Hindawi

Journal of Diabetes Research

Volume 2023, Article ID 9574563, 9 pages
https://doi.org/10.1155/2023/9574563

Research Article
Achievement of Target Glycemic Goal with Simple Basal Insulin
Regimen in Women with Gestational Diabetes: A Prospective
Cohort Study

Misao Fukuoka,1 Ichiro Yasuhi ,1 Hiroshi Yamashita,1 Kensuke Ashimoto,1,2 Nao Kurata,1
Junko Yamaguchi,1 Megumi Koga,1 So Sugimi,1 Sachie Suga,1 and Masashi Fukuda1
1
Department of Obstetrics and Gynecology, National Hospital Organization (NHO) Nagasaki Medical Center, Omura-City,
Nagasaki, Japan
2
Department of Obstetrics and Gynecology, Kameda General Hospital, Kamogawa, Chiba, Japan

Correspondence should be addressed to Ichiro Yasuhi; iyasuhi0414@gmail.com

Received 29 November 2022; Revised 10 May 2023; Accepted 11 May 2023; Published 29 May 2023

Academic Editor: Eusebio Chiefari

Copyright © 2023 Misao Fukuoka et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

There is little evidence concerning the need to treat gestational diabetes (GDM) in the same way as pregestational diabetes. We
evaluated the efficacy of the simple insulin injection (SII) regimen for achieving the target glucose goal without increasing
adverse perinatal outcomes in singleton pregnant women with GDM. All subjects underwent self-monitoring of blood glucose
(SMBG), and insulin therapy was indicated according to the SMBG profile. Insulin was initially started with the SII regimen, in
which one daily injection of NPH insulin before breakfast was used, and another NPH injection was added at bedtime, if
necessary. We used the target glucose as <95 mg/dL at fasting and <120 mg/dL postprandial and accepted <130 mg/dL for the
latter. If the target glucose did not reach with the regimen, we switched to the multiple daily injection (MDI) with additional
prandial insulin aspart. We compared the SMBG profile before delivery as well as the perinatal outcomes between the SII and
MDI groups. Among 361 women (age 33.7 years, nullipara 41%, prepregnancy body mass index 23.2 kg/m2) with GDM, 59%,
18%, and 23% were in the diet-alone, SII, and MDI groups, respectively. Consequently, regarding women requiring insulin
therapy, 43% were treated with the SII regimen throughout pregnancy. The severity of baseline hyperglycemia according to the
SMBG data at baseline was the MDI>the SII>the diet group. The rate of achieving target glucose levels before delivery in the
SII group at fasting, postprandial < 120 mg/dL and <130 mg/dL were 93%, 54% and 87%, respectively, which were similar to
that in the MDI group (93%, 57%, and 93%, respectively), with no significant differences in perinatal outcomes. In conclusion,
more than 40% of women with GDM requiring insulin therapy achieved the target glucose goal with this simple insulin
regimen without any increase in adverse effects.

1. Introduction insulin therapy is indicated if diet-based treatment fails to

maintain normoglycemia.
Even though gestational diabetes (GDM) is considered a Various types of insulin, including human insulin and
milder form of hyperglycemia than pregestational diabetes insulin analog, can be safely administered to pregnant
[1], various perinatal complications, including macrosomia, women. As with pregestational diabetic women, most physi-
primary cesarean section, neonatal hypoglycemia, and pre- cians prefer to administer intensive insulin therapy, such as
eclampsia, develop without intervention [2–4]. To prevent the multiple daily insulin injection (MDI) regimen, in which
such perinatal complications, strict glycemic control equiva- basal and rapid acting insulin injections are combined, to
lent to that in normal pregnant women is required, and women with GDM. However, women with GDM develop
2 Journal of Diabetes Research

