Malika LENEUVE-DORILAS

Background and objective :

French Guyana, an overseas department and region, has almost 8,000 births a year.
In French Guyana, since 1992, the proportion of premature births has been high at around 13.5%, almost double that of France (7%) (data from the Registre d'Issue de Grossesses and national perinatal survey). Unlike most countries where an increase in prematurity has been observed, the rate in French Guyana is stable. Admittedly, we could be satisfied with this non-increase, which would be a reflection of effective measures in French Guyana, but perinatal-related deaths remain one of the main causes of premature mortality in this department. This would partly explain the gap with mainland France in terms of life expectancy at birth.

While the rate of prematurity is not increasing in French Guyana, it is not decreasing either. Given this lack of regression, it seems important to understand the factors that make prematurity so frequent in French Guyana, and so difficult to curb. This thesis therefore set out to identify the factors that predict premature birth in French Guiana, with the ultimate aim of contributing not only to improving the care of pregnant women, but also to reducing the rate of prematurity.

Methodology :

This research work has four lines of investigation: A retrospective descriptive study, based on data from the RIGI (Registre d'Issue de Grossesses Informatisé) 2013-2014 of 12,983 viable births in all four health establishments in the department. The development of a predictive score for prematurity based on the 2013-2014 RIGI of 12,983 viable births, compared with data from the 2015 RIGI of 6,914 viable births. A single-centre case-control etiological study of extreme prematurity at the Centre Hospitalier Andrée Rosemon de Cayenne (CHAR), from February 2016 to January 2017. CHAR is the only type III health facility in French Guyana. Finally, analysis of the average term at birth and morbidity and mortality based on the RIG (Registre d'Issue de Grossesses) 2002-2007 of 3,5648 viable births and the RIGI (Registre d'Issue de Grossesse Informatisé) 2013-2014 of 12,983 viable births.

Results :

Over the study period, the proportion of premature births was 13.5% (1,755/12,983). The proportion of spontaneous prematurity was 51.3% (901/ 1,755) according to RIGI 2013-2014, compared with 48.7% (854/ 1,755) of induced prematurity (induction of labour and caesarean section before labour).

More than half (57.2% or 7,421/12,983) of the study population were covered by social security, but 9.3% (1,211/12,983) had no social security cover. The absence of social security coverage was a risk factor for prematurity, with an adjusted OR of 1.9 (CI 95% [1.6-2.3] p=0.0001).

Similarly, in terms of pregnancy management, the absence of antenatal care and birth preparation doubled the risk of preterm birth. The respective adjusted ORs were 2 [CI 95% =1.2-3.5] p=0.007 and 2.4 [CI 95% =1.5-3.7] p=0.0001.

In French Guyana, in the general population, prematurity is attributable to the absence of social security cover in 21.5% of cases, in 69% of cases to the absence of preparation for birth and in 72.2% of cases to the absence of a prenatal interview.

On the other hand, in terms of pathologies associated with pregnancy, pre-eclampsia syndrome was the main dysgravidity associated with the risk of prematurity (adjusted OR 6.7 [CI 95% =5.6-8.1] p=0.001). Finally, the fairly widespread hypothesis, suggesting that part of the high rate of prematurity is linked to the fact that 'black' babies are more mature and that 'black' mothers of Afro-Caribbean descent give birth physiologically earlier, did not emerge in our analyses. In fact, there was no statistically significant difference in morbidity and mortality between newborns of mothers of Afro-Caribbean origin and those of Caucasian women.

Analysis in the "black" population showed that being born at 36 days' gestation compared with 37 days' gestation was an adjusted risk factor for morbidity and mortality, with adjusted ORs for the RIG 2002-2007 and RIGI 2013-2014 of 1.9 [CI 95% =1.3-2.9] p=0.001 and 2.6 [CI 95% =1.8- 3.9] p=0.0001 respectively.

Conclusion:

The studies carried out identified a number of factors associated with prematurity, some of which had already been described elsewhere. Although at the individual level it was impossible to predict who would give birth prematurely, the weight of social factors and poor pregnancy monitoring suggested that a population-based approach might be relevant. The most vulnerable women often lived in well-identified areas where targeted action could be taken to improve monitoring and detect complications. The problem of social inequalities in health goes well beyond prematurity and is found in almost all diseases, which suggests that there are synergies to be sought and that the population level is undoubtedly strategic. The weight of the pre-eclamptic syndrome as a risk factor for induced prematurity in French Guyana raises questions: in fact, it appears to be much greater than elsewhere, for reasons that remain to be clarified.

