SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objectives Long term follow up women with gestational diabetes mellitus (GDM).

Design Case–control study.

Setting Academic obesity unit.

Population Women earlier identified as having gestational diabetes mellitus.

Method Twenty-eight women diagnosed with GDM in 1984–1985, and a control group (n= 52) who gave birth at the same time performed a 2-h oral glucose tolerance test 15 years later. Basic anthropometry and questions about various aspects of eating and exercise habits were furthermore obtained.

Results Ten women (35%) in the GDM group were diagnosed with type 2 diabetes mellitus and none in the control group (P < 0.001). Mean BMI in the diabetic group was 27.4 kg/m2 and in the non-diabetic GDM group 24.6 kg/m2 (P < 0.05). The mean weight gain since the first child was 8.4 kg in all GDM versus 8.1 kg in controls (ns). The women who developed type 2 diabetes mellitus, however, gained 15.1 kg since the birth of their first child (P < 0.05).

Conclusions Women who are diagnosed with GDM have a considerably higher risk of developing type 2 diabetes mellitus later in life. Despite a close medical monitoring during pregnancy, the further follow up within the health care system and information about long term consequences of GDM for later type 2 diabetes mellitus development seems to be generally lacking. More active strategies for future weight control and lifestyle advice after delivery might therefore be indicated for women with GDM.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Many women, seeking help for their weight problem, describe their pregnancies as major life events triggering sustained weight increase. A retrospective study of such women revealed that 74% had maintained more than 10 kg after each pregnancy1. The Stockholm Pregnancy and Women's Nutrition (SPAWN) study was set up to monitor—on a long term basis—the weight development and associated lifestyle factors of 1423 women, who gave birth in 1984–1985. Numerous factors associated with weight development have been assessed2–6. In the original SPAWN protocol, women who developed gestational diabetes mellitus (GDM) were excluded. It was assumed that weight development of this subgroup could differ from that of a normal pregnancy. At follow up 15 years later, however, it was realised that data from this subsample could provide interesting information about weight development over time in GDM women.

Information about the natural history of GDM is much needed. Although it is generally stated that 1–3% of women develop GDM during pregnancy7,8, the further development into a clinically manifest type 2 diabetes mellitus seems to follow different routes. The risk depends on numerous factors. Some of these are technical, such as the diagnostic criteria, which have been the cause of much international debate9. Some factors clearly cannot be modified, such as ethnicity, pre-pregnancy, weight, age, parity, family history of diabetes, degree of hyperglycaemia in pregnancy and immediately postpartum10. Other risk factors can be modified after a pregnancy, such as future weight development and further pregnancies10.

In earlier studies, GDM women have been reported to be heavier than matched pregnant women with a normal glucose tolerance11. Weight gain during pregnancy has, on the other hand, been found to be about 1.5–2 kg less in GDM women than in controls11–13. A similar pattern was found in that GDM women are heavier at conception but gain less during pregnancy. Because some of these studies were retrospective and weight at conception, as is generally the case, was self-reported, there remains some uncertainty, but the pattern seems to be consistent.

Because this group of women is at high risk for type 2 diabetes mellitus, it is pertinent to also monitor long term weight development in parallel to the natural history of type 2 diabetes mellitus. In their study, Sepe et al.11 concluded that “a well-designed population-based study of GDM would go a long way to assess these issues and their public health significance and impact”. Our information concerning the SPAWN women with GDM to some extent provided such an opportunity.

The aim of this study was to assess to what extent GDM women developed clinical type 2 diabetes later in life, to what extent that development could be predicted and if the impaired carbohydrate metabolism was related to other aspects of the metabolic syndrome.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

For this study, all women who were diagnosed with GDM in 1984–1985 and gave birth in southern Stockholm were sampled from the Swedish medical register14. The diagnosis of GDM was set after a general screening of all pregnant women by urine-glucose test. Women with pathologic urine test went through a 2 hour oral glucose tolerance test (OGTT) with 75 g glucose. The diagnosis of GDM was made if the women had a 2 hour value over 9 mmol/L. The GDM women were than followed by the maternal specialist clinic. Sixty-four such women were identified, six of them as type 1 diabetes mellitus and were thus excluded. From the SPAWN study3,4, 150 women at random (every fifth woman) were identified to form a control group. Twenty-eight subjects in the GDM group who volunteered to participate in our study did not differ significantly in age compared with the 30 GDM women, identified through the national register who did not wish to participate in the study. The mean ages of the 52 SPAWN control women who volunteered to participate were not significantly different in age, weight before pregnancy or weight gain during pregnancy compared with 98 who did not wish to participate in the study.