less severe hyperglycemia than pregnant women with preg- Regarding the management of GDM, dieticians started
estational diabetes [1]. Furthermore, there is little evidence diet therapy soon after the diagnosis. Patients also under-
concerning the need to treat GDM in the same way as preg- went self-monitoring of blood glucose (SMBG) four times
estational diabetes. In addition, from the patient’s perspec- daily (at fasting (before breakfast) and 2 h after each meal)
tive, the MDI regimen can cause both physical and mental using the same SMBG equipment (OneTouchVerioVue®;
stress and reduce the quality of life [5]. Recent studies have LifeScan Japan). We used a target glucose levels < 5:27 mmol/L
shown that 20%-60% of women with GDM require insulin (95 mg/dL) at fasting and <6.66 mmol/L (120 mg/dL) at 2 h
therapy to achieve the target glucose goal [3, 6–8]. Thus, after each meal [9, 10]. Because postprandial capillary glucose
for women with GDM, a simpler insulin regimen may help values are known to be 20%-25% higher than the venous glu-
improve compliance of insulin therapy during pregnancy. cose values [11], we accepted <130 mg/dL as the postprandial
We hypothesized that some women with GDM who did target on the SMBG profile in the clinical setting. Fasting and
not reach target glucose levels with nutrition therapy alone postprandial blood glucose levels were assessed separately.
and required insulin therapy had less severe hyperglycemia Insulin therapy was indicated if the SMBG value did not
than those requiring the MDI regimen and could therefore reach approximately ≥80% of the fasting and postprandial
be treated with a simpler regimen than the regimen. For this target glucose values, respectively [7]. Patients received die-
reason, we developed the simple insulin injection (SII) regi- tary education at weekly to monthly intervals according to
men, in which we use single or twice daily injection of NPH the characteristics of the subjects. We measured HbA1c
insulin as a treatment option for women with GDM. and random glucose values at least once a month. Women
The present study evaluated the efficacy of the SII regi- being treated with insulin therapy were encouraged to con-
men and demonstrated whether it successfully achieved the tinue SMBG until labor onset. In women with diet therapy
target glucose goal comparable to the MDI regimen without alone, the attending physician decided whether the measure-
increasing the risk of adverse perinatal outcomes in women ments should be abandoned or continued until labor onset.
with GDM. Attending physicians checked the SMBG memories in
women whose SMBG report was discrepant with other clin-
ical features, including the glucose values measured at every
2. Materials and Methods perinatal visit and sonographic fetal growth and amniotic
fluid assessments.
At a single tertiary perinatal care center (National Hospital At the beginning of insulin therapy, most patients
Organization (NHO) Nagasaki Medical Center, Omura, received the SII regimen, which consisted of one to two daily
Nagasaki) in Japan, we prospectively included Japanese sin- injections of NPH insulin (Novolin N®). We used one daily
gleton pregnant women with GDM diagnosed according to injection of NPH insulin (10 to 20 units) before breakfast
the JSOG criteria at 24–32 weeks of gestation who had deliv- and added another at bedtime, if necessary. If the target
ered in the period of January 1, 2015, to December 31, 2019. values were still not reached, we switched to the MDI regi-
In 2012, we introduced the SII regimen as a treatment men with the combination of NPH insulin (at fasting and/
option for women with GDM. Because we trained physicians or bedtime) and prandial injections of insulin aspart
to standardize the regimen in women with GDM from 2012 (NovoRapid®) at each meal. In the insulin regimens, insulin
to 2014, we excluded cases encountered in this period, as it doses were adjusted according to the SMBG data in each
was considered a transitional period for the introduction of case throughout pregnancy.
the regimen. Women with overt diabetes in pregnancy were We interviewed baseline characteristics at the first peri-
also excluded owing to the possibility of their having preges- natal visit during first trimester before the diagnosis or at
tational diabetes. We also excluded patients who were the reference visit to the institute after the diagnosis. Pre-
treated with insulin outside of the center, those with a pregnancy obesity was defined as a BMI ≥ 25 kg/m2. Gesta-
non-Japanese ethnicity, and those with a steroid medication tional weight gain (GWG) throughout pregnancy and
or insulin allergy. before and after the diagnosis of GDM was examined. A
The present study was approved by the NHO Nagasaki family history (FH) of diabetes was defined as unspecified
Medical Center. Patients provided their informed consent diabetes among second-degree relatives.
for the collection of their clinical data for the purpose As indices of metabolic control, we used the SMBG mea-
of research. surements and HbA1c value at the diagnosis and the latest
All pregnant women at the study institution underwent measurement before delivery (within four weeks before
universal screening using a 50 g glucose challenge test at 24 to delivery). Regarding the SMBG values, we used the mean
28 weeks’ gestation; those with test values of ≥7.49 mmol/L values of 7-day measurements at fasting (before breakfast)
(135 mg/dL) underwent a diagnostic 75 g oral glucose toler- and those of 7-day measurements of three daily postprandial
ance test (OGTT) after overnight fasting. According to the values, during the introduction period of the SMBG and 7
JSOG criteria [9], women with one or more abnormal values days before the last perinatal visit. As indices to assess the
above the cutoff (5.1 (92), 10.0 (180), and 8.5 (153) mmol/L achievement of the target glucose goal, we used the mean
(mg/dL) for fasting, 1 h and 2 h after a glucose load, respec- values of fasting blood glucose < 95 mg/dL and the mean
tively) were diagnosed as having GDM. Glycated hemoglobin values of postprandial blood glucose < 130 mg/dL. Because,
(HbA1c) levels were also measured at the time of the diag- regarding the postprandial glucose, a value < 120 mg/dL is
nostic OGTT. generally used as a target value worldwide, the achieving rate
Journal of Diabetes Research 3