Key words :
Prematurity, Risk factors, Predictive score, Neonatal morbidity and mortality, Social inequalities in health, Spontaneous prematurity, Induced prematurity

Context and objective:

French Guiana, an overseas department and region, has nearly 8,000 births per year.
Since 1992, the proportion of premature births, although stable, has remained high at around 13.5%, almost double that of France (7%) (data from the Pregnancy Outcome Register and national perinatal survey). While in most countries we see an increase in prematurity, we could, wrongly, be satisfied with a non-increase in the prematurity rate that would reflect progress. However, deaths from perinatal causes remain one of the main causes of premature mortality in Guyana and partly explain the gap with France in terms of life expectancy at birth.
Given this lack of improvement in the prematurity rate, it seems important to better understand the factors that make prematurity so frequent and so difficult to control in Guyana. The thesis work thus focused on identifying the predictive factors of different definitions of prematurity in the Guyanese context with the ultimate aim of contributing to improving the care of pregnant women and curbing the curve of the prematurity rate.

Methodology :

This research work is divided into 4 areas of investigation:

• A descriptive retrospective study, based on data from the RIGI (Register of Computerized Pregnancy Outcomes) 2013-2014 of 12,983 viable births in all four health establishments in the department,
• The development of a predictive prematurity score from the 2013-2014 RIGI of 12,983 viable births, compared to the 2015 RIGI data of 6,914 viable births,
• A case-control etiological study of extreme prematurity, monocentric, at the Andrée Rosemon Hospital in Cayenne, from February 2016 to January 2017. The only type III health facility in the French Guiana Region,
• Analysis of the average term at birth and morbidity and mortality from the RIG (Register of Pregnancy Outcomes) 2002-2007 of 35,648 viable births and the RIGI 2013-2014 of 12,983 viable births.

Results :

Over the study period, the proportion of preterm births was 13.5% (1,755/12,983). The proportion of spontaneous prematurity was 51.3% (901/ 1,755) according to the 2013-2014 RIGI, compared to 48.7% (854/ 1,755) of induced prematurity (induction of labour and cesarean section before labour).

More than half (57.2% or 7 421/12 983) of the study population had social security, but 9.3% (1 211/12 983) had no social security coverage. The lack of social security coverage was a risk factor for prematurity with an adjusted OR of 1.9 CI at 95%[1.6-2.3] p=0.0001. Similarly, with regard to pregnancy management, the absence of prenatal care as well as that of birth preparation would double the risk of premature birth. The respective adjusted ORs were 2[95% CI =1.2-3.5] p=0.007 and 2.4[95% CI =1.5-3.7] p=0.0001.

Thus, in Guyana as a general population, prematurity in Guyana would be due to the lack of social security coverage in 21.5% of cases, and in 69% of cases (in the general population) to the lack of birth preparation. According to the RIGI 2013-2014, in the general population, prematurity is also attributable, in 72.2% of cases, to the absence of prenatal care.

On the other hand, for pathologies associated with pregnancy, pre-eclampsia syndrome was the main dysgravidia associated with the risk of prematurity (OR adjusted by 6.7 [95% CI =5.6-8.1] p=0.0001). Finally, the fairly common hypothesis that part of the high prematurity rate is related to the fact that black babies are more mature and black mothers give birth physiologically a little earlier did not emerge in our analyses. Indeed, there was no statistically significant difference in morbidity and mortality for infants born to Afro- Caribbean mothers and Caucasian women.

Intra-ethnic analysis, in the Black population, showed that being born at 36 SA compared to the term 37 SA would be an adjusted risk factor for morbidity and mortality with respective adjusted ORs for the 2002-2007 RIG and 2013-2014 RIGI of 1.9[95% CI =1.3-2.9] p=0.001 and 2.6[95% CI =1.8-3.9] p=0.000.

Conclusion:

The work carried out has identified many factors associated with prematurity, factors already described elsewhere. Although at the individual level it was impossible to predict who would give birth prematurely, the weight of social factors and poor follow-up suggested that a population-based approach might be appropriate. Thus, the most vulnerable women often reside in well-identified areas that could be the subject of targeted actions to improve follow- up and identify complications. This problem of social inequalities in health goes well beyond prematurity and is found for almost all pathologies, suggesting that there are synergies to be sought and that the population scale is undoubtedly strategic. The weight of preeclampsia as a risk factor for induced prematurity in French Guiana raises questions: indeed, it seems much more important than elsewhere for reasons that remain to be clarified.

Keywords :

Prematurity, Risk factors, Predictive score, Neonatal morbidity and mortality, Social inequalities in health, Spontaneous prematurity, Induced prematurity