© 2001 by the American Diabetes Association, Inc.
Pregnancy Outcome in Type 1 Diabetic Women With Microalbuminuria
1 Department of Endocrinology, the
OBJECTIVETo determine the influence of microalbuminuria on pregnancy outcome in women with type 1 diabetes. RESEARCH DESIGN AND METHODSThis prospective cohort study took place at the Obstetric Clinic at National University Hospital, Copenhagen, from January 1996 to February 2000. All Caucasian women with type 1 diabetes, unselected from the eastern part of Denmark, with a living fetus before 17 weeks of gestation on admission were asked to participate. For women with more than one delivery in the study period, only the first pregnancy was included. Of the remaining 246 women, 240 (98%) entered the study. They were categorized according to their urinary albumin excretion (normal urinary albumin excretion, <30 mg/24 h; microalbuminuria, 30300 mg/24 h; or diabetic nephropathy, >300 mg/24 h) before pregnancy or in the first trimester. RESULTSA total of 203 women (85%) had normal urinary albumin excretion, 26 (11%) had microalbuminuria, and 11 (5%) had diabetic nephropathy. Mean HbA1c at 26 weeks was 7.5% (SD 1.1), 8.1 (0.9), and 8.8 (1.3) (P < 0.001), respectively. Of all deliveries in women with normal urinary albumin excretion, microalbuminuria, and diabetic nephropathy, 35, 62, and 91% (P < 0.001), respectively, were preterm, and 2, 4, and 45% (P < 0.001), respectively, were small-for-gestational-age infants. Preeclampsia developed in 6, 42, and 64% of the women (P < 0.001), respectively. Category of urinary albumin excretion (P < 0.01) and HbA1c at 26 weeks (P < 0.05) were independently associated with preterm delivery. CONCLUSIONSThe prevalence of preterm delivery is considerably increased in women with microalbuminuria, mainly caused by preeclampsia. Classification according to urinary albumin excretion and metabolic control around the time of conception are superior to the White classification in predicting preterm delivery in women with type 1 diabetes.
Abbreviations: BP, blood pressure
Microalbuminuria (urinary albumin excretion in the range of 30300 mg/24 h) is an early manifestation of diabetic microvascular disease (1,2) and leads to diabetic nephropathy in type 1 diabetes. It is associated with slightly elevated blood pressure (BP) within normal range and subclinical edema due to universal vascular leakage of albumin, and it predicts overt diabetic nephropathy with persistent proteinuria and hypertension (1,3). Increased perinatal morbidity associated with preterm delivery is a major problem in pregnancy complicated by type 1 diabetes. Up to one-third of infants of mothers with type 1 diabetes are delivered preterm (4,5,6,7), and the prevalence of preeclampsia characterized by hypertension, proteinuria, and edema is 1020%. Diabetic nephropathy (urinary albumin excretion >300 mg/24 h) present at conception is a major contributor to increased perinatal morbidity and mortality (8). Superimposed preeclampsia, which is associated with preterm delivery and intrauterine growth retardation (4,9,10), is a significant problem in women with diabetic nephropathy (4,8,11). Diabetic women with high early pregnancy proteinuria of 190499 mg/day have been reported to have an increased risk of developing preeclampsia, similar to that in women with diabetic nephropathy (11). In a previous study, we have demonstrated that type 1 diabetic women with microalbuminuria before pregnancy have a substantially increased risk of developing preeclampsia (12). However, whether pregnancy in type 1 diabetic women with microalbuminuria is associated with increased preterm delivery and perinatal morbidity has not been described in a cohort of women. The aim of the present study was to determine the influence of microalbuminuria on fetal outcome and maternal complications in pregnant women with type 1 diabetes.
All Caucasian women with type 1 diabetes before gestation who were admitted to our obstetric clinic before 17 weeks of gestation with a living fetus were asked to participate. The women were admitted unselected to the obstetric clinic from January 1996 to November 1999 from a well-defined geographical area covering 2,000,000 inhabitants. Women who had miscarriages ( 22 weeks of gestation; n = 25) were excluded. If a woman had more than one delivery (n = 19) during the inclusion period, only the first pregnancy was included. Of the remaining 246 consecutive pregnancies, 240 (98%) were entered into the study after informed consent. The research was in accordance with the Helsinki Declaration and approved by the local ethic committees. Values of urinary albumin excretion and BP within 2 years before pregnancy and the HbA1c value closest to the time of conception (26 weeks of gestation) were obtained from local records. Urinary albumin excretion, BP, and HbA1c were measured by various methods in the various local diabetes centers. The women were categorized according to their level of urinary albumin excretion, calculated as geometric mean of two to three measurements before pregnancy. Normal urinary albumin excretion was defined as <30 mg/24 h, microalbuminuria was defined as urinary albumin excretion of 30300 mg/24 h (13), and diabetic nephropathy was defined as urinary albumin excretion >300 mg/24 h. Of the 240 women, 195 (81%) had at least two available measurements of urinary albumin excretion both before and early in pregnancy. When comparing urinary albumin excretion early in pregnancy with urinary albumin excretion before pregnancy, 182 of these women remained in the same category, 6 women decreased one category, and 7 women increased one category. Consequently, if values of urinary albumin excretion were not available before pregnancy, urinary albumin excretion in the first trimester was used to classify the women (n = 31). Data regarding kidney function and BP status at baseline are summarized in the following sections.