of the value was also compared between the groups. Hypo- GDM (n = 361)
glycemic events were derived from medical records. SMBG+diet therapy
As perinatal outcomes, we compared the primary cesarean
section rate, the incidence of hypertensive disorders of preg-
Requiring
nancy (HDP), birthweight (BW), and BW z-score. Regarding insulin therapy
the BW assessment, we used large-for-gestational age (LGA) (n = 148, 41%)
and small-for-gestational age (SGA) infants, defined as those
with a BW > 90th and <10 percentile, respectively, according
to parity- and gender-specific Japanese BW curves [12]. Diet therapy alone The SII regimen Requiring the MDI
Regarding neonatal complications, we defined clinical hypo- throughout pregnancy throughout pregnancy regimen
(n = 213, 59%) (n = 64, 18%) (n = 84, 23%)
glycemia and hyperbilirubinemia as that requiring glucose
infusion and phototherapy, respectively. Diet group SII group MDI group
We used Tukey’s HSD test and Chi-square test to
compare numerical and categorical variables, respectively, Figure 1: Flow diagram of patient inclusion in each group. GDM:
between the groups and P values < 0.05 were considered to gestational diabetes; SMBG: self-monitoring of blood glucose; SII:
indicate statistical significance. simple insulin injection; MDI, multiple daily insulin injection.

