Q J Med 1999; 92: 451-454
© 1999 Association of Physicians
Pregnancy outcome in women with insulin-dependent diabetes mellitus complicated by nephropathy
1 From the Department of Diabetic Medicine and 2 Clinical Trials Unit, University Hospital Trust, and 3 Department of Obstetrics, Birmingham Womens Hospital Trust, Birmingham, UK
Received 27 October 1998 and in revised form 7 May 1999
Dr F.P. Dunne, Department of Diabetic Medicine, University Hospital Trust (Selly Oak), Raddlebarn Road, Birmingham B29 6JD
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
We retrospectively analysed pregnancy complicated by diabetic nephropathy in patients attending a University teaching hospital (1990–97), to examine fetal/maternal outcomes. Fetal outcomes included early intrauterine deaths, stillbirths, neonatal/perinatal mortality, size for gestational age, malformations, and need for neonatal unit care. Maternal outcomes included change in frequency of hypertension or severe proteinuria, serum creatinine data, and caesarean section rate. There were 21 pregnancies in 18 women, resulting in 21 live infants. Neonatal mortality (RR 10, 95%CI 0–3.9), perinatal mortality (RR 5, 95%CI 0–3.3) and congenital malformations (RR 5.0, 95%CI 0.3–26.3) were greater than in the background population. At delivery, 76% of babies were appropriate in size for gestational age; 57% were preterm, all of whom required neonatal unit care. The caesarean section rate was 90.5% vs. 20% in the background population (RR 4.5, 95%CI 3.4–5.0) (p<0.05). Hypertension frequency (p<0.001) and high-grade proteinuria (p<0.05) increased from booking to delivery. Although the take-home baby rate was 90%, perinatal/neonatal mortality, congenital malformations and caesarean sections, in addition to maternal morbidity, were significantly higher in women with diabetic nephropathy than in the background population.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Diabetic nephropathy is the leading cause of end-stage renal disease in the Western world, and is also linked to cardiovascular disease as the major cause of deaths related to diabetes mellitus.1,2 Diabetic nephropathy is present in 5–10% of diabetic pregnancies,3 and is a challenging and serious complication with regard to maternal health and fetal well-being. In the past, fetal and maternal outcomes were poor,4 but the introduction of perinatology and better treatment of hypertension and diabetes has lead to an improvement.5,6 Pregnancies complicated by nephropathy continue to be associated with a high complication rate. Intrauterine growth restriction (IUGR) occurs almost twice as commonly as in the general population, and its incidence is related to the severity of nephropathy and hypertension. Caesarean section rates are greater compared to the general population, and preterm deliveries are frequent, with almost 30% occurring before 34 weeks.5–8 Although neonatal mortality from prematurity continues to decline, preterm delivery still carries considerable short-term morbidity from respiratory distress, hyperbilirubinaemia and other complications. The maternal hazards include renal failure during or after pregnancy, superimposed pre-eclampsia, and the risk of eventual morbidity or death from macrovascular disease. The objectives of this retrospective study were to determine the effect of diabetic nephropathy on fetal outcome and the incidence of maternal obstetric complications and to evaluate pregnancy-related changes in serum creatinine in women with diabetic nephropathy.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Information was collected on pregnancy outcome in all women with insulin-dependent diabetes mellitus (IDDM) complicated by diabetic nephropathy who attended our joint diabetic-antenatal clinic over a 7-year period. Only women with pre-existing diabetic nephropathy and a pregnancy that continued beyond the first trimester were included. The renal disease was diagnosed on the basis of three separate dipstick-positive urine tests for protein in the absence of menstruation and urinary tract infection prior to pregnancy, and confirmed by 24-h urinary protein excretion prior to pregnancy, or by biopsy in two cases.
