Letters

Outcome of pregnancy in women with insulin dependent diabetes

Centralisation of care leads to better outcome

Editor–The finding of a persistently poor outcome of pregnancy in women with insulin dependent diabetes in two (northern) English regions is an important statement of the problem.^{1 2} Both studies provide figures and show outcomes that are no different from those widely reported in the past. Unfortunately, neither give evidence of any degree of centralisation of obstetric or diabetic care, with on-site neonatal intensive care, although this is a proved means of improving the outcome of pregnancy for diabetic mothers.³ The St Vincent declaration guidelines on the outcome of pregnancy, referred to in the accompanying editorial (p 263), are based on the Scandinavian reports held up as examples of good practice and state that "an interdisciplinary team should provide centralized diabetic pregnancy care in a hospital treating at least 20-30 cases a year. Pregnant diabetic patients should regularly visit the centre, before, during, and after the pregnancy."

The combined diabetes pregnancy clinic at the Royal Maternity Hospital in Belfast has existed for over 40 years, and outcome audit has shown the value of this approach.⁴ In an audit of over 800 pregnancies in diabetic mothers identified in Northern Ireland over the past 10 years the perinatal mortality in mothers cared for throughout at this centralised clinic during 1985-95 was 27/1000, compared with 70/1000 for patients referred later in pregnancy and 33/1000 for those mothers cared for in other maternity hospitals in Northern Ireland.

Overall perinatal mortality for the whole population of Northern Ireland (1.5 million) during this decade was 9.3/1000 total births, so that even at a centralised clinic there is still an increased risk in diabetic pregnancy. Centralisation of care and improved cooperation among the obstetricians and diabetes physicians within a health region will lead to a better outcome.⁵

David Hadden, Honorary professor of endocrinology,^a Anthony Traub, Consultant obstetrician ^a

^a Diabetes Pregnancy Clinic, Royal Maternity Hospital, Belfast BT12 6BA

1. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, et al. Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study. BMJ 1997;315:275-8. (2 August.) [Abstract/Free Full Text]
2. 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. BMJ 1997;315:279-81. (2 August.) [Abstract/Free Full Text]
3. Harley JMG, Montgomery DAD. Management of pregnancy complicated by diabetes. BMJ 1965;i:14-6.
4. Traub AI, Harley JMG, Cooper TK, Maguiness S, Hadden DR. Is centralized hospital care necessary for all insulin-dependent pregnant diabetics? Br Obstet Gynaecol 1987;94:957-62.
5. Hadden DR. Diabetes in pregnancy: past, present and future. In: Dornhorst A, Hadden DR. Diabetes and pregnancy: an international approach to diagnosis and management. Chichester, Wiley, 1996:3-21.

Rate of congenital malformations is almost certainly gross underestimate

Editor–Casson et al make several misrepresentations in their paper on the outcomes of pregnancy in women with insulin dependent diabetes.¹ In the United Kingdom the confidential inquiry into the outcome in babies of diabetic mothers was conducted over the period 1979-80 and included Scotland and Northern Ireland. Thus the perinatal mortality was for the whole of the United Kingdom. The Office for National Statistics (formerly the Office of Population Censuses and Surveys) provides data for only England and Wales. When data for Scotland and Northern Ireland were added the population perinatal mortality was 14.9/1000 for 1979-80.

The rate of congenital malformations available from the Office for National Statistics is almost certainly a gross underestimate, since reporting is voluntary. During 1960-77 the Liverpool and Bootle Congenital Abnormalities Registry (Eurocat) surveyed for congenital malformations and detected 7580 malformed babies (3.2%) out of a population of 236 443; for 1981 the Office of Population Censuses and Surveys gave a figure of 2.1%. The congenital malformation rate for the current study was 9.7%, a figure not very different from that found in the Rigshospitalet in Copenhagen up to 1978 (7.6%)² or the United Kingdom study for 1979-80 (7.1%).

The definition of a malformation is difficult, especially when it is minor and causes minimal interference with the individual's life. When malformations that alter quality of life, require corrective surgery, or cause death are considered, comparisons between different series show that the malformation rate has not changed either in the United Kingdom or in other European countries.^{3 4 5}

While the malformation rate among babies of mothers who had been diagnosed as diabetic before becoming pregnant remains high, it probably is not 10 times higher than that among babies in our background population. The United Kingdom, unlike the Scandinavian countries, has no obligatory reporting system for babies of diabetic mothers, and this makes analysis and examination of trends extremely difficult. Thus in the United Kingdom it is difficult to assess whether the targets set out in the St Vincent declaration are being approached, let alone met.