Normal urinary albumin excretion
Microalbuminuria
Diabetic nephropathy
The women were classified according to the White classification, which is traditionally used in obstetrics to grade the severity and duration of diabetes. White class B was defined as onset of type 1 diabetes after 19 years of age and duration of diabetes <10 years. White class C was defined as onset of type 1 diabetes between 10 and 19 years of age or duration of diabetes 1019 years. White class D was defined as onset of type 1 diabetes before 10 years of age or duration of diabetes Management of the diabetic pregnancies was performed according to the routine procedures of the Obstetric Clinic at the National University Hospital (Rigshospitalet) (16). The women were asked to perform home blood glucose measurements at least four times daily during their pregnancy and to adjust insulin doses accordingly in order to maintain preprandial blood glucose levels between 3 and 6 mmol/l. They visited the obstetric clinic and/or the local diabetes center every 1 or 2 weeks during pregnancy. Labor was routinely induced at 3740 weeks of gestation based on individual evaluation. Single measurements of 24-h urinary albumin excretion, HbA1c, and office BP were performed at least five times throughout the pregnancy (weeks 10, 14, 20, 28, and 34). Urine samples were analyzed for albumin by enzyme-linked immunosorbent assay (13) or by a turbidimetric method using the same antibodies and buffers. HbA1c was analyzed by high-performance liquid chromatography (17) or by antibody immunoassay (normal range 4.16.4%). Office BP was measured in a sitting position with the arm at heart level after 510 min rest.
The diagnosis of preeclampsia in women with normal urinary albumin excretion or microalbuminuria was based on the presence of office BP >140/90 mmHg (three measurements) accompanied by proteinuria >0.3 g/24 h (two urine samples) later than 20 weeks of gestation (proteinuria of 0.3 g/24 h is equivalent to a urinary albumin excretion of 190 mg/24 h). The diagnosis of preeclampsia in women with diabetic nephropathy was based on the same findings as well as a sudden increase of In addition to insulin and hypertensive drugs, seven women were taking levothyroxine, one was taking lithium, two were taking antidepressive drugs, one was taking hydrocortisone because of Addisons disease, and one was taking an antiepileptic drug during pregnancy. Preterm delivery (<37 weeks of gestation completed) included spontaneous delivery and delivery based on obstetric indications such as uncontrolled hypertension, severe symptoms of preeclampsia, or macrosomia. Small for gestational age was defined as <10th centile for gestational age and gender for Danish standard population. Perinatal mortality was defined as fetal death later than 22 weeks of gestation or within 1 week after delivery. Major congenital malformations were considered those responsible for death, those causing a significant future disability, or those requiring major surgery for correction (18). During the study period, no fetuses with congenital malformations were aborted as late abortions.
Statistical analysis
A total of 203 women (85%) had normal urinary albumin excretion, 26 (11%) had microalbuminuria, and 11 (5%) had diabetic nephropathy at baseline (Table 1). There was a trend toward longer duration of diabetes and higher BMI, HbA1c, and BP with increasing category of urinary albumin excretion.
Preterm delivery and preeclampsia occurred with substantially higher frequency in women with microalbuminuria or diabetic nephropathy (Table 2). When excluding the group with diabetic nephropathy from analysis, women with microalbuminuria still had a higher prevalence of preterm delivery compared with women with normal urinary albumin excretion (P < 0.05). The increased prevalence of preterm delivery was mainly due to the higher prevalence of preeclampsia in women with microalbuminuria or diabetic nephropathy (Fig. 1). A total of 35% of the women with microalbuminuria and 54% of the women with diabetic nephropathy delivered preterm due to preeclampsia. Preterm delivery due to other causes (spontaneous or for obstetrical reasons) was comparable in the three groups.
A substantial number of women with microalbuminuria delivered before 34 weeks of gestation. A total of 23 women with normal urinary albumin excretion at baseline developed a urinary albumin excretion >30 mg/24 h without hypertension late in pregnancy. The prevalence of preterm delivery before 34 weeks of gestation in these women was 4%. Increased prevalence of intrauterine growth retardation was seen in women with diabetic nephropathy but not in women with microalbuminuria (Table 2). Diabetic nephropathy was also associated with increased prevalence of jaundice requiring treatment, and there was a tendency toward a higher proportion of development of tachypnea requiring assisted ventilation (Table 2). Perinatal mortality and major congenital malformations were comparable in the three groups. Using multivariate logistic regression analysis, the baseline variables of urinary albumin excretion, systolic BP, HbA1c, White classification, age, BMI, parity, and smoking were tested as predictors of preterm delivery. Increased category of urinary albumin excretion (P < 0.01) and high HbA1c at 26 weeks of gestation (P < 0.05) were independently associated with preterm delivery. The same baseline variables were analyzed as predictors of preeclampsia. Increased category of urinary albumin excretion (P < 0.0001), high systolic BP within normal range (P < 0.001), and nulliparity (P < 0.05) were independently associated with development of preeclampsia. The incidences of preterm delivery in normoalbuminuric women according to tertiles of HbA1c at 26 weeks of gestation were 29, 37, and 48%, respectively (<0.05). The incidences of preterm delivery and preeclampsia in White classes B, C, and D + R were 36, 35, and 41% and 6, 6, and 14%, respectively.