Serum creatinine was used as an index of renal function. Renal insufficiency was regarded as mild when serum creatinine was <100 µmol/l, moderate when serum creatinine was 101–150 µmol/l and severe when serum creatinine was >151 µmol/l (normal pregnancy range 40–95 µmol/l). In all cases proteinuria was present. Proteinuria was classified by quantitative and semiquantitative measurements: low grade if dipstick testing for protein showed a value of 1+ or 2+ or the level was 300–3000 mg/24h; high grade if the dipstick showed a value of 3+ or 4+ or the level exceeded 3000 mg/24h. Hypertension was defined as a mean arterial blood pressure (MAP) [systolic pressure+ (diastolic pressurex2)/3] above 105 mmHg on three occasions. For the purposes of evaluating pregnancy-related changes in serum creatinine, blood pressure, and proteinuria, we used the first measure of these indices when performed at booking as the base-line value. HbA1 measurements using an electro-endosmosis method (before September 1995) and HbA1c measurements using a latex enhanced turbidometric immunoassay (after September 1995) were used as an index of glycaemic control (normal range 3.5–7.5% mean±2 SD). Delivery before 37 weeks of gestation was classified as preterm. An infant was considered small for gestational age (SGA) if he/she was <10th centile, large for gestational age (LGA) if >90th centile, and appropriate for gestational age (AGA) if between the 10th and 90th centiles for gestational age at delivery (using Gardner-Pearson charts corrected for fetal sex).
Fetal outcome parameters were rates of miscarriage (early fetal death before 24 weeks), stillbirth (late fetal death after 24 weeks), early neonatal deaths (deaths from delivery until 6 completed days of life) and perinatal mortality (stillbirths and deaths in the first week of life), prevalence of congenital malformations, size for gestational age and admission to neonatal unit (NNU) care.
Maternal outcomes were caesarean section rate, change in frequency of hypertension and severe proteinuria during pregnancy and follow-up data on effect of pregnancy on renal function (serum creatinine). Results were compared to that for the background hospital population over the same time period.
For the statistical analysis of neonatal and perinatal mortality rates, and the congenital malformation and caesarean section rates, relative risks (RR) and 95%CIs were used. For changes in MAP and proteinuria, a McNemar test was used, while for changes in serum creatinine a Sign test was used.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Data were recorded on 21 pregnancies in 18 women with pre-existing insulin-dependent diabetes complicated by diabetic nephropathy and retinopathy. The total number of live births in the background population during the same time was 40 176. The mean age of mothers at the time of delivery was 26.5 (range 21–40) years and the mean duration of diabetes was 19.5 (range 12–30) years. At the time of booking, 17 patients were classified as having mild, three as moderate and one as severe nephropathy. The first antenatal visit occurred during weeks 6–12, and the mean booking glycated haemoglobin was 9.7% (range 6.7–16.7%) (normal range 3.5–7.5%). Glycaemic control was worse in those with moderate to severe nephropathy in all three trimesters, as compared to those with mild nephropathy (Table 1
).
|
The mean weight of live infants was 2429 g (range 985–4140 g). Sixteen (76%) were AGA, two (9.5%) were LGA and three (14%) were SGA. Twelve (57.2%) were delivered preterm (including all those in moderate/severe group) because of antepartum haemorrhage due to placental abruption (one case) or due to worsening proteinuria and hypertension (11 cases). All preterm deliveries required admission to NNU care. There were two neonatal deaths, one following a severe placental abruption and one due to respiratory distress syndrome (both in those with mild nephropathy) but no stillbirths. The neonatal mortality rate was much greater than in the background hospital population (RR 10, 95%CI 0–3.9), as was the perinatal mortality (RR 5, 95%CI 0–3.3). Only one neonate was born with congenital anomalies (skeletal) which included a cleft palate, a hemivertebra and absent rib. The take-home baby rate was 90% (18/21).
There was a high frequency of obstetric complications. One woman had a sudden antepartum haemorrhage due to placental abruption, while 50% of pregnancies were complicated by worsening proteinuria and hypertension and/or fetal distress requiring preterm delivery. The caesarean section rate was high at 90.5%, 4.5 times higher than in the background population (RR 4.5, 95%CI 3.4–5.0) (p<0.05).
