GD: Basic Treatment Protocols: 

Research Citations

by KMom

Copyright 1998-2002 KMom@Vireday.Com. All rights reserved.


DISCLAIMER: The information on this website is not intended and should not be construed as medical advice. Consult your health provider. This particular web section is designed to present more than one view of a controversial subject, pro and con. It should be re-emphasized that nothing herein should be considered medical advice.

Contents

Purpose of This Web Section

The purpose of the section on Basic Treatment Protocols follows below. These are the medical study references for that section. The list of references was so long it had to have its own file, unlike in most FAQ sections on this site. However, for clarity, the 'purpose statement' of that section is also repeated here, somewhat abbreviated for space.

Many more references were used for the Treatment Protocols section than could be cited due to length considerations. The following listings represent some of the most important studies used to summarize the most common treatment regimens, plus protocol differences that can exist. A significant amount of detail was included for certain studies, especially those on particularly controversial topics, or those where a closer look at the data reveal important questions. This makes some entries rather long, but hopefully will help the reader clarify issues or determine which studies would be most important to look up themselves. Do your own readings on the subject; most medical studies are really not all that hard to read and interpret. Often, looking at the full medical study can be especially revealing; the conclusions and summaries in some abstracts are not always accurate or justified when the full research is examined, while others present an even more compelling case than their abstracts reflect.

Do be careful how you draw conclusions from medical research. Don't be too reverent; even medical research is subject to biases and sweeping generalizations and misjudgments. On the other hand, turn the same eye of doubt and questioning on claims of 'alternative' medical views. Whenever possible, review not just a few studies but a broad spectrum of studies, pro and con. It is the nature of medical research to have conflicting information and strong differences of opinion. Remember also that medical treatment evolves over time; procedures and approaches that used to be well-accepted are now totally discredited and unused, while new approaches may offer vast improvement in prognosis or might be found eventually to cause more harm than good. Take an overall, long-sighted view, and look carefully without jumping to preliminary conclusions. Happy researching!

Reminder of the purpose of the Basic Treatment Protocols section:

The Basic Treatment Protocols section is an introduction to some of the main issues that surround the treatment and care of gd pregnancies. However, it must be noted that some of this is especially controversial, and the authorities themselves are currently involved in trying to determine the most beneficial level of care and intervention in gd pregnancies. The way YOUR provider decides to handle your gd could differ immensely from the way another provider might handle the same case.

The top priority of this section of the gd faq is ONLY to give basic information on the most basic treatment issues and mention some of the controversies; another section will outline the different controversies in more detail and present arguments, pro and con (this section is under development). Kmom urges readers to do their own research, consult their providers and to become a PARTNER in their own care. Remember, NONE of this is medical advice, just a summary of the most common treatment courses and a brief mention of some of the controversies in treatment choices.

References

General GD Info/Treatment Issues

American Diabetes Association Position Statement: Gestational Diabetes Mellitus. Diabetes Care. Volume 21: Supplement 1, 1998. http://www.diabetes.org/diabetescare/supplement198/s60.htm

Official 1998 position statement on the definition, detection, diagnosis, and therapeutic strategies for gd. 

Carr, DB and Gabbe, S. Gestational Diabetes: Detection, Management, and Implications. Clinical Diabetes. Jan-Feb, 1998.  16(1):4-24. http://www.diabetes.org/clinicaldiabetes/v16n1j-f98/pg4.htm

Outstanding article summarizing gd testing, management, and even some of the controversies involved in gd, though from a traditional medical approach. Excellent overview, but very technical.

Stephenson, M.J. Gestational Diabetes Mellitus. Canadian Family Physician. 39:745-8, April 1993.

A must-read article for those serious about understanding gd treatment options. Covers fairly both philosophies of treatment, both the maximum and minimum schools of management. An excellent overview of the controversies. 

Diabetes and Pregnancy, ACOG Technical Bulletin (An Educational Aid to Obstetrician-Gynecologists), #200--Decemeber 1994.

The definitive summary from the American College of Obstetricians and Gynecologists for pre-existing diabetes in pregnancy as well as gestational diabetes. Technical but readable. Represents the current standard of care. 

Proceedings of the Fourth International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes Care. August 1998. Volume 21 (Supplement 2). http://www.diabetes.org/DiabetesCare/Supplement298/B161.htm

The entire issue of this Diabetes Care journal is devoted to the recommendations and papers from the 4th International Conference on GD. Many important issues are discussed. Very conservative views; tends to conveniently ignore the criticisms and studies that contradict their conclusions. One-sided but still valuable.

Jovanovic-Peterson, Lois, M.D. with Morton B. Stone. Managing Your Gestational Diabetes. Minneapolis: Chronimed Publishing, 1994. To order, write P.O. Box 59032, Minneapolis, MN 55459-9686, or call 1-800-848-2793.

A good introduction to gd issues by one of the leading researchers in the field, who also happens to be diabetic (Type I) and a mother herself. Treatment guidelines in this book are quite conservative and her writings contain some patronizing and fat-phobic statements. Good introduction to the conservative approach to gd. She is a very prolific writer in the field; many more items are available under her name.

Gestational Diabetes: What to Expect. The American Diabetes Association, Inc. Alexandria, Virginia: American Diabetes Association, 1992. To order, write to the ADA, 1970 Chain Bridge Road, McLean, VA 22109-0592, or call 1-800-232-3472. Third Edition (revised in 1997) now available also.

The standard intro to the subject, written by the leading authorities on diabetes. A good, easy-to-read summary of the standard medical approach to gd, very conservative in guidelines, with no discussion of gd controversies. 

Goer, Henci. Obstetric Myths vs. Research Realities. Westport, Connecticut: Bergin and Garvey, 1995. Can be ordered online from http://www.efn.org/~djz/birth/obmyth/

This excellent book reviews common obstetrical practices and analyzes which practices are truly justified by medical research. The Gestational Diabetes chapter examines the history of its treatment, and variations in protocols. She extensively reviews the medical research available on gd and concludes that many common gd protocols are questionable because they do not sufficiently alter outcome but did increase the incidence of cesarean sections and resulting complications.  A must-read.

Goer, Henci. "Gestational Diabetes: The Emperor Has No Clothes." Birth Gazette. Spring 1996: Volume 12, Number 2.

A shorter summary of the gd chapter from the above book. www.gentlebirth.org/archives/gdhgoer.html.

Walkinshaw S.A. Dietary Regulation for 'Gestational Diabetes'. In: Neilson JP, Crowther CA, Hodnett ED, Hofmeyr GJ (eds.) Pregnancy and Childbirth Module of The Cochrane Database of Systematic Reviews, [updated 01 September 1997]. Available in The Cochrane Library [database on disk and CDROM]. The Cochrane Collaboration; Issue 4. Oxford: Update Software; 1997. Updated quarterly.  Abstract available from www.cochrane.org

A review of previous studies that examined dietary treatment vs. no treatment for gd. Concludes that studies do not support the value of primary dietary treatment for gd, but cites a number of methodological problems with previous studies and notes the need for further research with better research design.

Enkin, Murray et al. A Guide to Effective Care in Pregnancy and Childbirth. Second Edition. Oxford: Oxford University Press (Oxford Medical Publications), 1995.

Based on the Cochrane Database of Systematic Reviews, which examined the research of 60 key journals, with an emphasis on the 'gold standard' of research, randomized controlled studies. "Evidence-Based Medicine" at its best. Found significant reason to question the current aggressive approach to gd. "There is no convincing evidence that treatment of women with an abnormal glucose tolerance test will reduce perinatal mortality or morbidity. Trials of dietary regulation... do not demonstrate a significant effect on any outcome, including macrosomia. Trials comparing the use of insulin plus diet with diet alone show a decrease in macrosomia, but no significant effect on other outcomes such as use of caesarean section, the incidence of shoulder dystocia...perinatal mortality...[or] neonatal jaundice or hypoglycaemia." They further go on to make the very strong statement that "The available data provide no evidence to support the wide recommendation that all pregnant women should be screened for 'gestational diabetes,' let alone that they should be treated with insulin. Until the risks of minor elevations of glucose during pregnancy have been established in appropriately conducted trials, therapy based on this diagnosis must be critically reviewed. The use of injectible therapy on the basis of the available data is highly contentious, and in many other fields of medical practice such aggressive therapy without proven benefit would be considered unethical."

Blank A., Grave G.D., Metzger B.E. Effects of Gestational Diabetes on Perinatal Morbidity Reassessed. Report of the International Workshop on Adverse Perinatal Outcomes of Gestational Diabetes Mellitus, December 3-4, 1992. Diabetes Care. 18(1):127-9, January 1995.

Report of some of the findings from the Third International Workshop on GDM. (The Fourth took place in 1997.) A quick summary of some of the perinatal morbidities associated with gd and the serious concerns they may present; notable for acknowledging some of the research controversies such as possible alternative causes for some of the problems, the problems with reproducibility of gd testing, the lack of cost-effectiveness of aggressively trying to reduce macrosomia on a wide scale, and the problems with research design and data of previous studies. Strongly promotes the need for further well-designed research done on a multi-center, multi-ethnic, and multi-national long-term scale.

O'Brien, ME and Gilson, G. Detection and Management of Gestational Diabetes in an Out-of-Hospital Birth Center. Journal of Nurse-Midwifery. 32(2):79-84. March/April, 1987.

Describes a pilot program to treat mild gestational diabetics in an out-of-hospital birth center. Treatment was by nurse-midwives in consultation with physician consultants (severe cases were transferred to the OB). The population served was 95% Hispanic; the incidence of gd was 10%. The c-section rate for the clients treated by midwives was 9%; if the clients transferred to OB care and had c-sections, the midwives' c/s rate was 11%. It is important to note that this rate was MUCH lower than the c-section rates reported by many other studies, which on average range between 20% and 35%, but in some studies have reached even higher.

Javanovic-Peterson, Lois, M.D. The Diagnosis and Management of Gestational Diabetes Mellitus. Clinical Diabetes. pp32-39, March/April 1995.

A very technical journal article covering basic gd information in great detail, including information on hormonal influences. Very conservative guidelines for insulin, and recommends strong caloric restrictions for 'morbidly obese' patients. Very dense reading, but good for those wanting more insight on the conservative view of gd.

Weller, KA. Diagnosis and Management of Gestational Diabetes. American Family Physician. 53(6):2053-7, 2061-2. May 1, 1996.

A review of gd treatment regimens, though it tends to be much less intervention-oriented than many treatment regimens found in endocrinology and obstetrics literature.

Jackson, E.A. et al. Management of Gestational Diabetes by Family Physicians and Obstetricians. Journal of Family Practice. 43(4):383-8, October 1996.

The charts of 813 women with gd were retrospectively examined to see if management practices and outcomes differed between family practice doctors and obstetricians. 33% of OB patients were placed on insulin, while only 24% of FP patients were placed on insulin, even though patients exhibited similar demographics of body mass index, weight gain, gestational week at entrance to care, etc. Even more striking was the difference in c-section rates----OB patients had a c-section rate of 33% while family practice patients had a c-section rate of only 11%. There were no significant differences in neonatal outcome (including macrosomia) between the two groups, despite treatment differences.

Hod, M et al. Gestational Diabetes: Is It a Clinical Entity? Diabetes Reviews. 3(4):602-613, 1995.

A review of the debate over whether gd is really a problem, with a strongly affirmative conclusion about the dangers of gd and the effectiveness of treatment. Advocates lowering treatment thresholds even further. Has the most astounding list of newborn treatment protocols, and some of the most extensive prenatal treatment protocols for the mother, too. Extremely conservative view of gd and gd treatment.

Keen, Harry. Gestational Diabetes: Can Epidemiology Help? Diabetes. December 1991. Volume 40, Supplement 2: 3-7.

Examines some of the history of diabetes and gd diagnosis, with special attention to the problems of a borderline range between normal and clearly diabetic. Notes that severe gd clearly needs treatment but that "there is much less certainty about the hazards, if any, associated with the lesser degrees of glucose intolerance...Attention is drawn to the paucity of evidence linking lesser degrees of glucose intolerance with significant disturbance of pregnancy outcome when confounding variables such as maternal age, adiposity, and parity are allowed for. It is in the area of the detection and treatment of these lesser degrees of glucose intolerance in pregnancy that serious questions of the detriment-to-benefit ratio arise. A population-based multiethnic multicultural inquiry into diagnostic methodology and criteria in pregnancy is proposed...extending...to a controlled clinical trial of the effects of intervention." A fair and unbiased overview of the controversy, history, and implications of gd. Adequately represents both sides. HIGHLY recommended reading.

Nordlander, E et al. Factors Influencing Neonatal Morbidity in Gestational Diabetic Pregnancy. British Journal of Obstetrics and Gynecology. June 1989. 96(6):671-678.

The influence of obstetric factors vs. maternal blood glucose control on neonatal morbidity was examined in 261 women with gd (plus a control group of 218 women without gd). Although the gd group had more morbidity (23%) than the control group (13%), it was found to be mostly due to gestational age at delivery. After correction for this factor, pre-pregnancy maternal weight was the only factor with added significance. "The present study clearly illustrates that other factors besides blood glucose control are of importance for neonatal outcome in gestational diabetic pregnancy."

Gestational Diabetes and High Rates of C-Sections

Naylor, CD et al. Cesarean Delivery in Relation to Birth Weight and Gestational Glucose Tolerance: Pathophysiology or Practice Style? Journal of the American Medical Association. April 17, 1996. 275(15):1165-70.

