Pre-pregnancy care for women with pre-gestational diabetes mellitus: a systematic review and meta-analysis

Pre-gestational diabetes mellitus (PGDM) and maternal hyperglycemia during the time of organogenesis is a known teratogen with detrimental effects on the fetal heart, renal, musculoskeletal and central nervous systems [1‐3]. Population based studies showed that there is a fivefold increase in the rate of cardiovascular malformations, and more than twofold increase in the rate of neural tube defects and urinary tract abnormalities in infants of diabetic mothers when compared to the background population [1, 2]. Moreover congenital malformations (GM) are associated with increased risk of stillbirth and perinatal mortality (PM) as they account for almost 50% of all deaths of infants born to mothers with PGDM [4, 5].

CM secondary to maternal diabetes can be prevented, in great part, by optimizing maternal health in the pre-pregnancy period. Glycemic control is one of the most important aspects of pre-pregnancy care (PPC) [6]; however other aspects of care such as folic acid supplementation, smoking cessation, screening and treatment of diabetes complications and discontinuing teratogenic medications, are as important for improving maternal and fetal outcomes and might be effective in reducing the rate of CM to the background level [7‐9].

The aim of this systematic review is to assess the effectiveness and safety of PPC in improving the CM and perinatal mortality for women with type 1 or type 2 PGDM.

The search retrieved 2452 potentially relevant titles of which the full text of 54 relevant reports were reviewed (Figure 1). A total of 25 reports of 21 studies were included in this review [7, 8, 11‐33]. (Three articles described the same cohort study with two interim [15, 16] and one final report [17], one study reported the outcomes for the same cohort in two articles [7, 27] and two articles report the outcomes of one cohort with one interim [29] and one final report [26]).

Twenty six studies were excluded because they did not meet the inclusion criteria; two reports were of conference proceedings and in one study data were not extractable (Additional file 2: Appendix 2).

Of the included studies, only one was a controlled trial, 13 studies were prospective cohort studies, six studies were retrospective cohort studies and one was a case control study (Tables 1 2 3 4). The included studies were conducted between 1983 and 2010 in Europe and the United States of America, except for one study conducted in Israel [19].

In this review PPC reduced the rate of CM from 7.4% to 1.9%; a rate similar to that reported for the background population (Figure 2).

Reports from practically implemented PPC programs support our findings of the effectiveness of PPC in improving the rate of CM of infants of mothers with PGDM [8, 33]. However, many uncertainties remain about the ability of these programs to reduce the malformation rate to that of the background population, particularly that a small but significant risk of CM remained [33, 34]. This observation might be due to many factors, for instance the preferential effect of the PPC on women with type 2 diabetes [8, 35]; the influence of other maternal risk factors for CM, such as maternal obesity [36]; and the degree of comprehensiveness; and coverage of services the program provides to prevent malformations beside glycemic control, such as folic acid supplementation and discontinuing teratogenic medication [8].

Population based studies showed a variable increase of 2.5 – 9 folds in the PM rate in diabetic mothers compared to the background population [37]. For women with PGDM 16-28% of PM is due to CM, and an additional 21-41% is due to preterm delivery [38, 39]. Since the rates of both complications improve with PPC [9], it is not surprising that the PM in women, who attended PPC, in this review, is reduced by 66% compared to those who did not (Figure 3).

During the period of organogenesis, maternal hyperglycemia is known to be associated with CM [34, 40]. Population based studies showed a linear relationship between maternal HbA1C level above 7%, in the first trimester, and both CM and PM [41]. It is estimated that an increase of 1% in the level of HbA1C in the first trimester increases the odds of an adverse pregnancy outcome by 5-6% [42].

The analysis of the pooled data in this review suggested that PPC is effective in reducing the level of HbA1C during the first trimester of pregnancy by 1.9% (Figure 4). However this result is associated with marked statistical heterogeneity (I² = 98%) (Figure 4). The reason for this heterogeneity is the difference in ‘effect size’ produced by the PPC between the included studies. Evers et al. reported a reduction of 8.5% in HbA1C between the intervention and the control groups [33], while Goldman et al. documented a reduction of 29% [19]. Another reason is the difference in the mean levels of HbA1C of the intervention groups (target level), which vary between 6.4% [33] and 8.5% [24, 26]. Nevertheless the overall estimate of the effect of PPC in reducing the level of HbA1C is evident by the same direction of effect in all included studies (Figure 4).