Hypertension (mean arterial pressure >105 mmHg) frequency rose from 11% at baseline (booking visit) to 83% prior to delivery (p=0.001), while high-grade proteinuria (3+ 4+ >3 g/24h) rose from 19% to 47% over the same time (p<0.05). Serum creatinine increased from a mean of 88.3 µmol/l (43–304 µmol/l) at baseline to a mean of 111.1 µmol/l (range 54–436 µmol/l) prior to delivery. Using paired samples in nine women, the levels prior to delivery were significantly greater than at booking (p<0.001). These women were followed for up to 48 months, and there was no statistical evidence of a difference between the follow-up serum creatinine and that prior to delivery (p=0.31) (Table 2). One patient has reached end-stage renal disease (ESRD) and is receiving haemodialysis.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
In this population, perinatal and neonatal mortality rates were all much greater than in the background population, as were the rates for caesarean section and congenital malformations, and the study highlights the problems confronting clinicians who care for pregnant women with IDDM and nephropathy. Despite this, successful pregnancy is feasible, with a take-home baby rate of 90%.
We have no information as to whether these pregnancies were planned, but clearly most women had not established good diabetic control before conception, as indicated by the booking glycated haemoglobin levels, or did not achieve normal control during the pregnancy. This poor glycaemic control in the first trimester is likely to be a contributing factor to the higher congenital malformation and perinatal mortality rates. In view of the poor control, it is surprising that congenital anomalies were not more frequent, although they were similar to those reported by Casson in 1997.9 We know that with preconceptual diabetes care we can reduce the rates for congenital malformations and perinatal mortality to that of the background population, and we must endeavour to promote it among this high-risk population. The neonatal and perinatal mortality rates in the background population are also higher than national figures, as this unit is a tertiary referral centre with a large fetal medicine department, and therefore receives referrals with known abnormalities for intra-uterine intervention or delivery and neonatal care. The mean weight of live infants was 2429 g (range 985–4140 g). Sixteen (76%) were AGA, two (9.5%) were LGA and three (14%) were SGA. Although 76% of babies were born AGA, this may reflect two disease processes, i.e. diabetes with a tendency to produce LGA babies, and renal impairment with a tendency to cause IUGR, resulting in SGA babies.
Obstetric complication rates were high, as illustrated by the development of probable pre-eclampsia (worsening hypertension and proteinuria) in 50% of cases necessitating delivery by caesarian section. Against a background of diabetic nephropathy, it may be difficult to diagnose pre-eclampsia confidently, as worsening diabetic nephropathy will produce similar degrees of hypertension and proteinuria. Significant hypertension and high-grade proteinuria increased from booking to delivery, as did serum creatinine. Despite this, the mean gestation was 34.7 (range 29–38) weeks, somewhat greater than the 31 weeks reported by Mackie et al. in 1995.10 The caesarian section rate was 4.5 times greater than the background population, but this is not unusual, and similar figures have been reported by other authors.11
Based on the follow-up data from nine women, pregnancy was associated with some deterioration in renal function prior to delivery (p<0.001) but no further change in the follow-up period (p=0.31) of up to 48 months (Table 2
). Other studies also suggest that pregnancy does not adversely affect the rate of decline of renal function.5,10,12–14 However, in view of the small numbers in this study, in addition to the lack of a control population and pre-pregnancy data, no firm conclusion can be reached. Twenty years ago medical opinion was very different, to the extent that Pederson concluded that diabetic nephropathy was a contraindication to pregnancy.4 With new technology and increasing experience, the outlook for these women and their offspring has improved.5,15 Indeed, the outcome of pregnancy in this series is similar to outcomes reported by Jones et al. in 1996 for pregnancies complicated by nephropathy due to other aetiologies.16
This study has demonstrated that a successful outcome is feasible in pregnancies complicated by diabetic nephropathy with a take-home baby rate of 90%. Despite the good take-home baby rate, diabetic nephropathy remains a high-risk state for mother and baby, with the risk of perinatal and neonatal mortality, congenital malformations and caesarean section all greater than in the background population.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Stephenson JM, Kenny S, Stevens LK, Fuller JH, Lee E. WHO Multinational Study Group. Proteinuria and mortality in diabetes: The WHO multinational study of vascular disease in diabetes. Diabetic Med 1995; 12:149–55.[ISI][Medline]
2. Rossing P, Rossing K, Jacobsen P, Parving HH. Unchanged incidence of diabetic nephropathy in IDDM patients. Diabetes 1995; 44:739–43.[Abstract]
