy and HIV status was included along with other potential confounders in the first models. An interaction term was created for antimalarial+/HIV+ status. The covariates we used for adjusted analysis were selected a priori based on previous studies10,11 and included HIV status, maternal age, gravida (1, >1 and < 5, ≥5), occupation, educational attainment, and birth by C-section. For the secondary analysis, logistic regression analysis (unadjusted and adjusted) was performed for the comparative analysis of antimalarial use with HIV status for the primary outcomes (NND, SGA, vSGA, SB, PTD, and vPTD) to determine if HIV status was an effect modifier. We handled missing data by listwise deletion. The missing data was minimal and occurred in the variables for HIV status (0.32%), maternal age (0.03%), occupation (3.99%), educational attainment (2.79%), birth by C-section (0.07%), and NND (2.23%). All statistical analysis was conducted using Stata, Version 15.
RESULTS
During the study period, 5,883 (25.5%) of 23,033 deliveries at the three sites were exposed to CP; 1,215 (9.1%) of 13,404 at Nyangabgwe, 851 (44.0%) of 1,933 at Ghanzi, and 3,817 (49.6%) of 7,696 at Maun. Baseline demographic characteristics of exposed and unexposed women were generally similar across the three delivery sites (Table 1). Most CP use (64.9%) occurred in 2016, with only 7.9% in 2015 and 27.3% in 2017. Sensitivity analyses comparing birth outcomes among women unexposed to CP in each calendar year, or before and after the change in CP guidelines, did not identify differences in adverse birth outcomes over time.
Table 1.
Baseline Demographic Characteristics Among All Births, by Hospital Site and CP Exposure.
| Characteristic | Botswana Hospital Site | |||||||
|---|---|---|---|---|---|---|---|---|
| Population | TOTAL, N=23,033 | Nyangabgwe n=13,404 (58.19%) | Ghanzi n=1,933 (8.39%) | Maun n= 7,696 (33.41%) | ||||
| Exposure | CP | No CP | CP | No CP | CP | No CP | CP | No CP |
| Exposure, (Year),Total, N | ||||||||
| (2015 – 2017), N=23,033 | n=5,883 (25.54%) | n=17,150 (74.46%) | n=1,215 (9.06%) | n=12,189 (90.94%) | n=851 (44.02%) | n=1,082 (55.98%) | n=3,817 (49.60%) | n=3,879 (50.40%) |
| (2015), N=7,656 | n=462 (6.03%) | n=7,194 (93.97%) | n=140 (2.99%) | n=4,544 (97.01 %) | n=0 (0.00%) | n=326 (100.00%) | n=322 (12.17%) | n=2,324 (87.83%) |
| (2016), N=8619 | n=3,817 (44.29%) | n=4,802 (55.71%) | n=760 (15.42%) | n=4,168 (84.58%) | n=707 (84.07%) | n=134 (15.93%) | n=2,350 (82.46%) | n=500 (17.54%) |
| (2017), N=6758 | n=1,604 (23.73%) | n=5,154 (76.27%) | n=315 (8.31%) | n=3,477 (91.6%) | n=144 (18.80%) | n=622 (81.20%) | n=1,145 (52.05%) | n=1,055 (47.95%) |
| Age, median in years (IQR) | 26 (21–32) | 26 (22–31) | 25 (22–31) | 26 (22–31) | 24 (20–30) | 25 (21–30) | 26 (21–32) | 26 (22–32) |
| None or Primary Education | 600 (9.69%) | 1,246 (7.27%) | 106 (8.72%) | 692 (5.68%) | 196 (23.03%) | 240 (22.18%) | 298 (7.81 %) | 314 (8.09%) |
| Never Married | 583 (9.91%) | 1815 (10.58%) | 156 (12.84%) | 1,405 (11.53%) | 52 (6.11%) | 57 (5.27%) | 375 (9.82%) | 353 (9.10%) |
| Gravida | ||||||||
| (>=5) | 729 (12.39%) | 2037 (11.88%) | 125 (10.29%) | 1,363 (11.18%) | 109 (12.81%) | 157 (14.51%) | 495 (12.97%) | 517 (13.33%) |
| (>1 & <5) | 3,217 (54.68%) | 9,110 (53.12%) | 685 (56.38%) | 6,335 (51.97%) | 447 (52.53%) | 563 (52.03%) | 2,085 (54.62%) | 2,212 (57.03%) |
| (=1) | 1,937 (32.93%) | 6,003 (35.00%) | 405 (33.33%) | 4,491 ( 36.84%) | 295 (34.67%) | 362 (33.46%) | 1,237 (32.41%) | 1,150 ( 29.65%) |
| Occupation | ||||||||
| Student | 328 (5.58%) | 1,185 (6.91%) | 88 (7.24%) | 905 (7.42%) | 27 (3.17%) | 37 (3.42%) | 213 (5.58%) | 243 (6.26%) |
| House wife/none | 3,551 (60.36%) | 9,312 (54.30%) | 690 (56.79%) | 6,324 (51.88%) | 571 (67.10%) | 737 (68.11%) | 2,290 (59.99%) | 2,251 (58.03%) |
| Salaried Employment | 1,793 (30.48%) | 5,946 (34.67%) | 409 (33.66%) | 4,518 (37.07%) | 203 (23.85%) | 235 (21.72%) | 1,181 (30.94%) | 1,193 (30.76%) |
| HIV status | 1,423 (24.19%) | 4,654 (27.14%) | 335 ( 27.57%) | 3,477 ( 28.53%) | 132 (15.51%) | 194 (17.93%) | 956 ( 25.05%) | 983 ( 25.34%) |
| Birth by C-section | 1,341 ( 22.79%) | 3,950 (23.03%) | 416 (34.24%) | 2,910 ( 23.87%) | 119 (13.98%) | 200 (18.48%) | 806 (21.12%) | 840 (21.66%) |
The data are n (%) or median (IQR). CP = Chloroquine and Proguanil.
