Article Type: Clinical Article

Accepted Article
Use of a clinical-laboratory score to guide treatment of gestational diabetes

Juliana B. do Valle *, Jean C. Silva, Daniela S. Oliveira, Lisiane Martins, Amanda

Lewandowski, Wagner Horst

Postgraduate Program in Health and the Environment, University of the Region of

Joinville – Univille, Joinville, Brazil

* Correspondence

Juliana Barros do Valle, Rua Barão do Rio Branco, 207 sala 7, Jaraguá do Sul, SC,

Brazil.

Email: julianabvalle@gmail.com

Keywords: Gestational diabetes; Neonatal outcome; Score; Treatment

Synopsis: The clinical-laboratory score could serve as a complement to physician

experience in planning the treatment of women with gestational diabetes.

Abstract

Objective: To assess the outcomes of implementing a clinical-laboratory score in

the treatment of pregnant women with gestational diabetes.

This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/ijgo.12326
This article is protected by copyright. All rights reserved.
 Methods: A retrospective before-and-after implementation analysis was undertaken
Accepted Article using data and neonatal outcomes for pregnant women with gestational diabetes

treated before (January 2011–December 2012; control group) and after (January

2013–December 2014; score group) introduction of a newly developed score. To

evaluate the effects of score adoption, odds ratios with 95% confidence intervals

were calculated after adjustment for confounding factors.

Results: The control group included a greater proportion of women treated with diet

alone (170/312 [54.5%]) than the study group did (122/391 [31.2%]; P<0.001). By

contrast, more women in the study group received metformin (172 [44.0%] vs 77

[24.7%]; P<0.001). The neonatal outcomes, including low Apgar scores at 1 minute

and at 5 minutes and neonatal intensive care unit admission, were similar in both

groups. Multivariate logistic regression analysis showed that the adoption of the

score did not significantly affect the choice of treatment or the birth weight rating.

Conclusion: The score served well as an orientation tool in therapeutic decision

making and had no negative effect on the treatment choice and perinatal outcomes.

1 INTRODUCTION

For 10% of women whose pregnancies are complicated by diabetes, the diagnosis

occurs before conception and as such they are deemed to have pregestational

diabetes, which is usually of type 2 [1]. Gestational diabetes mellitus (GDM) is a

metabolic disorder defined as any reduction in the degree of glucose tolerance that

begins during pregnancy. The prevalence of GDM in Brazil is 7%–10%, and a

significant proportion is diagnosed after 36 weeks of pregnancy[2].

This article is protected by copyright. All rights reserved.

 Maternal risk factors such as obesity and excessive weight gain during pregnancy
Accepted Article can have negative effects on the treatment of diabetes and on neonatal outcomes

[3–5]. Recommended total weight gain, based on pre-pregnancy body mass index

(BMI) has previously been reported in the literature; this varies from 7 kg to 11 kg

and from 5 kg to 9 kg for overweight and obese patients, respectively [6]. The

treatment of GDM initially consists of diet and exercise with the aim of reducing

maternal and fetal morbidity. Approximately 10%–20% of patients require

pharmaceutical intervention [1,3,4]. Further, the average blood glucose values in

pregnant women without diabetes differ throughout the day depending on whether or

not patients are obese [13]. Bertini et al. [7] reported that differences in the treatment

of pregnant women with diabetes were necessary depending on a patient’s BMI;

extreme obesity usually required a two-fold higher insulin dose to achieve the

glycemic goal. Both obesity and excessive weight gain can increase the risks of

having a large for gestational age (LGA) neonate and macrosomia [3,8]. Glycemic

control is achieved by considering the following variables: glucose value, fetal

abdominal circumference (FAC), BMI before pregnancy, and pregnancy duration.

The Brazilian Diabetes Society (SBD) and the Brazilian Federation of Gynecology

and Obstetrics Societies (FEBRASGO) recommend daily blood glucose

measurement, with the result considered excellent when the glucose value is 3.33–

5.83 mmol/L (60–105 mg/dL) before meals, at bedtime, and at dawn, and 5.55–

7.77 mmol/L (100–140 mg/dL) 1 hour after eating [8,9] .By comparison, the

American College of Obstetricians and Gynecologists and the American Diabetes

Association recommend the following blood glucose levels: 5.27 mmol/L (95 mg/dL)

or less before eating, 7.77 mmol/L (140 mg/dL) or less 1 hour after eating, and

This article is protected by copyright. All rights reserved.