3. Results degree of glycemia in the SMBG was lowest in the diet group
and highest in the MDI group (Table 3 and Figures 2(a) and
We included 361 patients; the mean maternal age and pre- 2(c)). As the baseline data, in the diet groups, 94% of women
pregnancy BMI were 33:8 ± 4:9 years and 23:2 ± 4:5 kg/m2, achieved the target glucose values at fasting and 80% and
respectively. 41%, 28%, and 48% were nulliparous, prepreg- 96% achieved <120 mg/dL and <130 mg/dL postprandial,
nancy obese (≥25 kg/m2), and with family history of diabe- respectively. On the other hand, only 63%, 20%, and 57%
tes, respectively. The mean gestational age at diagnosis of in the SII group and 39%, 6%, and 31% in the MDI group
GDM were 26:0 ± 1:9 weeks’ gestation. Among them, 213 achieved the target values, respectively.
(59%), 64 (18%), and 84 (23%) were treated with diet ther- Regarding the achievement of the target glucose goal on
apy alone (diet group), the SII regimen alone (SII group), SMBG before delivery, the mean and achievement rate of
and required the MDI regimen (MDI group), respectively fasting glucose values did not differ between the groups
(Figure 1). Forty-three percent of women requiring insulin (Table 3 and Figures 2(b) and 2(d)). Regarding the postpran-
therapy were treated with the SII regimen alone throughout dial glucose values before delivery, although the mean values
pregnancy. in the diet group were still significantly lower than the insu-
The maternal baseline characteristics and diagnostic lin groups, the differences were small and did not differ
OGTT results are summarized in Table 1. While the insulin within the insulin groups. Regarding the achievement rate
group showed a statistically significant association with obe- of <120 mg/dL postprandial, the rates in the insulin groups
sity, there were no significant differences in the baseline were still significantly lower than that in the diet group.
characteristics between the insulin groups. Regarding the On the other hand, the rates did not differ within the insulin
diagnostic OGTT results, the fasting and 1 h and 2 h PG groups. The achievement rates of the postprandial values of
values did not differ between the insulin groups; however, <130 mg/dL effectively rose in both insulin groups. Although
the HbA1c levels were higher in the MDI group (Table 1). the rate in the SII group was significantly lower than the diet
Women in the SII group required less than half of a daily group, the rates did not differ within the insulin groups
dose of total insulin and 24% less NPH insulin than the MDI (Table 3 and Figures 2(b) and 2(d)).
group (Table 2). Neither the GWG throughout pregnancy The perinatal outcomes are summarized in Table 4.
nor that before or after the diagnostic OGTT differed Although women in the insulin groups showed a signifi-
between the groups. The HbA1c levels at delivery were very cantly higher rate of HDP than those in the diet group, no
similar in each group. There were no hypoglycemic episodes marked difference was noted between the insulin groups.
requiring additional treatment during pregnancy in any of There were no significant differences in the perinatal out-
the insulin groups. comes other than the development of HDP between
The SMBG data are summarized in Table 3. The data at the groups.
the time of the diagnosis were missed in 15, 4, and 4 cases in
the diet, SII, and MDI groups, respectively. Regarding the 4. Discussion
SMBG data before delivery, one-third of women in the diet
group were allowed to quit the SMBG measurements In this study, we found that, among women with GDM
because of their good control, so we obtained the data mea- requiring insulin therapy, 43% were successfully treated with
sured in the 4 weeks before delivery in 142 cases in the the SII regimen throughout pregnancy, with the rate of
group. The data before delivery were missed in four and achieving target glucose values comparable to that in the
three cases in the SII and MDI groups, respectively. Both MDI regimen. Women in the SII group also demonstrated
fasting and postprandial values and the achievement rates similar perinatal outcomes to those in both diet and MDI
of target glucose values at the time of the diagnosis differed groups, without excessive maternal weight gain. These find-
significantly between the groups, demonstrating that the ings support the efficacy of the SII regimen for some women
4 Journal of Diabetes Research

Table 1: A comparison of the baseline characteristics and OGTT results between the groups.

Diet alone SII group MDI group p value p value p value

(n = 213) (n = 64) (n = 84) (diet vs. SII) (diet vs. MDI) (SII vs. MDI)
Maternal age (years) 33:5 ± 4:9 33:1 ± 5:2 34:9 ± 4:7 0.79 0.09 0.07
Nulliparous (%) 92 (43) 21 (33) 34 (40) 0.14 0.67 0.34
Family history of diabetes (%) 98 (46) 33 (52) 42 (50) 0.13 0.53 0.44
History of GDM (%) per parous women 39/121 (32) 13/43 (30) 19/50 (38) 0.80 0.47 0.43
Prepregnancy BMI (kg/m2) 22:0 ± 3:7 25:1 ± 4:9 24:9 ± 4:9 <0.0001 <0.0001 0.97
Prepregnancy obesity
40 (19) 27(42) n(40) 0.0001 <0.0001 0.83
(BMI ≥ 25 kg/m2)
Results of the diagnostic OGTT
Gestational age at OGTT (weeks) 26:0 ± 1:9 26:0 ± 2:0 25:8 ± 1:7 1.00 0.68 0.82
Fasting PG (mg/dL) 78 ± 9 85 ± 9 88 ± 10 <0.0001 <0.0001 0.21
1-h PG (mg/dL) 175 ± 24 179 ± 26 185 ± 22 0.55 0.0033 0.23
2-h PG (mg/dL) 155 ± 22 153 ± 26 163 ± 28 0.84 0.060 0.065
HbA1c (%) 5:2 ± 0:3 5:3 ± 0:3 5:5 ± 0:3 0.39 <0.0001 0.0021
Data were shown as mean ± SD or number (%). SII: simple insulin injection; MDI: multiple daily insulin injection; GDM: gestational diabetes; OGTT: oral
glucose tolerance test; PG: plasma glucose.