One of the few studies to adequately address the issue of physician style and perception on the treatment of gd patients. Macrosomia is usually used as the excuse for higher c-section rates in gd moms, but even when macrosomia rates are reduced through treatment, most (not all) studies have found a higher c/s rate anyhow, raising the question of the benefits vs. risks of treatment. This study found that treatment lowered macrosomia rates to the level found in normoglycemic populations, but that treated patients had twice the risk of a c/s anyhow. Only 10.5 % of gd women had macrosomic babies in this study, yet there was a 33% c/s rate. Furthermore, the study examined the outcome of 115 women who had gd by the stricter criteria of Carpenter and Coustan but did not qualify as gd by official NDDG standards and were not treated. The rate of macrosomia in these women was 28.7%, compared with 13.7% for normoglycemic women and 10.5% for 'official' gd moms. Shoulder dystocia, which should have been higher in the 'untreated' group, was not. Their overall c/s rate was higher, though---29.6%, compared to 20.2% for normoglycemic women. However, the c/s rate for non-macrosomic infants in this 'untreated' gd group was similar to that of the controls, but the c/s rate for the macrosomic babies in the 'untreated gd' group was a huge 45.5%! This could reflect labor problems due to untreated macrosomia, or it could reflect simply the tremendous bias of providers about delivery of macrosomic infants, diabetic or not---or it could be both. Among women with 'official' gd who had had a previous c/s, the VBAC rate was an atrocious 18%. Among normoglycemic controls, the rate was 31%; among those with 'untreated' gd, it was 40%. [VBAC success rates are usually about 70% for an overall rate of 30-40% VBACs.] An especially strong indication of the predisposition towards c/s that the simple label of 'gd' may give is shown in the rates of primary (first) c/s among women who'd had babies before. Among multiparous gd women, the primary c/s rate was strongly increased (23%) compared to 6% for women who were normoglycemic and 6% for women with 'untreated' gd. The authors concluded that "while detection and treatment of GDM normalized birth weights, rates of cesarean delivery remained inexplicably high. Recognition of GDM may lead to a lower threshold for surgical delivery that mitigates the potential benefits of treatment."

Coustan, D.R. Management of Gestational Diabetes Mellitus: A Self-Fulfilling Prophecy? [Editorial.] Journal of the American Medical Association. April 17, 1996. 275(15):1199-200.

Commentary on the above article. While still VERY strongly in favor of gd treatment (noting that it 'normalized' macrosomia in the gd population), he still comments on the very high c/s rate often accompanying treatment, even when macrosomia is lowered. "The reduction in macrosomia...was not accompanied by a reduction in cesarean deliveries, presumably because the caregivers viewed these patients differently...one...possible explanation of the increased cesarean delivery rate is our long-standing expectation that women with gestational diabetes mellitus are more likely to need cesarean deliveries, often because the infant is just too big to deliver...given our prior experience, we may be more prone to interpret the clinical situation as requiring cesarean delivery, with the subconscious belief that cesarean delivery is inevitable. This is the trap of the self-fulfilling prophecy. Cesarean delivery has inherent risks, such as infection, blood loss, and injury to adjacent organs. It also carries increased costs for the present pregnancy and possibly for future pregnancies...The authors appropriately warn that apparent increases in interventions, such as cesarean delivery rates, may be related more to the expectations of caregivers than to the clinical indications...The most significant step we can take is to anticipate a vaginal delivery in patients with appropriately treated gestational diabetes mellitus. We need to beware the self-fulfilling prophecy."

Santini, DL and Ales, KL. The Impact of Universal Screening for Gestational Glucose Intolerance on Outcome of Pregnancy. Surgery, Gynecology and Obstetrics. May 1990. 170: 427-436.

Looked at the debate over universal screening and its impact on treatment and pregnancy outcome. Retrospectively studied 1307 pregnancies at Cornell University Medical Center over 5 months, where some providers did universal screening and some did not screen at all. Compared the screened population vs. the unscreened population to see if screening and treatment helped reduce the number of large infants, and how treatment and outcome differed. "The process of screening not only failed to decrease the rate of large infants, but also failed to improve otherwise pregnancy outcomes and was associated with more intensive surveillance during pregnancy and a significantly higher rate of primary cesarean delivery." The c/s rate for those unscreened for gd was 21%; the c/s rate for those screened for gd was 27.6%, and the c/s rate for those screened and treated was 32.5%. The c/s rate for those screened and treated with diet alone was 30%; the c/s rate for those screened and then treated with diet plus insulin was 38.5%. "The process of screening is itself linked with more intensive surveillance during pregnancy...even in the absence of the diagnosis, labeling or treatment of gestational glucose intolerance." Furthermore, only about half of the women labeled as having gd actually met the criteria for having it. The rate of large infants was not significantly different between the screened and unscreened groups. No difference was found, either, in the rate of metabolic complications in newborns. A few explanations could be made; physician style probably impacted the difference in c/s rates some, and the unscreened population tended to be less obese, so perhaps that is why they had less large babies than expected. However, the unscreened population was more likely to be parous and to be older (factors which generally make for bigger babies) and to have private doctors (a factor which would tend to increase the c/s rate, not decrease it). C/S was also more common in women with larger infants in either group (31.5% vs. 23.9%), raising the question of how much is due to actual physical problems of having a larger baby and how much is due to physician bias and interventions for larger babies. Examines the difficulty in setting up a randomized, controlled trial with enough power to determine the value of universal screening and gd treatment to reduce infant size---it would have to have a very large amount of participants. Notes that up to now, screening and treatment both were assumed to have no significant adverse effects, but that this needs to be strongly questioned.

Goldman, M et al. Obstetric Complications with GDM: Effects of Maternal Weight. Diabetes. December 1991. 40 (Suppl. 2):79-82.

Retrospectively assessed 150 women with gd for obstetric complications with special attention to the influence of the mother's weight (305 normoglycemic controls were used for comparison, matched for size etc.). Found that hypertension and pre-eclampsia were more common in gdm, but it was not totally attributable to being overweight. Unlike in most studies, abnormalities of labor, shoulder dystocia, birth trauma, and macrosomia were not more common with gd. However, despite macrosomia not being more common with gd, there were more c-sections in gd patients (35% c/s with gd; 22% without gd). This clearly points out that c/s are more common with the 'gd' label, even when macrosomia (the usual 'cause' given to justify more c/s) is not present. However, the most interesting result was the comparison of the number of elective c/s without labor---there were twice as many in the gdm group (17% vs. 8%). The authors state that "the increase is due to more repeat cesarean sections and more operations performed due to anticipations of macrosomia." When the data was stratified by pre-pregnancy weight, the discrepancy in elective c-sections was even more pronounced. The rate of elective c/s in overweight and obese gd moms was 32.5% overall; 12% for overweight and obese nongd controls overall, vs. 13.7% in 'ideal weight' gd moms and 6.3% in 'ideal weight' controls. In other words, a fat gd mom had 2.4x the risk of being assigned an elective c/s that an ideal weight gd mom had. This shows the bias of the providers----they are less willing to 'allow' a trial of labor and quicker to resort to surgical intervention in women of size, probably because of their presumptions and fears of macrosomia, and the label of 'gd' strongly reinforces that tendency. Fully 1/3 of large gd moms in this study were not given a trial of labor.

Remsberg, KE et al. Diabetes in Pregnancy and Cesarean Delivery. Diabetes Care. September 1999. 22(9):1561-7.

Examined the records of 42, 071 South Carolina births in 1993 to determine the association, if any, between diabetes in pregnancy and c-sections, and whether this is mediated by birth weight (the usual defense for higher c/s rates in diabetic populations). After controlling for numerous confounding factors, cesarean delivery was found to be strongly associated with diabetic births, both when birth weight was taken into account and when it was not. Women with pre-existing diabetes were found to have 6.2x the risk for c/s, and women with gd were found to have 1.7x the risk for c/s. The estimates remained essentially unchanged with or without birthweight in the model.

Nutrition

Major, CA et al. The Effects of Carbohydrate Restriction in Patients with Diet-Controlled Gestational Diabetes. Obstetrics and Gynecology. April 1998. 91(4):600-4.

Compared diet-only gd moms on two different food plans, one low-carb (below 42%) and one higher-carb (45-50%). Higher proportions of protein (25%) and fat (35%) were included to make up the caloric difference. Significant reductions in post-prandial values were seen in the low-carb group, fewer needed insulin (5% vs. 33%), there were fewer large-for-gestational-age infants (9% vs. 42%), and there were fewer c-sections (3% vs. 48%) for 'cephalopelvic disproportion and macrosomia'. 2 subjects had shoulder dystocia, both in the high-carb group, and one child had birth trauma, but the incidence was so small that it did not reach statistical significance between groups. The low-carb group did have 2 subjects with ketones; it was managed by slightly increasing their carb levels. Study was quite small, so the power of its conclusions is small. Notes that while the percentage of LGA infants declined significantly, the overall rate of macrosomia (>4000g) was still high at 26%, with both groups experiencing it. Study speculates that this could be due to levels of fat or protein, or to insulin antibodies in those with insulin. (As usual, a role for simple genetics is completely ignored.) Although careful monitoring of ketones is important, a carb-restricted food plan might be of use in preventing need for insulin, since most previous gd food plans contained 50-60% carbs. Although the issue needs further replication, safety evaluation, and in-depth examination, this study indicates that a lower-carb approach is promising and might be an alternative to hypocaloric approaches. [Note that a too-low carb total can create problems, however. An extremely low-carb approach is NOT appropriate for pregnancy.]

Ratner, RE. Clinical Review 47. Gestational Diabetes Mellitus: After Three International Workshops Do We Know How to Diagnose and Manage It Yet? Journal of Clinical Endocrinology and Metabolism. July 1993. 77(1):1-4.

An excellent if conservative review of gd and many of its controversies. Notes that more than just glucose levels are implicated in the possible morbidities associated with gd. Discusses many of the treatment issues of gd, and notes the need to balance aggressive glycemic control with possible risks to the mother and fetus from overly aggressive treatment. Does favor moderate caloric restriction for obese women, though it notes that this is "one of the most contentious issues in the literature" and that "long-term effects, however, remain unknown and require additional investigation."

Home Glucose Monitoring

Langer O et al. Intensified versus conventional management of gestational diabetes. American Journal of Obstetrics and Gynecology. 1994. 170:1036-47.

Studied 'intensive management' (home monitoring on a glucometer 7x/day) versus 'conventional management' of gd (weekly lab tests and testing strips 4x daily) to see if more frequent and exact monitoring resulted in better outcome, or whether the old standard of lab testing at the OB's office every week or two was enough. Intensive management resulted in much lower rates of macrosomia, c-section, metabolic complications, shoulder dystocia, stillbirth, NICU days, and respiratory complications, to levels comparable to that of nondiabetic controls. Both conventional and intensified management had the *same* goals for treatment and insulin cutoffs, and both used diet and diet+insulin treatment arms; the main difference was the amount and type of monitoring used. This was not a study of diet-only vs. insulin, or regular insulin use vs. prophylactic insulin, although it is sometimes mistakenly cited as such. The more frequent monitoring with a more exact instrument (glucometer) led to 63% of the intensive group being put on insulin, as opposed to only 23% in the conventional management group. The more favorable results in intensive management could be because more insulin was used or simply because previously undiscovered high readings were found and treated--that too many questionable blood sugars were 'getting by' without appropriate treatment. "The intensified management approach is significantly associated with enhanced perinatal outcome. This management strategy clarifies the relationship between glycemic control and neonatal outcome." The study has been criticized by peers for not being truly randomized when it should have been, and that the delivering physicians weren't blinded to the treatment group the patients were in, which could have affected delivery decisions and affected c/s rates, etc. However, the main point of the study (that more frequent, exact measurement resulted in better outcomes) remains mostly intact, and is used to justify insurance coverage of home meters and supplies for gd.

Goldberg, JD et al. Gestational Diabetes: Impact of Home Glucose Monitoring on Neonatal Birth Weight. American Journal of Obstetrics and Gynecology. March 1986. 154(3):546-50.

Compared two groups of 58 gd patients each. One group's bG was measured simply by a weekly 2 hour postprandial blood draw at a clinic; the other group's bG was measured with a home monitor 4x daily (fasting, plus 3 one-hour postprandials). Home monitoring apparently 'caught' more patients needing insulin therapy (same standards for starting insulin were used with both groups); 50% of home monitoring group needed insulin as opposed to 21% of weekly testing group (whose fastings weren't taken either). A combination of the home monitoring and perhaps more insulin use lowered the macrosomia and Large-for-Gestational-Age infants (macrosomia from 24% to 9%; LGA from 41% to 12%). However, the purpose for reducing macrosomia was defeated; the c/s rate was INCREASED in the home monitoring group, despite the strong decrease in birthweight (32% c/s in home monitored group vs. 25% in weekly testing group). Still, if presumed physician bias towards c/s could be eliminated, the home monitoring showed it was very helpful in preventing macrosomia. Even more notable is the author's contention that "Home glucose monitoring and selective insulin therapy have made it possible to achieve an overall incidence of macrosomia equal to that achieved with prophylactic insulin treatment of all gestational diabetic women. Thus we recommend the use of home glucose monitoring for all gestational diabetic patients to individually tailor their management and to identify those needing insulin therapy."

Group B Strep

Ramos, E. et al. Group B Streptococcus Colonization in Pregnant Diabetic Women. Obstetrics and Gynecology. February 1997. 89:257-60.

Found a significantly higher proportion of diabetic women were positive for Group B Strep. Over 50% of women with pre-existing diabetes were positive for group B Strep, while about 35% of women with gd were positive. After adjusting for confounding factors, diabetic women as a group still had about 3x the risk for Strep B colonization. However, a weakness of the study is that it tested women at 20 weeks, 26-28 weeks, and 34-36 weeks; once a woman tested positive she was never tested again but permanently labeled as positive. It is therefore unknown whether these women were actually positive for colonization during labor and birth--they were simply assumed to be and treated as such. Neither was glycemic control evaluated in women who were positive; perhaps women with poorer control had more colonization. This study also contains an analysis of 3 previous studies on this subject; 2 found increased rates of group B colonization in diabetic women as a group, while the one which studied specifically women with gd did not find increased rates of group B colonization. More research remains to be done on this subject, but some feel that it might be prudent to test all women with gd for group B colonization, probably shortly before term. Since obesity also slightly increases the risk, obese women with gd should perhaps investigate this possibility even more. However, research and protocols continue to emerge and all this may need to be re-evaluated at some point.

Piper, JM et al. Group B Streptococcus Infection Rate Unchanged by Gestational Diabetes. Obstetrics and Gynecology. February 1999. 93(2):292-6.

In this study, the rate of Strep B infection did not differ between gd moms and non-gd moms (12% each). Prophylactic antibiotics was not given routinely either. Of the gd women who did have Strep B colonization, the perinatal morbidity was no different from those with Strep B but no gd (in other words, gd did not worsen the outcome if Strep B was found).

