Increased peripartum mortality associated with maternal subclinical malaria in Mozambique

The study was conducted in the Maternity Services Unit of Tete Provincial Hospital (HPT). The province of Tete, with an area of 100,724 km², is located in central Mozambique and is bound by Zambia to the north and Zimbabwe to the west (Fig. 1). The average temperature ranges between 23.4º and 32.9º. It is the third most populated province in the country and has a population of 2,829,594 inhabitants, 1,442,880 (50,1%) of whom are women [14].

Between 1 March 2017 and 30 May 2019, 232 women were recruited during delivery. For the selection, the criterion was only to recruit women when the researchers were present at the delivery. Women who refused to participate in the study or when the data could not be completed in the questionnaire were excluded. A cross-sectional study was carried out involving completion of a structured questionnaire that included demographic data and information about the current pregnancy, delivery, and the newborn. Each interview lasted between 10 and 15 min. The data collection instruments were the pregnancy cards of the National Health Service of the Republic of Mozambique and the clinical delivery records.

The Sysmex® autoanalyzer and Cellpack® reagents were used to provide hemoglobin values. Due to the cost of the study, haematometry was only performed when indicated by the obstetrician. HIV serostatus was assessed using a rapid test (Determine, Abbot Laboratories, USA) and positive results were confirmed using the Uni-gold rapid test (TM HIV, Trinity Biotech, Ireland). The Multistix® 10 SG reagent strip was used to measure protein in urine.

In parallel to data collection, samples of peripheral blood and placenta were collected at delivery for the detection of P. falciparum by multiplex PCR. Within 30 min of delivery, attending staff collected placental blood via the incision method and prepared filter paper samples [15]. In brief, staff made a shallow incision into the maternal side of the placenta using sterile scissors and collected the accumulated blood in the intervillous space with the help of a syringe. Peripheral blood was also collected by venipuncture and stored on filter papers. Filter paper samples were collected on FTA™ Classic Card (GE Healthcare UK Limited, UK) and stored in individual sterile plastic envelopes with desiccant.

Parasite DNA was isolated from the blood samples on filter paper with Chelex, as previously described [16], and samples were tested in duplicate by species-specific nested PCR for P. falciparum [17]. Amplifications were performed in 25 µl reactions containing 1 µl of template DNA, 0.4 µg of each primer, 2.5 U of Taq DNA polymerase (Promega, Madison, Wis.), and each deoxynucleotide triphosphate at a concentration of 0.2 mM. All amplifications were performed in a conventional thermocycler (Applied Biosystems™), the template DNA was denatured at 94 ºC for 7 min, followed by 40 cycles of amplification (melt at 94ºC for 20 s, anneal/extend at 62 ºC for 20 s, and 72 ºC for 30 s) and final extension at 72 ºC for 10 min. For the second round of amplification, 1 µl of the PCR product from the initial amplification was used as the template in 25 µl final volume. Amplified DNA was run on 2.5% agarose gels containing ethidium bromide and analyzed under UV light. A positive PCR of placental blood was defined as the presence of a species-specific band of amplified DNA.

Subclinical malaria: characterized by low levels of parasitemia and absence of symptoms. The adolescent population includes individuals in the 10–19 years group.

The following variables were analysed:

Maternal and neonatal outcomes (dependent variables in multivariate models): delivery type (vaginal versus caesarean); the haemoglobin level (g/dL); gestational age at delivery (defined as preterm when occurring before week 37), newborn weight (in kilograms) and as low birth weight (LBW) = 1 if weight was less than 2500 g at birth; Apgar score < 7 at 1 and 5 min after birth [17]; neonatal resuscitation (defined as cardiopulmonary resuscitation maneuvers); and stillbirth (defined as fetal death occurring after 28 complete weeks of gestation).

Categorical variables are summarized as absolute numbers and percentages, and continuous variables as means and 95% confidence intervals. In univariate analyses, categorical data were analysed for associations with subclinical P. falciparum infection using the Chi-square test, while means of continuous variables were compared using Student's t-tests. Multivariate analyses were performed to assess associations between subclinical P. falciparum infection and each maternal and neonatal outcome, after controlling for maternal comorbidity (presence of E, PE and HIV) and maternal and pregnancy characteristics available for most women (i.e., mother’s age, pregnancy number and pregnancy type). Logistic regressions were performed for binary outcomes (i.e., cesarean delivery, pre-term birth, LBW, Apgar < 7 at 1 and 5 min, neonatal resuscitation, and still birth) and linear regressions for continuous outcomes (i.e., hemoglobin level and neonatal weight). For each model, odds ratios and 95% confidence intervals are presented. Statistical significance was set at p-value < 0.05. Data were analysed with Stata version 15 (StataCorp®).