 6.66 mmol/L (120 mg/dL) or less 2 hours after eating[10]. Hyperglycemia can cause
Accepted Article macrosomia, but excessive adjustment of the blood glucose level (fasting glucose

<4.44 mmol/L [80 mg/dL]) can result in small-for-gestational-age (SGA) newborns

[11–13].

FAC can serve as an indicator of excessive fetal growth and an indirect marker of

fetal hyperinsulinism, primarily from 28 weeks of pregnancy [14]. Values between the

70th and 75th percentiles relative to normal pregnancies have been associated with

satisfactory maternal and fetal outcomes. Irrespective of blood glucose levels, the

FAC can be used to identify pregnancies at low risk for fetal macrosomia and can

help to avoid insulin use in approximately 38% of pregnant women without

increasing the neonatal morbidity [15].

As for BMI, both pre-pregnancy obesity (BMI ≥30, calculated as weight in kilograms

divided by the square of height in meters) and a substantial weight gain during the

prenatal period are associated with an increased risk of diabetes and a LGA neonate

[16]. Additionally, the average blood glucose value throughout the day appears to be

0.28–0.56 mmol/L (5–10 mg/dL) higher in obese women without diabetes than in

nonobese women[17].A two-fold higher insulin dose is typically required for pregnant

women with extreme obesity[7] .

The pregnancy duration at the onset of treatment and glycemic control values can

influence treatment and neonatal outcomes [1,9,18]. Early disease detection is

associated with a higher likelihood of requiring medication. Late diagnosis is

associated with a higher prevalence of complications such as polyhydramnios,

This article is protected by copyright. All rights reserved.

 premature birth, and an LGA newborn [18,19]. With regard to the pregnancy
Accepted Article duration, Bartha et al. [19] demonstrated that an early diagnosis of GDM could

reduce the risk of pre-eclampsia, requirement of insulin treatment, neonatal

hypoglycemia, and perinatal mortality owing to initiating therapy. By contrast, a late

diagnosis is associated with polyhydramnios, premature birth, an LGA newborn, and

an increased level of glycated hemoglobin[18,19].

Before January 2013, treatment decisions at Darcy Vargas Maternity Hospital,

Joinville, Brazil, were usually based on the clinical experience of the attending

clinician. However, the relevant measures are difficult to interpret in combination.

Therefore, an orientation tool based on a clinical-laboratory score was developed

that integrates the information from these variables (Table 1, Table 2). The aim of

the present study was to assess the outcomes of the implementation of this tool in

the treatment of GDM. Accordingly, physicians in training and medical students

would benefit from such a tool.

2 MATERIALS AND METHODS

The present study was a retrospective before-and-after implementation analysis of

treatment data and perinatal outcomes for women attending a clinic for pregnant

women with diabetes at a maternity hospital in Brazil. Institutional visits were

conducted by groups that included physicians, obstetrics and gynecology residents,

the supervisors of medical students, and other staff members. Data were collected

for patients who were treated between January 1, 2011, and December 31, 2014,

and whose prenatal care and birth occurred at Darcy Vargas Maternity Hospital. The

This article is protected by copyright. All rights reserved.

 clinical-laboratory score was introduced on January 1, 2013; patients treated before
Accepted Article this point formed the control group and those treated after formed the score group.

The inclusion criteria were diagnosis of GDM according to WHO criteria (fasting

blood glucose value ≥6.11 mmol/L [≥110 mg/dL] or blood glucose value

≥7.77 mmol/L [≥140 mg/dL] 2 hours after an oral glucose tolerance test), age of at

least 18 years, singleton pregnancy, absence of other conditions that might interfere

with perinatal outcomes or therapy, and availability of complete patient data. There

were no deleted records. The study and introduction of the Score were approved by

the Univille Ethics Committee. Written informed consent was obtained from all

participants upon first visit to the Diabetes Clinic and included use of patient data

and results for future studies.

All participants had undergone medical monitoring and nutritionist, physiotherapist,

psychologist, and nurse evaluations upon referral to the Diabetes Clinic. Diet was

adapted on an individual basis, and the calorie intake was calculated as 25 kcal/kg

body weight for obese patients and 35 kcal/kg body weight for nonobese patients.

Glycemic control was considered satisfactory when the fasting blood glucose value

was 5.00 mmol/L (90 mg/dL) and the 1-hour postprandial value was 7.77 mmol/L

(140 mg/dL).