Exercise and GD

Jovanovic-Peterson, L et al. Randomized Trial of Diet Versus Diet Plus Cardiovascular Conditioning on Glucose Levels in Gestational Diabetes. American Journal of Obstetrics and Gynecology. August 1989. 161(2):415-9.

19 women with gd (extremely small sample) were randomized into either a diet-only or a diet plus exercise group. Exercising women did 20 minutes of exercise 3x a week for 6 weeks; an arm ergometer was used to maintain their heart rates in the training range. Week 1 glycemic parameters were the same for both groups; the results began to diverge significantly at week 4. At the end of 6 weeks, however, the diet plus exercise group has much better numbers than the diet-only group (70 mg/dl fasting vs. 88 fasting; 106 vs. 187 on a 50g challenge test; and 4.2% vs. 4.7% glycohemoglobin test results). "We conclude that arm ergometer training is feasible in women with gestational diabetes mellitus and results in lower glycosylated hemoglobin, fasting, and 1-hour plasma glucose concentrations than diet alone. Arm Ergometer training may provide a useful treatment option for women with gestational diabetes mellitus and may obviate insulin treatment."

Jovanovic-Peterson, L and Peterson CM. Is Exercise Safe or Useful for Gestational Diabetic Women? Diabetes. December 1991. 40(Supplement 2):179-81.

Presumably a follow-up to the above study, with very similar results. The main difference was that this study examined the effect of exercise on contractions in the women; upper-extremity (arm-only) exercise did not produce any uterine contractions, but lower-extremity exercise tended to produce contractions. For women who find they have a great deal of contractions with walking or who are in danger of pre-term labor (i.e., women with twins or a history of pre-term labor), this finding that upper-extremity exercise is effective at improving glycemic levels while not producing contractions is important. Kmom has also tested upper-extremity exercise ('lifting' cans as dumbbells) and found that it does lower her blood sugar, though not quite as well as walking or other exercise. This study (and others like it) confirms that exercise options are available even to those who don't seem to be able to do regular exercise. (Of course, however, providers should be consulted first.)

Dye, TD et al. Exercise Cuts Rate of Diabetes in Pregnancy in Obese Women. American Journal of Epidemiology. December 1997. Summarized from a press release from Doctor's Guide to Medical and Other News, at www.pslgroup.com/dg/4a5ce.htm.

This study looked at the effect of exercise during pregnancy on the rate of the development of glucose intolerance. Little effect was found for women of average size or those somewhat 'overweight'. However, in women who were significantly obese (Body Mass Index of 33 or more; the usual recommendation for women is a BMI of <25), exercise had a definite preventive effect. Women with BMIs of 33+ who did not exercise were twice as likely to develop gd as their counterparts who did exercise. Curiously, the amount and frequency of exercise showed little difference in benefit; the important factor was the presence of absence of exercise.

Artal, R. Exercise: An Alternative Therapy for Gestational Diabetes. The Physician and Sportsmedicine. March 1996. 24(3):54-6.

A primer on things for doctors to consider when considering an exercise program for a woman with gd. Contains many cautions and caveats since some experts have expressed doubts at the lack of information confirming lack of harm to the fetus from maternal exercise (too bad they are not so demanding about proof of lack of harm from aggressive insulin treatments and early induction!). Notes that in a recent uncontrolled study, a small sample of patients with high fasting levels were placed on a mild exercise program. Only one needed insulin; the rest (who would have needed insulin without the program) were able to avoid it. None had any problems.

Bung, P et al. Exercise in Gestational Diabetes: An Optional Therapeutic Approach? Diabetes. 1991. 40(Supplement 2):182-185.

41 gd patients with abnormal fasting levels who failed a diet therapy trial of 1 week were randomized to either diet+exercise or diet+insulin. The exercise group did 45 minutes of exercise 3x per week in a lab with medical supervision. 17 patients in each group finished the study. No differences were seen between the groups in terms of blood sugar determinations, complication rates, or fetal health (remember that the exercise group did not have insulin; they achieved the same results as the insulin group WITHOUT insulin). The exercise group delivered slightly later on average (38.9 weeks vs. 38.2 weeks), perhaps because of differences in physician management (insulin-treated women are usually induced at 38-39 weeks, non-insulin women generally are induced a week or so behind this schedule).

Optimal Blood Sugar Readings During Labor

Carpenter, MW. Gestational Diabetes. In Current Therapy of Diabetes Mellitus by Ralph A. DeFronzo, MD. St. Louis: Mosby. 1998.

Chapter on managing gd in a medical textbook. Author is a leading gd researcher. Espouses "maternal glycemia between 80 and 100 mg per deciliter during labor" in order to minimize neonatal hypoglycemia, and notes that "values greater than 100 mg per deciliter are treated with insulin."

Carron Brown, S et al. Effect of Management Policy Upon 120 Type 1 Diabetic Pregnancies: Policy Decisions in Practice. Diabetic Medicine. July 1999. 16(7):573-8.

Re: managing Type I pregnancies. Among other things, espouses a "modest relaxation of the targets for blood glucose control during labor to minimize the risk of maternal hypoglycemia...the target range for blood glucose control in labour of 4-7 mmol/l [72-126 mg/dl] minimizes maternal hypoglycemia in labour and the data indicate that an upper limit of 8 mmol/l [144 mg/dl] would not increase the risk of neonatal hypoglycemia." This represents a much more liberal policy for bG in labor than many previous sources, which may have been unnecessarily strict.

Breastfeeding and GD/Diabetes

Pettitt, D.J. et al. Breastfeeding and Incidence of Non-Insulin-Dependent Diabetes Mellitus in Pima Indians. Lancet. July 19, 1997. 350(9072):166-168.

"Exclusive breastfeeding for the first 2 months of life is associated with a significantly lower rate of NIDDM in Pima Indians." The odds ratio for NIDDM in exclusively breastfed people, compared with those exclusively bottlefed, was .41 (less than half), even after adjusting for confounding factors. It should be noted that the effect of breastfeeding could be exaggerated in the study because the Pima Indians have the highest rates of diabetes in the world, or they might be under-represented because the study only examined the effect of 2 months of breastfeeding, whereas the current recommendation from the American Academy of Pediatrics is to breastfeed if possible for a full year or more. The study's effects need to be examined in other ethnic populations, with larger samples, and with longer periods of exclusive breastfeeding, but it is a striking finding nonetheless.

Cordero, L et al. Management of Infants of Diabetic Mothers. Arch Pediatr Adolesc Med. March 1998. 152(3):249-254.

Studied babies of gd and type I diabetic pregnancies, and the effect of 'routine care' instead of the more common intensive care and observation period usually used with babies of diabetic pregnancies. "Routine care failures were more common among infants whose mother had advanced diabetes, but less frequent among breast-fed infants...Breastfeeding among women with GDM and IDDM should be encouraged."

Kjos, S.L. et al. The Effect of Lactation on Glucose and Lipid Metabolism in Women with Recent Gestational Diabetes. Obstetrics and Gynecology. September 1993. 82(3):451-5.

Examined the effects of 4-12 weeks of breastfeeding (or not breastfeeding) on the blood sugar and cholesterol levels of recent gd mothers. 50% of women breastfed, 50% did not (did not examine the *amount* of nursing, just its presence or absence). Those who breastfed had lower fasting levels, better glucose metabolism, and better HDL cholesterol levels. Those who did not breastfeed had twice the rate of developing overt diabetes in the immediate postpartum period. "Lactation, even for a short duration, has a beneficial effect on glucose and lipid metabolism in women with gestational diabetes. Breast-feeding may offer a practical, low-cost intervention that helps reduce or delay the risk of subsequent diabetes in women with prior gestational diabetes."

The Breastfeeding Answer Book. La Leche League International. Revised Edition. Schaumburg, Illinois: La Leche League International. 1997. Can be ordered from www.lalecheleague.org online.

Excellent breastfeeding resource with all the latest research regarding questions on positioning, jaundice, birth control, weaning, pumping, weight gain, medications, breastfeeding problems, prematurity, nursing multiples, adoptive nursing, nutrition, tandem nursing, breast and nipple problems, etc. Highest Recommendation as a resource!

Prophylactic Insulin

Coustan DR and Lewis SB. Insulin therapy for gestational diabetes. Obstetrics and Gynecology. 1978. 51:306-10.

Small but randomized study that found that instituting daily insulin therapy reduced the incidence of macrosomia. However, according to other sources, no differences in rates of operative delivery or birth trauma could be demonstrated.

Coustan DR and Imarah J. Prophylactic insulin treatment of gestational diabetes reduces the incidence of macrosomia, operative delivery and birth trauma. American Journal of Obstetrics and Gynecology. 1984. 150:836-42.

The classic, must-read study that really intensified the prophylactic insulin movement. Retrospectively examined 445 charts of gd women delivered over a 5 year period. Examined the rate of macrosomic infants (>4000g), the rate of operative delivery (forceps, vacuum, c-section), and birth trauma (shoulder dystocia, injury) in the gd women, stratified by treatment modes (no treatment, diet alone, diet plus insulin). Found that insulin treatment significantly lowered the rates of macrosomia, operative delivery, and birth trauma. One of the few studies that found that lowering the macrosomia rate also lowered the c/s rate. The rates for macrosomia among gd women with no treatment, diet-only treatment, and diet+insulin treatment was 17.8%, 18.5%, and 7%. The rates for operative delivery for each group respectively were 28.5%, 30.4%, and 16.3%. The rates for birth trauma for each group respectively were 20.4%, 13.4%, 4.8%. Note that this was a retrospective study, not randomized. Still, one of the best results from prophylactic insulin use on the books yet. Critics note that using insulin reduced the birthweight on average, a total of 6 ounces (170g) and question whether that is clinically significant, but the percentage of very large babies *was* reduced significantly. Authors were fair enough to note at the end that it was appropriate to question the safety of prophylactic insulin use, but they only addressed a couple of areas of maternal safety and did not at all address fetal safety. Finally the authors note, "It would thus seem appropriate to offer prophylactic insulin therapy as a possible means of lessening the likelihood of a difficult delivery or cesarean section. The gestational diabetic woman should not be made to feel that this is a life-or-death issue for her fetus but rather a quality-of-delivery issue."

Thompson DJ et al. Prophylactic insulin in the management of gestational diabetes. Obstetrics and Gynecology. 1990. 75:960-4.

Randomized gestational diabetics into diet alone or diet plus insulin groups, with equal numbers of obese and average-weight women in each treatment group. Babies were delivered between 40-42 weeks, generally. Among 68 women successfully treated for a minimum of 6 weeks, the mean birth weight, macrosomia rate, and ponderal index were reduced significantly in the insulin group. Unlike in some studies, insulin significantly reduced the size of the babies of obese women, by an average level of about a pound and a half (very significant). However, the birth weights of the babies born to obese mothers on insulin were still heavier than those of non-obese women. Despite the big reduction in birth weight in the insulin group in both average-sized and obese mothers, their overall c/s rate was actually higher. The diet alone group had a c/s rate of 35% while the insulin group had a c/s rate of 41% (overall c/s rate for both groups was 38%, very high). Nor was morbidity reduced, raising the question of the value of treatment. No shoulder dystocia occurred in any patient, including the obese patients with macrosomic babies. Study advocates more aggressive insulin use in obese women based on the yet-unproven speculation that a decrease in macrosomia at birth may lead to less obesity and early diabetes among the offspring. Contains a good review of other prophylactic insulin studies.

Persson B, Stangenberg M, Hansson U, Nordlander E. Gestational Diabetes: Comparative Evaluation of Two Treatment Regimens, Diet Versus Diet and Insulin. Diabetes. 1985. 34:101-5.

Randomized 202 pregnant women with gd into diet-only and diet+insulin treatment groups. "The two treatment regimens disclosed no differences regarding achieved degree of maternal blood glucose control, hemoglobin A1c at delivery, obstetric or neonatal complications, infant's size at birth including skin-fold thickness, or C-peptide concentration in cord serum. Routine treatment of pregnant women with mild carbohydrate intolerance with insulin seems unnecessary."

Langer, O. et al. Glycemic Control in Gestational Diabetes Mellitus--How Tight is Tight Enough: Small for Gestational Age versus Large for Gestational Age? American Journal of Obstetrics and Gynecology. 161(3):646-53. September 1989.

334 gd mothers (and 334 controls) were studied to find out the relationship between optimal levels of glycemic control and perinatal outcome. Extremely tight control (mean blood glucose values <87 mg/dl) increased the number of small-for-gestational-age babies significantly, while high blood glucose values (mean > 104 mg/dl) had significantly more large-for-gestational-age babies. The middle group (mean blood glucose between 87-104 mg/dl) was similar to the control group. Critics have used this study to point out that excessively low bG goals may introduce risk and may lead to more SGA infants.

Garcia-Patterson, A et al. In Pregnancies with Gestational Diabetes Mellitus and Intensive Therapy, Perinatal Outcome is Worse in Small-For-Gestational-Age Newborns. American Journal of Obstetrics and Gynecology. August 1998. 179(2):481-5.

821 pregnancies of gd moms receiving intensive metabolic therapy were examined for the relationship between perinatal outcome and birth weight. 7% were Small-For-Gestational-Age (SGA), 85% were Appropriate-for-Gestational-Age (AGA), and 8% were Large-For-Gestational-Age. After adjustment for preterm delivery, rates of adverse fetal outcome were about 3x as likely in SGA babies than in AGA or LGA babies. "Among women with gestational diabetes mellitus who are receiving intensive therapy, perinatal outcome is worse for small for gestational age neonates than for appropriate and large for gestational age neonates." Critics have used this study as well to point out that excessively low bG goals may introduce risk and may lead to more SGA infants.

Kitzmiller, JL. Sweet Success with Diabetes: The Development of Insulin Therapy and Glycemic Control for Pregnancy. Diabetes Care. December 1993. 16(3):107-21.

