Does maternal diabetes delay fetal pulmonary maturity?

Does maternal diabetes delay fetal pulmonary maturity? Jeanna M. Piper, MD, and Oded Langer, MD

San Antonio, Texas OBJECTIVE: The purpose of our study was to determine whether a relationship exists between diabetic glucose control and the immaturity rate of fetal pulmonary maturity test results. STUDY DESIGN: One thousand consecutive women who had fetal pulmonary maturity testing before delivery were studied. Diabetic patients were categorized according to whether their diabetes was well controlled or poorly controlled, then compared to the nondiabetic population on the basis of amniocentesis results. RESULTS: Pregnant women with poorly controlled diabetes had a significantly higher risk of immature fetal lung profiles at amniocentesis than did pregnant women without diabetes. Pregnant women with wellcontrolled diabetes were not significantly different from the nondiabetic population. There was no difference between gestational and pregestational diabetic pregnancies. CONCLUSIONS: Timing of fetal pulmonary maturation is linked to the level of maternal glucose control in diabetic pregnancies. Adequate glucose control may lower the risk of fetal pulmonary immaturity to that seen in the nondiabetic population. (AM J OBSTET GVNECOL 1993; 168:783-6.)

Key words: Diabetes in pregnancy, fetal pulmonary maturity, amniocentesis Maternal diabetes has been associated with an increase in the risk of pulmonary immaturity, even in term gestations. J.g The mechanism of this delay, however, is poorly understood. The maternal hyperglycemia and resultant fetal hyperinsulinemia of diabetic pregnancies may be responsible for the delayed maturation. One recent study of pulmonary maturity testing in diabetic pregnancies reported a higher mean blood glucose in patients with an immature lung profile than in those with mature test results. JO That study was limited by sample size and did not categorize patients with diabetes by level of glycemic control to allow direct comparisons. We hypothesize that the rate of pulmonary maturation in diabetic pregnancies is related to the level of glucose control and that adequate glucose control will allow maturation to proceed at the same rate as in nondiabetic pregnancies.

Material and methods Between January 1986 and January 1992 all fetal pulmonary maturity test results at Medical Center Hospital, San Antonio, Texas, were reviewed. Antepartum, delivery, and neonatal information was obtained. Gestational age was based on the mother's last menstrual

From the Department of Obstetrics and Gynecology, The University of Texas Health Science Center at San Antonio. Received for publication May 28, 1992; revised August 4, 1992; accepted October 20, 1992. Reprint requests: Jeanna M. Piper, MD, Department of Obstetrics and Gynecology, The University of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78284-7836. Copyright © 1993 by Mosby-Year Book, Inc. 0002-9378/93 $1.00 + .20 6/1/43618

period confirmed by early exam, fetal heart tones by Doppler at 10 to 12 weeks' gestation, or ultrasonography (when available) or was revised based on ultra sono graphic examination. All patients without preexisting diabetes underwent screening for gestational diabetes during their pregnancy. Patients were classified as having diabetes if either gestational or pregestational diabetes was present. The patients without diabetes formed the control group with the exception that women who had chronic hypertension or preeclampsia were excluded because of possible alteration of pulmonary maturation by their hypertensive disorder. For the patients with diabetes, level of glucose control was established as described later in this article. Before October 1989 blood glucose tests were performed weekly at clinic visits with fasting and 2-hour postprandial glucose values obtained. Since that time glucose monitoring has been performed at home seven times daily with a memory reflectance glucometer. With both methods, all values were averaged to obtain an overall mean glucose value. A mean blood glucose level of :s; 105 mg/dl was considered to be well controlled, and a mean glucose level of > 105 was considered to be poorly controlled. II Diabetic pregnancies without glucose data were excluded. Fetal pulmonary maturity test results in patients with diabetes were considered mature only if phosphatidylglycerol was present. Nondiabetic pregnancies were considered mature with either phosphatidylglycerol present or a lecithin/sphingomyelin ratio ~ 2/1. Phosphatidylglycerol testing was performed by the immunologic technique (Amniostat FLM, Irvine Scientific, Santa Ana, Calif.) that recognizes the presence of phosphatidylglycerol at a concentration of 0.5 f1g/ml. Leci-