Table 2: A comparison of the insulin therapy, GWG, and HbA1c at delivery between the groups.

Diet alone SII group MDI group p value p value p value

(n = 213) (n = 64) (n = 84) (diet vs. SII) (diet vs. MDI) (SII vs. MDI)
Required insulin dose during pregnancy
— 19:4 ± 7:7 46:0 ± 22:9 — — <0.0001
(U/day)
NPH insulin (U/day) — 19:4 ± 7:7 25:6 ± 10:6 — — 0.0084
Insulin aspart (U/day) — — 20:4 ± 16:7 — — —
GWG (kg) 7:7 ± 4:2 6:6 ± 4:1 7:7 ± 5:3 0.25 1.00 0.32
Pre-OGTT GWG (kg) 4:5 ± 3:3 3:6 ± 3:5 4:2 ± 3:9 0.17 0.70 0.63
OGTT-delivery GWG (kg) 3:1 ± 3:0 3:0 ± 3:3 3:6 ± 4:6 0.99 0.58 0.62
HbA1c before delivery (%) 5:5 ± 0:3 5:5 ± 0:4 5:5 ± 0:4 1.00 1.00 1.00
Data were shown as mean ± SD or number (%). GWG: gestational weight gain; SII: simple insulin injection; MDI: multiple daily insulin injection; GDM:
gestational diabetes; BMI: body mass index; OGTT: oral glucose tolerance test.

with GDM in whom diet therapy alone failed to achieve tar- NPH insulin alone was used. Later, Coustan et al. [18] pro-
get glucose values. Women who could be successfully treated posed the concept of “prophylactic” insulin therapy to
with the SII regimen were characterized as being similarly reduce macrosomia in GDM pregnancy. They used a uni-
obese to and having less severe glycemia than those in the form regimen involving the combination of a fixed dose of
MDI group. 20 units of NPH and 10 units of insulin before breakfast
Regarding insulin regimens using a single daily dose of and found that the uniform regimen successfully reduced
NPH insulin alone for the treatment of GDM, some experts macrosomic infants compared with the diet-alone group.
have reported the efficacy of a bedtime NPH regimen for The prophylactic effect of the uniform combined fixed-
women with fasting hyperglycemia, as expert opinions with- dose insulin regimen was reproduced in a later study [19].
out clinical evidence [13–15]. There have been few studies In those studies, however, the uniform regimen with a fixed
addressing the efficacy of such NPH-only regimens for insulin dose was used without SMBG measurements, so
women with GDM with not only fasting but also postpran- technically, the concept of the regimen was “prophylactic”
dial hyperglycemia as a treatment option. Historically, and not “therapeutic.”
O’Sullivan et al. [16, 17] reported that insulin therapy in With the widespread use of the SMBG system during
women with GDM reduced macrosomic infants [16] and pregnancy in women with type 1 diabetes since the early
perinatal mortality [17] compared with diet therapy alone. 1980s, intensive tight glycemic control using the MDI regi-
Notably, in those reports, a fixed regimen with 10 units of men, which involves four daily injections instead of a single
Journal of Diabetes Research 5

Table 3: A comparison of the SMBG at baseline and before delivery between the groups.

Diet alone SII group MDI group p value p value p value

(n = 213) (n = 64) (n = 84) (diet vs. SII) (diet vs. MDI) (SII vs. MDI)
Baseline SMBG data
Number of cases 198 60 80
Mean fasting BG (mg/dL) 83 ± 8 94 ± 10 97 ± 13 <0.0001 <0.0001 0.055
Mean postprandial BG (mg/dL) 112 ± 11 130 ± 13 138 ± 14 <0.0001 <0.0001 <0.0001
Achieved TG of
186 (94) 38 (63) 31 (39) <0.0001 <0.0001 0.0040
fasting < 95 mg/dL
Achieved TG of
158 (80) 12 (20) 5 (6) <0.0001 <0.0001 0.014
postprandial < 120 mg/dL
Achieved TG of
190 (96) 34 (57) 25 (31) <0.0001 <0.0001 0.0026
postprandial < 130 mg/dL
SMBG before delivery
Number of cases 142 60 81
Mean fasting BG (mg/dL) 82 ± 8 85 ± 8 84 ± 7 0.11 0.13 0.97
Mean postprandial 113 ± 11 118 ± 11 119 ± 10 0.0027 0.0001 0.91
BG (mg/dL)
Achieved TG of
133 (94) 56 (93) 75 (93) 0.93 0.76 0.87
fasting < 95 mg/dL
Achieved TG of
114 (80) 39 (65) 46 (57) 0.026 0.0003 0.32
postprandial < 120 mg/dL
Achieved TG of
136 (95) 52 (87) 75 (93) 0.020 0.31 0.24
postprandial < 130 mg/dL
Data were shown as mean ± SD or number (%). SMBG: self-monitoring of blood glucose; SII: simple insulin injection; MDI: multiple daily insulin injection;
BG: blood glucose; TG: target glucose.