All women were invited to the Obesity Unit and came for an OGTT with 75 g glucose after a night of at least 8 hours of fasting15. A diagnosis of diabetes was made if the women had a 2 hour blood-glucose value over 10 mmol/L. Blood was sampled immediately before the glucose load for fasting glucose, insulin and liver tests and other routine lab samples. Weight was measured in light underwear on a scale which was regularly calibrated to the nearest 0.1 kg (Tarya TBF 305, Tokyo, Japan) and height was measured to the nearest centimetre. Fat mass was measured by bio-impedance (Tarya TBF 305). Glucose was sampled every 30 minutes for 2 hours; on the first and last occasion, blood insulin was also measured. Subjects were asked to fill in questionnaires about their eating behaviour, exercise habits, social conditions and medical history and questions about their knowledge of risk for type 2 diabetes mellitus.

Case records for the index pregnancy supervision were obtained from all maternity units after permission from all participating women had been obtained. Parametric statistics were used. Student's t test, χ2 test and Kaplan–Meier tests were performed as appropriate. The study was approved by the Ethics Committee of the Huddinge University Hospital and all women gave written informed consent.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The mean age in the GDM group did not differ significantly from that of SPAWN control women, nor did waist circumference or waist/hip ratio. The mean BMI did not differ between the GDM and the SPAWN control women. The number of children differed significantly between groups (P < 0.05) (Table 1). Four GDM of 28 and nine of 52 control women stated that they had reached menopause (ns).

Table 1.  Clinical data of 15 years of follow up of women who were diagnosed with GDM during a pregnancy 1984–1985 compared with matched control group. Values are mean (SD).
 GDM (n= 28)Control group (n= 52)P
  1. *Student's t test.

  2. **χ2 test.

Age (year)47.6 (4.2)45.6 (4.7) 
BMI (kg/m2)25.7 (1.11)24.7 (2.01)ns*
Waist/hip ratio (cm)0.81 (0.06)0.84 (0.01)ns*
Waist (cm)84.3 (9.6)84.4 (9.6)ns*
Children (n)2.6 (0.9)2.4 (0.6)<0.05*
No. of type 2 diabetes mellitus patients100< 0.001**

During the 15 years follow up period, six women out of 28 (21%) GDM women had been diagnosed with type 2 diabetes mellitus in general practice, and another four were identified as diabetics by the OGTT in our unit, but all in the control group had normal OGTT (P < 0.001) and none were diagnosed with diabetes. The mean fasting insulin concentration was 72.5 mmol/L in the seven who were not put on insulin compared with 49.5 mmol/L in the GDM women who did not develop type 2 diabetes mellitus (P < 0.05) The cumulative frequency of GDM women who remained free from type 2 diabetes mellitus during the 15 years follow up is shown in Fig. 1. At follow up, the group of GDM women who developed type 2 diabetes mellitus (GDMDM2) had a significantly higher BMI (P < 0.05) and almost significantly higher waist/hip ratio (P= 0.06) than GDM women who did not. However, blood pressure levels and serum lipoprotein concentrations did not differ between the groups (Table 2).

image

Figure 1. Cumulative frequency of GDM women who remain free from type 2 diabetes mellitus during a 15 year follow up (P < 0.001), Kaplan–Meier plot.

Download figure to PowerPoint

Table 2.  Clinical data for the group of GDM women who developed type 2 diabetes (GDMDM2) compared with those who did not (GDMNOT). Values are mean (SD).
 GDMDM2GDMNOTP*
  1. *Student's t test.

Age (year) at follow up45.7 (4.5)47.5 (4.6) 
Weight before first pregnancy (kg)53.4 (5.5)56.2 (16.9)ns
Weight (kg)71.3 (13.7)66.1 (7.8)<0.05
BMI (kg/m2)27.4 (3.5)24.6 (3.8)<0.05
Fat (%)35.9 (7.7)31.6 (5.3)ns
Waist (cm)88 (0.1)82 (8.9)0.06
WHR (m)0.85 (0.06)0.84 (0.12)ns
Blood pressure diastolic (mmHg)80 (8)74 (9)ns
Blood pressure systolic (mmHg)128 (17)120 (11)ns
Blood pressure mean (mmHg)104 (9)97 (11)ns
Cholesterol (mmol/L)5.1 (0.8)5.5 (1.14)ns
Triglycerides (mmol/L)2.6 (0.3)1.8 (1.6)ns
LDL (mmol/L)3.2 (0.9)2.6 (1.2)ns
HDL (mmol/L)1.3 (0.4)2.0 (0.8)ns
Fasting blood sugar (mmol/L)8.2 (4.9)6.3 (1.6)<0.05
HBA1c (mmol/L)5.5 (1.5)4.7 (1.6)<0.05
No. of children3.0 (0.8)2.5 (1.0)ns

