Predictors of upper gastrointestinal bleeding in patients with schistosomal periportal fibrosis

A case-control study was conducted between 1985 and 1987 in the Gezira-Managil area of central Sudan to assess the major predictors of haematemesis. Eighty-four patients who had suffered at least one attack of oesophageal bleeding and had schistosomal periportal fibrosis demonstrated by ultrasonography were compared with 173 subjects without bleeding but with ultrasonographic evidence of periportal fibrosis. A splenic longitudinal dimension of more than 11 cm, periportal fibrosis worse than grade I and varices more than grade I were independently associated with a significant risk of variceal bleeding. Age, sex, presence of a palpable liver and portal vein diameter were not associated with a significant risk of bleeding after adjustment for potential confounding variables. Factors identified in this study could be helpful in the prophylactic management of patients with complicated schistosomiasis.

Levels and determinants of perinatal mortality in central Sudan

OBJECTIVES

To determine levels and risk factors for perinatal mortality in Central Sudan.

METHODS

Hospital and community based studies were conducted during the period 1989-1990. Of 5328 births registered in the hospital, 197 stillborns and 812 live-born infants were included in a nested case-control study. In the community, a follow-up study was conducted on 1592 midwife-assisted home deliveries. Multivariate logistic regression analysis was used to identify predictors of perinatal mortality and adjusted population attributable risks were estimated to assess the contribution of each factor.

RESULTS

The perinatal mortality rate was 85.4/1000 births in the hospital population and 29.4/1000 births in the community population. The major risk factors for perinatal mortality in the hospital and the community studies were similar and modifiable.

CONCLUSIONS

To lower perinatal mortality, improvements in maternal nutrition, malaria treatment and control, avoidance of agricultural pesticides during pregnancy, and adequate antenatal and intrapartum care are recommended.

Agricultural pesticide exposure and perinatal mortality in central Sudan

Hospital- and community-based studies were conducted in central Sudan to investigate the association between pesticide exposure and perinatal mortality. The cases were 197 stillbirths in the hospital and 36 perinatal deaths in the community; the controls were 812 liveborn, normal-birth-weight infants in the hospital, and 1505 liveborn infants who survived for the first 7 days after birth in the community. The odds ratio (OR) of perinatal death associated with pesticide exposure was estimated using multiple logistic regression. There was a consistent and significant association between pesticide exposure and perinatal mortality in the hospital (adjusted OR = 1.9; 95% confidence interval (CI): 1.3-2.8) and the community populations (adjusted OR = 2.7; 95% CI: 1.1-6.4). The OR was significantly higher among women engaged in farming (3.6; 95% CI: 1.6-8.0), but not among women in nonfarming occupations (1.6; 95% CI: 0.8-3.3). The estimated attributable risks of perinatal death owing to pesticide exposure were 22.6% for hospital stillbirths and 15.7% for community perinatal deaths; but among women engaged in farming in the hospital population the attributable risks were substantially higher (34.5%).

Determinants of neonatal mortality in central Sudan

A follow-up study was conducted in six community health centres during the period April 1989 to March 1990 to determine the risk factors which influence neonatal survival in central Sudan. The estimated neonatal mortality rate ranged between 20.0 and 36.0 per 1000 live births per year, and the major cause of death was tetanus neonatorum (29% of neonatal deaths). The mortality rate was lowest when tetanus toxoid was received during pregnancy and the umbilical cord was cleaned by a modern hygienic method (mortality rate of 11 per 1000). In contrast, the mortality rate was highest when no tetanus toxoid was received and no or traditional cord cleaning was used (mortality rate of 62 per 1000; relative risk (RR) = 5.6, 95% confidence interval (CI) 2.0-14.9). The major predictors of neonatal mortality were tetanus, short birth-to-conception interval, multiparity, reported malaria during pregnancy, low birthweight, low maternal weight and low socio-economic status. The population attributable risks were high, and the preventable factors collectively accounted for 93.5% of neonatal mortality. Safe deliveries and wider immunization coverage are needed to control neonatal tetanus in this community. Other interventions to lower neonatal mortality in central Sudan should include accessible family planning programmes and measures to lower the incidence of low birthweight.