Technical history of the treatment of type I diabetic pregnancies, it does contain a few notes of interest here. Traces the modification of Pederson's Hypothesis (maternal hyperglycemia increases fetal insulin levels, causing macrosomia); it is now thought that other maternal substrates, notably Free Fatty Acids and Beta-Hydroxybutyrate, may be responsible for some fetal macrosomia. Notes that higher doses of insulin may decrease macrosomia by decreasing these levels. Examines the controversy of whether ketones impair IQ/cause learning problems, and notes that insulin requirements often decrease in the last weeks in type I pregnancies. Finally, notes the potential dangers of adding excess insulin to the mother and the lack of research on its safety. "An important point...is that hyperinsulinemia is produced by clinical methods of insulin treatment of diabetic women...There has been little investigation of the effects of this iatrogenic hyperinsulinemia on placental physiology, blood vessels, or the tendencies toward hypertension in pregnant diabetic women."

Garner, P et al. A Randomized Controlled Trial of Strict Glycemic Control and Tertiary Level Obstetric Care Versus Routine Obstetric Care in the Management of Gestational Diabetes: A Pilot Study. American Journal of Obstetrics and Gynecology 177(1):190-5, 1997.

One of the largest and best-designed studies of the effectiveness of gd care; is a pilot study designed to be followed up with a multicenter trial of sufficient sample size to confirm their findings. Criticizes the inadequacies of other clinical trials to date and points to the need for further prospective randomized controlled trials of larger size. Its preliminary findings based on the pilot study is that intensive treatment of gd (insulin started at fastings of 80!) may have little effect on birth weight, birth trauma, operative delivery, or neonatal metabolic disorders, but emphasizes that the sample size (though the largest of its kind so far) is insufficient to allow any recommendations on the effect of treatment vs. no treatment in gd. A must-read for anyone serious about researching gd.

Santini, DL and Ales, KL. The Impact of Universal Screening for Gestational Glucose Intolerance on Outcome of Pregnancy. Surgery, Gynecology and Obstetrics. May 1990. 170: 427-436.

Looked at the debate over universal screening and its impact on treatment and pregnancy outcome. Retrospectively studied 1307 pregnancies at Cornell University Medical Center over 5 months, where some providers did universal screening and some did not screen at all. Compared the screened population vs. the unscreened population to see if screening and treatment helped reduce the number of large infants, and how treatment and outcome differed. "The process of screening not only failed to decrease the rate of large infants, but also failed to improve otherwise pregnancy outcomes and was associated with more intensive surveillance during pregnancy and a significantly higher rate of primary cesarean delivery." The c/s rate for those unscreened for gd was 21%; the c/s rate for those screened for gd was 27.6%, and the c/s rate for those screened and treated was 32.5%. The c/s rate for those screened and treated with diet alone was 30%; the c/s rate for those screened and then treated with diet plus insulin was 38.5%. "The process of screening is itself linked with more intensive surveillance during pregnancy...even in the absence of the diagnosis, labeling or treatment of gestational glucose intolerance." Furthermore, only about half of the women labeled as having gd actually met the criteria for having it. The rate of large infants was not significantly different between the screened and unscreened groups. No difference was found, either, in the rate of metabolic complications in newborns. A few explanations could be made; physician style probably impacted the difference in c/s rates some, and the unscreened population tended to be less obese, so perhaps that is why they had less large babies than expected. However, the unscreened population was more likely to be parous and to be older (factors which generally make for bigger babies) and to have private doctors (a factor which would tend to increase the c/s rate, not decrease it). C/S was also more common in women with larger infants in either group (31.5% vs. 23.9%), raising the question of how much is due to actual physical problems of having a larger baby and how much is due to physician bias/interventions for larger babies. Examines the difficulty in setting up a randomized, controlled trial with enough power to determine the value of universal screening and gd treatment to reduce infant size---it would have to have a very large amount of participants. Notes that up to now, screening and treatment both were assumed to have no significant adverse effects, but that this needs to be strongly questioned.

Langer, O. Maternal Glycemic Criteria for Insulin Therapy in Gestational Diabetes Mellitus. Diabetes Care. August 1998. 21(Supplement 2). Available for viewing at www.diabetes.org/DiabetesCare/Supplement298/B91.htm.

Supports a very aggressive approach to insulin therapy, especially in obese women. Contends that many women qualifying for insulin are actually receiving too-small doses, and highlights the need for studies to compare optimal insulin dosage. Recounts a study showing that the majority of specialists studied were not instituting insulin at even the ACOG target range (105 fasting/120 post-prandial)--it's doubtful whether this is really true of all OBs. Blames the use of too-liberal insulin standards and the under-dosage of insulin for the lack of improvement of macrosomia rates in many gd studies. Advocates insulin use if fastings are >95, treatment goals of <95 fasting/<115 postprandial/<95 pre-meal, and mean blood glucose readings of 90-100 mg/dl. Also strongly advocates use of self-monitoring by patients and contends it is being vastly underused (questionable, but promotion of self-monitoring is a reasonable position). Theorizes that the rate of insulin use will be about 50-60% in this regimen but that providers should not shy away from this percentage--that normoglycemia is more important than avoiding insulin. Fails to address the issue of higher c/s rates in groups given insulin, or the issue of safety of aggressive insulin use. Makes a reasonable case for study of insulin dosage (are the doses being used too low?), and a decent argument for lowering the fasting requirement to 95, though the point is still debatable. However, he concludes that "patients with fasting plasma glucose on the OGTT of <96 mg/dl (and ideally nonobese) be assigned to diet therapy. Obese women or those with fasting plasma glucose >95 mg/dl on the OGTT should be referred to insulin therapy in order to minimize exposure of the fetus to a hyperglycemic environment." This seems to be saying that ALL obese women should be placed on insulin, regardless of favorable bG results on the OGTT or dietary treatment. If the results are <95, then by his definition they are not 'hyperglycemic', yet they would still be assigned to insulin. Does not make a case to justify the use of insulin for successfully treated obese moms, just makes the sweeping recommendation for it.

Simmons, D and Robertson, S. Influence of Maternal Insulin Treatment on the Infants of Women with Gestational Diabetes. Diabet Med. September 1997. 14(9):762-5.

Examined the long-term impact of insulin therapy on the adiposity of the offspring. Looked at the degree of fatness at about age 2.5 of babies of moms treated for gd. Babies of insulin-treated women had less subscapular fat and less biceps fat than diet-treated moms, despite insulin-treated moms being more obese, older, and more hyperglycemic. "Insulin therapy in gestational diabetes may reduce the incidence of obesity in the offspring of women with gestational diabetes and this should now be tested by a larger, randomized controlled trial." Note that sample size was extremely small, significantly limiting the power of this finding, and that children were only examined at the age of about 2.5 (difficult to make long-term generalizations from). Also, it is unclear if this is meaningful at all, i.e. will having less subscapular and biceps fat translate into less or more diabetes later in life? Still, it is an interesting finding and one of the few studies to study the results of insulin on the child. Its call for further studies with adequate methodology is very important.

O'Sullivan, JB and Mahan, CM. Insulin Treatment and High Risk Groups. Diabetes Care. May-June, 1980. 3(3):482-5.

Follow-up study of 615 gd mothers by the author whose classic works were most instrumental in early gd research. Followed these mothers over 16 years. Half of the group were randomly assigned to insulin therapy originally, half were not. Evaluated whether initial treatment with prophylactic insulin potentially reduced the rate of subsequent diabetes in mothers; it did not. However, a sub-analysis found that among women who bore a baby of large birthweight or who had a family history of diabetes, "subsequent decompensated diabetes was found to be significantly reduced" among those who had been treated prophylactically with insulin. "This finding suggests the possibility of long-term preventive benefits from insulin treatment in high risk subsets of women with gestational diabetes." However, it's important to note that O'Sullivan's original study groups were notable for having multiple risk factors that muddy his original analyses; it is difficult to know the significance of these early findings. Still, it's an interesting study.

Carr, DB and Gabbe, S. Gestational Diabetes: Detection, Management, and Implications. Clinical Diabetes. 16(1):4-24, 1998 Jan-Feb. http://www.diabetes.org/clinicaldiabetes/v16n1j-f98/pg4.htm

Outstanding article summarizing gd management and even some of its controversies, though from a traditional medical approach. See above. Re: prophylactic insulin, it says, "It is important to recognize the data suggesting that insulin therapy may achieve lower rates of macrosomia if initiated when fasting blood glucose is >95 mg/dl. However, prophylactic insulin treatment in patients whose fasting and postprandial values remain within the recommended range is not advised."

Ratner, RE. Clinical Review 47. Gestational Diabetes Mellitus: After Three International Workshops Do We Know How to Diagnose and Manage It Yet? Journal of Clinical Endocrinology and Metabolism. July 1993. 77(1):1-4.

An excellent if conservative review of gd and controversies. Notes that more than just glucose levels are implicated in the possible morbidities associated with gd. Discusses many treatment issues, and notes the need to balance aggressive glycemic control with possible risks to the mother and fetus from overly aggressive treatment. Does favor moderate caloric restriction for obese women, though it notes that this is "one of the most contentious issues in the literature" and that "long-term effects, however, remain unknown and require additional investigation." Regarding prophylactic insulin, it states that "controlled trials comparing diet plus insulin to diet alone in GDM reveal mixed results... preliminary data may support revision of these [traditional] goals to initiation of insulin when fasting plasma glucose levels exceed 95 mg/dL but prophylactic insulin treatment is not convincingly beneficial."

Sacks, DA. Fetal Macrosomia and Gestational Diabetes: What's the Problem? Obstetrics and Gynecology. May 1993. 81(5, part 1):775-81.

A truly outstanding, well-balanced review of the issue of macrosomia and gd pregnancies. Reviewed 79 articles through 1993 relevant to the subject, then summarizes the difficulties in drawing conclusions from them and suggests strategies for further investigation. For example, found that some studies clearly found a relationship between maternal glucose levels and fetal macrosomia, while others did not. Notes all the confounding variables possible, and the differences in study design and observational content---"because of these differences, meaningful comparison of data between studies is exceedingly difficult." Notes the problems even defining what macrosomia is, but treats seriously the potential for shoulder dystocia and injury. However, it still notes that this concern should be "placed in clinical perspective. Only a small proportion of LGA infants of diabetic mothers will develop shoulder dystocia. Brachial plexus palsy...usually resolves during the neonatal period. Physical and sonographic estimates of excessive fetal weight carry substantial margins of error. Furthermore, cesarean delivery introduces a significant risk of maternal morbidity." Regarding prophylactic insulin, the study has multiple observations. "Despite insulin treatment, obese women had a higher incidence of macrosomic neonates, even when the data were stratified by maternal glucose levels...[other] data suggest a potential benefit of routine insulin administration to certain gestational diabetic women. However...the lack of uniformity in reporting and control of variables that may influence birth weight makes it exceedingly difficult to draw definitive conclusions."

Metzger, BE. Treatment of Mild Gestational Diabetes: Is It Time for a Controlled Clinical Trial? Editorial in Diabetes Care. 11(10):813-16. Nov/Dec 1988.

Reviews a number of studies where intensive insulin therapy has been used and finds that "insulin therapy has not always been more successful than dietary treatment, and corresponding improvements in obstetrical outcomes and reductions in neonatal morbidities have not been found to be consistent." In some studies, intensive insulin use reduced the c-section rate, while in others it either had no effect or actually increased the c-section rate. Calls for a large-scale, multicenter controlled clinical trial with a large number of subjects and rigidly defined protocols.

Enkin, Murray et al. A Guide to Effective Care in Pregnancy and Childbirth. Second Edition. Oxford: Oxford University Press (Oxford Medical Publications), 1995.

Based on the conclusions and research from the Cochrane Database of Systematic Reviews, which examined the research of 60 key journals. Careful attention was paid to the methodology of the research design, with an emphasis on the 'gold standard' of research, randomized controlled studies. "Evidence-Based Medicine" at its best. Found significant reason to question the current aggressive approach to gd. "There is no convincing evidence that treatment of women with an abnormal glucose tolerance test will reduce perinatal mortality or morbidity. Trials of dietary regulation... do not demonstrate a significant effect on any outcome, including macrosomia. Trials comparing the use of insulin plus diet with diet alone show a decrease in macrosomia, but no significant effect on other outcomes such as use of caesarean section, the incidence of shoulder dystocia...perinatal mortality...[or] neonatal jaundice or hypoglycaemia." They further go on to make the very strong statement that "The available data provide no evidence to support the wide recommendation that all pregnant women should be screened for 'gestational diabetes', let alone that they should be treated with insulin. Until the risks of minor elevations of glucose during pregnancy have been established in appropriately conducted trials, therapy based on this diagnosis must be critically reviewed. The use of injectable therapy on the basis of the available data is highly contentious, and in many other fields of medical practice such aggressive therapy without proven benefit would be considered unethical."

Labor and Delivery Issues

(Selected references; more will appear in the section on GD: Delivery Protocols. Also note that some references appear more than once, since they may apply to more than one subcategory. For the reader's reference, birth weights of 4000g=8 lbs., 13 oz.; 4250g=9 lbs., 6 oz.; 4500g=9 lbs., 14 oz.; and 5000g=11 lbs.)

Management of Macrosomia in Non-Diabetic Pregnancies

Gregory, KD et al. Maternal and Infant Complications in High and Normal Weight Infants by Method of Delivery. Obstetrics and Gynecology. October 1998. 92(4 Pt 1): 507-13.

Estimated the population risks of maternal and infant complications with the birth of >4000g (almost 9 lbs.) babies in Washington in 1990. "This population-based study showed that most macrosomic infants are delivered vaginally with low rates of maternal and neonatal complications. Macrosomic infants have higher rates of Erb's Palsy, but similar rates of other serious complications of shoulder dystocia when compared with normal weight infants."

Lipscomb, KR et al. The Outcome of Macrosomic Infants Weighing at least 4500 grams: Los Angeles County + University of Southern California Experience. Obstetrics and Gynecology. April 1995. 85(4):558-564.