Fifty-four percent of the GDM women stated that they had never been informed that they had a higher risk of developing type 2 diabetes mellitus than others. However, no statistically significant relationship between weight development since the index pregnancy and the likelihood for type 2 diabetes mellitus was found (χ2 test). Figure 2 summarises the weight development for women in all three study groups from the birth of their first child until final follow up. (The index pregnancy was child no. 1 in 38% cases, no. 2 in 38% cases, no. 3 or above in 21% cases and no. 4 in 3%.) The index pregnancies were the mean 1.8 pregnancy. In Fig. 2, self-reported data of weight before and one year after pregnancy are also shown. The GDMDM2 group had gained more weight since the birth of their first child compared with the other groups (P < 0.05). By definition, all women had one child born in 1984–1985 but for the whole group, the mean age of the first child was 19 (range 15–30), second child 15 (range 2–28), third child 12 (range 2–20) and fourth child 12 (6–16) years, respectively.

image

Figure 2. Weight gain for women in the three groups from first child until year 2000. Self-reported data of weight before and one year after each pregnancy are shown in the figure (six women had more than three children).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We do not know to whether our study represents a biased population. Women with GDM might be more anxious to be checked or might be unwilling to appear for further screening.

With these limitations, as far as we are aware, this is the longest follow up study of the natural development of GDM over time. Women who were diagnosed with GDM had developed type 2 diabetes mellitus in about one-third of all subjects in a more or less linear fashion over more than a 15 year period. Even if each and every of the 58 women had participated and none of the remaining women turned out to have type 2 diabetes mellitus, the percentage of the type 2 diabetes mellitus had still been 17%versus 0% in the control group. In the literature, incidence data vary from 3% up to 50% with follow up periods ranging from a few months up to 12 years10,16–18. Numerous factors can explain this variation, and in addition to design differences, genetic variation or diagnostic criteria and clinical treatment policies may play a part. This makes a critical comparison between results difficult. GDM women who developed type 2 diabetes mellitus had a significantly higher BMI and almost significantly higher waist/hip ratio at follow up, which indicates an association with a metabolic syndrome19,20. Insulin concentrations, taken before and after 2 hours under our conditions did not yield any significant additional information as one-third of the subjects had already developed type 2 diabetes mellitus and three of these 10 women were already on insulin treatment.

For design reasons, we do not have information about the anthropometry of the GDM women and the controls before the index pregnancy and do not know whether such information would have had predictive value.

With menopause there is a shift in the fat distribution and a development towards a metabolic syndrome21,22. GDM and non-GDM women were of the same mean age and the percentage of women who had entered menopause did not differ significantly between groups.

Women who develop GDM during their pregnancy will be monitored in close contact with their maternity clinic and under specialist supervision. The advice given about diet control and need to prevent excessive weight gain may result in a lower weight gain during pregnancy than in other pregnant women23,24. However, when the child is delivered, most GDM women will not be followed up, neither at the maternity units nor in the primary health care system, once blood glucose levels have returned to normal. It has been argued however that screening for GDM is not justified, mainly because it causes psychosocial problems and will not result in any change in behaviour25. Although weight control of GDM women during pregnancy is well monitored, our study demonstrates that GDM women will gain more weight in the long run than controls, once the child has been delivered. For these women, it is unlikely that they will contact the medical health care system until they develop overt signs of type 2 diabetes mellitus.

Although all GDM women were probably well informed about their diabetic condition during the pregnancy, 54% of them claimed that they had no idea that they might run a higher risk than others to develop clinically manifest type 2 diabetes mellitus at any time after delivery. We do not know whether this represents a selective memory, inadequate follow up within the health care system or lack of appropriate clinical routines. In some countries, such as Sweden, there are recommendations that these women should be followed up annually by glucose tests, but in clinical reality, this does not seem to happen often as most women will not seek medical care until they develop clinical symptoms and rapidly develop other priorities, once their child has been born.

