Impact of oxidative stress in response to malarial infection during pregnancy: Complications, histological changes, and pregnancy outcomes.

Background and
Objectives: Pregnancy malaria is a major underestimated global public health problem. To understand the involvement of oxidative stress (OS) in the pathophysiology of placental malaria, OS biomarkers, malondialdehyde (MDA), uric acid (UA), and superoxide dismutase (SOD) levels were analyzed and correlated to placental histopathological changes and pregnancy outcomes.
Methods: A hospital-based study was conducted in Mangaluru, Karnataka, India, to analyze the changes in hematological parameters and the serum OS biomarker levels. Histological analysis of placenta, associated complications, and pregnancy outcomes were compared using Kruskal-Wallis test, and pairwise comparison between two groups was made by Mann-Whitney U-test. Correlations were calculated by Pearson's and Spearman's rank correlations.
Results: Among 105 pregnant women, 34 were healthy controls and the infected group comprised of Plasmodium Vivax (Pv) (n = 48), Plasmodium falciparum (Pf) (n = 13), and mixed (n = 10) malaria infections. Of 71 infected cases, 67.6% had mild malaria, whereas 32.4% had severe malaria. The white blood cell and C-reactive protein levels were found to increase, whereas hemoglobin, red blood cell, and platelet levels decreased during both types of malarial infections. The MDA and UA values increased and SOD levels decreased particularly during severe Pf infections. Histological changes such as syncytial knots, syncytial ruptures, and fibrinoid necrosis were observed particularly during Pf infections and leukocyte infiltration was observed in Pv malaria.
Conclusion: Evaluation of MDA, UA, and SOD levels can serve as an indicator of OS during pregnancy malaria. The OS during pregnancy may lead to complications such as severe anemia, pulmonary edema, intra uterine growth retardation, premature delivery, and low birth weight, not only during Pf but also in Pv malaria. It is important to create awareness among rural and immigrant population residing in Mangaluru and its surroundings about required preventive measures and free government-supported antenatal care services. Copyright:

INTRODUCTION

Malaria is one of the common and deadliest infectious diseases in the world. According to World Health Organization (WHO) report, during 2017, malarial infections resulted in an estimated over 219 million infections and ~435,000 deaths worldwide. The majority of malarial burden is in Africa (92%), followed by Southeast Asia (5%) and Eastern Mediterranean region (2%).[1] In 2017, about 9.6 million cases and ~16.7 thousand malarial deaths were estimated in India.[1] In India, the state of Karnataka, located in the south-western region of India, is a significant contributor of malarial infections. Overall, among thirty districts in Karnataka state, the majority of malarial burden is reported from only two districts, namely, Dakshina Kannada and Udupi.[2] During 2017, among a total of 8075 reported malaria cases in Mangaluru, the administrative capital of Dakshina Kannada district, 6452 cases were due to Plasmodium vivax (Pv) and 1623 cases were due to Plasmodium falciparum (Pf) infections.

Malaria in pregnancy (MIP) is an underestimated public health problem in endemic countries. Globally, each year, MIP results in about 20% of stillbirths and 11% of new born deaths in sub-Saharan Africa and 10,000 maternal deaths.[3] Serious complications arise during MIP due to sequestration of Pf parasite-infected red blood cells (IRBCs) in the placenta mediated by placental chondroitin sulfate A. Parasite sequestration leads to leukocyte infiltration and production of inflammatory cytokines and chemokines.[45] IRBC accumulation in the placenta also leads to oxidative stress (OS). Altogether, these events lead to significant maternal and fetal morbidity and mortality.[6] During MIP, complications observed are anemia; pulmonary edema; hypoglycemia; cerebral malaria; puerperal sepsis; and death and complications in fetus such as abortion, still birth, intrauterine growth retardation (IUGR), premature delivery, and low birth weight (LBW).[6]

The OS arises due to imbalance between reactive oxygen species (ROS) and antioxidant levels in the body that damage the cells, tissues, and organs.[78] In response to Plasmodium infection, the innate immune cells such as neutrophils and macrophages produce increased levels of ROS, such as anion superoxide radical (O2−•), hydrogen peroxide (H2O2), hypochlorous acid (HOCl−), hydroxyl radical (•OH), and reactive nitrogen species (RNS) such as nitric oxide radical (NO•), nitrogen dioxide radical (NO2•), and anion peroxynitrite (ONOO−).[910] In vivo studies with Plasmodium Berghei-infected pregnant mice showed increase in lipid peroxidation markers such as malondialdehyde (MDA) and decreased antioxidants such as reduced glutathione (GSH) and superoxide dismutase (SOD) levels, suggesting that accentuated OS plays a major role in increased severity during MIP.[1112]

MDA is a reactive aldehyde, formed by the degradation of polyunsaturated lipids by ROS, and is considered to be mutagenic.[1112] MDA and other lipid peroxidation products are also recognized by immune receptors, triggering inflammatory responses, thus playing an important role in the immunopathology of malaria. Measurement of MDA levels is most widely used as a biomarker of OS in several disease conditions.[1314] Further, uric acid (UA) is an important parasite-derived factor that contributes to malaria pathogenesis.[1415] Studies have shown that UA increases the production of cytokines such as tumor necrosis factor-alpha, interleukin (IL)-1β, IL-6, and IL-10 from innate immune cells.[1617] The antioxidants counteract to limit the damage caused by increased ROS levels.[1218] Lower levels of antioxidants such as Vitamins A, C, and E; β-carotene, the reduced form of GSH, catalase; and SOD in malaria-infected patients result in augmented levels of oxidative products, resulting in OS. Recent studies also suggests SOD as a surrogate marker of severe vivax malaria.[111219]

Although several pathophysiological mechanisms have been suggested that contribute to MIP, there is very little evidence implicating OS to poor pregnancy outcomes. In this study, to gain better understanding of MIP, we analyzed the levels of OS biomarkers, including MDA, UA, and SOD in pregnant women suffering from Pf, Pv, and mixed infections. We also studied the possible contribution of OS to malarial anemia, placental histopathological changes, and pregnancy outcomes.