Overall, among the 5,883 (25.5%) pregnant women who were exposed to CP at the three sites, 998 (17.0%) had PTD, 345 (5.9%) had vPTD, 59 (1.0%) had NND, 117 (2.0%) had SB, 962 (16.4%) had SGA, and 410 (7.0%) had vSGA. In univariate analyses, the odds of adverse birth outcomes was reduced among pregnant women who used CP compared to those who were unexposed to CP: vPTD (OR 0.85, 95%CI 0.75–0.97) and NND (OR 0.65, 95% CI 0.50–0.87). Conversely, we did not observe associations between CP and PTD, SB, SGA, or vSGA (Table 2). In adjusted analyses, CP receipt was associated with reduction in vPTD (aOR 0.85, 95%CI 0.75–0.97) and NND (aOR 0.66, 95%CI 0.49–0.88), compared with the women unexposed to CP. However, we did not observe associations between CP and PTD, SB, SGA, or vSGA (Table 2).
Table 2.
Adverse birth outcomes among infants born to women in Botswana, by CP exposure.
| Adverse Birth Outcomes | CP (n=5,883) | No CP (n=17,150) | Unadjusted Odds Ratio (95% CI) | Adjusted Odds Ratio (95% CI) |
|---|---|---|---|---|
| Preterm Delivery | 998 (16.96) | 3,031 (17.62) | 0.96 (0.88 – 1.03) | 0.95 (0.87 – 1.03) |
| Very Preterm Delivery * | 345 (5.86) | 1,167 (6.80) | 0.85 (0.75 – 0.97) | 0.85 (0.75 – 0.97) |
| Neonetal Death * | 59 (1.03) | 260 (1.55) | 0.65 (0.50 – 0.87) | 0.66 (0.49 – 0.88) |
| Stillbirth | 117 (1.99) | 346 (2.02) | 0.99 (0.79 – 1.21) | 1.04 (0.84 – 1.29) |
| SGA | 962 (16.35) | 2754 (16.06) | 1.02 (0.94 – 1.10) | 1.03 (0.95 – 1.12) |
| Very SGA | 410 (6.97) | 1,097 (6.40) | 1.10 (0.97 – 1.23) | 1.08 (0.96 – 1.22) |
The data are n (%), unadjusted odds ratio (95% CI) or adjusted odds ratio (95% CI). All models were adjusted for HIV status, maternal age, gravida, occupation, and educational attainment. CP = Chloroquine and Proguanil. Adverse birth outcomes among women from three (Nyangabwe, Ghanzi, Maun, Total, N=23,033) government hospitals in Botswana.
indicates a statistically significant reduction in adjusted odds ratio (p-value < 0.05). CI = Confidence Interval.
Women who received CP had reduced odds for PTD (aOR 0.85, 95% CI 0.76–0.96), vPTD (aOR 0.83, 95% CI 0.68–1.01) and NND (aOR 0.65, 95% CI 0.42– 1.00), compared with unexposed women, after adjusting for HIV status, maternal age, educational attainment, gravida, occupation (Table 3). We did not observe any statistically significant associations at Maun on SB, SGA and vSGA.
Table 3.