784

Piper and Langer

March lYY3

Am

Table I. Gestational age breakdown by study group Women with diabetes Weeks' gestation

Well controlled

<34 34-36.9 37-37.9 38-38.9 ;?:39

12 27 35 31 35 140

TOTAL

I

Poorly controlled

Women without diabetes

15 44 15 37 29 149

204 176 48 32 80 540

thin/sphingomyelin ratios were determined by the method of Gluck and Kulovich,12 with values reported as 1/1, l.5/1, or ~2/l. Neonatal respiratory status was defined as presence or absence of hyaline membrane disease. The diagnosis of hyaline membrane disease was based on chest x-ray, oxygen, and positive pressure requirements beyond 12 hours of life with no other attributable lung problem. For the purpose of analysis, in women with multiple gestations the first infant was uniformly selected. In patients with multiple amniocenteses, only the one closest to delivery was entered into the analysis. Gestational ages were categorized into five groups for analysis «34, 34 to 36.9, 37 to 37.9,38 to 38.9, and ~39 weeks). That categorization was selected to give the most information possible on the early term gestations (37 to 39 weeks). Statistical analysis was performed with X2 and Fisher exact test, with p < 0.05 considered significant. Odds ratio analysis with 95% confidence limits was also performed. Results Over this 73-month period, 1000 women had one or more amniocenteses with fetal pulmonary maturity testing. Of those, 112 (11%) were exclud(·d because of inadequate amniotic fluid sample, delivery outside our system, or incomplete data. Fifty-seven (6%) were excluded for maternal hypertensive disorders. Two were excluded for lack of glucose data. Eight hundred twenty-nine pregnancies remained for analysis: 289 women were classified as having diabetes (106 pre gestational, 183 gestational), and the remaining 540 women without diabetes or hypertension were used as the control group. The mean interval from amniocentesis to delivery was 5.7 days for the diabetic group and 5.0 days for the nondiabetic group. The amniocenteses in the preterm patients were most commonly performed for preterm labor or premature rupture of membranes. In the term gestations the main indication for amniocentesis was for elective delivery: in the diabetic group, poor glucose control and macrosomia were other indications.

J Obslet Gynecol

One hundred forty of the women with diabetes had good glucose control with an overall mean of 96.2 ± 5.8 mg/dl. The other 149 women with diabetes had poor glucose control with an overall mean of 124.6 ± 18.2 mg/dl. The breakdown of the group with well-controlled diabetes, poorly controlled diabetes, and no diabetes by gestational age is shown in Table I. The patients in the self-monitored group were compared with the clinic-monitored group with no differences found in their pulmonary maturity rates; based on this finding the groups were combined for the remainder of the study. Furthermore, analysis of the self-monitored group yielded results comparable with those of the combined group. The rates of immature fetal lung profiles at amniocentesis were analyzed by gestational age categories. There were no differences in immature lung profile rates between pre gestational and gestational diabetic pregnancies. The rates of immature test results for patients with poorly controlled diabetes, well-controlled diabetes, and no diabetes are displayed in Fig. 1. The asterisks designate immature lung profile rates in the diabetic groups, which are significantly different from the gestational age-matched, nondiabetic pregnancy category. Significant differences were found in the poorly controlled diabetic group as compared with the nondiabetic group at 34 to 36.9, 37 to 37.9, and ~ 39 weeks, with a trend toward significance in the other two groups (p < 0.075). The only other significant difference was between the well-controlled diabetic and poorly controlled diabetic pregnancies in the 37 to 37.9-week category. Odds ratio analysis for relative likelihood of immature profiles at amniotic fluid analysis is displayed in Table II. Significant 95% confidence limits were found for comparisons of pregnant women with poorly controlled diabetes to pregnant women without diabetes at 34 to 37.9 weeks and between poorly controlled diabetic and well-controlled diabetic pregnancies at 37 to 37.9 weeks. In the preterm gestations « 37 weeks) hyaline membrane disease occurred in 30 of 384 nondiabetic pregnancies (8%) and in 6 of 97 diabetic pregnancies (6%). There were no significant differences found between pregnancies with well-controlled diabetes, poorly controlled diabetes, and no diabetes at any gestational age. There was a trend toward higher rates of hyaline membrane disease at < 34 weeks' gestation in the poorly controlled diabetic pregnancies (4 of 14, 29%) when compared with the nondiabetic pregnancies (25 of 207, 12%). Comment Our study analyzed the relationship between the results of fetal pulmonary maturity testing and diabetic

Piper and Langer

Volume 168, Number 3, Part I Am J Obstet Gynecol

785

% IMMATURE _ ...