It is obvious that a strict programme to identify women at high risk to develop type 2 diabetes mellitus, as indicated by their GDM, could be of importance to postpone the well-known complications of type 2 diabetes mellitus. Two recent studies indicate that a modest weight reduction has a profound effect in prevention of the development of clinical manifest type 2 diabetes mellitus 26,27. Future type 2 diabetes could not be predicted by our data on family history (data not shown) and although the GDM women who developed type 2 diabetes mellitus had statistically significant higher BMI (27.4 versus 24.6 kg/m2), it is difficult to use this for a clinical decision. The diabetic women in our study had a mean BMI of 27.4 kg/m2, a level which already indicates a pronounced risk for obesity-related diabetic complications. Preventive long term follow up programmes for GDM women thus seem essential to develop10.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors would like to thank Lena Mannström Svensson for her enthusiastic support and help in this study.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Rössner S. Short communication: pregnancy, weight cycling and weight gain in obesity. Int J Obes 1992;16: 145147.
  • 2
    Öhlin A, Rössner S. Development on body weight during and after pregnancy. In: BjörntorpP, RössnerS, editors. Obesity in Europe 88, Proceedings of the 1st European Congress on Obesity, 5–6 June 1988, Stockholm, Sweden. Paris , London : John Libbey, 1989: 115120.
  • 3
    Öhlin A, Rössner S. Maternal body weight development after pregnancy. Int J Obes 1990;14: 159173.
  • 4
    Öhlin A, Rössner S. Trends in eating patterns, physical activity and sociodemographic factors in relation to postpartum body weight development. Br J Nutr 1994;71: 457470.
  • 5
    Rössner S, Öhlin A. Pregnancy as a risk factor for obesity: lessons from the Stockholm Pregnancy and Weight Development Study. Obes Res 1995;3(Suppl 2):267275.
  • 6
    Öhlin A, Rössner S. Factors related to body weight changes during and after pregnancy: the Stockholm Pregnancy and Weight Development Study. Obes Res 1996;4(3):271276.
  • 7
    Gabbe SG. Gestational diabetes mellitus. N Engl J Med 1986;315: 10251026.
  • 8
    Guttorm E. Practical screening for diabetes mellitus in pregnant women. Acta Endocrinol (Copenhagen) 1974;75(182):1124.
  • 9
    Metzger BE, Coustan DR. Summary and recommendations of the fourth international workshop-conference on gestational diabetes mellitus. Diabetes Care 1998;21: B161B167.
  • 10
    Dornhorst A., Rossi M.. Risk and prevention of Type 2 diabetes in women with gestational diabetes. Diabetes Care 1998;21(2):B43B49.
  • 11
    Sepe SJ, Connell FA, Geiss LS, Teutsch SM. Gestational diabetes: incidence, maternal characteristics and perinatal outcome. Diabetes 1985;34(2):1316.
  • 12
    Catalano PM, Roman NM, Tyzbir ED, Merritt AO, Driscoll P, Amini SB. Weight gain in women with gestational diabetes. Obstet Gynecol 1993;81: 523528.
  • 13
    Jacobson JD, Cousins L. A population-based study of maternal and perinatal outcome in patients with gestational diabetes. Am J Obstet Gynecol 1989;161: 981986.
  • 14
    Cnattingius S, Ericson A, Gunnarskog J, Kallen B. A quality study of a medical birth registry. Scand J Soc Med 1990;18(2):143148.
  • 15
    WHO Study GroupDiabetes mellitus. Word Health Organ Tech Rep Ser 1985;727: 1113.
  • 16
    Damm P, Kuhl C, Bertelsen A, et al. Predictive factors for the development of diabetes in women with previous gestational diabetes mellitus. Am J Obstet Gynecol 1992;167: 607.
  • 17
    Lam KS, Li DF, Lauder IJ, et al. Prediction of persistent carbohydrate intolerance in patients with gestational diabetes. Diabetes Res Clin Pract 1991;12: 181186.
  • 18
    Metzger BE, Cho NH, Roston SM, et al. Prepregnancy weight and anteparum insulin secretion predict glucose tolerance five years after gestational diabetes mellitus. Diabetes Care 1993;16: 15981605.
  • 19
    Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37: 15951607.
  • 20
    Meigs JB, D'Agostino RB, Wilson PWF, Cupples LA, Nathan DM, Singer DE. Risk variable clustering in the insulin resistance syndrome: the Framingham Offspring Study. Diabetes 1997;46: 15941600.
  • 21
    Matthews KA, Meilahn E, Kuller LH, Kelsey SF, Caggiula AW, Wing RR. Menopause and risk factors for coronary heart disease. N Engl J Med 1989;321: 641646.
  • 22
    Poehlman ET, Toth MJ, Gardner AW. Changes in energy balance and body composition at menopause: a controlled longitudinal study. Ann Intern Med 1995;123: 673675.
  • 23
    Snyder J, Grey Donald K, Koski KG. Predictors of infants' birth weight in gestational diabetes. Am J Clin Nutr 1994;59: 1040910414.
  • 24
    Rea A, Bond D, Evans S, Noth F, Roberman B, Walters B. A randomised controlled trial of dietary energy restriction in the management of obese women with gestational diabetes. Aust N Z J Obstet Gynaecol 2000;40(4):416422.
  • 25
    Jarret RJ. Should we screen for gestational diabetes?. BMJ 1997;315(7110):736739.
  • 26
    Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346(6):393403 (February 7).
  • 27
    Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344(18):13431350.May 3