Adverse birth outcomes among pregnant women in Botswana by hospital site and CP exposure.
| Botswana Hospital Site | ||||||||
|---|---|---|---|---|---|---|---|---|
| TOTAL, N=23,033 | Nyangabgwen n=13,404 | Ghanzi n=1,933 | Maun n= 7,696 | |||||
| Adverse Birth Outcome | CP | No CP | CP | No CP | CP | No CP | CP | No CP |
| (n=5,883) | (n=17,150) | (n=1,215) | (n=12,189) | (n=851) | (n=1,082) | (n=3,817) | (n=3,879) | |
| Preterm Delivery | ||||||||
| No. (%) | 998 (16.96) | 3,031 (17.62) | 223 (18.35) | 2.121(17.40) | 97 (11.40) | 116 (10.72) | 678 (17.76) | 784(20.21) |
| aOR* (95% CI) | 0.95 (0.87 – 1.03) | 1 [REF] | 1.05 (0.90–1.23) | 1 [REF] | 1.08 (0.80 – 1.46) | 1 [REF] | 0.85 (0.76–0.96) | 1 [REF] |
| Very Preterm Delivery | ||||||||
| No. (%) | 345 (5.86) | 1,167 (6.80) | 100 (8.23) | 875 (7.18) | 33 (3.88) | 32 (2.96) | 212 (5.55) | 260 (6.70) |
| aOR* (95% CI) | 0.85 (0.75 – 0.97) | 1 [REF] | 1.15 (0.92–1.43) | 1 [REF] | 1.24 (0.74–2.09) | 1 [REF] | 0.83 (0.68–1.01) | 1 [REF] |
| Neonetal Death | ||||||||
| No. (%) | 59 (1.03) | 260 (1.55) | 14 (1.19) | 193 (1.62) | 9 (1.08) | 11(1.04) | 36 (0.96) | 56 (1.48) |
| aOR* (95% CI) | 0.66(0.49 – 0.88) | 1 [REF] | 0.73 (0.42 – 1.26) | 1 [REF] | 1.04 (0.43 – 2.53) | 1 [REF] | 0.65 (0.42 – 1.00) | 1 [REF] |
| Stillbirth | ||||||||
| No. (%) | 117 (1.99) | 346 (2.02) | 35 (2.88) | 240(1.97) | 18(2.12) | 28(2.59) | 64(1.68) | 78(2.01) |
| aOR* (95% CI) | 1.04 (0.84 – 1.29) | 1 [REF] | 1.54 (1.08 – 2.22) | 1 [REF] | 0.81(0.44 – 1.52) | 1 [REF] | 0.94 (0.66 – 1.34) | 1 [REF] |
| SGA | ||||||||
| No. (%) | 962(16.35) | 2754(16.06) | 233 (19.18) | 1983 (16.27) | 155 (18.21) | 194 (17.93) | 574 (15.04) | 577 (14.87) |
| aOR* (95% CI) | 1.03 (0.95 – 1.12) | 1 [REF] | 1.24 (1.06 – 1.44) | 1 [REF] | 1.09 (0.85 – 1.39) | 1 [REF] | 0.98 (0.86 – 1.12) | 1 [REF] |
| Very SGA | ||||||||
| No. (%) | 410 (6.97) | 1,097 (6.40) | 105 (8.64) | 781 (6.41) | 67 (7.87) | 70 (6.47) | 238 (6.24) | 246 (6.34) |
| aOR* (95% CI) | 1.08 (0.96 – 1.22) | 1 [REF] | 1.42 (1.14 – 1.77) | 1 [REF] | 1.25 (0.86 – 1.81) | 1 [REF] | 0.94 (0.77 – 1.13) | 1 [REF] |
The data are n (%, 95% CI), or adjusted odds ratio (95% CI). All models were adjusted for HIV status, maternal age, gravida, occupation, and educational attainment. CP = Chloroquine and Proguanil. aOR = adjusted odds ratio. CI = Confidence Interval.
But at Nyangabgwe, where malaria is infrequent14 and prophylaxis with CP was employed more selectively (9.1%), we observed an association between CP use and increased SB (aOR 1.54, 95% CI 1.08–2.22), SGA (aOR 1.24, 95% CI 1.06–1.44), and vSGA (aOR 1.42, 95% CI 1.14–1.77) compared with unexposed women, after controlling for HIV status, maternal age, educational attainment, gravida, and occupation (Table 3). We did not observe any statistically significant associations at Nyangabgwe on PTD, vPTD, or NND. Unlike Nyangabgwe, in Ghanzi, a more moderate to high malaria risk site (Figure 1),14 with fewer overall deliveries but more CP use (44.0%), we did not observe statistically significant associations between CP use and any of the six birth outcomes after adjusting for HIV status, maternal age, educational attainment, gravida, occupation (Table 3).
Figure 1.
Map of Botswana, adapted from the World Health Organization, World Malaria Report 2019. This map shows the 2017 epidemiologic profile and confirmed malaria cases at the 3 government hospital delivery sites in Botswana for the Tsepamo Study, a birth outcomes surveillance study.
In Maun, we observed a statistically significant interaction (p-value = 0.008) between CP exposure and HIV status for PTD; >95% of women living with HIV (WLHIV) were also receiving ART in pregnancy, precluding an independent assessment by ART status. Here, we found that the association between CP and PTD was lower among WLHIV (aOR 1.09, 95% CI 0.78–1.35), than women without HIV (aOR 0.77, 95% CI 0.67–0.89) (Table 6).
Table 6.