Clara Lowy, Reader in medicine ^b

^b Department of Endocrinology, Diabetes and Metabolic Medicine, UMDS, St Thomas's Hospital, London SE1 7EH

1. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, et al. Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study. BMJ 1997;315:275-8. (2 August.)
2. Molsted-Pedersen L. Pregnancy and diabetes, a survey. Acta Endocrinol 1980;94(suppl 238):13-9.
3. Hanson U, Persson B. Outcome of pregnancies complicated by type 1 insulin-dependent diabetes in Sweden. Am J Perinatol 1993;10:330-5. [Medline]
4. Nielsen GL, Nielsen PH. Outcome of 328 pregnancies in 205 women with insulin-dependent diabetes mellitus in the county of Northern Jutland from 1976-90. Eur J Obstet Gynecol Reprod Biol 1993;50:33-8. [Medline]
5. Cnattingius S, Berne C, Nordstrom ML. Pregnancy outcome and infant mortality in diabetic patients in Sweden. Diabetic Med 1994;11:696-700. [Medline]

Any improvement in metabolic control during pregnancy reduces risk of adverse fetal outcome

Editor–Unexpectedly high perinatal mortality and morbidity have been reported recently in studies of pregnant women with insulin dependent diabetes.^{1 2} Metabolic control as measured by haemoglobin A1c concentration was suboptimal, and better periconceptional regulation might have improved outcome. Neither of the studies, however, specifically addressed the predictive value of various levels of haemoglobin A1c concentration on the outcome of pregnancy.

We have reported a similar outcome in pregnancies of women with insulin dependent diabetes,³ but our data included haemoglobin A1c concentrations before conception and during the first trimester for 60 and 171 pregnancies respectively. We performed logistic regression analyses, including haemoglobin A1c values as predictor variables for adverse outcome (spontaneous abortion and lethal or severe malformations).⁴ We observed a consistent increasing risk of adverse outcome with increasing haemoglobin A1c concentration. The associations were most pronounced in the high range, but there was no indication of a threshold below which metabolic control seemed to be of minor importance.

Comparison of outcome in pregnancies in which metabolic control remained unchanged or deteriorated with that in pregnancies in which metabolic control improved yielded an odds ratio of an adverse outcome in the former group of 3.1 (95% confidence interval 0.99 to 9.6). When the outcome in pregnancies in which haemoglobin A1c values remained above 7.9% was compared with that in pregnancies in which the values either changed from above to below 7.9% or remained under 7.9% the odds ratio of an adverse outcome increased to 3.8 (1.2 to 12).

Our data strongly indicate a clinically significant and consistent relation between haemoglobin A1c concentration and adverse outcome in pregnancies of women with insulin dependent diabetes, without any indication of a cut off value below which further improvement in the haemoglobin A1c concentration is of minor importance. Although our study may have been subject to confounding by indication, the message to clinicians and pregnant diabetic women seems clear: any improvement in metabolic control at any time during pregnancy seems to reduce the risk of adverse fetal outcome.

Gunnar Lauge Nielsen, Consultant,^c Henrik Toft Sørensen, Associate professor,^d Jørn Olsen, Professor,^e Per Hostrup Nielsen, Consultant,^f Svend Sabroe, Assistant professor ^g

^c Department of Obstetrics, 9000 Aalborg, Denmark, ^d Medical Department V, Aarhus University Hospital, 8000 Aarhus, Denmark, ^e Danish Epidemiology Science Centre, Steno Institute of Public Health, University of Aarhus, ^f Department of Obstetrics, Aalborg Hospital, 9000 Aalborg, ^g Institute of Epidemiology and Social Medicine, University of Aarhus

1. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, et al. Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study. BMJ 1997;315:275-8. (2 August.)
2. 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. BMJ 1997;315:279-81. (2 August.)
3. Nielsen GL, Nielsen PH. Results of 312 pregnancies among white class B-F mothers in northern Jutland from 1976-1992. Dan Med Bull 1994;41:115-8. [Medline]
4. Nielsen GL, Sørensen HT, Nielsen PH, Sabroe S, Olsen J. HbA1c in IDDM pregnancies as predictor of adverse fetal outcome. Acta Diabetol 1997;34:217-22. [Medline]

Management of impaired glucose tolerance in pregnancy needs study

Editor–Casson et al report the rates of pregnancy loss and congenital malformation and measures of fetal growth in a population of insulin dependent diabetic women.¹ Hawthorne et al conclude that diabetic pregnancy remains a high risk state, with perinatal mortality and fetal malformation rates much higher than those in the background population despite intensive management of diabetes.²

In diabetic antenatal clinics most of the women seen are those with impaired glucose tolerance in pregnancy (gestational impaired glucose tolerance) rather than those with insulin dependent diabetes. This is an important group of women, whose management should also be evaluated. Abnormal glucose tolerance in pregnancy is widely reported to be associated with increased perinatal morbidity and major congenital abnormality.³ Various management regimens have been advocated for women with impaired glucose tolerance which aim to return glucose control to normal and thereby improve perinatal outcome.