METHODS

Study site, design, and population

A prospective hospital-based study was conducted during 2014–2017 at Government Lady Goschen hospital, K. S. Hegde Medical Hospital, Mangalore Nursing Home and Spoorthi Polyclinic in Mangaluru, Karnataka, India. A total of 105 pregnant women (18–40 years) seeking medical attention were included in the study upon obtaining signed informed consent. The study protocol was approved by the Ethical Committee of Kuvempu University, Shivamogga, Karnataka; the Central Ethics Committee of NITTE University, Mangaluru, Karnataka; and the Institutional Review Board of Pennsylvania State University College of Medicine, Hershey, PA, USA. A structured questionnaire was used by the attending physician and staff to collect information on age, sociodemographic profile, economic status, education level, occupation, prior pregnancies, history of previous infections, and clinical presentation of patients. The inclusion criterion of the study was pregnant woman with Pv, Pf, and mixed (Pv and Pf) infections with mild and severe malaria (SM). Exclusion criteria were pregnant women using any prophylaxis prior to diagnosis; those with any other known diseases such as diabetes, autoimmune diseases, stressful life event, and poor psychosocial status or using any antidepressant medications; and participants testing positive for HIV, typhoid, dengue, and hepatitis B and C infections.

Malarial diagnosis and sample collection

Pregnant women experiencing common symptoms such as fever, headache, or chills were examined for malarial infections by rapid diagnostic test kits (RDT kit, SD Bioline, India) and by microscopic examination of blood smears. Thick and thin blood slides were prepared, stained with 4% Giemsa stain (Sigma Aldrich, St. Louis, MO 63118, United States), and were microscopically observed for the presence of Plasmodium parasite and the species type. The parasite densities were calculated by counting the number of parasites and expressed as parasites/μL of blood (number of parasites counted/number of white blood cells [WBCs] counted × total number of WBCs/μL of blood) or (number of parasites counted/number of RBCs counted × total number of RBCs/μL of blood). Percentage parasitemia was determined as (number of parasites/μL of blood/number of RBCs/μL of blood) ×100. The infected patients were treated by physicians as per National Vector Borne Disease Control Programme (NVBDCP) recommendations. The healthy pregnant women attending antenatal care (ANC) of Lady Goschen Hospital for general checkups were included as healthy controls (HC).

Venous blood (~1 mL) was collected into sterile vacutainers to determine the levels of hemoglobin (Hb), RBCs, hematocrit, mean cell volume, mean cell hemoglobin (MCH), and MCH concentration using an automated hematology analyzer (Mind Ray-Biomedical, Shenzhen, China). The serum (clot activator tubes) and plasma (heparin coated tubes) samples were centrifuged, and the supernatant was collected and stored at −70°C. Comorbid infections such as HIV, typhoid, dengue, and hepatitis B and C were screened using commercially available ELISA kits (J. Mitra and Co, New Delhi, India).

Classification of study participants and case definitions

The study participants were classified as follows:

Healthy controls

The healthy pregnant women seeking ANC services at the hospital and testing negative for malaria by peripheral blood smear and RDT kits and testing negative for any other comorbid conditions, with Hb levels >11 g/dL, were used for comparison of data with various infected groups.

As per the recommendations of the WHO and NVBDCP, the infected study participants were classified into mild malaria (MM) and SM.

Mild malaria

Pregnant women patients, with low-grade clinical symptoms, such as fever, headache, or chills and positive for the presence of Plasmodium parasite by peripheral blood smear and RDT kit.

Severe malaria

Infected pregnant women with severe anemia (SA) (i.e., Hb levels <5 g/dL), acute renal failure (serum creatinine ≥3 mg/dL), jaundice/icterus (serum bilirubin ≥3 mg/dL), metabolic acidosis (plasma bicarbonate <15 mmol/l), spontaneous bleeding, hypoglycemia (plasma glucose <40 mg/dL), hyperparasitemia (≥10% parasitemia), acute respiratory distress syndrome and pulmonary edema, LBW (a birth weight of <2500 g), prematurity (gestational age of <37 weeks as assessed by Ballard examination,[20]) and stillbirth (death of a fetus before delivery at 28 weeks of gestation).

Based on varying Hb levels(g/dL), the pregnant women enrolled in the study were grouped into three groups:(i) non-anemic(NA): study participants with Hb ≥11 g/dL; (ii) mild-to-moderate anemia (MMdA): Hb ranged between 10.9 and 5/dL; and (iii) SA: Hb levels of Gestational age was established by obstetric dating by last menstrual period. Based on the trimester, the pregnant women were classified into three groups: (i) first trimester was defined as conception through 13 weeks; (ii) second trimester as 14–26 weeks; and (iii) third trimester as 27–40 weeks.[21]

Depending on the number of previous pregnancies, the pregnant women were classified into three groups (i) primigravide (first-time pregnant), (ii) second gravidae (second-time pregnant), and (iii) multigravide (>2 previous pregnancies).