Adverse birth outcomes among infants born to women in Maun, Botswana, by CP exposure and HIV status for Interaction Model.
| Botswana Hospital Site | |||
|---|---|---|---|
| Maun n= 7,696 | |||
| Adverse Birth Outcome | Yes Antimalarials | No Antimalarials | p value |
| (n=3,817) | (n=3,879) | ||
| Preterm Delivery | |||
| No. (%) | 678 (17.76) | 784 (20.21) | |
| OR (95% CI) | 0.85 (0.76–0.96) | 1 [REF] | 0.006 |
| aOR* (95% CI) | 0.85 (0.76–0.96) | 1 [REF] | 0.009 |
| M-H Combined | 0.85 (0.76–0.96) | 1 [REF] | 0.007 |
| aOR (−) (95% CI) | 0.77 (0.67 – 0.89) | 1 [REF] | 0.008 |
| aOR (+) (95% CI) | 1.09 (0.78 – 1.35) | 1 [REF] | 0.008 |
| β3** (95% CI) | 0.34 (0.07 – 0.60) | 1 [REF] | 0.012 |
| pvalue for TOH | 0.008 | ||
DISCUSSION
We utilized a change in Botswana’s guidelines to evaluate the effect of prophylactic CP use on six birth outcomes. We found that the overall odds of vPTD and NND were lower with CP use, and that this association was driven entirely by the most malaria-endemic site, where PTD was also lower with CP use. HIV infection (or ART use) appeared to reduce the protective association with CP and PTD. To our knowledge, this is the first study to evaluate the effect of prophylactic antimalarials on birth outcomes by HIV status in Botswana.
Our study is in accord with data from several others regarding the benefits of antimalarial prophylaxis to reduce adverse birth outcomes.15–20 A parallel-group, open-label, individually randomized controlled superiority trial conducted in Mali, at an area with highly seasonal malaria transmission, reported that a 3-dose regimen of IPTp-SP significantly reduced the risk of preterm births compared to the previously recommended 2-dose regimen.15 The fact that CP appeared most protective at the only malaria-endemic site in our study supports the plausibility of a true biologic effect of PTp-CP on birth outcomes, in accord with these prior findings. Hence, the endemicity of the geographical area may play a role in determining the effect of the preventative agent on birth outcomes.21 The mechanism of protection for IPTp generally, and of CP specifically, is presumed to be decreased parasite burden in the placenta, which improves vascular function.22 The relative impact of specific antimalarial regimens and of malaria endemicity on adverse birth outcomes warrant further study.
The association between CP exposure and adverse birth outcomes varied significantly by region. At Maun, where malaria transmission14 is high, and also very close in proximity to the locations where malaria is endemic, and where most CP was used (49.60%), we observed associations between CP and three birth outcomes in adjusted analyses (Table 3). In contrast to the findings at the Maun site, the significant association between CP and three adverse birth outcomes at the Nyangabgwe site was concerning to us. This association may be understood as a true effect from the CP itself, which might be unmasked in a low-malaria setting. However, we believe it is more likely to be the result of unmeasured confounding, since the Nyangabgwe site had only 9% of women who received CP and these may have differed from other women in their distance to the delivery center (most women in the Francistown region are not exposed to malaria and do not receive CP) or in other demographic characteristics that were not measured.
We identified HIV infection status or ART use, which could not be disentangled from HIV status given that >95% of pregnant women in Botswana receive ART as an effect modifier of the relationship between antimalarials and PTD. HIV or ART appeared to reduce the association between CP and PTD, perhaps because ART is known to have a direct impact on PTD, which may have mitigated any protective benefit of the CP. However, HIV status did not impact the relationship between CP and the other adverse birth outcomes, even though HIV (and ART use when separately evaluated) is an independent risk factor for several adverse birth outcomes, as previously reported.10,11 It would be interesting to find out if use of antimalarials in WLHIV is more likely to be associated with adverse birth outcomes when used with specific ART regimens, including modern dolutegravir-based regimens. Further studies are required to validate and better understand this interaction.
Our findings may have specific public health relevance in some settings that may consider CP use. It is known that IPTp significantly decreases risk of maternal anemia and neonatal mortality,1–3,23–28 but the ideal application of this intervention is unclear in regions with lower or uncertain burdens of malaria. WHO recommends use of IPTp-SP in regions of moderate to high malaria transmission in Africa.1–3 However, in regions of high HIV prevalence, management of pregnant women with HIV-malaria co-infection poses challenges due to the emergence of SP resistance, drug-drug interactions, and toxicity,29–34 hence alternatives to IPTp-SP to prevent malaria infection during pregnancy are needed for this population, and the use of CP may remain a consideration in these settings.