The presence of microalbuminuria before pregnancy and higher HbA1c at 26 weeks of gestation were associated with increased risk of preterm delivery in women with type 1 diabetes. The substantially increased prevalence of preterm delivery with the increasing degree of albuminuria was caused by higher prevalence of preeclampsia. The incidences of development of jaundice or transitory respiratory insufficiency in infants and of small-for-gestational-age infants were considerably increased in women with diabetic nephropathy but not in women with microalbuminuria. This is the first study to demonstrate that presence of microalbuminuria before pregnancy in an unselected cohort of women with type 1 diabetes is associated with increased risk of preterm delivery. Our findings are in accordance with Combs et al. (11), who reported an increased prevalence of preeclampsia and preterm delivery in women with early pregnancy proteinuria of 190499 mg/24 h. However, measurement of urinary albumin excretion is much more sensitive compared with measurement of proteinuria. A substantial number of the women with microalbuminuria increased their protein excretion to the nephrotic range during pregnancy. This is in accordance with Biesenbach et al. (19), who found that nephrotic proteinuria developed in 4 of 12 women with microalbuminuria during pregnancy. One might speculate whether antihypertensive treatment during pregnancy should be indicated with increasing albumin excretion. We chose to initiate antihypertensive treatment in normotensive women with proteinuria exceeding 3 g/24 h during pregnancy. After delivery, the albumin excretion has been shown to return to prepregnancy levels (12,19). The prevalence of preterm delivery in our study was comparable to the literature (4), although some authors have reported a higher prevalence of women delivering at term (5,6,7). Preterm delivery before 34 weeks of gestation is associated with increased prevalence of perinatal morbidity, especially in diabetic pregnancy, in which fetal lung maturation is often delayed. In our study, as much as 25% of the infants of women with microalbuminuria were delivered before 34 weeks of gestation. Increased urinary albumin excretion and high HbA1c level were independent predictors of preterm delivery. The association between high levels of HbA1c and development of preeclampsia has been described earlier in type 1 diabetes (11,20). However, our study is the first to document a correlation between elevated HbA1c at 26 weeks of gestation and increased prevalence of preterm delivery. Increased urinary albumin excretion, together with early high systolic BP within the normal range and nulliparity, was also an independent predictor of preeclampsia. The presence of hypertension is of importance for the prediction of preeclampsia (21,22). However, the clinical value of BP as an early predictor of preeclampsia has been demonstrated to be of minor importance because of a great overlap in BP in women in whom preeclampsia developed and in women who remained normotensive (23). The White classification was not an independent predictor of preterm delivery or preeclampsia. Others have demonstrated a strong association between both the prevalence of preterm delivery and the prevalence of preeclampsia and the increasing severity of diabetes according to White classification (4,6,7). However, our study is the first to include urinary albumin excretion and HbA1c at 26 weeks of gestation in addition to the White classification. In women with microalbuminuria or diabetic nephropathy, development of preeclampsia was the most important single cause of preterm delivery. In women with normal urinary albumin excretion, we also found a surprisingly high prevalence of preterm delivery associated with poor metabolic control. This major clinical problem requires further investigation. Treatment with ACE inhibitors before pregnancy along with tight metabolic control in women with diabetic nephropathy, resulting in urinary albumin excretion <500 mg/24 h, has been reported to have a prolonged protective effect on maternal renal function and results in a favorable maternal-fetal outcome without development of preeclampsia (24). Unfortunately, we are unable to support this report, because all three women with diabetic nephropathy and four of the seven women with microalbuminuria treated with ACE inhibitors before pregnancy demonstrated progression of proteinuria during pregnancy. The fetal outcome was comparable to that of the remaining women in the two groups. One explanation for this difference could be that the metabolic control in our unselected material was significantly poorer than the highly selected material of eight women obtaining both excellent metabolic control and a good response of ACE inhibition on proteinuria (24). In conclusion, the prevalence of preterm delivery is considerably increased in women with microalbuminuria, mainly due to preeclampsia. Classification according to urinary albumin excretion and metabolic control around conception are superior to the White classification in predicting preterm delivery in type 1 diabetes.
This study received funding from The Danish Diabetes Association.
Address correspondence and reprint requests to Dr. Elisabeth R. Mathiesen, Department of Endocrinology, PE-2132, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. E-mail: em{at}rh.dk. Received for publication 14 March 2001 and accepted in revised form 24 May 2001. A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.
This article has been cited by other articles:
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||