Despite global partnerships that aim to eradicate malaria, few studies have investigated susceptibility to severe placental malaria infection (apart from in primigravida1). Reduced fetal growth1 and maternal anaemia sequelae2 are known, however, to have serious consequences. We measured mothers at antenatal booking and later explored severe placental malaria as part of a study on factors affecting fetal growth.
Participants, methods, and results
We investigated a cohort of 436 consecutive rural Bantu women with a singleton pregnancy, with a gestation of less than 24 weeks at booking (determined by ultrasonography), and planning to deliver in Kasaji Hospital in this remote but settled rural area of Katanga province in the Democratic Republic of Congo. A third (31%, 135) were primigravida.3 RMN followed them all through pregnancy to delivery, between 1996 and 1998 (unaware of their placental malaria status at this stage). RMN took placental impression smears using a rigorous protocol. HC later examined them and counted parasites per field (without antenatal measurements) at the Tropical Diagnostic Laboratory (Dublin). RMN graded parasite density (0 = no parasites in 150 fields; 1 = 1-49 parasites per 50 fields; 2 = more than 50 parasites per 50 fields). Because supplies were limited, women were not given chemoprophylaxis for malaria, but prompt treatment with chloroquine was available for symptoms. We found placental infection in 322 women (74%; 125 (93%) primigravida, 197 (65%) multigravida; and 85 (58%) gravida six or more), mainly Plasmodium falciparum.
Women later found to have severe (grade 2) placental malaria were shorter and thinner at booking than those with no placental malaria (table 1; statistically significant decreasing gradient with placental malaria severity groups; analysis of variance P < 0.05), for three measures of skeletal smallness (height, interiliac crests, and intergreater trochanters) and three measures of thinness (body mass index, mid-arm circumference, and weight). Foot size and triceps skinfold thickness showed similar but non-significant trends.
Logistic regression using these continuous variables and comparing grade 0 with grade 2 placental malaria showed that only 63% (131 of the 204 complete cases analysed) of women correctly allocated to the grades, with mid-arm circumference the most important variable, indicating that additional variables also need consideration. None the less, we found the highest frequency of severe placental malaria (36/83, 43%) with combined shortness (< 155 cm) and thinness (body mass index < 19.8 kg/m2); intermediate rates in women shorter but fatter (54/154, 35%) or taller but thinner (20/65, 31%); and the lowest rates (29/124, 23%) in taller fatter women (χ2 for the trend P < 0.01).
Table 1.
Maternal measurements at the antenatal booking clinic by grade of placental malaria*
|
Grade
|
||||
|---|---|---|---|---|
| Maternal measurement at antenatal booking | 0 (non-infected placenta) | 1 (infected placenta) | 2 (infected placenta) | P value† |
| Height (cm) | ||||
| No | 111 | 176 | 139 | |
| Median | 155.0 | 154.5 | 153.0 | |
| Mean (standard deviation) | 155.1 (6.2) | 154.7 (6.2) | 153.2 (6.5) | P≤0.020 |
| Interiliac crests (cm) | ||||
| No | 98 | 150 | 113 | |
| Median | 24.8 | 24.5 | 24.0 | |
| Mean (standard deviation) | 24.7 (1.6) | 24.5 (1.5) | 24.0 (1.5) | P≤0.001 |
| Inter-greater trochanter (cm) | ||||
| No | 98 | 150 | 113 | |
| Median | 27.8 | 27.3 | 27.0 | |
| Mean (standard deviation) | 27.6 (1.8) | 27.4 (1.6) | 26.9 (1.5) | P≤0.002 |
| Foot size (Clarks' foot gauge) | ||||
| No | 98 | 153 | 116 | |
| Median | 5.0 | 5.0 | 4.5 | |
| Mean (standard deviation) | 4.9 (1.5) | 4.9 (1.5) | 4.5 (1.6) | P≥0.060‡ |
| Weight (kg) | ||||
| No | 114 | 177 | 140 | |
| Median | 49.8 | 49.3 | 47.1 | |
| Mean (standard deviation) | 51.2 (8.4) | 50.0 (7.4) | 48.0 (6.6) | P≤0.001 |
| Mid-arm circumference (mm) | ||||
| No | 113 | 177 | 140 | |
| Median | 244 | 238 | 232 | |
| Mean (standard deviation) | 249 (27) | 243 (23) | 235 (21) | P≤0.001 |
| Triceps skinfold (mm) | ||||
| No | 55 | 107 | 69 | |
| Median | 11.6 | 11.3 | 11.2 | |
| Mean (standard deviation) | 12.6 (4.6) | 11.9 (3.9) | 11.4 (3.4) | P≥0.270‡ |
| Body mass index (kg/m2) | ||||
| No | 111 | 176 | 139 | |
| Median | 20.8 | 20.6 | 20.0 | |
| Mean (standard deviation) | 21.3 (2.7) | 20.8 (2.3) | 20.4 (2.2) | P≤0.006 |
All variables were tested and all except height and triceps skinfold showed non-normal distributions, log transformation was not successful so non-parametric statistics were used.