The strengths of our study included the large sample size used to evaluate the six adverse birth outcomes, a study population with homogenous sociodemographic characteristics, and restriction to women who received CP from their first ANC visit (to reduce immortal time bias). However, our study had some limitations. Because our study design was observational, we could not establish causation, and unmeasured confounding and bias was possible. Additional limitations were in the area of exposure for evaluation of medication adherence and compliance during pregnancy and inability to distinguish between the effect of HIV and specific ART regimen. For the outcomes, factors such as unmeasured early outcome – e.g. loss of pregnancy and NND that was not in the delivery site – were limitations of the study. Moreover, there were possible changes in administration and guidelines from the Botswana Malaria Control Program for the specific antimalarials used during study and potential variation in practice, management, and region/distance at the different ANC clinics and hospital delivery sites.
Table 4.
Adverse birth outcomes among infants born to women in Botswana, by CP exposure, hospital site, and HIV status
| Botswana Hospital Site | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Nyangabgwe | Ghanzi | Maun | ||||||||||
| Adverse Birth Outcome | HTV Exposed | HTV Unexposed | p value | HTV Exposed | HTV Unexposed | p value | HTV Exposed | HTV Unexposed | p value | HTV Exposed | HTV Unexposed | p value |
| (n=6,077) | (n=16,883) | (n=3,812) | (n=9,561) | (n=326) | (n=1,602) | (n=1939) | (n=5,720) | |||||
| Preterm Delivery | ||||||||||||
| No. (%) | 1,321 (21.74) | 2,673 (15.83) | 822 (21.56) | 1,513 (15.82) | 41 (12.58) | 170 (10.61) | 458 (23.62) | 990 (17.31) | ||||
| OR (95% CI) | 1.48 (1.37 – 1.59) | 1 [REF] | < 0.001 | 1.46 (1.33 –1.61) | 1 [REF] | < 0.001 | 1.21 (0.84 – 1.74) | 1 [REF] | 0.301 | 1.47 (1.30 – 1.67) | 1 [REF] | < 0.001 |
| aOR§ (95% CI) | 1.48 (1.37 – 1.59) | 1 [REF] | < 0.001 | 1.46 (1.33 –1.61) | 1 [REF] | < 0.001 | 1.22 (0.85 – 1.75) | 1 [REF] | 0.291 | 1.48(1.30 – 1.67) | 1 [REF] | < 0.001 |
| Very Preterm Delivery | ||||||||||||
| No. (%) | 529 (8.70) | 972 (5.76) | 357 (9.37) | 614 (6.42) | 17 (5.21) | 47 (2.93) | 155 (7.99) | 311 (5.44) | ||||
| OR (95% CI) | 1.55 (1.39 – 1.74) | 1 [REF] | < 0.001 | 1.51 (1.31 – 1.73) | 1 [REF] | < 0.001 | 1.82 (1.03 – 3.21) | 1 [REF] | 0.039 | 1.51 (1.23 – 1.85) | 1 [REF] | < 0.001 |
| aOR§ (95% CI) | 1.55 (1.39 – 1.74) | 1 [REF] | < 0.001 | 1.50 (1.31 – 1.73) | 1 [REF] | < 0.001 | 1.85 (1.05 – 3.26) | 1 [REF] | 0.034 | 1.51 (1.23 – 1.84) | 1 [REF] | < 0.001 |
| Neonetal Death | ||||||||||||
| No. (%) | 98 (1.66) | 219 (1.32) | 74 (2.00) | 132 (1.41) | 4 (1.27) | 16 (1.02) | 20 (1.06) | 71 (1.26) | ||||
| OR (95% CI) | 1.26 (0.99 – 1.60 ) | 1 [REF] | 0.061 | 1.42 (1.07 –1.90) | 1 [REF] | 0.016 | 1.24 (0.41 – 3.74) | 1 [REF] | 0.700 | 0.84 (0.51 – 1.38) | 1 [REF] | 0.489 |
| aOR§ (95% CI) | 1.24 (0.98 – 1.58 ) | 1 [REF] | 0.074 | 1.42 (1.06 –1.90) | 1 [REF] | 0.016 | 1.25 (0.41 – 3.75) | 1 [REF] | 0.697 | 0.84 (0.51 – 1.38) | 1 [REF] | 0.484 |
| Stillbirth | ||||||||||||
