Table 1 Characteristics of included prospective cohort studies

Garcia-Patterson 1997 [18] (Spain)

66 participants with type I and type II who attended the pre-pregnancy clinic and 119 participants with type I and type II diabetes who did not.

PPC included intensive insulin therapy, self-monitoring of blood glucose and dietary advice

The HbA1C was significantly better in the PPC group than for the NPPC group (p = 0.01). There was no significant difference between the two groups in rate of congenital malformations.

Medium (The baseline characteristics in relation to the vasculopathy are different. No blinding for the outcome assessment).

Herman 1999 [20] (USA)

24 women with type I diabetes who attended the pre-pregnancy clinic, and 74 women with type I diabetes who did not attend the pre-pregnancy clinic.

PPC included education, counseling, glycemic control, and assessment of complications of diabetes such as nephropathy and retinopathy

Women who had PPC had significantly lower level of HbA1C at booking and throughout pregnancy. There was no significant difference between the two groups in the frequency of infants with congenital malformations.

High (The study was not designed to assess the clinical outcomes of the pre-pregnancy care but the differences in the socio-demographic features between the groups who attend the pr-pregnancy care and those who did not. The target level for the glycemic control was not clear and the absolute level of Hb A1C at booking and all through pregnancy for the study and the control groups was not mentioned)

Jaffiol 2000 [21] (France)

21 IDDM attended the pre-conception care and 40 did not attend

PPC included education, glycemic control, self monitoring of blood glucose and contraception

There were significant reductions in the PM and congenital malformations in the PPC group as well as in the level of maternal HbA1C in the 1^st trimester compared to the NPPC group.

Low (good report, clear intervention description, the comparative groups received same antenatal intervention. No blinding for outcome assessment)

Jensen 1986 [22] (Denmark)

9 women with insulin dependent diabetes had pre-pregnancy care and 11 women with insulin dependent diabetic who did not receive pre-pregnancy care.

PPC included continuous insulin infusion initiated 2 months prior to conception.

No significant difference in congenital malformations and HAb1C level, between the two groups

High (small number of study and control group, many differences in the baseline characteristics in the severity of diabetes, 5 of the 11 control women were treated in the diabetic clinic in the hospital before pregnancy so they knew about the importance of glycemic control both groups have the same HA1C levels in early pregnancy)

Kitzmiller 1991 [23] (USA)

84 women in pre-pregnancy care and 110 women had no pre-pregnancy care.

PPC included glycemic and dietary control education, exercise and contraception.

The frequency of congenital abnormalities in the PPC group was significantly higher than in the NPPC group (p <0.05).

Low (good report clear methodology)

Rosenn 1991 [24] (USA)

28 women in the pre-pregnancy group and 71 in the control group

PPC included dietary advice and glycemic control

HbA1C concentration in the PPC group was lower than in the NPPC group (p <0.0008). There were no congenital malformations in either group.

Medium (52% of pre-pregnancy care patients dropped out, no blinding in the assessment of the outcome)

Temple 2006a [7] 2006b [27] (UK)

110 women with type I diabetes attended the pre-pregnancy care clinic and 180 women with type I diabetes did not attend the pre-pregnancy care clinic

PPC included: Glycemic control, folic acid supplementation, smoking cessation, education.

The rate of congenital malformations was lower in PPC group compared to the NPPC group (p < 0.065). PM was significantly more in the latter group than the former one (p < 0.026)

Low (Baseline characteristics in both groups were similar; the prospective nature of the study ascertained the completeness of the follow up, the completeness of the baseline and the outcome data. Use of appropriate statistical tests such as logistic regression analysis confirmed the association between the pre-pregnancy care and outcomes).

Willhoite 1993 [28] (USA)

62 women with either type I or type II diabetes who received pre-pregnancy counseling and 123 women witheither type I or type II diabetes who did not receive pre-pregnancy counseling

PPC included counseling by health professional the control group received no counseling.

PPC group had significantly less PM than the NPPC group (OR3.9 CI 1.2-13.9) and insignificantly less congenital malformations (OR 4.2 CI 0.5-29.7)

High (Base line characteristics of the two groups were significantly different in age, duration of diabetes and smoking all are confounding factors for the outcomes. The two groups did not receive the same antenatal intra-partum and postnatal care. The assessor of the congenital malformation was not blinded)

Boulot 2003 [31] (France)

172 women with either type I or typeII diabetes who received PPC and 260 women with either type I or type II diabetes who did not receive PPC

PPC included education, assessment of diabetes complications glycemic control self monitoring of blood glucose and contraception

PPC group had significantly less PM than the NPPC group, (p <0.005) for type 1 diabetics and significantly less congenital malformations, (p <0.005) for type 1 diabetics

Low (cases and control were well defined and comparable, selection bias is unlikely as consecutive cases were enrolled, the prospective nature of the study ascertained the completeness of the follow up, the completeness of the baseline and the outcome data)

Galindo 2006 [30] (Spain)

15 women with pre-gestational diabetes received PPC and 112 women with pre-gestational diabetes did not receive PPC.

PPC included education, glycemic control self monitoring of blood glucose

The frequency of congenital abnormalities in the PPC group was 3/15 compared to 14/112 in the NPPC group.

Low (cases and control were well defined and comparable, selection bias is unlikely as consecutive cases were enrolled, the prospective nature of the study ascertained the completeness of the follow up, the completeness of the baseline and the outcome data)

Garcia Ingelmo 1998 [32] (Spain)

12 women with pre-gestational diabetes received PPC and 12 women with pre-gestational diabetes did not receive PPC.

PPC glycemic control.

The frequency of congenital abnormalities in the PPC group was 3/12 compared to 2/12 in the NPPC group. HbA1c was significantly lower in the first trimester in the PPC group compared to the NPPC group, (p <0.01)

High (Both the study population and the control were not representative of the general diabetic population with frequency of diabetic vascular complications approaching 50%. The PPC components were not defined neither the target blood glucose)

Murphy 2010 [8] (U.K)

181 participants with type I and type II who received PPC and 499 participants with type I and type II diabetes who did not receive PPC

PPC included: glycemic control, folic acid supplementation, smoking cessation, education.

The frequency of congenital abnormalities in the PPC group was 0.7% compared to 5.6% in the NPPC group (p <0.02). The PM in the PPC group was 0.7% which was similar to 2.2% in NPPC (p <0.4).

Low (good report clear methodology)

Evers 2004 [33] (Netherland)

271 women with type I diabetes had planned pregnancy and 152 women did not plan their pregnancy.

PPC had planned pregnancy and folic acid supplementation

The frequency of congenital malformation in PPC group was 11/271 (4.1%) compared to 18/152 (12.2%) in the NPPC group. The mean of HbA1C concentration of the PPC group was significantly lower than the NPPC group.

Medium (confounding factors such as smoking, education level and social class were not examined.

The results of HbA1C during the first trimester were not available for 29% of the whole study group)