or double injection of insulin, has been widely used in the Weight gain induced by intensive insulin therapy has
management of type 1 diabetic pregnancy [20]. Tight glyce- long been recognized as a major problem in diabetes therapy
mic control using the MDI regimen with the SMBG system [24, 25], especially with NPH insulin [26]. Weight gain
has been introduced for not only type 1 but also type 2 dia- causes adverse effects, including the manifestation of insulin
betes as well as GDM during pregnancy [21]. Nachum et al. resistance, hyperlipidemia, and blood pressure elevation, not
[22] concluded that giving insulin four times rather than only for nonpregnant adults but also for pregnant women. It
twice daily in pregnancy improved glycemic control and occurs in the early stages of insulin introduction [25], and
the perinatal outcomes in their randomized control trial common reasons for the weight gain include caloric reten-
(RCT). However, in the study, more than one-third of the tion from a reduced urinary excretion of glucose and a
patients were pregestational diabetes. More recently, two reduction in the metabolic rate due to decreased hepatic glu-
landmark RCTs [3, 4] demonstrated the efficacy of interven- cose output [24, 25]. No report has yet described whether
tion in women with mild GDM using regimens consisting of insulin therapy, especially regimens using NPH insulin
either short-acting insulin alone or short-acting insulin and alone, is associated with excessive GWG, although some
NPH in a basal/bolus regimen [23]. In these contexts, the investigators have reported that maternal GWG in women
MDI regimen appears to have become the standard insulin treated with insulin was greater than that in the metformin
treatment in women with GDM, so a simple regimen involv- group [27, 28]. Because weight gain during pregnancy is a
ing NPH alone was abandoned. physiological phenomenon, it is difficult to distinguish
However, while tighter glycemic control undoubtedly insulin-mediated excessive weight gain from physiological
protects the health of both mother and infant, few studies weight gain during pregnancy. In our current study, we
have concluded that the MDI regimen is the best practice demonstrated that women treated with NPH alone were
for all women with GDM who require insulin therapy. The not associated with excessive GWG, with this population
MDI regimen consists of multiple injections of two different actually showing the least GWG among the three groups
types of insulin—namely, intermediate/long-acting and (Table 2). To our knowledge, this is the first report to com-
rapid/ultrarapid insulin—and is thus more complicated than pare the GWG between women treated with NPH insulin
the SII regimen. In addition, such a complicated regimen alone and those who received MDI insulin therapy among
can cause maternal anxiety during pregnancy [5]. Our find- women with GDM. In addition, a recent study reported
ing that more than 40% of women requiring insulin therapy the mean GWG in uncomplicated pregnancies in Japanese
were successfully treated with the SII regimen alone may population was 10:1 ± 3:7 kg [29]. We believe that the results
help reduce this anxiety. of GWG (Table 2) were adequate, as the diet regimen
6 Journal of Diabetes Research

180 ⁎⁎⁎
180 ⁎⁎
⁎⁎⁎ ⁎⁎⁎
160 160
⁎⁎⁎ ⁎⁎ NS
140 140
⁎⁎⁎ NS 138±14
120 130±13
120
(mg/dL)