100%

POOR CONTROL

~ GOOD CONTROL _

NON-DIABETIC

SO%

60%

40%

20%

0%

<34

34-36.9

37-37.9

3S-3S.9

>39

GESTATIONAL AGE IN WEEKS Fig. 1. Rates of immature fetal lung profiles at amniocentesis are shown for poorly controlled diabetic, well-controlled diabetic, and nondiabetic pregnancies. Asterisks, Diabetic group significantly different!.p < 0.05) from corresponding nondiabetic gestational age category.

Table II. Likelihood of immature fetal lung profiles at amniocentesis Poorly controlled diabetes vs no diabetes

Well-controlled diabetes vs no diabetes

Poorly controlled diabetes vs well-controlled diabetes

Weeks' gestation

Relative risk

95% Confidence interval

Relative risk

95% Confidence interval

Relative risk

95% Confidence interval

<34 34-36.9 37-37.9 38-38.9 >38.9

5.97 2.91 16.50 3.58 3.96

0.79-36.9 1.40-6.08* 3.21-95* 0.78-18.6 0.83-19.2

4.69 2.33 3.81 1.43 2.45

0.61-31.10 0.94-5.76 9.94-16.60 0.23-9.10 0.48-12.70

1.27 1.25 4.33 2.50 1.62

0.02-107 0.43-3.7 1.02-19.3* 0.61-10.9 0.33-8.2

*Significant value.

glucose control in 289 pregnant women with diabetes as compared with 540 pregnant women without diabetes. Statistical analysis showed significantly higher rates of pulmonary immaturity in the poorly controlled diabetic pregnancies compared with nondiabetic controls. In contrast, there were no significant differences between pregnant women with well-controlled diabetes and the nondiabetic population. Previous studies have attempted to evaluate the issue of the effect of diabetic control on fetal pulmonary maturity test results. 9. 10. 13·17 These studies evaluated a group of women whose diabetes was aggressively managed either alone,9. 10. 13·15 compared with a nondiabetic group, 16 or compared with a diabetic group not aggres-

sively managed. 17 Although two of these studies reported mean blood glucose data, 10. 14 there was no attempt to classifY patients' diabetes as being well or poorly controlled. A unique feature of our study was the use of a large group of women with diabetes with available blood glucose data that allowed us to categorize the pregnancies having good or poor glycemic control based on a defined threshold (105 mg/dl) and to analyze the effects of this threshold of glucose control on fetal pulmonary maturity test results. The importance of our findings are twofold. One aspect is whether diabetic pregnancies require more aggressive use of amniocentesis for confirmation of fetal pulmonary maturity than nondiabetic pregnancies. The

786

Piper and Langer

studies that have mandated this testing considered all diabetic pregnancies together, with a resultant spectrum of glucose control. I, 18 Our study used a strict definition of adequate glucose control to identity the patient with well-controlled diabetes as having risks similar to the patients without diabetes, In light of these findings the well-controlled diabetic pregnancy might be managed by the same protocol as the nondiabetic pregnancies with regard to the need for fetal pulmonary maturity testing, The other aspect of these findings relates to the mechanism of timing of pulmonary maturation in diabetic pregnancies, There appears to be an association between elevated maternal mean blood glucose level and immature profiles at amniotic fluid analysis, Although it did not reach significance, there also appears to be an association between the level of maternal glucose control and the rate of hyaline membrane disease. This association represents another area in which stringent glucose control may have a significant impact on the perinatal outcome of diabetic pregnanCles. In summary, our study suggests that delay in fetal pulmonary maturation is linked to maternal glucose control in diabetic pregnancies and that adequate glucose control will reduce the risk of pulmonary immaturity to that seen in nondiabetic pregnancies. Even so, almost 20% of pregnant women with well-controlled diabetes will have immature pulmonary maturity testing at 37 to 37.9 weeks; 10% remain immature beyond 38 weeks. Thus the physician must be aware that there is still a risk of fetal lung immaturity at term gestation. Further evaluation is needed to clarity the relationship between the lung profile and respiratory distress syndrome in diabetes pregnancies. Further study is needed to clarity the mechanism that controls the timing of pulmonary maturation in pregnancy and to determine the means by which that timing is altered in pregnant women with poorly controlled diabetes. REFERENCES