Obstetric evaluations occurred every 14–21 days at the hospital's Diabetes Clinic

after the initial consultation and diagnosis, and other professional evaluations

occurred on medical request. At the first obstetric visit after diagnosis, the glycemic

values were assessed during the day at the hospital (fasting blood glucose and

glucose levels 1 hour after breakfast, lunch, and afternoon coffee). At subsequent

This article is protected by copyright. All rights reserved.

 visits, the fasting glucose value and the value 1 hour after breakfast were assessed.
Accepted Article If the two measurements were not satisfactory, a new profile was obtained by

repeating the measurements at the four timepoints described earlier.

If glycemic control had not been achieved by the subsequent visit, medication

treatment was initiated with oral metformin (Glifage, Merck, Rio de Janeiro, Brazil),

insulin (Novolin R or N, Novo Nordisk, Bagsvaerd, Denmark), or both. The decision

to prescribe insulin or metformin was made using score points (the score group) or

was made on a case-by-case basis by physicians according to experience.

The dose of metformin was 500 mg at lunch and dinner; it was increased by 500–

1000 mg weekly up to a maximum of 2500 mg/day if necessary. The insulin dose

was 0.7 UI/kg/day, administered before each meal (approximately one third before

breakfast, one quarter before lunch, one quarter before the evening meal, and one

sixth administered before going to sleep at night) according to glucose profile results

(adjustments up to 20% above the daily dose could be made on a case-by-case

basis), with regular insulin being used preprandially and neutral protamine Hagedorn

insulin used at bedtime. The medications were provided by the public health system.

Delivery was recommended at 40 weeks for patients treated with diet only, and at

39 weeks for those treated with medications. If the fetal weight was greater than the

97th percentile for the gestational age, delivery was medically indicated at 37 weeks.

From January 1, 2013, the clinical-laboratory score was applied at every medical

appointment. The patient received a score for each measure included in the tool

This article is protected by copyright. All rights reserved.

 (Table 1). The final score was calculated by summing the individual values and used
Accepted Article as basis for the therapeutic recommendation (Table 2).

For the present study, medical records were reviewed for maternal age, pre-

pregnancy BMI, parity, total weight gain during pregnancy, pregnancy duration on

arrival at the clinic, fasting and postprandial blood glucose levels, and treatment

applied (diet only, metformin, insulin, or both medications). Women with a BMI

between 26 and less than 30 were classified as overweight, and those with a BMI of

30 or more were classified as obese. Women with BMI below 18.5 were considered

to be underweight, BMI of 18.5–25.9 was considered normal weight; these

classifications were not included in the present statistical analysis.

The evaluated outcomes included the following newborn characteristics: gestational

age at birth, premature birth, route of delivery, birth weight, Apgar score, and

neonatal intensive care unit (NICU) admission. Newborns whose weight was below

the 10th percentile on growth curves were considered SGA; those with a weight

above the 10th percentile and below the 90th percentile were considered appropriate

for gestational age; and those with a weight above the 90th percentile were

considered LGA. Macrosomia was classified as a birth weight of 4 kg or more.

The primary outcome measures were the choice of treatment (diet alone, metformin,

insulin, or both metformin and insulin combined), and the neonatal outcomes.

Data were tabulated using Excel 2010 (Microsoft, Redmond, WA, USA) and

analyzed using SPSS version 21 (IBM, Armonk, NY, USA). All variables were

This article is protected by copyright. All rights reserved.

 analyzed descriptively. For quantitative variables, the Kolmogorov–Smirnov
Accepted Article normality test was performed. Normally distributed data were analyzed using the

parametric t test, and non-normally distributed data were analyzed using the

nonparametric Mann–Whitney U test. For nominal variables, the χ2 test was used;

the Fisher exact test was used when the number of datapoints was less than five.

Multivariate logistic regression analysis was performed to calculate odds ratios with

95% confidence intervals to evaluate the effect of adopting the clinical-laboratory

score on the treatment choice and neonatal outcomes. P<0.05 was considered

statistically significant.

3 RESULTS

Of 703 pregnant women with diabetes treated over the entire study period, 391

patients (55.6%) were in the score group and 312 (44.4%) in the control group. No

patients were excluded from the study.