3. Landon MB, Gabbe SG. Diabetes mellitus and pregnancy. Obstet Gyn Clin North Am 1992; 19:633–48.
4. Pedersen J, Molsted-Pedersen L, Andersen B. Assessors of fetal perinatal mortality in diabetic pregnancy. Diabetes 1974; 23:302–5.[Medline]
5. Kitzmiller JL, Brown ER, Phillippe M, Stark AR, Acker D, Kaldang A, Singh S, Hare JW. Diabetic nephropathy and perinatal outcome. Am J Obstet Gynecol 1981; 141:741–51.[Medline]
6. Kimmerle R, Zass RP, Cupisti S, Somville T, Bender R, Pawlowski B, Berger M. Pregnancies in women with diabetic nephropathy; long term outcome for mother and child. Diabetologia 1995; 38:227–35.[Medline]
7. Grenfell A, Brudenell JM, Doddridge MC, Watkins PJ. Pregnancy in diabetic women who have proteinuria. Q J Med 1986; 228:379–86.
8. Reece E, Coustan DR, Hayslett JP, Holford T, Coulehan J, O'Connor TZ, Hobbins JC. Diabetic nephropathy: pregnancy performance and fetomaternal outcome. Am J Obstet Gynecol 1988; 159:56–66.[Medline]
9.
Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, Platt MJ, Stanisstreet M, van Velszen D, Walkinshaw S. Outcome of pregnancy in insulin-dependent diabetic women: results of a five year population cohort study. Br Med J 1997; 315:275–8.
10. Mackie ADR, Doddridge MC, Gamsu HR, Brudenell JM, Nocolaides KH, Drury PL. Outcome of pregnancy in patients with insulin dependent diabetes mellitus and nephropathy with moderate renal impairment. Diabetic Med 1996; 13:90–6.[Medline]
11.
Hawthorne G, Robson S, Ryall EA, Sen D, Roberts SH, Ward Platt MP. Prospective population based survey of outcome of pregnancy in diabetic women: results of the Northern Diabetic Pregnancy Audit 1994. Br Med J 1997; 315:279–281.
12. Reece EA, Winn HN, Hayslett JP, Coulehan J, Wan M, Hobbins JC. Does pregnancy alter the rate of progression of diabetic nephropathy? Am J Perinatol 1990; 7:193–7.[Medline]
13. Kaaja R, Sjoberg L, Hellsted T, Immonen I, Sane T, Teramo K. Long term effects of pregnancy on diabetic complications. Diabetic Med 1996; 13:165–9.[ISI][Medline]
14. Miodovnik M, Rosenn BM, Khoury JC, Grigsby JL, Siddiqi TA. Does pregnancy increase the risk for development and progression of diabetic nephropathy. Am J Obstet Gynecol 1996; 174:1180–91.[Medline]
15. Jovanovic R, Jovanovic L. Obstetric management normoglycaemia is maintained in diabetic pregnant women with vascular compromise. Am J Obstet Gynecol 1984; 149:617–23.[ISI][Medline]
16.
Jones DC, Hayslett JP. Outcome of pregnancy in women with moderate or severe renal insufficiency. N Engl J Med 1996; 335:226–32.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||