IUGR was defined as a binary outcome of <10th percentile of fetal weight for attained gestational age, and LBW was defined as babies having weight of <2500 g.[22]

Measurement of oxidative stress and antioxidant biomarkers

The serum MDA levels were quantified by thiobarbituric acid-reactive substances method.[18] Briefly, serum (100 μL) MDA was measured by incubating with 0.6% thiobarbit UA (150 μL), 15% trichloroacetic acid (150 μL), and 0.2N HCl (100 μL) at 90°C for 10 min; allowed to cool at room temperature; and centrifuged for 15 min at 1200 × g. The supernatant was collected and the absorbance was measured at 535 nm. The concentration of MDA (nmoles/100 mL) was calculated by: OD of sample × total reaction volume/nanomolar extinction coefficient × sample volume.

Measurement of uric acid levels

The UA levels were quantified by commercially available kit (Agape, Agape Hills, Pattimattom PO, Kochi, Kerala, India). Briefly, serum (25 μL) and reagent mixture (1 mL) were mixed, incubated for 5 min at 37°C, and the absorbance was measured at 546 nm.

Measurement of superoxide dismutase levels

The plasma SOD levels were measured using commercially available kit (Cayman chemical, Ann Arbor, MI, USA). Briefly, radical detector and standards/samples were added to 96-well plates (Thermo Fisher Scientific Ltd., Waltham, MA, USA). The reaction was initiated by the addition of xanthine oxidase and incubated at room temperature for 30 min, and the absorbance was read at 450 nm (iMark Micro plate Reader, Bio-Rad, CA, USA).

Histological evaluation of placental samples

Pieces of placenta (~2 cm3) were aseptically collected after delivery, approximately between the region of umbilical cord insertion and the placental edge, and stored in 10% neutral buffered formalin at 4°C until processed. The samples were paraffin-embedded, 5-μm sections on glass slides which were processed and stained with hematoxylin-eosin (H and E) stain. The blinded stained slides were evaluated by two histologists. Differences in the observed results were re-examined by a third histopathologist for confirmation. A Labomed LX 300 microscope IVu5100 camera, [Los Angeles, CA 90034, U.S.A] was used to capture the images of slides. The parameters were assessed using Image J software (Image J 1.46c Wayne Rasband, NIH, USA,http://imagej.nih.gov/ij).

Statistical analysis

Statistical analysis was performed using GraphPad Prism (GraphPad, Inc., San Diego CA, USA). Quantitative variables were presented as number (percentages) and mean ± standard deviation. The comparison of nonparametric data between various groups was performed using the Kruskal–Wallis test, and significance between any of the two groups was determined by the Mann–Whitney U-test with a 95% confidence interval. Correlations between two continuous variables were analyzed by Spearman's rank correlation. P < 0.05 was considered statistically significant.

RESULTS

Characteristics of the study participants

A total of 105 pregnant women were enrolled after obtaining informed consent. Upon malarial diagnosis, 71 pregnant women were found to be infected. Among the various infecting species, Pv (67.6%, n = 48) was the most prevalent, followed by Pf (18.3%, n = 13) and mixed (Pv + Pf) (14.1%, n = 10) infections. For comparison, a total of 34 healthy pregnant women visiting hospital for ANC were enrolled as HC. The average age of the study participants was 25.7 years (range: 18–40 years). Major proportion of infected pregnant women were immigrants (n = 55, 77.5%) from the northern and north-eastern parts of India. Most infected pregnant women (n = 57, 80.3%) belonged to low-income status, whereas 14 (19.7%) women had a middle class socioeconomic status. The average birth weight of babies delivered by the infected group (2.48 kg ± 0.26) was found to be statistically significantly lower than the babies delivered by HC group (2.64 kg ± 0.11, P = 0.004) [Table 1].

Table 1

Characteristics of the study participants

Parameter	HC	Plasmodium vivax	Plasmodium falciparum	Mixed	Overall (infected)
Number of study participants	34	48 (67.6)	13 (18.3)	10 (14.1)	71 (100)
Mean age in years (range)	27.0 (17-40)	24.5 (18-33)	27.3 (20-36)	25.1 (19-34)	25.7 (17-40)
Residence, n (%)
Natives	25	11 (68.8)	3 (18.8)	2 (12.5)	16 (22.5)
Immigrants	9	37 (67.3)	10 (18.2)	8 (14.5)	55 (77.5)
Socioeconomic status, n (%)
Low income	8	41 (71.9)	8 (14.0)	8 (14)	57 (80.3)
Middle income	26	7 (50)	5 (35.7)	2 (14.3)	14 (19.7)
High income	0	0	0	0	0
Delivery outcome
Mean birth weight	2.64±0.11	2.5±0.31	2.5±0.20	2.4±0.32	2.48±0.26
Normal	34	30 (66.7)	9 (20)	6 (13.3)	45 (63.4)
LBW	0	18 (69.2)	4 (15.4)	4 (15.4)	26 (36.6)

Data expressed as number of study participants (percentages with respect to infected groups). LBW: Low birth weight, HC: Healthy controls, Pv: Plasmodium vivax, Pf: Plasmodium falciparum

Changes in blood cell parameters

The changes in blood cell parameters of infected pregnant women and HC groups were analyzed and compared. The percent parasitemia was significantly higher in Pf patients compared with Pv and mixed (Pv + Pf) infection patients [Table 2]. The WBC levels, in comparison to HC, were considerably higher irrespective of the infecting species. The hemoglobin levels, in comparison to HC, were found to be decreased during infection, especially during Pf infection. The RBC levels were also decreased, particularly during Pf infections. Further, regardless of the infecting species, platelet levels were also lower to a great extent. The level of C-reactive protein (CRP) (an acute-phase protein) was found to be increased in all the infected cases, mostly in women with Pf infection [Table 2].