| No. (%) | 156 (2.57) | 303 (1.79) | 99 (2.60) | 175 (1.83) | 10 (3.07) | 35 (2.19) | 47 (2.42) | 93 (1.63) | ||||
| OR (95% CI) | 1.44 (1.19 – 1.75) | 1 [REF] | < 0.001 | 1.43 (1.11 – 1.83) | 1 [REF] | 0.005 | 1.41 (0.69 – 2.89) | 1 [REF] | 0.339 | 1.50 (1.05 – 2.14) | 1 [REF] | 0.024 |
| aOR§ (95% CI) | 1.44 (1.19 – 1.75) | 1 [REF] | < 0.001 | 1.43 (1.12 – 1.84) | 1 [REF] | 0.005 | 1.41 (0.69 – 2.87) | 1 [REF] | 0.349 | 1.50 (1.05 – 2.14) | 1 [REF] | 0.025 |
| SGA | ||||||||||||
| No. (%) | 1,278 (21.03) | 2,427 (14.38) | 800 (20.99) | 1,413 (14.78) | 81 (24.85) | 265 (16.54) | 397 (20.47) | 749 (13.09) | ||||
| OR (95% CI) | 1.59 (1.47 – 1.71) | 1 [REF] | < 0.001 | 1.53 (1.39 – 1.69) | 1 [REF] | < 0.001 | 1.67 (1.25 – 2.21) | 1 [REF] | < 0.001 | 1.71 (1.49 – 1.95) | 1 [REF] | < 0.001 |
| aOR§ (95% CI) | 1.59 (1.47 – 1.71) | 1 [REF] | < 0.001 | 1.53 (1.39 – 1.69) | 1 [REF] | < 0.001 | 1.67 (1.26 – 2.22) | 1 [REF] | < 0.001 | 1.71 (1.49 – 1.95) | 1 [REF] | < 0.001 |
| Very SGA | ||||||||||||
| No. (%) | 524 (8.62) | 979 (5.80) | 326 (8.55) | 560 (5.86) | 38 (11.66) | 98 (6.12) | 160 (8.25) | 321 (5.61) | ||||
| OR (95% CI) | 1.53 (1.37 – 1.71) | 1 [REF] | < 0.001 | 1.50 (1.30 – 1.73) | 1 [REF] | < 0.001 | 2.02 (1.36 – 3.01) | 1 [REF] | < 0.001 | 1.51 (1.24 – 1.84) | 1 [REF] | < 0.001 |
| aOR§ (95% CI) | 1.51 (1.31 – 1.74) | 1 [REF] | < 0.001 | 1.51 (1.31 – 1.74) | 1 [REF] | < 0.001 | 2.05 (1.38 – 3.04) | 1 [REF] | < 0.001 | 1.51 (1.24 – 1.84) | 1 [REF] | < 0.001 |
aOR§ adjusted for antimalarial use.
aOR = adjusted odds ratio. CI = Confidence Interval.
Table 5.
Variables: Exposures and Outcomes
| EXPOSURE | DEFINITION |
| Primary Exposure | |
| Prophylaxis Antimalarial Use (Chloroquine and Proguanil (C-P)) | Documentation of combined C-P regimen use during pregnancy in obstetrical record |
| Administration throughout pregnancy | Combined C-P regimen: Proguanil 200mg daily and Chloroquine 5mg/kg of body weight weekly -i.e. 300mg base |
| Secondary Exposure | |
| HIV Positive | Documentation of positive maternal HIV diagnostic test result during or before pregnancy |
| HIV Negative | Documentation of negative maternal HIV diagnostic test result during or before pregnancy were considered HIV unexposed |
| On Antiretroviral (ART) | Births to Women Living with HIV with ART exposure before delivery |
| ART naive | Births to Women Living with HIV with no ART exposure before delivery |
| BIRTH OUTCOME | DEFINITION |
| Primary Outcome | |
| Preterm Delivery (PTD) | Birth at gestational age less than 37 weeks |
| Very Preterm Delivery (vPTD) | Birth at gestational age less than 34 weeks |
| Neonatal death (NNTD) | Deaths before 28 days of age among neonates who had never left the hospital |
| Still birth (SB) | Fetal death - Apgar Scores of 0,0, and 0 |
| Small for Gestational Age (SGA) | Birth weight below the 10th percentile for gestational age |
| Very Small for Gestational Age (vSGA) | Birth weight below the 3rd percentile for gestational age |
| Secondary Analysis | |
| PTD, vPTD, NNTD, SB, SGA, and vSGA | Birth outcomes by HIV status and ART use |
Funding
National Institutes of Health (NIH/NICHD, R01 HD080471).