0=no parasites in 150 fields; 1=1-49 parasites per 50 fields; 2=more than 50 parasites per 50 fields.
One way (denotes parametric test, equal variances) analysis of variance (for three groups) or Joncheere-Terpstra for non-parametric variables.
Not significant.
What is already known on this topic
Severe placental malaria is associated with primigravida
What this study adds
Maternal shortness (“stunting”), and small skeletal size and thinness (“wasting”) are associated with a greater likelihood and severity of placental malaria; the dose-response relationship indicates a causative biological mechanism
Comment
Both shortness and skeletal smallness and thinness are associated with increased likelihood and severity of placental malaria. The severity of placental malaria could be influenced by effects of the immune system in early life and childhood, leading to stunting, probably reflecting poor nutrition and family poverty in childhood. Studies in Gambia show that nutritional programming in fetal and early life may compromise the immune system in a way that only becomes apparent in adult life.4 This susceptibility combined with a currently depressed immune system due to current undernutrition could explain the higher rates for short thin pregnant women.
Few other studies have been found, height measurement may lack placental severity information,5 or placental severity subjects were recruited only at delivery.1 However, increased rates of severe placental malaria were found with rural maternal residence, as well as the primigravid.1 If the rural population in Gambia were shorter than the urban population, the data would confirm our observations in Congo.
Would triaging prophylaxis to the short and thin at pregnancy booking be better than the current treatment only protocol to reduce prevalence of severe placental malaria?
Contributors: All the authors designed the study after a pilot investigation by RMN, supervised by HJL. RMN collected the data. All authors, with extensive statistical input from VFH, analysed the data. RMN developed an early draft to which all authors contributed. HJL developed this topic separately and wrote the paper. Hilary Cran, Tropical Diagnostic Laboratory, Dublin, examined the placental slides. Bernard Brabin, Liverpool School of Tropical Medicine; Jason Gardosi, Perinatal Unit, Birmingham; Caroline Fall and Clive Osmond at the Medical Research Council Environmental Epidemiology Unit, Southampton; Henry Kitchener, Department of Obstetrics and Gynaecology, University of Manchester; and Ann Thomson, School of Nursing and Midwifery, University of Manchester; gave valuable input. HJL is guarantor.
Funding: ACE Trust, Dublin, funded the microscopy. RMN was funded by the University of Manchester Scholarship Fund and the British Commonwealth Nurses War Memorial Fund.
Competing interests: None declared.
Ethical approval: Kasaji Hospital, Congo. Also, the University of Manchester ethically inspected the protocol in 1994-5; at that time separate approval was not needed.
References
- 1.McGregor IA, Wilson ME, Billewicz WZ. Malaria infection of the placenta in the Gambia, West Africa: its incidence and relationship to stillbirth, birthweight and placental weight. Trans R Soc Trop Med Hyg 1983;77: 232-44. [DOI] [PubMed] [Google Scholar]
- 2.Matteelli A, Donato F, Shein A, Muchi JA, Leopardi O, Astori L, et al. Malaria and anaemia in pregnant women in urban Zanzibar, Tanzania. Ann Trop Med Parasitol 1994;88: 475-83. [DOI] [PubMed] [Google Scholar]
- 3.Newby RM. Symphysis-fundal height and the influence of placental malaria and poverty on pregnancy outcomes in rural Democratic Republic of Congo. PhD thesis. Manchester: School of Primary Care, University of Manchester, 2000.
- 4.Moore SE. Nutrition, immunity and the fetal and infant origins of disease hypothesis in developing countries. Proc Nutr Soc 1998;57: 241-7. [DOI] [PubMed] [Google Scholar]
- 5.Reinhardt MC, Ambroise-Thomas P, Cavallo-Serra R, Meylan C, Gautier R. Malaria at delivery in Abidjan. Helv Paediatr Acta Suppl 1978;(41): 65-84. [PubMed]