Table 2

Effect of malarial infection on hematological parameters and oxidative stress biomarkers across various infecting species

Variable	HC	Pv	Pf	Mixed	P*(between groups)
					Pv versus Pf	Pv versus mixed	Pf versus mixed
Parasitemia (%)	0.0±0.0	0.2±0.2	0.6±0.4	0.4±0.3	0.0004	0.1708	0.0539
WBC (×103/µL)	5.6±0.7	7.3±2.5	7.9±4.0	6.8±2.0	0.9896	0.6369	0.7725
Hb (g/dL)	12.2±0.7	8.0±1.9	7.2±3.3	8.3±2.7	0.018	0.9313	0.0368
HCT (%)	40.0±4.2	29.3±8.6	28.0±2.8	26.1±4.3	0.6602	0.3516	0.2767
Total RBC (×103/µL)	4.7±0.5	3.9±3.2	2.4±0.5	3.0±0.5	<0.0001	0.0356	0.0127
Total platelet count (×104/µL)	2.1±0.5	1.5±0.6	1.3±0.7	1.4±0.7	0.3046	0.5663	0.7055
CRP	3.1±2.1	40.2±45.5	86.6±36.9	48.1±25.9	<0.0001	0.2146	0.0044
MDA (IU/L)	0.8±0.5	2.2±1.4	3.4±2.9	1.7±1.2	0.0354	0.2822	0.0215
Uric acid (mg/dL)	2.5±1.2	6.5±2.7	7.9±4.0	5.4±3.3	0.0803	0.4434	0.0342
SOD (U/mL)	45.0±10.5	30.6±10.4	28.0±10.9	33.1±10.3	0.0283	0.5271	0.0481

Data represented as mean±SD, comparison of two groups by Mann–Whitney U-test and Kruskal–Wallis test for comparison between three groups. *P<0.05 was considered to be significant and are highlighted in bold. WBC: White blood cell, Hb: Hemoglobin, HCT: Hematocrit, RBC: Red blood cell, CRP: C-reactive protein, MDA: Malondialdehyde, UA: Uric acid, SOD: Superoxide dismutase, SD: Standard deviation, Pv: Plasmodium vivax, Pf: Plasmodium falciparum

The circulatory levels of various OS biomarkers were analyzed and compared. In comparison to HC, a significant increase in MDA levels was found in the infected group. Among the varying infecting species, women with Pf infections had higher circulatory MDA levels. The UA levels, in comparison to HC, were also elevated during malaria. Upon comparison between the various infecting species, women with Pf infection had increased UA levels. The SOD levels decreased during malarial infections, especially during Pf infection [Table 2]. Further, there was no statistically significant association between the status of malarial infection, age, blood cell parameters, MDA, UA, and SOD levels analyzed (P > 0.05).

Oxidative stress among mild and severe malarial patients

Of 71 infected women, 48 (67.6%) had MM, whereas 23 (32.4%) suffered from SM infections [Table 3]. In comparison to pregnant women with MM, percent parasitemia was significantly higher in SM patients, especially during Pf infections. Irrespective of the infecting species, in comparison to MM patients, the levels of MDA were found to be significantly higher in SM groups. Upon comparison between the SM groups, women with Pf infections had significantly greater levels of MDA. The serum UA levels, in comparison to MM patients, were found to be significantly higher in SM groups during Pv and Pf infections. The UA levels between SM groups were found to be particularly higher in women suffering from severe Pf infections. Upon comparison with MM groups, significantly lower SOD levels were observed in SM patients across all the infecting species. Among the SM patients, the SOD levels were considerably decreased in Pf and mixed infection groups [Table 3 and Figure 1].

Table 3

Malondialdehyde, uric acid, and superoxide dismutase levels in mild and severe malarial patients during Plasmodium vivax, Plasmodium falciparum, and mixed infections

Parameter	Pv (n=48)		Pf (n=13)		Mixed (n=10)		P*(between SM groups)
	MM	SM	MM	SM	MM	SM	Pv versus Pf	Pv versus mixed	Pf versus mixed
Number of study participants	35 (72.9)	13 (56.5)	7 (14.6)	6 (26.1)	6 (12.5)	4 (17.4)	0.031	0.254	0.232
Parasitemia (%)	0.2±0.1	0.5±0.4	0.4±0.2	0.9±0.4	0.4±0.2	0.5±0.3	0.0285	0.1697	0.0376
MDA	1.9±1.0	3.5±2.1	1.5±1.1	6.9±0.8	1.3±1.0	3.9±1.0	0.0015	0.7209	0.0095
UA	6.2±2.0	7.8±1.6	5.9±3.4	11.7±2.0	3.7±2.9	8.0±2.0	<0.0001	0.5008	0.0571
SOD	29.9±9.4	19.0±6.3	32.4±10.3	13.3±4.2	38.9±7.8	16.9±5.6	0.045	0.5008	0.281

Data represented as mean±SD; comparison of two groups by Mann–Whitney U-test and Kruskal–Wallis test for comparison between three groups. *P<0.05 was considered to be significant and are highlighted in bold. MDA: Malondialdehyde, UA: Uric acid, SOD: Superoxide dismutase, SD: Standard deviation, Pv: Plasmodium vivax, Pf: Plasmodium falciparum

Figure 1

Levels of oxidative stress biomarkers in pregnant women with mild and severe malaria. The levels of malondialdehyde (a), uric acid, (b) and superoxide dismutase (c) in mild malaria (gray boxes) and severe malaria (line and chequered gray boxes) during. vivax, Plasmodium falciparum, and mixed infections. The data represented as mean ± standard deviation were analyzed by one-way nonparametric Kruskal–Wallis test for multiple comparisons, and Mann–Whitney U test for comparison between two groups; ns indicates nonsignificant, *, **, *** and **** indicate significance at P< 0.05, P <0.01, P< 0.001, and < 0.0001, respectively