A randomised controlled study of the management of women with impaired glucose tolerance compared standard antenatal care with intensified care including dietary advice, capillary glucose monitoring, and serial ultrasonography.⁴ Although the babies in the group receiving standard care had a significantly higher birth weight, this was offset by delivery occurring one week later than in the intensified care group. Thus this study, rather than supporting intensified antenatal care, did quite the opposite. It therefore seems that intensive management regimens for impaired glucose tolerance have been introduced prematurely.

A prospective randomised study has now been initiated at Bradford Royal Infirmary and the united Leeds teaching hospitals to test the hypothesis that managing women with impaired glucose tolerance without monitoring their glucose concentrations will not lead to a deterioration in perinatal outcome. This regimen is being compared with one in which impaired glucose tolerance is monitored and treated, with the aim of achieving normal plasma glucose concentrations. Fetal outcome in this study will be assessed by factors including length of stay on the special care baby unit, premature delivery, birth trauma, and number of capillary samples obtained. Measurements of maternal outcome include the incidence of caesarean section and induction of labour, number of antenatal visits, number of capillary samples obtained, and requirements for introduction of insulin.

We hope that this study will go some way towards answering the question of whether impaired glucose tolerance in pregnancy needs to be managed as aggressively as it tends to be and whether the St Vincent declaration applies to this group of patients as well as to those with insulin dependent diabetes mellitus.

Lynne Rogerson, Specialist registrar,^h Karen Bancroft, Senior registrar ^h

^h Bradford Royal Infirmary, Bradford, West Yorkshire BD9 6RJ

1. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, et al. Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study. BMJ 1997;315:275-8. (2 August.)
2. 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. BMJ 1997;315:279-81. (2 August.)
3. Hod M, Merlob P, Friedman S, Schoenfield A, Ovadia J. Gestational diabetes mellitus. A survey of perinatal complications in the 1980s. Diabetes 1991;40:74-8.
4. Li DF, Wong VCW, O'Hoy KMKY, Yeung CY, Ma HK. Is treatment needed for mild impairment of glucose tolerance in pregnancy? A randomised controlled trial. Br J Obstet Gynaecol 1987;94:851-4. [Medline]

Authors' reply

Editor–Although the perinatal mortalities that Hadden and Traub cite show the benefit of centralised care for pregnant women with diabetes, the figure from the combined diabetes pregnancy clinic at the Royal Maternity Hospital in Belfast (2.7%) and that for other maternity hospitals in Northern Ireland (3.3%) both fall within the 95% confidence interval for our study (1.7% to 5.5%). Our study included some hospitals that ran combined clinics and others that did not. The geographical distribution of these hospitals is such that it would be logistically difficult for all women in the area to be cared for at a combined clinic with the criteria for experience quoted by Hadden and Traub.

We fully acknowledge Lowy's point, that data on congenital malformations from the Office for National Statistics are underestimates, but, as discussed in our paper, these are the only contemporary figures available for comparison. For our study we used the criteria given by the Office for National Statistics to exclude minor malformations from the analysis.

Nielsen et al draw attention to the possibility of using haemoglobin Alc concentrations as predictors of the outcome of pregnancy in diabetic women. Information on diabetic control, treatment regimens, and complications of diabetes was collected as part of our study, and these data will form the basis of further analyses.

Rogerson and Bancroft make the point that infants of women with impaired glucose tolerance during pregnancy may also be at increased risk of poor outcome. We look forward to hearing the outcome of the study they describe and the contribution it will make to evidence based practice in this area.

I F Casson, Consultant diabetologist,ⁱ C A Clarke, Emeritus professor,^j M Stanisstreet, Senior lecturer,^j C V Howard, Head of research group,^k O McKendrick, Research associate,^k S Pennycook, Research nurse,^k D van Velzen, Professor,^k P O D Pharoah, Professor of public health,^l M J Platt, Senior lecturer,^l S Walkinshaw, Consultant in maternal and fetal medicine ^m

ⁱ Broadgreen Hospital, Liverpool L14 3LD, ^j School of Biological Sciences, University of Liverpool, Liverpool L69 3BX, ^k Fetal and Infant Pathology, University of Liverpool, ^l Department of Public Health, University of Liverpool, ^m Women's Hospital, Liverpool L8 7NJ


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Relevant Article

Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study

I F Casson, C A Clarke, C V Howard, O McKendrick, S Pennycook, P O D Pharoah, M J Platt, M Stanisstreet, D van Velszen, and S Walkinshaw
BMJ 1997 315: 275-278. [Abstract] [Full Text]