Examined the maternal and neonatal outcome of macrosomic infants weighing at least 4500g (9lbs, 14 oz.) from non-diabetic pregnancies. 227 mother/infant pairs were examined; 35 had an elective c-section and 192 had a trial of labor. Of those undergoing a trial of labor, 82% delivered vaginally. There were 29 cases of shoulder dystocia (doesn't note how many were induced or had forceps or other interventions applied, nor in what position these women labored) for a rate of 18.5%; however, only 7 cases had birth trauma injuries and none of these were permanent. Mothers had increased rates of lacerations when shoulder dystocia occurred. There was no statistically significant difference in hemorrhage or hospital stay for women who had a vaginal delivery (with or without shoulder dystocia) vs. women who had an elective c/s. Infections in moms who had a c/s after a 'failed' trial of labor were higher than in moms who had an elective c/s; however, only 18% of moms who had a trial of labor in this situation ended up with a c/s. The authors conclude that "Vaginal delivery is a reasonable alternative to elective cesarean for infants with estimated birth weights of at least 4500 g, and a trial of labor can be offered."

Blickstein, I et al. Antepartum Risks of Shoulder Dystocia and Brachial Plexus Injury for Infants Weighing 4200 g or More. Gynecol Obstet Invest. 1998. 45(2):77-80.

236 vaginally delivered babies weighing >4200g were evaluated for rate of shoulder dystocia and birth injury. 11.4% of these deliveries had shoulder dystocia (again, doesn't note how many were induced or had forceps, nor in what position these women pushed); 1.3% had injuries. "A policy of cesarean section for all infants weighing >4200g would result in at least 5- to 6-fold increase in cesarean rate in this group of patients. Our data reconfirm that shoulder dystocia and brachial plexus injury are unpredictable, even in macrosomic infants. It is a matter of policy whether to accept the expected 1:9 and 1:79 respective risks associated with vaginal births."

Bryant, DR et al. Limited Usefulness of Fetal Weight in Predicting Neonatal Brachial Plexus Injury. American Journal Of Obstetrics and Gynecology. September 1998. 179(3 Pt 1):686-9.

Examined the rate of neonatal injuries from vaginal birth and fetal weight-based injury prevention strategies over a period of 12 years. The overall rate of injury was .13%; in infants from 4500-4999g the rate was 3%, and over 5000g the rate was 6.7% for non-diabetic mothers. "To prevent a single case of permanent injury, 155 to 588 cesarean deliveries are required at the currently recommended cutoff weight of 4500g. The rates of lasting morbidity do not justify routine cesarean delivery for infants without diabetic complications weighing <5000g." [There were not enough cases among babies of diabetic mothers in this study to allow meaningful birth weight-specific analysis; it does not make any recommendations on the issue for diabetic mothers other than the need for further study.] At another point in the study, the authors note that "Any recommendation for the route of delivery of a macrosomic infant that is based on birth weight assumes that birth weight can be preducted accurately before delivery...Because these techniques more often overestimate fetal weight, the overall tendency caused by use of ultrasonographically estimated fetal weight would be further increased rates of unwarranted operative intervention...Our study confirms that excessive rates of operative intervention would be required to prevent a small amount of morbidity, as reported by others."

Diani, F et al. Fetal Macrosomia and Management of Delivery. Clin Exp Obstet Gynecol. 1997. 24(4):212-214.

Over a 22 year period, there were 737 pregnancies in which the infant birth weight was >4000g. Normal vaginal delivery occurred in 79%; 3.3% needed vacuum extraction but were still delivered vaginally. 5% had an elective c-section, and 12.5% of those who had a trial of labor later had a c-section, for a total of a 17.6% c/s rate overall. Birth injuries occurred in 5.3% of infants, while 82% of moms had tears etc., but remember that this occurred over a wide range of time and delivery protocols (episiotomies used to be routine in nearly everybody, and these often extend to large tears) so this is not really applicable to predict future rates of perineal tears or damage. The authors state that "because the normal outcome of neonatal births actually encourages the preference for normal vaginal delivery, we concluded that mothers with macrosomic fetuses can safely be managed expectantly unless there is high maternal and fetal risk."

Davis, R et al. The Role of Previous Birthweight on Risk for Macrosomia in a Subsequent Birth. Epidemiology. November 1995. 6(6):607-11.

Assessed the risk for delivery of a macrosomic infant after delivery of a previous macrosomic infant (doesn't specify diabetic vs. nondiabetic cases). Used a population-based cohort study, studying 1793 infants weighing >4500g in Washington State between 1984-90 that were linked to a subsequent birth, and comparing them with 3596 randomly selected infants >4500g also linked to a subsequent livebirth. "The overall prevalence of macrosomic infants subsequent to a previous macrosomic birth was 22%, a proportion that did not vary notably with parity, or when paternity changed between successive births." The authors then conclude that women with one macrosomic infant are a markedly increased risk for repeat macrosomic births, an interesting conclusion when 78% did NOT go on to have another macrosomic child. Since the overall population risk of having an infant >4500g is very low, 22% recurrence is an increased risk, but still 3/4 of them did not go on to have another similarly sized baby. If early inductions or elective c-sections were routinely done for women who had previously had a large baby, then 78% of those inductions and c-sections would have been unnecessary, based on size alone. If shoulder dystocia or birth injury were the outcome considered instead, the rates of unnecessary inductions and unnecessary c-sections would be even higher.

Kolderup, LB et al. Incidence of Persistent Birth Injury in Macrosomic Infants: Association with Mode of Delivery. American Journal of Obstetrics and Gynecology. July 1997. 177(1):37-41.

Deliveries of 2924 macrosomic infants (>4000g) were reviewed to analyze the association between persistent injury and delivery method. Macrosomic infants had a 6-fold increase in significant injury compared to controls. Risk of trauma correlated wtih delivery mode: forceps were associated with fourfold risk of persistent injury compared to spontaneous injury or c-section. However, the overall incidence of persistent cases remained low (.3%!). "A policy of elective cesarean section for macrosomia would necessitate 148 to 258 cesarean sections to prevent a single persistent injury. Avoidance of operative [forceps] vaginal delivery would require 50-99 cesarean sections per injury prevented. These findings support a trial of labor and judicious operative vaginal delivery for macrosomic infants."

Berard, J et al. Fetal Macrosomia: Risk Factors and Outcome. A Study of the Outcome Concerning 100 cases >4500g. Eur J Obstet Gynecol Reprod Biol. March 1998. 77(1):51-9.

Evaluated the common recommendation of elective c-section of babies over 4500g, and analyzed maternal and fetal complications according to the mode of delivery. 100 infants >4500g delivered over a 5 year period in France were retrospectively analayzed. 19 patients of the 100 had gd; 3 had overt pre-existing diabetes. Of the 100 total births analyzed, 13 had an elective c/s, while 26% had a c/s after a trial of labor. 73% of the 87 women who had a trial of labor delivered vaginally. There was a 16% rate of shoulder dystocia; however, only half of these had injuries, and none of the injuries were permanent. "Vaginal delivery is a reasonable alternative to elective cesarean section for infants with estimated birth weights of less than 5000g and a trial of labor can be offered. For the fetuses with estimated birth weight >5000g, an elective cesarean section should be recommended, especially in primiparous women."

Linnet, KM et al. Delivery of Extremely Large Infants. Ugeskr Laeger. February 23, 1998. 160(9):1312-4.

Over ten years, 67 mothers delivered an infant >5000g at a certain hospital in Denmark. Their results were compared with a matched group of infants of normal weight. The macrosomic group showed significantly more c-sections, shoulder dystocia and injuries, but Apgar scores were not significantly lower. Except for one, all the affected infants showed complete recovery. "In conclusion, elective cesarean section can not be generally recommended for an estimated birthweight exceeding 5000g (11 lbs.), if a trained obstetrician is present at delivery."

Boyd, ME et al. Fetal Macrosomia: Prediction, Risks, Proposed Management. Obstetrics and Gynecology. June 1983. 61(6):715-22.

The outcome of macrosomic (>4000g) infant delivery during two time periods 15 years apart (63-65 vs. 78-80) were studied retrospectively. Rate of macrosomia remained the same (10% overall) between the 2 periods, but the c/s rate rose from 8% in the 1960s to 21% from 1978-80. However, this rise is c/s did NOT improve perinatal outcome as it was supposed to; in fact there was a higher rate of birth trauma in the latter period, "suggesting that decreased experience with difficult vaginal deliveries may have reduced the skills of the accoucheur [person 'delivering' the baby] as cesarean section delivery became more frequent. If the increased cesarean section rate has not benefited macrosomic infants studied here...the value of a high cesarean rate must be questioned." Note that use of midforceps delivery was strongly associated with birth injury or morbidity in large babies. The authors conclude that perhaps elective induction of labor shortly before term for pregnancies with the most rapid growth might avoid many cases of macrosomia. [See next studies.]

Induction for Macrosomia in Non-Diabetic Pregnancies

Summers, L. Methods of Cervical Ripening and Labor Induction. Journal of Nurse-Midwifery. March/April, 1997. 42(2):71-85.

An excellent overview of induction practices, both traditional and alternative. Superb review of the various types of inductions and cervical ripening methods. Highly recommended.

Goer, Henci. The Thinking Woman's Guide to a Better Birth. New York: Berkeley Publishing Group (Perigee Book). 1999.

Outstanding review of childbirth issues, especially induction. "Studies [on macrosomia] comparing induced women with women allowed to begin labor on their own all show that induced women have more cesareans and equal numbers of shoulder dystocias...shoulder dystocia isn't very tightly tied to weight, and while it's a dangerous situation, handled properly it rarely results in permanent injury."

Combs, CA et al. Elective Induction versus Spontaneous Labor After Sonographic Diagnosis of Fetal Macrosomia. Obstetrics and Gynecology. April 1993. 81(4):492-496.

Compared elective induction of labor with spontaneous labor for effect on c-section rate and shoulder dystocia rate in 159 non-diabetic cases where macrosomia was predicted by ultrasound. Found that c/s rate was doubled in the induction group and the shoulder dystocia rate was higher too, though not by a great deal. The induced group had a c/s rate of *57%*! The spontaneous labor group had a c/s rate of 31%, still very high but certainly much less than the induced group! After correcting for potential confounders, elective induction was associated with a 2.7x risk of c/s. "Because elective induction of labor increased the cesarean rate and did not prevent shoulder dystocia, we conclude that mothers with macrosomia fetuses can safely be managed expectantly unless there is a medical indication for induction."

Leaphart, WL, et al. Labor Induction with a Prenatal Diagnosis of Fetal Macrosomia. J Maternal Fetal Med. March-April 1997. 6(2):99-102.

Studied 53 non-diabetic patients who underwent induction for fetal macrosomia, and compared their c-section rate to the same number of women delivering a child of same or greater weight entering labor spontaneously. Theorized that since their institution has a low c/s rate, their induction c/s rate would not be different from their spontaneous labor c/s rate in women with babies of similar size. However, they were surprised to learn that the c/s rate in the induction group was double the rate in the spontaneous labor group (36% vs. 17%). "An increased risk of cesarean delivery was observed in subjects undergoing induction for the indication of fetal macrosomia. These data support a plan of expectant management when fetal macrosomia is suspected."

Gonen O, et al. Induction of Labor versus Expectant Management in Macrosomia: A Randomized Study. Obstetrics and Gynecology. June 1997. 89(6):913-917.

The aim of this study was to determine whether or not induction of labor in cases of macrosomia at term improves maternal and neonatal outcome, as many have proposed. 273 non-diabetic patients at term with an estimated fetal weight of 4000-4500g were randomized into an induction of labor group and an expectant management group. The rates of c/s delivery and neonatal outcome were not significantly different between groups. "In this prospective, randomized study, induction of labor for suspected macrosomia at term did not decrease the rate of cesarean delivery or reduce neonatal morbidity. Ultrasonic estimation of fetal weight between 4000 and 4500 g should not be considered an indication for induction of labor."

Weeks, JW et al. Fetal Macrosomia: Does Antenatal Prediction Affect Delivery Route and Birth Outcome? American Journal of Obstetrics and Gynecology. October 1995. 173(4):1215-1219.

Examines the psychological influence on delivery route and birth outcome of a clinical or ultrasonographic prediction of macrosomia, even in settings where macrosomia is not considered an indication for c/s. 504 patient charts of non-diabetic women delivering babies over 4200g over 5 years were restrospectively examined. Statistical comparisons were made between patients in whom the macrosomia was predicted and those in whom it was not. In those pregnancies where macrosomia was predicted, 42% were induced, and 52% ended with a c/s. In those pregnancies where macrosomia was not predicted, 27% were induced, and 30% ended with a c/s. There were no significant differences in shoulder dystocia or birth trauma, however. "The antenatal prediction of fetal macrosomia is associated with a marked increase in cesarean deliveries without a significant reduction in the incidence of shoulder dystocia or fetal injury. Ultrasonography and labor induction for patients at risk for fetal macrosomia should be discouraged."

Delpapa, EH and E. Mueller-Heubach. Pregnancy Outcome Following Ultrasound Diagnosis of Macrosomia. Obstetrics and Gynecology. September 1991. 78(3 pt 1):340-43.

Studied 242 nondiabetic women with suspected macrosomic pregnancies (by ultrasound weight estimation). 77% of predictions exceeded actual birthweight; only 48% were even within 500g of the actual birth weight. 23% were more than 500g overestimated. 50% of the babies predicted to be macrosomic weren't. A trial of labor resulted in the the vaginal delivery of 72% of all of these cases. There were 5 cases of shoulder dystocia but no birth trauma. In order to prevent these 5 cases of shoulder dystocia (from which no persistent morbidity occurred), 76 additional c/sections would have had to have been done. "Our study does not support the contention that elective cesarean is justified in those women with fetuses suspected to be macrosomic as a means of preventing persistent infant morbidity. A very large number of unnecessary cesareans would be performed without much preventive effect...early induction does not appear indicated as a means of preventing persistent infant morbidity...it would seem that prevention of morbidity would be best accomplished by proper and immediate management of shoulder dystocia once it occurs...Because shoulder dystocia rarely causes birth trauma, intervention protocols for women with fetuses suspected to be macrosomic include a very large number of patients who are not at risk for infant morbidity. For this reason, intervention--either cesarean delivery or early induction--does not appear to be indicated for fetuses with macrosomia diagnosed by ultrasound."