Oxidative stress during varying intensity of malarial anemia

In this study, 10 (14.08%) women were NA, while 54 (76.05%) had mild-to-moderate levels of anemia (MMdA) and 7 (9.85%) showed SA [Table 4]. Among the SA groups, Hb levels were significantly decreased during Pf infection. The parasitemia (%) levels among SA women were considerably higher during Pf infections. The parasitemia levels were found to be greatly increased with increasing anemic intensity, as a significant negative correlation was observed between increase in parasitic burden and decreased Hb levels during Pv (r = −0.119, P = 0.0021), Pf (r = −0.0152, P = 0.0028), and mixed (r = −0.0187, P = 0.0032) infections. The MDA levels, irrespective of infecting species, were found to be significantly increased with an increase in anemic intensity (P ≤ 0.001). Among the various SA groups, women with Pf infections resulted in significantly higher circulatory MDA levels [Table 4 and Figure 2]. A significant negative correlation was observed between increased MDA and decreased Hb levels during Pv (r = −0.0232, P = 0.0051) and Pf (r = −0.0151, P = 0.0023) infections. A statistically significant positive correlation was observed between the MDA levels and increase in parasitic density during Pv (r = 0.121, P = 0.0042) and Pf (r = 0.168, P = 0.0018) infections. The UA levels, irrespective of the type of infecting species, were also found to be increased with increasing anemic severity. A statistically significant positive correlation was observed between the UA levels and increase in parasitic density during Pv (r = 0.152, P = 0.0124) and Pf (r = 0.211, P = 0.002) infection. An inverse relationship was observed between increased UA levels and decreased Hb levels during Pv (r = −0.157, P = 0.0027) and Pf (r = −0.167, P = 0.0031) infection.

Table 4

Oxidative stress biomarker levels across increasing degree of anemia across various infecting species

Parameter	Pv			Pf			Mixed			P value (between SA groups)
	NA	MMdA	SA	NA	MMdA	SA	NA	MMdA	SA	Pv versus Pf	Pv versus mixed	Pf versus mixed
n (%)	3 (6.3)	42 (87.5)	3 (30.0)	3 (23.1)	7 (53.8)	3 (23.1.0)	4.0 (40)	5.0 (50)	1.0 (10.0)
Hb	12.0±0.8	8.1±1.3	4.5±0.4	12.0±0.2	7.5±1.6	3.2±0.2	11.9±0.4	7.1±0.8	4.2±0.1	0.0571	0.5721	0.043
RBC	3.7±0.7	4.0±3.4	3.5±0.6	3.1±0.3	2.3±0.4	2.1±0.1	3.4±0.2	2.8±0.6	2.7±0.4	0.0286	0.0212	0.121
Parasitemia (%)	0.1±0.1	0.2±0.1	0.8±0.4	0.2±0.1	0.6±0.2	1.4±0.1	0.2±0.1	0.4±0.2	0.8±0.1	0.0271	> 0.9999	0.012
MDA (IU/L)	0.6±0.2	1.8±0.7	5.6±0.4	0.2±0.1	2.0±0.7	6.7±2.0	0.5±0.1	1.9±0.6	4.3±1.2	0.0203	0.1429	0.052
UA (mg/dL)	1.2±0.1	6.4±2.1	10.7±1.2	1.1±0.1	7.9±0.9	11.3±1.9	1.1±0.1	6.6±0.7	10.9±0.1	0.2662	0.7679	0.5357
SOD (U/mL)	43.9±4.4	30.9±10.0	20.1±4.3	46.8±1.7	27.9±4.6	18.6±3.2	45.1±1.0	30.1±5.2	15.3±1.1	0.329	0.25	0.1667

Data represented as mean±SD; comparison of three groups by Kruskal–Wallis test; comparison of two groups by Mann–Whitney U-test; values of P<0.05 was considered to be significant and are highlighted in bold. Hb: Hemoglobin, RBC: Red blood cells, MDA: Malondialdehyde, UA: Uric acid, SOD: Superoxide dismutase, SD: Standard deviation, Pv: Plasmodium vivax, Pf: Plasmodium falciparum

Figure 2

Levels of oxidative stress biomarkers among pregnant women with varying levels of anemia. The levels of malondialdehyde (a-c), uric acid, (d-f) and Superoxide dismutase (g-i) among non-anemic (open boxes) and mild-to-moderate anemia (closed gray boxes) and severe anemia (closed black boxes) patients during Plasmodium vivax(a, d and g) and Plasmodium falciparum(b, e, and h), and mixed (c, f, and i) infections. A comparison of malondialdehyde (J), uric acid (k) and superoxide dismutase (l) among severe anemic (SA) patients during Plasmodium vivax(Closed horizontal lines), Plasmodium falciparum(Closed vertical lines) and mixed (Closed chequered box) Data represented as mean ± standard deviation and analyzed by one way nonparametric Kruskal–Wallis test for multiple comparisons, and Mann–Whitney U test for comparison between two groups; ns indicates nonsignificant, *, **, ***and **** indicate significance at P< 0.05, P < 0.01, P< 0.001, and P< 0.0001, respectively

Further, in comparison to NA groups, the SOD levels were found to be decreased in MMdA and SA groups (P ≤ 0.001) across Pv, Pf, and mixed infections [Table 4 and Figure 2]. However, between the SA groups, the SOD levels showed no significant differences. An inverse relationship was observed between increased parasitemia and decreased SOD levels especially during Pf (r = −0.156, P = 0.0036) infection. An opposite trend was observed between increased MDA and SOD levels during Pv (r = −0.146, P = 0.0042) and Pf (r = −0.233, P = 0.0027) infection. An inverse trend was also observed between increased UA and SOD levels during Pf (r = −0.159, P = 0.0038) infection. A significant positive correlation was observed between the decreased SOD levels and decreased Hb levels during Pv (r = 0.199, P = 0.0024) and Pf (r = 0.56, P = 0.004) infections. UA levels were found to be increased with an increase in parasitic density during Pf (r = 0.252, P = 0.0030) infection.