(mg/dL)
112±11 NS 118±11 119±10
100 100
113±11
94±10 97±13
80 83±8 80 85±8
82±8 84±7
60 60
40 40
20 20
0 0
Fasting BG Postprandial BG Fasting BG Postprandial BG
Diet Diet
SII SII
MDI MDI
(a) Mean blood glucose on SMBG at baseline (b) Mean blood glucose on SMBG before delivery
⁎⁎⁎ ⁎⁎⁎ NS
⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎ ⁎ NS
100 100 NS
⁎⁎⁎ 96 ⁎
94 ⁎⁎ ⁎⁎ 94 93 93 95 93
80 80 NS 87
80 ⁎ 80
(%)

60
(%)

63 60 65
57 57
40 40
39
20 31
20 20
6
0 0
Fasting TG Postprandial TG Postprandial TG Fasting TG Postprandial TG Postprandial TG
< 95 mg/dL < 120 mg/dL < 130 mg/dL < 95 mg/dL < 120 mg/dL < 130 mg/dL
Diet Diet
SII SII
MDI MDI
(c) The rate of achieved TG on SMBG at baseline (d) The rate of achieved TG on SMBG before delivery

Figure 2: A comparison of the mean blood glucose values and the achieving rate of target glucose goals on SMBG at baseline and before
delivery. ∗ p < 0:05, ∗∗ p < 0:005, and ∗∗∗ p < 0:001. SMBG: self-monitoring of blood glucose; BG: blood glucose; SII: simple insulin
injection; MDI: multiple dairy injection; TG: target glucose; NS: not significant.

generally includes 30% of caloric restriction for women target glucose levels before delivery in the SII group were
with GDM. comparable to those in the MDI group (Table 3). Second,
The strength of this study is that we used SMBG data to because the cases experienced during the first two years after
evaluate glucose control. The SMBG measurements are used the introduction of the SII regimen were excluded, we
not only to decide the indication of insulin therapy but also believe that, as a standardized protocol, this effectively min-
to evaluate the tight glucose control required in women with imized physician bias.
GDM. Thus, in the clinical setting, SMBG is an essential and However, several limitations associated with the present
integral tool in the care of GDM. While the PG (but not the study also warrant mention. First, the decision to switch the
HbA1c) values at the diagnostic test differed slightly regard- treatment from diet therapy alone to insulin therapy and
ing the basal glycemia between the groups (Table 1), the from the SII to MDI regimen was left to the attending phy-
SMBG indices at baseline clearly demonstrated that women sicians, although the physicians shared the standard of
with the SII regimen were more significantly hyperglycemic achieving the target glucose value, with this achievement
and had lower rates of achieving control than women in the defined as reaching approximately 80% of the fasting and
diet group (Table 3), suggesting that the insulin therapy was nonfasting target glucose values. Again, the baseline SMBG
rationally indicated. The baseline SMBG data also clearly data plainly demonstrated that difference in the glycemic
indicated that the levels of hyperglycemia were significantly severity between the groups. In addition, the rate of insulin
higher in the MDI group than in the SII group. Conse- therapy in our study was 41% for women with GDM. The
quently, it is obvious that the level of baseline hyperglycemia rate was relevant in the high-risk population managed in a
in the SII group fell between that in the diet group and the tertiary perinatal care center [7, 8, 30–32]. Therefore, even
MDI group. In addition to the perinatal outcomes, the though physicians’ biases may have affected the results, such
SMBG indices clearly showed that the achievement rates of an influence was likely minimal. Regarding the postprandial
Journal of Diabetes Research 7

Table 4: A comparison of the perinatal outcomes between the groups.