In the two studies that evaluated severe maternal hypoglycemia as an adverse effect of PPC [7, 26] the pooled data showed an increased risk of hypoglycemia in women who attended the PPC as compared to those who did not (Figure 5). However, this result should be approached with caution due to the marked statistical heterogeneity associated with the meta-analysis (I² = 85%) (Figure 5) and the small number of included studies. The most likely explanation of the statistical heterogeneity is the variable effect size of PPC on maternal hypoglycemia in the two studies. The study by Temple et al. [7] showed no effect of PPC on maternal hypoglycemia, RR 1.18 (95% CI 0.88-1.58), while the study by Steel et al. showed a significant increase in hypoglycemia among women who attended PPC RR 3.19 (95% CI 1.56-6.53) (Figure 5). We believe that the statistical heterogeneity is a reflection of a true clinical heterogeneity. During the 16 year interval between the two studies (Steel et al. 1990, Temple et al. 2006) many innovations in the treatment of diabetes in pregnancy have been developed. Such innovations included patients’ education and counseling, intensive self-monitoring of blood glucose and functional insulin therapy [43]. Although meta-analysis suggested an increased risk of severe hypoglycemia with PPC, we believe this is an unlikely adverse effect with modern treatment and monitoring of diabetes during pregnancy.

In this review some isolated interventions in the pre-pregnancy period were not as effective as a ‘program of interventions’. One case control study addressed the effectiveness of multivitamin supplementation in the pre-pregnancy period, as an isolated intervention, in reducing the rate of CM and showed no effectiveness [12].

One study conducted an economic evaluation of PPC and found that it is associated with considerable savings, and reduced resource utilization [20]. However, population based studies showed that only 34-38% of eligible women receive PPC [3, 28]. Hence further studies should focus on how to increase utilization and uptake of PPC and reduction of the rate of unplanned pregnancies among diabetic women.

This systematic review provides a moderate to high level of evidence for the effectiveness of PPC in the improvement of maternal and fetal outcomes [44]. It confirms previous findings by Ray et al. [6]; nevertheless, the strength of our review is based on the comprehensive evaluation of the available evidence on the effectiveness and safety of PPC.

We are aware of the limitations of the observational studies as the main source of evidence and the inherent bias associated with the design; however, randomized controlled trials to assess the effectiveness of PPC are neither ethical nor feasible. Nevertheless the nature of the intervention lent strength to the observational studies by avoiding certain biases known to occur in such study designs. Lack of allocation concealment and blinding of participants were avoided by recruiting the intervention and the control groups at different times during the course of the study (pre-pregnancy period and antenatal period). Additionally, and due to the relatively short duration of the pregnancy, attrition bias was noted in only one study, [24] all other studies had complete follow-up of both groups. The possible bias due to confounding factors such as social class, level of education, subjects’ motivation, smoking, maternal age, obesity, parity and vascular complications of diabetes, was noted by most of the studies but only two studies used the appropriate statistical tests to quantify the effect of the PPC apart from the confounders [7, 8]. It is worth noting that all but one of the studies included in this review were conducted in Europe and the United States of America [19], which limits the generalizability of this review.

The review carries important implications for practice and research as it highlights the importance of the integration of PPC in the routine care of diabetic women during their reproductive age. and have practical implication considering the recent report of the CEMCH [3] which showed that CM rate in infants of diabetic mothers in England, Wales and Northern Ireland is twofold, and the PM was nearly fourfold, that of the background population. These findings are also of a paramount importance to many communities in the Middle East [45], North Africa [46] and some communities in Asia [47] where the rate of CM is very high due to many causes including maternal diabetes.

One of the main obstacles to the full implementation of PPC programs is the failure of the target population to utilize provided services [8]. We suggest that more research is needed in methods of encouraging diabetic women to utilize PPC.

PPC for women with type 1 or type 2 PGDM is effective in improving rates of CM, PM and in reducing maternal HbA1C in the first trimester of pregnancy. PPC might cause maternal hypoglycemia in the first trimester of pregnancy.