Level of oxidative stress biomarkers across various trimester and gravidity

To assess the possible influence of OS during pregnancy, the circulatory levels of MDA, UA, and SOD were analyzed based on trimester and gravidity status across varying infecting species.

Among 71 infected, 22 (31.0%) women were in their first trimester, 32 (45.1%) were in second trimester, and 17 (23.9%) were in third trimester of pregnancy. The Hb levels, in comparison to women in their 1sttrimester, were found to be significantly lower during the 3rd trimester across all the infecting species. However, the parasitemia levels did not differ considerably across different trimesters. The serum MDA levels, in comparison to levels during the 1sttrimester, were significantly increased, especially during Pf infection. A similar trend of higher circulatory UA levels, irrespective of the infecting species, was observed. The SOD levels were found to be lower during the last trimester, especially in women with Pf infection [Table 5 and Figure 3].

Table 5

Oxidative stress biomarker levels in pregnant women patients during various trimesters across different infecting species

Parameter	Pv			Pf			Mixed infection			P (comparison between III trimester infected species)
	Trimester			Trimester			Trimester
	I	II	III	I	II	III	I	II	III	Pv versus Pf	Pv versus mixed	Pf versus mixed
	n (%)	18 (37.5)	17 (35.4)	13 (27.1)	2 (15.4)	9 (69.2)	2 (15.4)	2 (20.0)	6 (60.0)	2 (20.0)	0.412	0.417	>0.9999
	Hb	8.7±2.1	8.1±1.5	6.4±1.5	11.0±4.0	7.5±2.1	3.9±1.0	8.6±5.9	7.9±1.8	6.3±0.7	0.0219	0.4667	0.0467
Parasitemia (%)	0.3±0.3	0.2±0.1	0.4±0.2	0.4±0.20	0.7±0.4	0.6±0.6	0.5±0.5	0.4±0.2	0.4±0.2	0.0143	0.9333	0.02
MDA (IU/L)	1.8±0.7	1.5±0.7	3.7±1.6	0.6±0.8	1.5±0.7	7.0±1.5	1.0±0.8	1.8±0.8	2.8±2.0	0.0044	0.4762	0.0452
UA (mg/dL)	6.0±2.3	6.6±1.6	9.6±3.5	2.4±2.3	7.1±2.6	10.5±1.7	4.6±0.8	5.7±2.4	8.9±2.9	0.0451	0.9333	0.8
SOD (U/mL)	30.3±7.9	33.0±11.0	21.9±8.4	44.6±5.0	30.6±9.9	14.0±4.3	45.4±3.2	33.3±7.2	20.3±7.0	0.0183	0.9333	0.04

Data represented as mean±SD; comparison of three groups by Kruskal–Wallis test; comparison of two groups by Mann–Whitney U-test; values of P<0.05 was considered to be significant and are highlighted in bold. Hb: Hemoglobin, MDA: Malondialdehyde, UA: Uric acid, SOD: Superoxide dismutase, SD: Standard deviation, Pv: Plasmodium vivax, Pf: Plasmodium falciparum

Figure 3

Levels of oxidative stress biomarkers in patients across different trimesters. The levels of malondialdehyde (a-c), uric acid, (d-f) and superoxide dismutase (g-i) during 1st trimester (open boxes), 2nd trimester (gray boxes), and 3rd trimester (black boxes) during Plasmodium vivax(a, d and g), Plasmodium falciparum (b, e and h), and mixed (c, f and i) infections. Data represented as mean ± standard deviation and analyzed by one-way nonparametric Kruskal–Wallis test for multiple comparisons, and Mann–Whitney U test for comparison between two groups; ns indicates nonsignificant, *, ** and *** indicate significance at P< 0.05, P< 0.01, and P< 0.001, respectively

The changes in OS biomarkers were assessed based on gravidity of pregnancy. In this study, 32 (45.1%) were primigravidae, the more susceptible to malarial infections, followed by second-gravidae (25, 35.2%) and multi-gravidae women (14, 19.7%) [Table 6 and Figure 4]. In comparison to second-gravidae and multi-gravidae women, the Hb levels were found to be significantly decreased among primigravidae. The parasitemia was found to be significantly higher in primigravidae compared to secondigravidae and multigravidae women. The MDA and UA levels were found to be significantly increased in women among primigravidae groups, whereas the SOD levels were found to be decreased, especially during Pf infections.