1. Kulovich MV, Gluck L. The lung profile. II. Complicated pregnancy. AM J OBSTET GYNECOL 1979;135:64-70. 2. Gluck L, Kulovich MB. Lecithin/sphingomyelin ratios in amniotic fluid in normal and abnormal pregnancy. AM J OBSTET GYNECOL 1973; 115:539-46.

March 1993 Am J Obstet Gyneco1

3. Cunningham MD, Desai NS, Thompson SA, Green JM. Amniotic fluid phosphatidylglycerol in diabetic pregnancies. AM J OBSTET GYNECOL 1978;131:719-24. 4. Obladen M, Merritt TA, Gluck L. Acceleration of pulmonary surfactant maturation in stressed pregnancies: a study of neonatal lung effluent. AM J OBSTFT GYNECOL 1979; 135: 1079-85. 5. Tsai MY, Shultz EK, Williams PP, et al. Assay of disaturated phosphatidylcholine in amniotic fluid as a test of fetal lung maturity: experience with 2000 analyses. Clin Chern 1987; 33:1648-51. 6. Singh EJ, Mejia A, Zuspan FP. Studies of human amniotic fluid phospholipids in normal, diabetic, and drug-abuse pregnancies. AM J OBSTET GYNECOL 1974;199:623-9. 7, Tsai MY, Shultz EK, NeisonJA. Amniotic fluid phosphatidylglycerol in diabetic and control pregnant patients at different gestational lengths. AM J OBSTET GYNECOL 1984; 149:388-92. 8. Aubry RH, Rourke JE, Almanza R, Cantor RM, Van Doren JE. 'Ibe lecithin/sphingomyelin ratio in a high-risk obstetric population. Obstet Gynecol 1976;49:21-7. 9. Ylinen K. High maternal levels of hemoglobin Alc associated with delayed fetal lung maturation in insulin-dependent diabetic pregnancies. Acta Obstet Gynecol Scand 1987;66:263-6. 10. Ojomo EO, Coustan DR. Absence of evidence of pulmonary maturity at amniocentesis in term infants of diabetic mothers. AM J OBSTET GYNECOL 1990;163:954-7, 11. Langer 0, Levy J, Brustman L, Anyaegbunam A, Merkatz R, Divon M. Glycemic control in gestational diabetes mellitus - how tight is tight enough: small for gestational age versus large for gestational age? AM J OBSTET GYNECOL 1989; 161 :646-53. 12. Gluck L, Kulovich MY, The lung profile. I. Normal pregnancy. AM J OBSTET GYNECOL 1979;135:57-63. 13. Curet LB, Olson RW, Schneider JM, Zachman RD. Effect of diabetes mellitus on amniotic fluid lecithin/sphingomyelin ratio and respiratory distress syndrome. AM J OBSTET GYNECOL 1979;135:10-3. 14. Ferroni KM, Gross TL, Sokol RJ, Chik L. What affects fetal pulmonary maturation during diabetic pregnancy? AM J OBSTET GYNECOL 1984;150:270-4. 15. Dudley DK, Black DM. Reliability of lecithin/sphingomyelin ratios in diabetic pregnancy. Obstet GynecoI1985;66: 521-4. 16. McMahan MJ, Mimouni F, Miodovnik M, Hull WM, WhitsettJA. Surfactant associated protein (SAP-35) in amniotic fluid from diabetic and nondiabetic pregnancies. Obstet Gynecol 1987;70:94-8. 17. Mimouni F, Miodovnik M, Whitsett JA, Holroyde JC, Siddiqi TA, Tsang RC. Respiratory distress syndrome in infants of diabetic mothers in the 1980s: no direct adverse effect of maternal diabetes with modern management. Obstet Gynecol 1987;69:191-5. 18, Hallman M, Teramo K. Amniotic fluid phospholipid profile as a predictor of fetal maturity in diabetic pregnancies. Obstet Gynecol 1979;54:703-7.