Both groups had similar characteristics regarding age and parity (Table 3). A pre-

pregnancy BMI classified as obese and weight gain below the recommendation were

more frequent in the score group than in the control group (both P<0.001) (Table 3).

Compared with the score group, the control group included a higher percentage of

women with a normal pre-pregnancy BMI and with adequate or higher-than-

recommended weight gain (all P<0.001) (Table 3). The total weight gain in the

control group was higher than that in the control group (P<0.001) (Table 3).

With respect to data related to diabetes during pregnancy, the score group had a

longer pregnancy duration at arrival at the clinic and included a higher percentage of

This article is protected by copyright. All rights reserved.

 women receiving metformin than did the control group (both P<0.001) (Table 4). The
Accepted Article control group included a greater number of patients whose treatment was based on

diet only (P<0.001) (Table 4).

With regard to the characteristics of the newborns, the gestational age at delivery

and the Apgar scores at 1 minute and 5 minutes were higher in the control group

than in the score group (all P<0.01) (Table 5). There were more cesarean deliveries

and premature births in the score group (P<0.01) (Table 5).

The adoption of the score did not affect the type of treatment or the classification of

the newborn weight, as indicated by multivariate analysis adjusted for confounding

factors such as maternal age, pre-pregnancy BMI, weight gain, pregnancy duration

at admission, mean fasting glucose value, and mean postprandial glucose value

(Table 6).

4 DISCUSSION

The present study showed that the implementation of a clinical-laboratory score to

guide treatment of gestational diabetes had no effect on treatment choice and did not

negatively affect perinatal outcomes.

The clinical-laboratory score was constructed using six measures (Table 1) that

included fasting blood glucose, postprandrial blood glucose, FAC, BMI, and duration

of pregnancy. Each variable contributed to an overall score used to guide treatment.

Hyperglycemia is related to maternal complications such as pre-eclampsia and

perinatal complications that can include a birth weight above the 90th percentile,

This article is protected by copyright. All rights reserved.

 premature delivery, dystocia, NICU admission, and hyperbilirubinemia [1,20]. The
Accepted Article glycemic goal varies depending on the protocol used. The glycemic targets for

pregnant women at Darcy Vargas Maternity Hospital are similar to those

recommended in the literature [4,10]. The Hyperglycemia and Adverse Pregnancy

Outcomes study [21] demonstrated that the risks of maternal, fetal, and neonatal

adverse events increase if glucose values are increased at approximately 24–28

weeks of pregnancy, even if the actual values remain within the range considered

normal. The relationships between glycated hemoglobin values and maternal–fetal

complications have been described in the literature but remain controversial [5].

These observations emphasize the importance of timely diagnosis and blood

glucose monitoring during pregnancy. It was also noted, in the current study, that the

control group included a greater proportion of patients who received dietary

treatment only, whereas the score group included a greater proportion of patients

who received metformin. However, these group-dependent treatment biases were

not further statistically investigated and are only mentioned as descriptive

characterization of the groups. Preterm birth, birth weight, and NICU admission are

mainly related to postprandial blood glucose values [21]. There is a strong

association between an infant weight above the 90th percentile and increased

maternal blood glucose levels. However, pursuing strict control and narrow glycemic

targets in diabetic pregnant women without considering other parameters of

evaluation increases the incidence of SGA newborns. Kjos et al. [15] noted a

reduced use of insulin and a lower incidence of SGA newborns if pregnant women

with diabetes were managed on the basis of fetal growth instead of strict glycemic

control. Also, ensuring appropriate weight gain during pregnancy could improve

perinatal outcomes. Gante et al. [22] observed a higher frequency of SGA newborns

This article is protected by copyright. All rights reserved.

 among obese pregnant women with diabetes and below-recommended weight gain;
Accepted Article by contrast, the prevalences of macrosomia, LGA newborns, a low Apgar score, and

cesarean delivery were higher among obese pregnant women with diabetes and

above-recommended weight gain. In addition, one of the measures included in the

score was the FAC. Some studies have employed the FAC as a follow-up tool for

managing diabetes beyond the conventionally used glycemic values; an FAC at or

above the 75th percentile frequently reflects a state of fetal hyperinsulinemia and

thus requires a more careful approach [3, 22].