REFERENCES
- 1.World Health Organization. WHO policy brief for the implementation of intermittent preventive treatment of malaria in pregnancy using sulfadoxine-pyrimethamine (IPTp-SP). Geneva: World Health Organization. April 2013. (revised January 2014). https://www.who.int/publications/i/item/WHO-HTM-GMP-2014.4. [Google Scholar]
- 2.World Health Organization. Malaria in pregnant women. Geneva: World Health Organization. 2017. http://www.who.int/malaria/areas/high_risk_groups/pregnancy/en/. [Google Scholar]
- 3.World Health Organization. A strategic framework for malaria prevention and control during pregnancy in the African region. Brazzaville: WHO Regional Office for Africa. 2004. http://www.afro.who.int/sites/default/files/2017-06/malaria_in_pregnancy_092004.pdf. [Google Scholar]
- 4.Briand V, Badaut C, Cot M. Placental malaria, maternal HIV infection and infant morbidity. Ann Trop Paediatr. 2009;29:71–83. [DOI] [PubMed] [Google Scholar]
- 5.Ayisi JG, van Eijk AM, Newman RD, ter Kuile FO, Shi YP, Yang C, Kolczak MS, Otieno JA, Misore AO, Kager PA, Lal RB, Steketee RW, Nahlen BL. Maternal malaria and perinatal HIV transmission, western Kenya. Emerg Infect Dis. 2004;10:643–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Brahmbhatt H, Kigozi G, Wabwire-Mangen F, Serwadda D, Sewankambo N, Lutalo T, Wawer MJ, Abramowsky C, Sullivan D, Gray R. The effects of placental malaria on mother-to-child HIV transmission in Rakai, Uganda. AIDS. 2003;17:2539–41. [DOI] [PubMed] [Google Scholar]
- 7.Ayouba A, Nerrienet E, Menu E, Lobe MM, Thonnon J, Leke RJ, Barre-Sinoussi F, Martin P, Cunin P, Yaounde MG. Mother-to-child transmission of human immunodeficiency virus type 1 in relation to the season in Yaounde, Cameroon. Am J Trop Med Hyg. 2003;69:447–9. [PubMed] [Google Scholar]
- 8.ter Kuile FO, Parise ME, Verhoeff FH, Udhayakumar V, Newman RD, van Eijk AM, Rogerson SJ, Steketee RW. The burden of co-infection with human immunodeficiency virus type 1 and malaria in pregnant women in sub-saharan Africa. Am J Trop Med Hyg. 2004;71:41–54. [PubMed] [Google Scholar]
- 9.Flateau C, Le Loup G, Pialoux G. Consequences of HIV infection on malaria and therapeutic implications: a systematic review. Lancet Infect Dis. 2011;11:541–56. [DOI] [PubMed] [Google Scholar]
- 10.Chen JY, Ribaudo HJ, Souda S, Parekh N, Ogwu A, Lockman S, Powis K, Dryden-Peterson S, Creek T, Jimbo W, Madidimalo T, Makhema J, Essex M, Shapiro RL. Highly active antiretroviral therapy and adverse birth outcomes among HIV-infected women in Botswana. J Infect Dis. 2012;206:1695–705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zash R, Jacobson DL, Diseko M, Mayondi G, Mmalane M, Essex M, Petlo C, Lockman S, Makhema J, Shapiro RL. Comparative Safety of Antiretroviral Treatment Regimens in Pregnancy. JAMA Pediatr. 2017;171:e172222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Botswana Ministry of Health. Guidelines for the diagnosis and treatment of malaria in Botswana. Gaborone: National Malaria Control Programme, Department of Public Health, The Republic of Botswana Ministry of Health. 2007. [Google Scholar]
- 13.Botswana Ministry of Health. Revised guidelines for the diagnosis and treatment of malaria in Botswana. Gaborone: National Malaria Control Programme, Department of Public Health, The Republic of Botswana Ministry of Health. 2015. [Google Scholar]
- 14.World Health Organization. Malaria 2017 country profile: Botswana. Geneva: World Health Organization. 2017. https://www.who.int/publications/m/item/malaria-2017-country-profile-bwa. [Google Scholar]
- 15.Diakite OS, Kayentao K, Traore BT, Djimde A, Traore B, Diallo M, Ongoiba A, Doumtabe D, Doumbo S, Traore MS, Dara A, Guindo O, Karim DM, Coulibaly S, Bougoudogo F, Ter Kuile FO, Danis M, Doumbo OK. Superiority of 3 over 2 doses of intermittent preventive treatment with sulfadoxine-pyrimethamine for the prevention of malaria during pregnancy in mali: a randomized controlled trial. Clin Infect Dis. 2011;53:215–23. [DOI] [PubMed] [Google Scholar]
- 16.Manyando C, Njunju EM, Mwakazanga D, Chongwe G, Mkandawire R, Champo D, Mulenga M, De Crop M, Claeys Y, Ravinetto RM, van Overmeir C, Alessandro UD, Van Geertruyden JP. Safety of daily co-trimoxazole in pregnancy in an area of changing malaria epidemiology: a phase 3b randomized controlled clinical trial. PLoS One. 2014;9:e96017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mikomangwa WP, Oms M, Aklillu E, Kamuhabwa AAR. Adverse birth outcomes among mothers who received intermittent preventive treatment with Sulphadoxine-Pyrimethamine in the low malaria transmission region. BMC Pregnancy Childbirth. 2019;19:236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Desai M, Hill J, Fernandes S, Walker P, Pell C, Gutman J, Kayentao K, Gonzalez R, Webster J, Greenwood B, Cot M, Ter Kuile FO. Prevention of malaria in pregnancy. Lancet Infect Dis. 2018;18:e119–e32. [DOI] [PubMed] [Google Scholar]