Accuracy and Impact of Estimated Fetal Weight

Pollack, RN et al. Macrosomia in Postdates Pregnancies: The Accuracy of Routine Ultrasonographic Screening. American Journal of Obstetrics and Gynecology. July 1992. 167(1):7-11.

519 postdates pregnancies with estimation of fetal weight that occurred within 1 week of delivery were analyzed. Only 56% of fetuses that were actually macrosomic were predicted accurately; only 64% of the fetuses that were estimated to be macrosomic actually were. Notes the dilemma of the doctor presented with a fetus estimated to be macrosomic; they can opt for a trial of labor and risk problems, or they can choose abdominal delivery (c/s) with its attendant morbidity "as long as it is appreciated a priori that in 36% of cases the antenatal diagnosis of macrosomia will not be substantiated" (i.e., one-third of the elective c/s for estimated macrosomia would be unnecessary!!). "At a birth weight of >3750 gm, the Hadlock model (which uses abdominal circumference and femur length) systematically overestimated the birth weight...Routine ultrasonographic screening for macrosomia in postdates pregnancies is associated with a relatively low positive predictive value...Therefore implementation of such screening is of limited use."

Weeks, JW et al. Fetal Macrosomia: Does Antenatal Prediction Affect Delivery Route and Birth Outcome? American Journal of Obstetrics and Gynecology. October 1995. 173(4):1215-1219.

Examines the psychological influence on delivery route and birth outcome of a clinical or ultrasonographic prediction of macrosomia, even in settings where macrosomia is not considered an indication for c/s. 504 patient charts of non-diabetic women delivering babies over 4200g over 5 years were restrospectively examined. Statistical comparisons were made between patients in whom the macrosomia was predicted and those in whom it was not. In those pregnancies where macrosomia was predicted, 42% were induced, and 52% ended with a c/s. In those pregnancies where macrosomia was not predicted, 27% were induced, and 30% ended with a c/s. There were no significant differences in shoulder dystocia or birth trauma, however. "The antenatal prediction of fetal macrosomia is associated with a marked increase in cesarean deliveries without a significant reduction in the incidence of shoulder dystocia or fetal injury. Ultrasonography and labor induction for patients at risk for fetal macrosomia should be discouraged."

Levine, AB et al. Sonographic Diagnosis of the Large for Gestational Age Fetus at Term: Does It Make A Difference? Obstetrics and Gynecology. January 1992. 79(1):55-8.

Retrospective study whose purpose was to determine the accuracy of ultrasound in the diagnosis of the Large-for-Gestational-Age (LGA) fetus and to see whether this influenced obstetric management. 22% of the study group (those with estimated LGA babies) had diabetes, mostly gd. An equal amount of diabetic women were among the controls. Found that the sonographic prediction was incorrect in HALF the cases. Analyzed the management of labor based on LGA prediction: women diagnosed with an LGA fetus were diagnosed by their doctors as having more labor abnormalities (30% vs. 19% controls), had more epidurals (74% vs. 57%), and more cesarean deliveries (53% vs. 32%). To determine whether it was the prediction of LGA vs. the actual birth weight causing the differences in management and outcomes, they stratified the study population prediction vs. actual weight. For babies predicted to be LGA but who were, in fact, of average size, the incorrect diagnosis of LGA had a statistically significant effect on both the diagnosis of labor abnormalities and the incidence of elective cesareans, raising the question of how much physician bias and management of supsected macrosomia is to blame for the problems actually associated with macrosomia. "Because this is a retrospective study, only limited conclusions can be drawn. We observed an association between sonographic estimation of fetal weight at term and the management of labor and delivery. Whether a true cause and effect relationship exists cannot be determined from this study, but, based on our findings, we urge caution in the use of sonographic estimations of fetal weight to guide obstetric decisions concerning labor and delivery."

Spellacy, WN et al. Macrosomia--Maternal Characteristics and Infant Complications. Obstetrics and Gynecology. August 1985. 66(2):158-61.

Found that macrosomia was found most often in women with diabetes, obese women, or women with postdates pregnancies (>42 weeks). "It would seem prudent to scan all women in labor who are obese (over 90 kg), have diabetes mellitus, or are postmature, to determine an estimated fetal weight. Because there is an error of about 10% in this estimate, it would seem reasonable to deliver all infants estimated to weigh more than 5000g by cesarean section to avoid fetal trauma...This plan could reduce the perinatal injuries for the macrosomic infants without greatly increasing the number of cesarean deliveries." (Note that it does not recommend an even lower cutoff for diabetic moms.)

Delpapa, EH and E. Mueller-Heubach. Pregnancy Outcome Following Ultrasound Diagnosis of Macrosomia. Obstetrics and Gynecology. September 1991. 78(3 pt 1):340-43.

Studied 242 nondiabetic women with suspected macrosomic pregnancies (by ultrasound weight estimation). 77% of predictions exceeded actual birthweight; only 48% were even within 500g of the actual birth weight. 23% were more than 500g overestimated. 50% of the babies predicted to be macrosomic weren't. A trial of labor resulted in the the vaginal delivery of 72% of all of these cases. There were 5 cases of shoulder dystocia but no birth trauma. In order to prevent these 5 cases of shoulder dystocia (from which no persistent morbidity occurred), 76 additional c/sections would have had to have been done. "Our study does not support the contention that elective cesarean is justified in those women with fetuses suspected to be macrosomic as a means of preventing persistent infant morbidity. A very large number of unnecessary cesareans would be performed without much preventive effect...early induction does not appear indicated as a means of preventing persistent infant morbidity...it would seem that prevention of morbidity would be best accomplished by proper and immediate management of shoulder dystocia once it occurs...Because shoulder dystocia rarely causes birth trauma, intervention protocols for women with fetuses suspected to be macrosomic include a very large number of patients who are not at risk for infant morbidity. For this reason, intervention--either cesarean delivery or early induction--does not appear to be indicated for fetuses with macrosomia diagnosed by ultrasound."

Alsulyman, OM et al. The Accuracy of Intrapartum Ultrasonographic Fetal Weight Estimation in Diabetic Pregnancies. American Journal of Obstetrics and Gynecology. September 1997. 177(3):503-6.

Compared the accuracy of ultrasonographic fetal weight estimation in pregnant diabetic women with that of matched non-diabetic controls. Significantly greater error in size prediction was observed in babies above 4500g. "When matched for maternal body mass and birth weight, the accuracy of ultrasonographic fetal weight estimation was similar among diabetic and nondiabetic women. Birth weights > or = 4500 g rather than maternal diabetes seem to be associated with less accurate ultrasonographic fetal weight estimates."

Johnstone, FD et al. Clinical and Ultrasound Prediction of Macrosomia in Diabetic Pregnancy. Br J Obstet Gynaecol. August 1996. 103(8):747-754.

Examined serial ultrasounds in diabetic pregnancies (most type I and II, with a few gd pregnancies too) to determine prediction power. "All measurements are poor predictors of eventual standardised birthweight...There is no difference in the prediction power for macrosomia between clinical and ultrasound measurements. Even regular serial scanning and clinical examination will not always diagnose the macrosomic fetus in diabetic pregnancy. In our hands, clinical examination is as predictive as ultrasound measurements. Ultrasound does add to clinical prediction power but only to a small extent. Ultrasound should be used in a selective way, as defined by clinical findings, and with recognition and understanding of the errors and biases involved."

Cohen, B et al. Sonographic Prediction of Shoulder Dystocia in Infants of Diabetic Mothers. Obstetrics and Gynecology. July 1996. 88(1):10-13.

Predicting shoulder dystocia is very tricky, even in macrosomic infants of diabetic mothers. This study retrospectively looks at a specific technique of ultrasound examination (comparing the difference between the abdominal diameter and biparietal diameter, or "AD-BPD difference") to see if it is better at predicting shoulder dystocia in this group, since these infants sometimes experience preferential growth in the truncal area as opposed to the head. Eligibility requirements included diabetic pregnancy, ultrasound with the above measurements within 2 weeks of delivery, estimated fetal weight of 3800-4200g, and vaginal delivery. Found 31 patients who fit this criteria; 6 had shoulder dystocia (rate of 19%). Rate of injury and conditions of laboring (position, stirrups, etc.) not noted. The mean AD-BPD difference for the shoulder dystocia group was 3.1, whereas the mean for the non-shoulder dystocia group was 2.6. Therefore, the trunks of those babies who experienced shoulder dystocia were asymmetrically larger. Shoulder dystocia occurred in 6 of 20 patients (30%) with a AD-BPD difference of at least 2.6, but not in any of the 11 patients where it was <2.6. Therefore the authors propose using this test and this cutoff to identify "those fetuses at high risk for birth injury." This sounds promising, but it's also important to note that if 2.6 were used as the cutoff for doing an elective c-section, only 30% of those c-sections would have been necessary to prevent shoulder dystocia; 70% would have been unnecessary! And not all of those infants with shoulder dystocia have injuries; how many more c-sections would have been unnecessary by the criteria of actually preventing injury?

Shoulder Dystocia in Non-Diabetic and Diabetic Pregnancies

Wiznitzer, A. Obstructed Labor and Shoulder Dystocia. Curr Opin Obstet Gynecol. December 1995. 7(6): 486-91.

"Good clinical judgment can reduce the rate of shoulder dystocia. However, in some cases it remains a problem for the obstetrician and because it occurs so rarely, the care provider may have limited skills to manage this condition." Recommends c/section for diabetic women carrying fetuses estimated to be >4250 g and for non-diabetic women carrying fetuses estimated to be >4500g.

Delpapa, EH and E. Mueller-Heubach. Pregnancy Outcome Following Ultrasound Diagnosis of Macrosomia. Obstetrics and Gynecology. September 1991. 78(3 pt 1):340-43.

Studied 242 nondiabetic women with suspected macrosomic pregnancies (by ultrasound weight estimation). 77% of predictions exceeded actual birthweight; only 48% were even within 500g of the actual birth weight. 23% were more than 500g overestimated. 50% of the babies predicted to be macrosomic weren't. A trial of labor resulted in the the vaginal delivery of 72% of all of these cases. There were 5 cases of shoulder dystocia but no birth trauma. In order to prevent these 5 cases of shoulder dystocia (from which no persistent morbidity occurred), 76 additional c/sections would have had to have been done. "Our study does not support the contention that elective cesarean is justified in those women with fetuses suspected to be macrosomic as a means of preventing persistent infant morbidity. A very large number of unnecessary cesareans would be performed without much preventive effect...early induction does not appear indicated as a means of preventing persistent infant morbidity...it would seem that prevention of morbidity would be best accomplished by proper and immediate management of shoulder dystocia once it occurs...Because shoulder dystocia rarely causes birth trauma, intervention protocols for women with fetuses suspected to be macrosomic include a very large number of patients who are not at risk for infant morbidity. For this reason, intervention--either cesarean delivery or early induction--does not appear to be indicated for fetuses with macrosomia diagnosed by ultrasound."

Acker, DB et al. Risk Factors for Shoulder Dystocia. Obstetrics and Gynecology. December 1985. 66(6):762-8.

One of a number of articles in the 80s advocating a policy of elective c-section for fetuses estimated to be macrosomic. This study retrospectively examined both nondiabetic and diabetic mothers; no differentiation is made between pre-existing diabetics and gestational diabetics in this study. Of 67 diabetics whose babies weighed >4000g, nearly half delivered by c/s. Of the 36 diabetic women who delivered vaginally with babies >4000g (a success rate of 54%!), nearly a third experienced shoulder dystocia. Nothing is mentioned of how the women were forced to labor or push, but being the early 80s, it was probably on their backs and in stirrups, something which quite probably increases the rate of shoulder dystocia. Use of forceps also raised the rate of shoulder dystocia, but the small number of cases of this made this difficult to draw conclusions from. Of the babies in both diabetic and non-diabetic groups who experienced shoulder dystocia, there were significantly more rates of low Apgar scores and fractures and brachial palsy. Long-term followup information was not available; it is not known how many of these injuries persisted. For non-diabetic women, shoulder dystocia was particularly associated with post-term pregnancies with a very large baby, especially those experiencing an arrest disorder during labor. For diabetic pregnancies, it was most associated with a very large infant (>4500g) . "Cesarean section is recommended as the delivery method for diabetic gravidas whose estimated fetal weight is 4000+g. If others confirm the risk, the authors advise serious consideration of cesarean section for gravidas who are carrying fetuses estimated to be 4500+g and who experience an abnormal labor."

Nesbitt, TS et al. Shoulder Dystocia and Associated Risk Factors with Macrosomic Infants Born in California. American Journal of Obstetrics and Gynecology. August 1998. 179(2):476-80.

Examined the one-year incidence of shoulder dystocia and associated risk factors in California. Macrosomia defined lower than usual as >3500 g (usual definition is >4000g, often higher). Found an increased risk of shoulder dystocia associated with diabetes (1.7x risk), assisted delivery--i.e. vacuum or forceps assistance (1.9x risk), and induction of labor (1.3x risk). Of special note here is that the use of forceps/vacuum increased the risk of shoulder dystocia in non-diabetic births by 35-45%. Shoulder dystocia was also strongly increased in diabetic births 'assisted' by vacuum or forceps. The highest risk for shoulder dystocia appears to be in induced diabetic labors with infants over 3500g where the OB uses vacuum or forceps to 'help' things along. Whether this is an argument for elective c/section in these cases or an argument against excessive interventions like routine induction and forceps use from the OB is debatable. Of special note is their statement that "The inaccuracy of estimating fetal weight is a severe limitation in attempting to establish guidelines designed to prevent shoulder dystocia."

Lewis, DF et al. Can Shoulder Dystocia Be Predicted? Preconceptive and Prenatal Factors. Journal of Reproductive Medicine. August 1998. 43(8):654-8.