The aim of the present study was not to compare one treatment to another, but

rather to use an appropriate score to guide the choice of treatment from among

those available at the study institution. However, as mentioned previously, we noted

that the score group included a greater proportion of women undergoing metformin

treatment, as opposed to treatments including diet and/or insulin, compared with the

control group. Perinatal studies [23, 24] have revealed similar results for metformin

and insulin in pregnant women with diabetes. Moreover, the high cost, complexity,

storage requirements, and route of administration of insulin increase the

attractiveness of metformin as a therapeutic option. Silva et al. observed a lower

birth weight among pregnant women with diabetes who received metformin

compared with those receiving glyburide [12, 25] A meta-analysis [23] concluded that

metformin reduced the incidence of maternal and neonatal adverse effects, including

hypoglycemia and NICU admission, compared with insulin. However, additional

studies are required to evaluate the potential risks associated with metformin therapy

for pregnant women with diabetes [23].

This article is protected by copyright. All rights reserved.

 Among the main outcomes, inclusion in the score group has no substantial effect on
Accepted Article the choice of treatment, demonstrated by the adjusted odds ratios (Table 6).

Although a dietary treatment approach would be always preferable, not all patients

benefit from a diet-only treatment. The score aims, in this respect, to guide decisions

regarding treatment choice. The odds ratios and confidence intervals produced

indicated that the score did not influence on the choice of treatment, with the

confidence intervals including “1”, demonstrating a lack of statistical significance.

Similarly, the score had no effect on neonatal outcomes. Results including birth

rating, low apgar scores, and neonatal care unit admission were similar in both

groups.

The present study was limited by not taking the time between diagnosis, referral to

the ambulatory clinic, and patient arrival into account, although the protocol for

diabetes screening used by the primary care network followed the recommendations

of the Brazilian Ministry of Health. Accordingly, these intervals could represent a

source of bias. Another potential limitation was the difference in the number of

participants in the two groups. Because the availability of complete data was an

inclusion criterion, greater care was taken completing the records once the score

was implemented; hence, the score group included a larger number of patients.

Moreover, differences were observed in maternal characteristics such as BMI,

weight gain, pregnancy duration at the beginning of treatment, type of treatment, and

route of delivery. The nature of the present study must be considered; the two

samples were studied at different times and there may have been seasonal

influences on service. However, in the multivariate analysis, possible confounders

(maternal age, BMI, weight gain during pregnancy, pregnancy duration at the

This article is protected by copyright. All rights reserved.

 beginning of treatment, fasting and postprandial blood glucose levels) were taken
Accepted Article into consideration.

To summarize, the present analysis revealed that the adoption of the clinical-

laboratory score had no negative effects on the choice of treatment and neonatal

outcomes. Thus, the score could serve as a complement to clinical experience in the

planning of treatment for GDM. This tool could be particularly useful for guiding

prenatal care professionals at academic centers that train doctors and medical

students. The adoption of measures and protocols that facilitate therapeutic decision

making is attractive, particularly at teaching hospitals or medical centers that provide

clinical training to current and future health professionals. Accordingly, the score was

not designed to be the sole determinant of the therapeutic choice, but to serve as a

tool that is used in combination with the attending physician’s clinical experience.

Author contributions

JBdV and JCS supervised the conduct of the study. JCS designed and planned the

study. DSO, LM, and AL collected the data. JBdV and WH analyzed the data and

wrote the manuscript. All authors edited, revised, and approved the manuscript.

Conflicts of interest

The authors have no conflicts of interest.

References

[1] Coustan DR, Lowe LP, Metzger BE, Dyer AR. The Hyperglycemia and
Adverse Pregnancy Outcome (HAPO) study: paving the way for new
diagnostic criteria for gestational diabetes mellitus. Am J Obstet Gynecol
2010;202. doi:10.1016/j.ajog.2010.04.006.

This article is protected by copyright. All rights reserved.

 [2] Santos EMF, de Amorim LP, Costa OLN, Oliveira N, Guimarães AC. Profile of
Accepted Article gestational and metabolic risk in the prenatal care service of a public maternity
in the Brazilian Northeast . Perf Risco Gestacional E Metabólico No Serviço
Pré-Natal Matern Pública Do Nord Do Bras 2012;34:102–6.
doi:10.1590/S0100-72032012000300002.

[3] Schaefer-Graf UM, Kjos SL, Fauzan OH, Buhling KJ, Siebert G, Buhrer C, et
al. A Randomized Trial Evaluating a Predominately Fetal Growth-Based
Strategy to Guide Management of Gestational Diabetes in Caucasian Women.
Diabetes Care 2004;27:297–302. doi:10.2337/diacare.27.2.297.