Table 6

Oxidative stress biomarker levels across varying gravidity during Plasmodium vivax, Plasmodium falciparum, and mixed infections

Parameter	Pv			Pf			Mixed			P (comparison between primigravidae of infected species)
	Primi-gravidae	Second-gravidae	Multi-gravidae	Primi-gravidae	Second gravidae	Multi-gravidae	Primi-gravidae	Second-gravidae	Multi-gravidae	Pv versus Pf	Pv versus mixed	Pf versus mixed
n (%)	23 (47.9)	17 (35.4)	8 (16.7)	4 (30.8)	5 (38.5)	4 (30.8)	5 (50.0)	3 (30.0)	2 (20.0)	>0.999	0.747	0.924
Hb	6.5±1.1	8.6±2.2	7.9±1.4	4.2±0.9	9.7±2.4	7.1±3.5	5.4±0.9	10.3±2.0	9.9±4.0	0.0228	0.7111	0.0333
Parasitemia (%)	0.4±0.2	0.2±0.2	0.3±0.2	0.8±0.4	0.7±0.5	0.7±0.5	0.5±0.2	0.2±0.0	0.3±0.2	0.254	0.547	0.784
MDA (IU/L)	2.9±1.8	1.8±0.8	1.5±0.8	5.4±1.8	2.0±1.1	1.7±1.0	3.7±1.3	0.8±0.6	1.5±1.6	0.0121	0.8081	0.0471
UA (mg/dL)	7.2±2.4	7.0±2.7	4.9±1.6	11.2±2.0	7.4±1.7	5.9±3.2	7.0±1.1	4.6±0.5	3.5±3.3	0.0121	0.8081	0.0267
SOD (U/mL)	25.0±10.8	33.8±8.0	35.5±5.9	18.8±5.4	35.6±8.8	35.9±12.2	21.4±5.7	39.9±6.8	42.1±7.7	0.0213	0.4	0.0333

Data represented as mean±SD; comparison of three groups by Kruskal–Wallis test; comparison of two groups by Mann–Whitney U-test; values of P<0.05 was considered to be significant and are highlighted in bold. Hb: Hemoglobin, MDA: Malondialdehyde, UA: Uric acid, SOD: Superoxide dismutase, SD: Standard deviation, Pv: Plasmodium vivax, Pf: Plasmodium falciparum

Figure 4

Levels of oxidative stress biomarkers in patients across gravidity. The levels of malondialdehyde (a-c), uric acid, (d-f) and superoxide dismutase (g-i) among primigravidae (open boxes), second gravidae (gray boxes), and multi gravidae (black boxes) during Plasmodium vivax(a, d and g), and Plasmodium falciparum(b, e and h) and mixed (c, f and i) infections. Data represented as mean ± standard deviation and analyzed by one-way nonparametric Kruskal–Wallis test for multiple comparisons, and Mann–Whitney U test for comparison between two groups; ns indicates nonsignificant, *, **, *** and **** indicate significance at P< 0.05, P< 0.01, P< 0.001, and P< 0.0001, respectively

Histopathological changes observed among the study participants

The H and E-stained placental tissue sections were examined for any morphological and pathological changes and for immune cell infiltration during Pv, Pf, and mixed infections. Upon comparison with HC, malarial infections caused significant changes in the placenta. Among various infecting species, Pf infections were found to have caused significant histological changes such as syncytial knots, syncytial ruptures, and fibrinoid necrosis. Irrespective of the infecting species, in comparison to HC, the leukocyte infiltration was observed to be significantly increased in the infected groups. Within various infecting species, the leukocyte infiltration was statistically significantly higher in Pv infections [P < 0.05, Figure 5].

Figure 5

Placental histological changes observed in pregnant women suffering from severe malaria during Plasmodium vivax, Plasmodium falciparum, and mixed infections. The number of syncytial knots (a), syncytial ruptures (b), fibrinoid necrosis, (c) and infiltration of intervillous leukocytes (d) in HC (open boxes), Plasmodium vivax(gray closed boxes), Plasmodium falciparum(black gray closed boxes), and mixed (black closed boxes) malaria infections. Data represented as mean ± standard deviation and analyzed by one-way nonparametric Kruskal–Wallis test for multiple comparisons, and Mann–Whitney U test for comparison between two groups; ns indicates nonsignificant, *, **, *** and **** indicate significance at P< 0.05, P< 0.01, P< 0.001, and P< 0.0001, respectively

DISCUSSION

It is evident from these results that, increased OS plays an important role in MIP not only during Pf but also during Pv infections. Even though Pv and Pf infections are prevalent in Mangaluru city and its surrounding areas, Pv was found to be the major infecting species.[21923] The majority of infected pregnant women were immigrants belonging to a low socioeconomic status.[21924] These immigrants from northern and north-eastern states such as Bihar, Assam, and Odisha were employed as laborers in building and road construction sector and hotels. They were found to be living near these construction sites and due to increased mosquito breeding at these sites, this population was at increased risk of infections. Most of these pregnant women reported not to have regularly availed any ANC. As these pregnant women belonged to a low socioeconomic status and lacked basic education, their families were not aware of government-sponsored ANC services which can be availed free of charge in government primary health-care centers or hospitals.[2] Upon malarial symptoms such as fever, headache, and chills, these infected pregnant women were primarily dependent on local birth attendants (LBAs) for advice, who were prescribing medications such as paracetamol and multivitamins. However, there were also incidences of noncompliance to medications prescribed by LBA as it sometimes resulted in worsening the situation leading to SM. These women were not aware about the serious problem of malaria in this region and necessary preventive measures to be taken. Overall, the data suggest that the major cause for high prevalence of malaria among pregnant women is the lack of awareness about protective and preventive measures, including mosquito repellents, bed nets, insecticide treated nets, indoor residual spraying, and limited access to antimalarials.[31925]