Examined data from 1622 term singleton, vertex pregnancies who delivered in one year. The study group was comprised of those patients (99) who experienced shoulder dystocia; the control group was made up of those patients who did not experience it. Factors associated with shoulder dystocia in this study included macrosomia (>4000g), previous shoulder dystocia in a prior pregnancy, concurrent diabetes, prior delivery of a macrosomic fetus, and excessive weight gain in pregnancy (>44 lbs.). Surprisingly, some factors traditionally associated with shoulder dystocia in other studes (obesity, history of diabetes, many children, short stature, postdatism, and advanced maternal age) did not turn out to be associated with it in this study, emphasizing the difficulty of predicting it by risk factors alone. In fact, the study makes special note that 1/4 of the 1622 patients had at least one significant risk factor for shoulder dystocia, yet only 13% of those with risk factors actually developed it. To have performed elective c-sections on those with a risk factor for shoulder dystocia would have caused 354 unnecessary c-sections. Even among the patients with more than one risk factor, only 25% of them actually had shoulder dystocia occur. To have performed an elective c/s on those with more than one risk factor would have caused 75% of them to have had unnecessary c-sections. The study further points out the weaknesses of trying to predict macrosomia, and also notes the "even if macrosomia could be accurately predicted, we must remember that most patients can safely deliver a macrosomic infant without difficulty. In our series, 68% of the macrosomic infants were delivered vaginally without difficulty, making elective abdominal deliveries in this group unnecessary." Finally, they conclude that "physicians cannot reliably predict shoulder dystocia by using preconceptive and prenatal risk factors...any such rigidly held recommendations can result only in unnecessary cesarean sections and can continue to confuse the issue."

Cohen, B et al. Sonographic Prediction of Shoulder Dystocia in Infants of Diabetic Mothers. Obstetrics and Gynecology. July 1996. 88(1):10-13.

Predicting shoulder dystocia is very tricky, even in macrosomic infants of diabetic mothers. This study retrospectively looks at a specific technique of ultrasound examination (comparing the difference between the abdominal diameter and biparietal diameter, or "AD-BPD difference") to see if it is better at predicting shoulder dystocia in this group, since these infants sometimes experience preferential growth in the truncal area as opposed to the head. Eligibility requirements included diabetic pregnancy, ultrasound with the above measurements within 2 weeks of delivery, estimated fetal weight of 3800-4200g, and vaginal delivery. Found 31 patients who fit this criteria; 6 had shoulder dystocia (rate of 19%). Rate of injury and conditions of laboring (position, stirrups, etc.) not noted. The mean AD-BPD difference for the shoulder dystocia group was 3.1, whereas the mean for the non-shoulder dystocia group was 2.6. Therefore, the trunks of those babies who experienced shoulder dystocia were asymmetrically larger. Shoulder dystocia occurred in 6 of 20 patients (30%) with a AD-BPD difference of at least 2.6, but not in any of the 11 patients where it was <2.6. Therefore the authors propose using this test and this cutoff to identify "those fetuses at high risk for birth injury." This sounds promising, but it's also important to note that if 2.6 were used as the cutoff for doing an elective c-section, only 30% of those c-sections would have been necessary to prevent shoulder dystocia; 70% would have been unnecessary! And not all of those infants with shoulder dystocia have injuries; how many more c-sections would have been unnecessary by the criteria of actually preventing injury?

Hod, M et al. Antepartum Management Protocol. Timing and Mode of Delivery in Gestational Diabetes. Diabetes Care. August 1998. 21(Supplement 2):B113-B117.

Examined longitudinally past policies of therapy intensity, timing and mode of delivery compared to these policies presently. Found that extremely strict metabolic goals and surveillance, in combination with policies of early induction and elective c-section, significantly lowered rates of macrosomia, shoulder dystocia, c-sections, and neonatal problems. The first period (80-89) had no set goal for glycemia, elective c/s when estimated fetal weight (EFW) was >4500g, and no elective early induction. The second period (90-92) had desired mean glycemia <105 mg/dl, elective c/s when EFW was >4000g, and elective early induction at 40 weeks for LGA fetuses. The third period (93-95) had desired mean glycemia at <95 mg/dl, elective c/s for EFW of >4000g, and elective early induction at 38 weeks for LGA fetuses. Macrosomia declined (period 1=18%, 2=15%, 3=9%), and LGA fetuses declined (1=24%, 2=21%, 3=12%). The c/s rate had a gradual decrease from 21% to 18% to 16% but this decrease did not achieve statistical significance, curious since such a large decrease of macrosomia and LGA fetuses should theroetically produce a greater decline in c/s. Shoulder dystocia also showed a decline (1.5 to 1.2 to .6%), though this also did not achieve statistical significance (again, a surprise that more difference was not achieved). Several important points should be considered, though. The design was longitudinal, comparing different protocols from different periods; this design is inherently weaker because of so many possible confounding variables. Second, elective inductions in this protocol (a 35% rate by the last period, triple the previous rate) involved prostaglandin gel, not pitocin. Most US programs use pitocin, which may result in a higher c/s rate. Third, the average age at delivery in the latest period was only 38 weeks; no wonder their rates of 'large' babies declined! (To be fair, rates declined more than can be attributed to simply earlier delivery, but a good portion of the decline must be due to earlier delivery.) Fourth, metabolic complications (hypoglycemia, jaundice, etc.) were not significantly different between periods; tighter control and fewer large infants should have improved metabolic performance in neonates but did not. This is very significant. Fifth, the difference between the overall c/s rate and the primary c/s rate narrowed between periods; perhaps this is attributible to the more liberal VBAC policy common to the 90s. In other words, the lowering of the overall c/s rate may be at least partly due to less use of repeat c-sections, a policy significantly different in most institutions from the early 80s. Data further stratifying the c/s rates and causes is needed. Sixth, the authors do not distinguish between the results for insulin-requiring gd mothers vs. diet-only gd mothers, an important distinction. Earlier induction may be more justified with insulin-requiring moms, but those results should not be generalized to diet-only moms indiscriminately. The authors conclude that "Our data show that maintaining strict control of maternal diabetes and adhering to an active management protocol for early elective delivery based on the estimated fetal weight have a significant effect on reducing the rate of macrosomia, thereby affecting the incidence of both traumatic births and cesarean deliveries...However, at this stage, there is still a considerable lack of unbiased knowledge regarding the appropriate levels for intervention and a lack of appropriate and accurate tools to predict fetal birth weight. Thus further large prospective clinical trials are warranted to establish appropriate management protocols, timing of delivery, the EFW at which to intervene, and the desired mode of delivery."

Keller, JD et al. Shoulder Dystocia and Birth Trauma in Gestational Diabetes: A Five-Year Experience. American Journal of Obstetrics and Gynecology. October 1991. 165:928-30.

One of the few studies to examine shoulder dystocia in gd only; many studies mix type I and II insulin-dependent diabetes results with gd results. Of the babies estimated to be above 4000g, 66% delivered vaginally; 8 of these experienced shoulder dystocia but only 1 showed any permanent problems and it was mild. To have automatically done a c/s for weights estimated >4000g would have almost doubled the number of c/s in order to prevent one permanent injury. Also, almost half the incidences of shoulder dystocia and birth trauma occurred at levels <4000g, and the accuracy of estimated fetal weight is seriously limited. "Consequently, we continue to believe that to deliver by cesarean section all fetuses estimated to weigh >4000g would considerably increase the number of cesarean sections performed but not eliminate the risk of shoulder dystocia." Shoulder dystocia was associated significantly with use of forceps; the authors recommend great caution in using forceps in gd pregnancies with possibly macrosomic infants.

Lurie, S et al. Outcome of Pregnancy in Class A1 and A2 Gestational Diabetic Patients Delivered Beyond 40 Weeks' Gestation. Amercian Journal of Perinatology. September 1992. 9(5-6):484-488.

Retrospective study (5 yrs.) evaluating the outcomes of those A1 (diet-only) and A2 (insulin-dependent) gd patients delivering between 40-42 weeks compared to class A1 and A2 gd moms delivering before 40 weeks and nondiabetic controls delivering after 40 weeks. Unless fetal health was compromised, the test group was evaluated at 40 weeks with a non-stress test/evaluation of cervical status and twice weekly thereafter. If indicated, labor was induced; labor was also induced if fetus was estimated to be >4000g at that time. If fetal weight was estimated to be >4500g, elective c-section was done. Of the group (both A1 and A2) allowed to go beyond 40 weeks, 75.8% delivered vaginally, 6.5% needed vacuum extraction, and 17.7% needed a c-section. 3% of these experienced shoulder dystocia. The shoulder dystocia rate of diabetics delivering before 40 weeks was 2%; the shoulder dystocia rate of nondiabetic controls was 3%. Apgar scores and neonatal morbidity (hypoglycemia, etc.) was not significantly different between the group delivered after 40 weeks and the group before 40 weeks. The mean gestational age at delivery for the test group (40+ weeks) was 40.9 weeks in A1 and 40.49 weeks in A2 patients. "The low incidence of cesarean sections can be attributed to lowering the rate of failed induction of labor because the patients were allowed to proceed to spontaneous labor...By allowing the pregnancies of gestational diabetic patients class A1 and class A2 to proceed beyond 40 weeks of gestation, we did not increase the incidence of perinatal mortality and morbidity rate. The cesarean rate was low (10.76% in class A1 and 22.03% in class A2). We suggest that not only elective intervention prior to 40 weeks of gestation is to be avoided, but an attempt should be made to allow the gestational diabetics class A1 and class A2 to proceed to spontaneous labor."

Rouse, DJ and Owen, J. Prophylactice Cesarean Delivery for Fetal Macrosomia Diagnosed by Means of Ultrasonography--A Faustian Bargain? American Journal of Obstetrics and Gynecology. August 1999. 181(2):332-8.

Examined the reasonableness of prophylactic c-sections for estimated macrosomia in both normoglycemic and diabetic populations. "Under the most plausible assumptions, a prophylactic cesarean policy with either a 4000- or 4500-g macrosomia threshold would require more than 1000 cesarean deliveries and millions of dollars to avert a single permanent brachial plexus injury. In the pregnancies of diabetic women, although such policies would be expected to perform appreciably better, their use would nevertheless entail considerable intervention for any benefit achieved. Under most assumptions, hundreds of cesarean deliveries and hundreds of thousands of dollars would be required to avert a single permanent brachial plexus injury. In the light of the available data, optimizing the management of shoulder dystocia seems at present to be the most immediate and tenable approach to the prevention of birth-related brachial plexus injury."

Sacks, DA. Fetal Macrosomia and Gestational Diabetes: What's the Problem? Obstetrics and Gynecology. May 1993. 81(5, part 1):775-81.

A truly outstanding, well-balanced review of the issue of macrosomia and gd pregnancies. Reviewed 79 articles through 1993 relevant to the subject, then summarizes the difficulties in drawing conclusions from them and suggests strategies for further investigation. For example, found that some studies clearly found a relationship between maternal glucose levels and fetal macrosomia, while others did not. Notes all the confounding variables possible, and the differences in study design and observational content---"because of these differences, meaningful comparison of data between studies is exceedingly difficult." Notes the problems even defining what macrosomia is, but treats seriously the potential for shoulder dystocia and injury. However, it still notes that this concern should be "placed in clinical perspective. Only a small proportion of LGA infants of diabetic mothers will develop shoulder dystocia. Brachial plexus palsy...usually resolves during the neonatal period. Physical and sonographic estimates of excessive fetal weight carry substantial margins of error. Furthermore, cesarean delivery introduces a significant risk of maternal morbidity."

Elective C-Section for Macrosomia in Diabetic Pregnancies

Conway, DL and Langer, O. Elective Delivery of Infants with Macrosomia in Diabetic Women: Reduced Shoulder Dystocia versus Increased Cesarean Deliveries. American Journal Of Obstetrics and Gynecology. May 1998. 178(5):922-925.

Examined 2604 diabetic women (all types; doesn't distinguish results between gd and others) to see if elective delivery of infants diagnosed with macrosomia by ultrasound would significantly reduce the rate of shoulder dystocia without significantly increasing the c-section rate. Diabetic women with fetuses estimated by ultrasound to be >4250 g (9 lb. 6 oz. range) were delivered by elective c-section; the ones with fetuses estimated to be <4250 g but large-for-gestional-age were electively induced. The c-section rate increased modestly from 21.7% to 25.1% during this protocol; shoulder dystocia dropped modestly from 2.4% to 1.1%. Ultrasonography correctly identified the presence or absence of macrosomia in 87% of patients, a higher accuracy than in many studies. The rate of shoulder dystocia in macrosomic infants delivered vaginally was 7.4%; the rate in the control group was 18.8%. Notes that an intensified management approach to diabetes prevented many women from needing this protocol; only 10.6% of diabetic patients at term required intervention. Notes that others have chosen a 4500g cutoff for elective section, and still others a 4000g cutoff. Feels that 4250g is a better cutoff, since a cutoff of 4000g would cause a 16.8% bigger c/s rate, and did not allow enough room for ultrasonographic error. On the other hand, while a cutoff of 4500g lowered the c/s increase somewhat, it would fail to prevent approximately 40% of shoulder dystocia cases. "An ultrasonographically estimated weight threshold as an indication for elective delivery in diabetic women reduces the rate of shoulder dystocia without a clinically meaningful increase in cesarean section rate. This practice, in conjunction with an intensified management approach to diabetes, improves the outcome of these high-risk women and their infants."

Ecker, JL et al. Birth Weight as a Predictor of Brachial Plexus Injury. Obstetrics and Gynecology. May 1997. 89 (5 pt 1):643-47.

Examined the relationship between birth weight and brachial plexus injury in both diabetic and non-diabetic pregnancies and to estimate the number of c-sections needed to reduce such injuries. Emphasizes the need to use *birth injury* as the outcome proper to evaluating protocols, not shoulder dystocia, since it represents real, not potential, morbidity and are not subjective. "The incidence of brachial plexus injury [although small] increased with increasing birth weight, operative vaginal delivery, and the presence of glucose intolerance. In the group of women without diabetes, between 19 and 162 cesarean deliveries would have been necessary to prevent a single immediate brachial plexus injury. Among women with diabetes, between 5 and 48 additional cesareans would have been required. Although birth weight is a predictor of brachial plexus injury, the number of cesarean deliveries necessary to prevent a single injury is high at most birth weights...In our opinion, the number of cesareans necessary to prevent a single birth injury in a normoglycemic population precludes our recommending mandatory cesarean delivery at any weight cutoff...In the population with diabetes, findings in the subgroup with birthweights greater than 5000g suggest that prophylactic cesarean may be a reasonable management plan, although the small number of cases in this category precludes any definitive recommendation." [And keep in mind that most brachial plexus injuries are not permanent; the estimated numbers of c/s to prevent a *permanent* injury are much, much higher (23-100 more c/s for EFW of >5000g, 91-400 more for >4500g, 219-962 more for >4000g in diabetic mothers; between 85-3226 for nondiabetic mothers of various wt. categories).]