[4] Walker JD. NICE guidance on diabetes in pregnancy: management of diabetes

and its complications from preconception to the postnatal period. NICE clinical
guideline 63. London, March 2008. Diabet Med 2008;25:1025–7.
doi:10.1111/j.1464-5491.2008.02532.x.

[5] Lowe LP, Metzger BE, Dyer AR, Lowe J, McCance DR, Lappin TRJ, et al.
Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study: associations
of maternal A1C and glucose with pregnancy outcomes. Diabetes Care
2012;35:574–80. doi:10.2337/dc11-1687.

[6] Metzger BE, Gabbe SG, Persson B, Buchanan T a, Catalano P a, Damm P, et

al. International association of diabetes and pregnancy study groups
recommendations on the diagnosis and classification of hyperglycemia in
pregnancy. Diabetes Care 2010;33:676–82. doi:10.2337/dc09-1848.

[7] Bertini AM;Taborda W; Silva JC. Diabetes mellitus: Clinical aspects and
diagnosis. Rev Med Mat Fetal 2011;2(1)4-1 2011.

[8] Reichelt AJ, Oppermann MLR, Schmidt MI. Guidelines of the 2nd Meeting of
The Diabetes and Pregnancy Task Force. Arq Bras Endocrinol Metabol
2002;46:574–81. doi:10.1590/S0004-27302002000500012.

[9] Negrato CA, Montenegro RM, Mattar R, Zajdenverg L, Francisco RP V,

Pereira BG, et al. Dysglycemias in pregnancy: from diagnosis to treatment.
Brazilian consensus statement. Diabetol Metab Syndr 2010;2:27.
doi:10.1186/1758-5996-2-27.

[10] American Diabetes Association . Diagnosis and classification of diabetes

mellitus. Diabetes Care 2014;37:81–90. doi:10.2337/dc14-S081.

[11] Vandorsten JP, Dodson WC, Espeland MA, Grobman WA, Guise JM, Mercer
BM, et al. NIH consensus development conference: diagnosing gestational
diabetes mellitus. NIH Consens State Sci Statements 29:1–31.

[12] Nicholson WK, Wilson LM, Witkop CT, Baptiste-Roberts K, Bennett WL, Bolen
S, et al. Therapeutic management, delivery, and postpartum risk assessment
and screening in gestational diabetes. Evid Rep Technol Assess (Full Rep)
2008:1–96.

This article is protected by copyright. All rights reserved.

 [13] Middleton P, Crowther CA, Simmonds L, Muller P. Different intensities of
Accepted Article glycaemic control for pregnant women with pre-existing diabetes. Cochrane
Database Syst Rev 2010:CD008540. doi:10.1002/14651858.CD008540.pub2.

[14] Metzger BE, Buchanan T a, Coustan DR, de Leiva A, Dunger DB, Hadden DR,
et al. Summary and recommendations of the Fifth International Workshop-
Conference on Gestational Diabetes Mellitus. Diabetes Care 2007;30 Suppl
2:S251-60. doi:10.2337/dc07-s225.

[15] Kjos SL, Schaefer-Graf U, Sardesi S, Peters RK, Buley A, Xiang AH, et al. A
randomized controlled trial using glycemic plus fetal ultrasound parameters
versus glycemic parameters to determine insulin therapy in gestational
diabetes with fasting hyperglycemia. Diabetes Care 2001;24:1904–10.

[16] Bowers K, Laughon SK, Kiely M, Brite J, Chen Z, Zhang C. Gestational

diabetes, pre-pregnancy obesity and pregnancy weight gain in relation to
excess fetal growth: variations by race/ethnicity. Diabetologia 2013;56:1263–
71. doi:10.1007/s00125-013-2881-5.
[17] Harmon KA, Gerard L, Jensen DR, Kealey EH, Hernandez TL, Reece MS, et
al. Continuous glucose profiles in obese and normal-weight pregnant women
on a controlled diet: metabolic determinants of fetal growth. Diabetes Care
2011;34:2198–204. doi:10.2337/dc11-0723.

[18] Correa PJ, Vargas JF, Sen S, Illanes SE. Prediction of Gestational Diabetes
Early in Pregnancy: Targeting the Long-Term Complications. Gynecol Obstet
Invest 2014:145–9. doi:10.1159/000357616.