MDA is one of the most widely used biomarkers for lipid peroxidation. Our findings corroborate with the previously published reports of increased MDA levels among pregnant women infected from Pv, Pf, and mixed infections.[62627] Our findings of a linear relationship between increased parasitic burden and elevated MDA levels are in agreement with that of previously published reports.[2122232425262728] The direct correlation between increased levels of CRP and higher MDA levels suggests that increased OS can lead to increase in inflammation leading to systemic complications.[26272829] It is also evident that increased lipid peroxidation might play an important role in the progression of MM patients to SM. Here, in comparison to pregnant women with MM, we found that serum MDA levels were remarkably higher in SM women, particularly in patients with severe Pf infections.[2930]

During malarial infections, UA is generated by hemolysis and oxidation of nucleic acids.[1831] Studies have shown that UA has the ability to induce inflammatory responses via NALP3 inflammasome pathway.[1432] Findings from our study also support the hypothesis that UA plays a major role in malarial pathogenesis. We found that pregnant women across all infecting species had elevated circulatory levels of UA, particularly during Pf infections. Increase in parasitic burden had a direct relationship with elevated UA levels, especially in Pf infections. It is well known that UA can induce inflammatory responses and as such, we found a direct influence between higher UA levels and CRP levels.[7183334] In this study, upon comparison with MM groups, women with SM had significantly increased serum UA levels, especially during Pf infections. This suggests that there exists a possible strong inflammatory response by UA which might play an important role in augmenting the malarial severity from MM to its severe forms, leading to pregnancy complications.

During Plasmodium infections, to avoid tissue damage, it is essential that large quantities of toxic redox-active by-products such as heme resulting from high metabolic rate of rapidly multiplying parasites are effectively neutralized. In accordance with previous reports, we also found lower SOD levels in SM patients, especially during Pf infections. This supports the hypothesis that, during malarial infections, insufficient levels of host antioxidant defense mechanism fails to adequately neutralize the increased ROS generated, thus aggravating MM into SM.[62735] A negative relationship between decreased SOD levels and increased parasitemia, Hb levels, and CRP suggests that inadequate levels of SOD are responsible for worsening of anemic severity. In this study, confounders such as body mass index (BMI), dietary intake, obesity, and physical activity may have minimal influence on OS and thus have not been adjusted.

Anemia is one of the most common complications during malaria, especially in younger children and pregnant women.[3436] In this study, a significant proportion of patients (87.3%) experienced malarial anemia, of which 11.2% suffered from SA.[2] Elevated free radical levels due to hemoglobin digestion in the Plasmodium IRBC lead not only to the damage of the erythrocyte membrane of IRBC but also damage of the uninfected RBC, resulting in the development of malarial anemia.[3637] Here, we found that the MDA and UA levels increased with an increase in parasitic burden and severity of anemia, mostly during Pf infections. Inefficient antioxidant defenses might lead to mild form of anemia into SA.[182738] In this study, women suffering from SA had significantly lower SOD levels in comparison to MMdA groups. We also observed a significant positive relationship between the decreased RBC and Hb levels among pregnant women typically during Pf infections. From our study, it appears that increased OS might be one of the mechanisms by which maternal anemia may worsen not only during Pf but also during Pv and mixed infections.

It is known that pregnant women are more vulnerable to Pf infection, and this susceptibility is known to diminish with successive pregnancies.[38] We also found increased MDA and UA levels and decreased SOD levels in primigravidae during last trimester.[3940] The decreased immunity in pregnant women might have resulted in augmented OS particularly among women in their primigravidae and those in their last trimester of pregnancy which had resulted in unfavorable outcomes such as in IUGR and babies with LBW. Interestingly, among the various infecting groups, 69.2% of women with Pv infections delivered LBW babies.

Studies have shown that increased OS plays a crucial role in the development of various systemic complications of malaria.[1341] The Plasmodium parasite generates free radicals such as ROS/RNS in the tissues which can lead to local tissue damage. Various studies on pregnant mice and human strongly suggest a pivotal role of OS leading to complications and placental tissue damage. It is clearly evident from our results that, SM infections can result in significant changes in placenta such as syncytial knots, syncytial ruptures, and fibrinoid necrosis, principally during Pf infection. However, the increased leukocyte infiltration during Pv infections might suggest that Pv can induce severe inflammatory responses, leading to adverse outcomes.[42] These placental changes can in turn result in adverse outcomes such as IUGR and LBW babies.[43]

Limitations

Although the study is extensive, there are some limitations such as: (i) the malarial diagnosis was performed by microscopic examination of Giemsa-stained blood smears and no molecular methods such as polymerase chain reaction were used and (ii) confounders such as, BMI, dietary intake, obesity, and physical activity might have a minimal influence on OS, and thus they have not been adjusted.

CONCLUSIONS

The OS caused by malarial infections can play a major role in the severity of both falciparum and vivax malaria. OS-induced malaria conditions might include IUGR, LBW, and even still births. Thus, it is important to create awareness among rural and immigrant population residing in Mangaluru area about the preventive measures and free government-sponsored ANC services in order to reduce pregnancy malaria incidences in this region.

Ethical statement

The study protocol was approved by the Ethical Committee of Kuvempu University, Shivamogga, Karnataka; the Central Ethics Committee of NITTE University, Mangaluru, Karnataka, and the Institutional Review Board of Pennsylvania State University College of Medicine, Hershey, PA, USA. All the study participants were recruited into the study only after obtaining informed consent.

Financial support and sponsorship

This work was supported by the Grant D43TW008268 (to DCG) from the Fogarty International Center of the National Institutes of Health, USA, under the Global Infectious Diseases Program. Effort by DCG was also partly supported by the grant R01AI104844 (to DCG) from the National Institutes of Allergy and Infectious Diseases, National Institutes of Health. Financial support (to ANR) from VGST, Govt. of Karnataka, India is also greatly acknowledged.

Conflicts of interest

There are no conflicts of interest.