Lurie, S et al. Outcome of Pregnancy in Class A1 and A2 Gestational Diabetic Patients Delivered Beyond 40 Weeks' Gestation. American Journal of Perinatology. September 1992. 9(5-6):484-488.

Retrospective study (5 yrs.) evaluating the outcomes of those A1 (diet-only) and A2 (insulin-dependent) gd patients delivering between 40-42 weeks compared to class A1 and A2 gd moms delivering before 40 weeks and nondiabetic controls delivering after 40 weeks. Unless fetal health was compromised, the test group was evaluated at 40 weeks with a non-stress test/evaluation of cervical status and twice weekly thereafter. If indicated, labor was induced; labor was also induced if fetus was estimated to be >4000g at that time. If fetal weight was estimated to be >4500g, elective c-section was done. Of the group (both A1 and A2) allowed to go beyond 40 weeks, 75.8% delivered vaginally, 6.5% needed vacuum extraction, and 17.7% needed a c-section. 3% of these experienced shoulder dystocia. The shoulder dystocia rate of diabetics delivering before 40 weeks was 2%; the shoulder dystocia rate of nondiabetic controls was 3%. Apgar scores and neonatal morbidity (hypoglycemia, etc.) was not significantly different between the group delivered after 40 weeks and the group before 40 weeks. The mean gestational age at delivery for the test group (40+ weeks) was 40.9 weeks in A1 and 40.49 weeks in A2 patients. "The low incidence of cesarean sections can be attributed to lowering the rate of failed induction of labor because the patients were allowed to proceed to spontaneous labor...By allowing the pregnancies of gestational diabetic patients class A1 and class A2 to proceed beyond 40 weeks of gestation, we did not increase the incidence of perinatal mortality and morbidity rate. The cesarean rate was low (10.76% in class A1 and 22.03% in class A2). We suggest that not only elective intervention prior to 40 weeks of gestation is to be avoided, but an attempt should be made to allow the gestational diabetics class A1 and class A2 to proceed to spontaneous labor."

Rouse, DJ et al. The Effectiveness and Costs of Elective Cesarean Delivery for Fetal Macrosomia Diagnosed by Ultrasound. Journal of the American Medical Association. November 13, 1996. 276(18):1480-1486.

Used an analytical model to help quantify the potential effectiveness and costs of a policy of elective cesarean delivery for fetal macrosomia diagnosed by ultrasound. Considered both diabetic and non-diabetic scenarios. Found that the rate of shoulder dystocia and permanent injury would go down, but that c-section rates would go up a lot. Very technical article, but interesting. "For the 97% of pregnant women who are not diabetic, a policy of elective cesarean delivery for ultrasonographically diagnosed fetal macrosomia is medically and economically unsound. In pregnancies complicated by diabetes, such a policy appears to be more tenable, although the merits of such an approach are debatable."

Rouse, DJ and Owen, J. Prophylactic Cesarean Delivery for Fetal Macrosomia Diagnosed by Means of Ultrasonography--A Faustian Bargain? American Journal of Obstetrics and Gynecology. August 1999. 181(2):332-8.

Examined the reasonableness of prophylactic c-sections for estimated macrosomia in both normoglycemic and diabetic populations. "Under the most plausible assumptions, a prophylactic cesarean policy with either a 4000- or 4500-g macrosomia threshold would require more than 1000 cesarean deliveries and millions of dollars to avert a single permanent brachial plexus injury. In the pregnancies of diabetic women, although such policies would be expected to perform appreciably better, their use would nevertheless entail considerable intervention for any benefit achieved. Under most assumptions, hundreds of cesarean deliveries and hundreds of thousands of dollars would be required to avert a single permanent brachial plexus injury. In the light of the available data, optimizing the management of shoulder dystocia seems at present to be the most immediate and tenable approach to the prevention of birth-related brachial plexus injury."

Langer, O et al. Shoulder Dystocia: Should the Fetus Weighing Greater Than or Equal to 4000g Be Delivered by Cesarean Section? American Journal Of Obstetrics and Gynecology. October 1991. 165 (4 Pt 1):831-837.

A total of nearly 76,000 pregnancies over 15 years (1970-1985) were stratified into diabetic and nondiabetic groups. The incidence of macrosomia [>4000g] was 21% in the diabetic group and 7.6% in the nondiabetic group (no distinction made between gd and other diabetes). 8% of shoulder dystocia occurred in the diabetic group at weights over 4250g. In contrast, 20% of shoulder dystocia could have been prevented by elective c/s in the nondiabetic group when fetal weight was >4500g. Birth weight, diabetes, and labor abnormalities were the principal contributors to shoulder dystocia, according to analysis. (And physician practices contributing to shoulder dystocia?) "Elective cesarean section is strongly recommended for diabetics with fetal weights greater than or equal to 4250 gm, and trial of vaginal delivery for nondiabetic fetuses with weights greater than or equal to 4000 gm is recommended. In all cases the clinician must be watchful for labor abnormalities in macrosomic fetuses."

Lurie, S et al. Induction of Labor at 38 to 39 Weeks of Gestation Reduces the Incidence of Shoulder Dystocia in Gestational Diabetic Patients Class A2. American Journal of Perinatology. July 1996. 13(5):293-6.

Studied whether a policy of elective induction of labor at 38-39 weeks for Class A2 (insulin-requiring gestational diabetics) would lower the incidence of shoulder dystocia. This protocol, studied over a period of 4 years, was compared to a previous protocol allowing spontaneous labor unless fetal health became compromised. "The incidence of shoulder dystocia in patients in whom labor was electively induced at 38 to 39 weeks gestation was 1.4% as compared to 10.2% in patients who delivered beyond 40 weeks' gestation. No increase in cesarean section rate was demonstrated. We conclude that elective induction of labor is suggested for insulin-requiring diabetic women in order to reduce the incidence of shoulder dystocia." [emphasis Kmom's]

Berard, J et al. Fetal Macrosomia: Risk Factors and Outcome. A Study of the Outcome Concerning 100 cases >4500g. Eur J Obstet Gynecol Reprod Biol. March 1998. 77(1):51-9.

Evaluated the common recommendation of elective c-section of babies over 4500g, and analyzed maternal and fetal complications according to the mode of delivery. 100 infants >4500g delivered over a 5 year period in France were retrospectively analayzed. 19 patients of the 100 had gd; 3 had overt pre-existing diabetes. Of the 100 total births analyzed, 13 had an elective c/s, while 26% had a c/s after a trial of labor. 73% of the 87 women who had a trial of labor delivered vaginally. There was a 16% rate of shoulder dystocia; however, only half of these had injuries, and none of the injuries were permanent. "Vaginal delivery is a reasonable alternative to elective cesarean section for infants with estimated birth weights of less than 5000g and a trial of labor can be offered. For the fetuses with estimated birth weight >5000g, an elective cesarean section should be recommended, especially in primiparous women."

Rasmussen, MJ et al. The Timing of Delivery in Diabetic Pregnancy: a 10-Year Review. Australian and New Zealand Journal of Obstetrics and Gynecology. November 1992. 32(4):313-317.

Studied the timing of delivery in 276 pregnancies of diabetic women (doesn't specify which type) in Dublin over 9 years. The mean gestation at delivery was 39 weeks. 83% delivered at or beyond 38 weeks, 41% at or beyond 40 weeks. The overall c/s rate was 28%; 19% elective and 9% in labor. 67% of the diabetics had a normal labor and 5% needed forceps. The induction rate was 27%. There were 5 deaths (1.8%) of normally formed infants at or beyond 38 weeks; all had been preceded by polyhydramnios or macrosomia and recognized poor control.

Enkin, Murray et al. A Guide to Effective Care in Pregnancy and Childbirth. Second Edition. Oxford: Oxford University Press (Oxford Medical Publications), 1995. (also referenced above and below)

[See other entries for more info on book's opinion on other issues.] Examined the research of 60 key journals (emphasis on randomized controlled studies) to promote "evidence-based medicine". Does not address the issue of how to best deliver macrosomic gd pregnancies. This is the authors' evaluation of macrosomia, weight estimation, and prevention of shoulder dystocia in type I and II diabetic pregnancies: "Pregnancies in diabetic women should have additional ultrasound surveillance of growth in late second and third trimester [to look for macrosomia]. Absolute size is not a good predictor of outcome and attention should be paid to growth velocity. As in the non-diabetic population, the precise value of screening for growth problems is not established. Macrosomia can be assessed by ultrasound, but most formulae utilized to calculate fetal weight perform poorly in the larger fetus, and such estimates should be interpreted with caution...Caesarean section has been suggested for diabetic women with a fetal weight estimated to be above 4000 grams and for non-diabetic women with an estimated fetal weight above 4500 grams and slow progress of labour. Such a policy has not been subjected to prospective evaluation. Even if fetal weight estimation were accurate, this would prevent only a small proportion of potential cases of shoulder dystocia, while the majority of caesarean sections would be unnecessary. Almost half of the cases of shoulder dystocia occur in infants weighing less than 4000 grams. There is at present no reliable method of antenatal prediction of shoulder dystocia, and efforts to reduce the problem should be directed towards ensuring that birth attendants are skilled in the management of this condition when it occurs."

Landon, MB and Gabbe, SG. Fetal Surveillance and Timing of Delivery in Pregnancy Complicated by Diabetes Mellitus. Obstetrics and Gynecology Clinics of North America. March 1996. 23(1):109-23.

A summary of delivery and testing protocols and their rationale in diabetic pregnancy, concentrating almost entirely upon type I and II diabetic pregnancies. Contains some info on prenatal testing recommendations for gd moms. "It is debatable whether routine antepartum fetal heart rate testing should be employed in uncomplicated diet-controlled GDM. There is some consensus that women who require insulin for treatment of GDM should undergo twice-weekly heart rate testing at 32 weeks' gestation [and also in women with prior stillbirth and hypertension]...It appears that the third-trimester stillbirth rate in these patients is no higher than that in the general obstetric population...the benefit of testing all women with GDM remains in question. Without a large prospective study comparing outcomes in monitored and nonmonitored women with GDM without other risk factors, it is not possible to determine whether any benefit exists to antepartum fetal surveillance in this seemingly low-risk population." Regarding elective delivery for macrosomic infants of regular diabetic pregnancy (they don't address it for the context of gd), the authors state: "The decision to proceed with an elective delivery for suspected macrosomia is controversial. In our survey of obstetricians and perinatologists, a wide range of estimated birth weights was cited as a threshold for performing a cesarean section to avoid shoulder dystocia and birth trauma. Although we have suggested a weight of 4000g in the diabetic population, others have endorsed birth weight thresholds of 4200 and 4500g."

Weller, KA. Diagnosis and Management of Gestational Diabetes. American Family Physician. May1, 1996. 53(6):2053-7.

An overview of proper treatment protocols for gd in general, done by and for family physicians, a group that tends to be less interventive in gd pregnancies than OBs. Regarding delivery protocols, it states, "The majority of patients whose glucose levels are well controlled with diet and exercise do not need intensive antenatal monitoring. The risk of intrauterine fetal death in a patient with adequately controlled gestational diabetes is not significantly higher than that for all pregnancies...Diet-controlled gestational diabetes is not an indication for early induction of labor or elective cesarean section. As in nondiabetic patients, induction of labor by 42 weeks of gestation is recommended. Fetal macrosomia may complicate the delivery of patients with gestational diabetes, increasing the risk of shoulder dystocia and postpartum hemorrhage. As in pregnancies not complicated by gestational diabetes, labor should be managed expectantly if macrosomia is suspected."

Carpenter, MW. Gestational Diabetes. In Current Therapy of Diabetes Mellitus by Ralph A. DeFronzo, MD. St. Louis: Mosby. 1998.

A general treatise about gd treatment for a medical textbook by one of the leading researchers in the gd field (of the Coustan and Carpenter gd diagnosis guidelines). Many of his protocols tend toward the conservative, but about delivery protocols for gd, he writes, "There is a limited consensus about the timing of delivery. Most authors do not suggest intervention for delivery until 40 weeks' gestation is reached. Generally we await spontaneous labor in patients with GDM. Aside from spontaneous labor, our criteria for delivering women with GDM [by induction include]...poor patient compliance, elevated maternal glucose concentrations not thought to be amenable to outpatient care, other pregnancy complications, such as pregnancy-induced hypertension, pregnancy duration greater than 42 weeks, elective induction when the cervix is favorable and when fetal pulmonic maturity is assured." He further goes on to state, "Generally, no attempt at vaginal delivery is made if the estimated fetal weight is above 4500 g because of the risk of shoulder dystocia at the time of birth. There are no prospective data, however, to establish the 'safe' threshold for estimated weight. The use of forceps or other means of operative vaginal delivery in mothers with GDM is controversial. Appropriately designed trials of forceps application in these cases have not been performed. Our practice is generally to avoid all but outlet forceps deliveries in these cases."

Shachar, IB and Weinstein, D. High Risk Pregnancy Outcome by Route of Delivery. Curr Opin Obstet Gynecol. December 1998. 10(6):447-52.

"This review examines the preferred route of delivery in accordance with neonatal and maternal outcome of three high risk pregnancy conditions: multiple pregnancy; delivery after cesarean section; and delivery of infants with macrosomia to mothers with gestational diabetes mellitus. The most common feature of all these conditions is the lack of information, based on large prospective controlled studies, available to the treating physician for choosing the delivery route of choice and for minimizing morbidity and mortality of both infant and mother."


Copyright 1998-2002 KMom@Vireday.Com. All rights reserved. No portion of this work may be reprinted, reproduced, or sold, either by itself or as part of a larger work, without the express written permission of the author; this restriction covers all publication media, electrical, chemical, mechanical or other such as may arise over time.


[ Back to Kmom Area ]