[19] Bartha JL, Martinez-Del-Fresno P, Comino-Delgado R. Early diagnosis of

gestational diabetes mellitus and prevention of diabetes-related complications.
Eur J Obstet Gynecol Reprod Biol 2003;109:41–4. doi:10.1016/S0301-
2115(02)00480-3.

[20] Crowther CA, Hiller JE, Moss JR, McPhee AJ, Jeffries WS, Robinson JS.
Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N
Engl J Med 2005;352:2477–86. doi:10.1056/NEJMoa042973.

[21] Metzger BE, Lowe LP, The HAPO Study Cooperative Research Group.
Hyperglycemia and Adverse Pregnancy Outcomes. N Engl J Med
2008;358:1991–2002. doi:10.1056/NEJMoa0707943.

[22] Gante I, Amaral N, Dores J, Almeida MC. Impact of gestational weight gain on
obstetric and neonatal outcomes in obese diabetic women. BMC Pregnancy
Childbirth 2015;15:249. doi:10.1186/s12884-015-0692-z.

[23] Li G, Zhao S, Cui S, Li L, Xu Y, Li Y. Effect comparison of metformin with

insulin treatment for gestational diabetes: a meta-analysis based on RCTs.
Arch Gynecol Obstet 2014;292:111–20. doi:10.1007/s00404-014-3566-0.

[24] Bertini AM, Silva JC, Taborda W, Becker F, Lemos Bebber FR, Zucco Viesi

This article is protected by copyright. All rights reserved.

 JM, et al. Perinatal outcomes and the use of oral hypoglycemic agents. J
Perinat Med 2005;33:519–23. doi:10.1515/JPM.2005.092.
Accepted Article
[25] Silva JC, Pacheco C, Bizato J, De Souza BV, Ribeiro TE, Bertini AM.
Metformin compared with glyburide for the management of gestational
diabetes. Int J Gynecol Obstet 2010;111:37–40.
doi:10.1016/j.ijgo.2010.04.028.

Supplementary file S1 Portuguese translation of abstract.

Table 1 Clinical-laboratory score for the treatment of diabetes during pregnancy.

Measure Score
–2 –1 0 +1 +2
Fasting blood glucose, mmol/L ≥4.44 (≥80)
≥5.00 (≥90) to ≥5.55
(mg/dL) — <4.44 (<80) to <5.00
<5.55 (<100) (≥100)
(<90)
Postprandial blood glucose, ≥5.55 (≥100) ≥6.66 (≥120)
≥7.77
mmol/L (mg/dL) — <5.55 (<100) to <6.66 to <7.77
(≥140)
(<120) (<140)
Fetal abdominal circumference,
≤10 >10 to <25 ≥25 to <75 ≥75 to <90 ≥90
percentile
Body mass index Below normal Normal Above normal
— —
(<18.5) (18.5–25.9) (>25.9)
Pregnancy duration at
<28 ≥28 to <32 ≥32 to <36 ≥36 —
consultation, wk

Table 2 Recommendations for the treatment of diabetes during pregnancy based on the clinical-
laboratory score.
Total score a Suggested therapy
<0 New consultation with nutritionist
≥0 to <2 Maintenance of diet and physical activity
≥2 to ≤4 Oral hypoglycemic drug
>4 Insulin concomitant with diet and/or oral hypoglycemic drug
a
Sum of scores for each individual measure, as shown in Table 1.

This article is protected by copyright. All rights reserved.

 a
Table 3 Characteristics of the pregnant women with diabetes.
Accepted Article Characteristic Score group (n=391) Control group (n=312) P value
Age, y 30.5 ± 6.4 30.6 ± 6.3 0.757 b
c b
Pre-pregnancy body mass index 30.2 ± 6.0 28.2 ± 6.1 <0.001
d
Low weight 3 (0.8) 2 (0.6) 0.895
Normal weight 80 (20.5) 114 (36.5) <0.001 d
Overweight 119 (30.4) 80 (25.6) 0.308 e
Obese 189 (48.3) 116 (37.2) <0.001
b
Parity 2.6 ± 1.7 2.6 ± 1.6 0.772
Pregnancy (with respect to this
study)
1st 106 (27.1) 90 (28.8) 0.592 d
d
2nd 108 (27.6) 83 (26.6) 0.767
d
≥3rd 177 (45.3) 139 (44.6) 0.667
Total weight gain, kg 5.1 ± 5.9 9.3 ± 7.5 <0.001 b
Below recommendation 270 (69.1) 130 (41.7) <0.001 e
e
Adequate 61 (15.6) 86 (27.6) <0.001
e
Above recommendation 60 (15.3) 96 (30.8) <0.001
Abbreviation: BMI, body mass index.
a
Values are given as mean ± SD or number (percentage), unless indicated otherwise.
b
Nonparametric Mann–Whitney U test.
c
Calculated as weight in kilograms divided by the square of height in meters.
d 2
χ test.
e
Q2 test.