- 19.Desai M, Gutman J, L’Lanziva A, Otieno K, Juma E, Kariuki S, Ouma P, Were V, Laserson K, Katana A, Williamson J, ter Kuile FO. Intermittent screening and treatment or intermittent preventive treatment with dihydroartemisinin-piperaquine versus intermittent preventive treatment with sulfadoxine-pyrimethamine for the control of malaria during pregnancy in western Kenya: an open-label, three-group, randomised controlled superiority trial. Lancet. 2015;386:2507–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kakuru A, Jagannathan P, Muhindo MK, Natureeba P, Awori P, Nakalembe M, Opira B, Olwoch P, Ategeka J, Nayebare P, Clark TD, Feeney ME, Charlebois ED, Rizzuto G, Muehlenbachs A, Havlir DV, Kamya MR, Dorsey G. Dihydroartemisinin-Piperaquine for the Prevention of Malaria in Pregnancy. N Engl J Med. 2016;374:928–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hamer DH, Mwanakasale V, Macleod WB, Chalwe V, Mukwamataba D, Champo D, Mwananyanda L, Chilengi R, Mubikayi L, Mulele CK, Mulenga M, Thea DM, Gill CJ. Two-dose versus monthly intermittent preventive treatment of malaria with sulfadoxine-pyrimethamine in HIV-seropositive pregnant Zambian women. J Infect Dis. 2007;196:1585–94. [DOI] [PubMed] [Google Scholar]
- 22.Newman RD, Parise ME, Slutsker L, Nahlen B, Steketee RW. Safety, efficacy and determinants of effectiveness of antimalarial drugs during pregnancy: implications for prevention programmes in Plasmodium falciparum-endemic sub-Saharan Africa. Trop Med Int Health. 2003;8:488–506. [DOI] [PubMed] [Google Scholar]
- 23.World Health Organization. World malaria report 2019. Geneva: World Health Organization. 2019. https://www.who.int/publications/i/item/9789241565721. [Google Scholar]
- 24.Radeva-Petrova D, Kayentao K, ter Kuile FO, Sinclair D, Garner P. Drugs for preventing malaria in pregnant women in endemic areas: any drug regimen versus placebo or no treatment. Cochrane Database Syst Rev. 2014:CD000169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kayentao K, Garner P, van Eijk AM, Naidoo I, Roper C, Mulokozi A, MacArthur JR, Luntamo M, Ashorn P, Doumbo OK, ter Kuile FO. Intermittent preventive therapy for malaria during pregnancy using 2 vs 3 or more doses of sulfadoxine-pyrimethamine and risk of low birth weight in Africa: systematic review and meta-analysis. JAMA. 2013;309:594–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Garner P, Gulmezoglu AM. Drugs for preventing malaria-related illness in pregnant women and death in the newborn. Cochrane Database Syst Rev. 2003:CD000169. [DOI] [PubMed] [Google Scholar]
- 27.Eisele TP, Larsen DA, Anglewicz PA, Keating J, Yukich J, Bennett A, Hutchinson P, Steketee RW. Malaria prevention in pregnancy, birthweight, and neonatal mortality: a meta-analysis of 32 national cross-sectional datasets in Africa. Lancet Infect Dis. 2012;12:942–9. [DOI] [PubMed] [Google Scholar]
- 28.Elphinstone RE, Weckman AM, McDonald CR, Tran V, Zhong K, Madanitsa M, Kalilani-Phiri L, Khairallah C, Taylor SM, Meshnick SR, Mwapasa V, Ter Kuile FO, Conroy AL, Kain KC. Early malaria infection, dysregulation of angiogenesis, metabolism and inflammation across pregnancy, and risk of preterm birth in Malawi: A cohort study. PLoS Med. 2019;16:e1002914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.World Health Organization. Malaria in HIV/AIDS patients. Geneva: World Health Organization. 2017. [Google Scholar]
- 30.Whitworth J. Malaria and HIV. HIV InSite Knowledge Base Chapter. San Francisco: University of California San Francisco. 2006. https://hivinsite.ucsf.edu/InSite?page=kb-05-04-04. [Google Scholar]
- 31.World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach. 2nd ed. Geneva: World Health Organization. 2016. https://www.who.int/publications/i/item/9789241549684. [PubMed] [Google Scholar]
- 32.Brentlinger PE, Behrens CB, Micek MA. Challenges in the concurrent management of malaria and HIV in pregnancy in sub-Saharan Africa. Lancet Infect Dis. 2006;6:100–11. [DOI] [PubMed] [Google Scholar]
- 33.Uneke CJ, Ogbonna A. Malaria and HIV co-infection in pregnancy in sub-Saharan Africa: impact of treatment using antimalarial and antiretroviral agents. Trans R Soc Trop Med Hyg. 2009;103:761–7. [DOI] [PubMed] [Google Scholar]
- 34.Harrington WE, Mutabingwa TK, Muehlenbachs A, Sorensen B, Bolla MC, Fried M, Duffy PE. Competitive facilitation of drug-resistant Plasmodium falciparum malaria parasites in pregnant women who receive preventive treatment. Proc Natl Acad Sci U S A. 2009;106:9027–32. [DOI] [PMC free article] [PubMed] [Google Scholar]