Table 4 Diabetes-related variables during pregnancy. a

Variable Score group (n=391) Control group P-value
(n=312)
b
Pregnancy duration at clinic arrival, wk 30.0 ± 5.8 26.6 ± 8.0 <0.001
b
Fasting blood glucose, mmol/L (mg/dL) 5.01 ± 0.70 4.90 ± 0.70 0.009
(90.3 ± 12.7) (88.3 ± 12.7)
1-hour postprandial blood glucose, 6.52 ± 0.97 6.48 ± 0.98 0.584 b
mmol/L (mg/dL) (117.5 ± 17.5) (116.8 ± 17.6)
Type of treatment
c
Diet 122 (31.2) 170 (54.5) <0.001
c
Metformin 172 (44.0) 77 (24.7) <0.001
c
Insulin 63 (16.1) 43 (13.8) 0.391
Metformin + insulin 34 (8.7) 22 (7.1) 0.424 c
a
Values are given as mean ± SD or number (percentage), unless indicated otherwise.
b
Nonparametric Mann-Whitney U test.
c 2
χ test.

This article is protected by copyright. All rights reserved.

 a
Table 5 Neonatal characteristics.
Characteristic Score group Control group P-value
Accepted Article (n=391) (n=312)
Gestational age at delivery, wk 38.4 ± 1.4 38.7 ± 1.3 <0.001 b
c d
Premature birth 35 (9.0) 15 (4.8) 0.034
Type of delivery
Vaginal 167 (42.7) 169 (54.2) 0.003 d
Cesarean 224 (57.3) 143 (45.8) 0.003 d
b
Birth weight, g 3313.8 ± 495.2 3357.3 ± 523.9 0.278
Weight classification
Small for gestational age 13 (3.3) 8 (2.6) 0.556 d
Appropriate for gestational age 310 (79.3) 252 (80.8) 0.625 d
d
Large for gestational age 68 (17.4) 52 (16.7) 0.732
d
Macrosomia 28 (7.2) 23 (7.4) 0.915
Apgar at 1 min 8.0 ± 1.1 8.2 ± 1.2 <0.001 b
<7 23 (5.9) 17 (5.4) 0.805 d
b
Apgar at 5 min 8.9 ± 0.8 9.2 ± 0.7 <0.001
e
<7 5 (1.3) 3 (1.0) 1.00
Admission to neonatal intensive care 22 (5.6) 14 (4.5) 0.496 d
unit
a
Values are given as mean ± SD or number (percentage), unless indicated otherwise.
b
Nonparametric Mann–Whitney U test.
c
<37 weeks.
d 2
χ test.
e
Fisher exact test.

Table 6 Effect of adoption of the clinical-laboratory score on the type of treatment used and the
newborn weight rating.
Variable Unadjusted OR (95% CI) Adjusted OR (95% CI) a
Type of treatment
Diet 0.437 (0.240–0.798) 0.521 (0.242–1.119)
Metformin 1.405 (0.760–2.595) 1.502 (0.732–3.083)
Insulin 0.922 (0.472–1.799) 1.238 (0.570–2.689)
Metformin + insulin 1.260 (0.716–2.218) 0.825 (0.422–1.610)
Newborn weight rating
Small for gestational age 1.307 (0.535–3.194) 2.343 (0.790–6.946)
Appropriate for gestational age 0.911 (0.628–1.323) 0.936 (0.589–1.487)
Large for gestational age 1.071 (0.722–1.591) 0.899 (0.540–1.498)
Macrosomia 0.870 (0.409–1.851) 0.747 (0.351–1.592)
Abbreviations: OR, odds ratio; CI, confidence interval.
a
Adjusted for maternal age, pre-pregnancy body mass index, weight gain during pregnancy,
pregnancy duration at the start of treatment, mean fasting glucose value, and mean postprandial
glucose value.

This article is protected by copyright. All rights reserved.