Diet alone SII group MDI group p value p value p value

(n = 213) (n = 64) (n = 84) (diet vs. SII) (diet vs. MDI) (SII vs. MDI)
Gestational age at delivery (weeks) 38:6 ± 1:8 38:1 ± 2:5 38:7 ± 0:9 0.11 0.94 0.11
Cesarean section (%) 49 (23) 16 (25) 22 (26) 0.74 0.56 0.87
Primary cesarean section (%) 29 (14) 5 (8) 13 (15) 0.21 0.68 0.16
Hypertension disorders in pregnancy (%) 3 (1) 4 (6) 6 (7) 0.031 0.009 0.83
Birthweight (g) 2945 ± 463 2925 ± 483 2997 ± 375 0.94 0.65 0.60
Male infants (%) 108 (51) 33 (52) 39 (46) 0.90 0.51 0.54
Birthweight z-score −0:02 ± 1:03 0:06 ± 0:93 0:03 ± 1:01 0.82 0.93 0.97
LGA infants (%) 24 (11) 7 (11) 12 (14) 0.94 0.47 0.55
SGA infants (%) 24 (11) 4 (6) 6 (7) 0.24 0.29 0.83
Apgar score (1 min) 8:3 ± 1:0 8:3 ± 0:7 8:3 ± 0:7 0.96 0.99 0.93
Apgar score (1 min) ≤7 (%) 11 (5) 2 (3) 3 (4) 0.47 0.53 0.87
Apgar score (5 min) 9:0 ± 0:4 8:9 ± 0:5 9:0 ± 0:4 0.24 0.99 0.29
Apgar score (5 min) ≤7 (%) 0 (0) 1 (2) 0 (0) 0.07 — 0.25
Shoulder dystocia (%) 1 (0.6) 1 (1.7) 0 (0) 0.40 0.53 0.26
NICU admission (%) 33 (11) 13 (22) 12 (14) 0.40 0.59 0.25
Hypoglycemia 3 (2) 2 (3) 3 (4) 0.43 0.27 0.85
Hyperbilirubinemia (%) 21 (11) 7 (15) 12 (15) 0.99 0.38 0.49
Data were shown as mean ± SD or number (%). SII: simple insulin injection; MDI: multiple daily insulin injection; LGA: large-for-gestational age; SGA: small-
for-gestational age; NICU: neonatal intensive care unit.

target glucose, the achievement rate of <120 mg/dL in the contribute to their safety, and reduce anxiety compared with
insulin groups did not reach that in the diet group. Although the MDI regimen.
a value of <120 mg/dL is generally recommended worldwide
[11], the targets including both fasting and postprandial Data Availability
values were derived from empirical data of uncomplicated
pregnant women, not evidence-based in the association of The original data used to support the findings of this study
adverse perinatal outcomes [33]. In fact, the perinatal out- are available via private communication with the corre-
comes did not differ between the diet and insulin groups in sponding author upon request.
our study. In addition, the targets were derived from venous
sampling data, rather from SMBG data [33]. As described in
Materials and Methods, it is a characteristic that postpran-
Conflicts of Interest
dial values in the SMBG measurement are higher than those The authors report no conflict of interest.
in the venous sampling, while those at fasting are similar to
the venous values [11]. Therefore, we believe a value of
<130 mg/dL is an acceptable target in the clinical setting. Authors’ Contributions
Finally, on top of having a small sample size, this study MiF, IY, and HY wrote the initial research proposal and
was conducted at a single tertiary center including women manuscript and analyzed data. KA and NK contributed to
with only a Japanese ethnic background, so our results have analyze data. JY, MK, SoS, SaS, and MaF also contributed
limited adaptability to a general population. GDM is to the discussion and interpretation of data.
strongly associated with the development of type 2 diabetes
after delivery, and East Asian populations, including the
Japanese, have one of the highest incidences of type 2 diabetes References
and are among the least obese in the world [34, 35]. Therefore,
[1] International Association of Diabetes and Pregnancy Study
our findings may contribute to some extent to the manage-
Groups, “International Association of Diabetes and Pregnancy
ment of GDM, especially in East Asian populations. Study Groups Recommendations on the diagnosis and classifi-
In conclusion, we demonstrated that, among women cation of hyperglycemia in pregnancy,” Diabetes Care, vol. 33,
with GDM who required insulin therapy because they failed no. 3, pp. 676–682, 2010.
to achieve the target glucose values with nutritional therapy [2] D. R. Coustan, L. P. Lowe, B. E. Metzger, and A. R. Dyer, “The
alone, more than 40% were successfully treated with the SII Hyperglycemia and Adverse Pregnancy Outcome (HAPO)
regimen instead of the MDI regimen without any increase in study: paving the way for new diagnostic criteria for gesta-
adverse perinatal outcomes or excessive GWG. This simple tional diabetes mellitus,” American Journal of Obstetrics and
procedure may ensure patient compliance with treatment, Gynecology, vol. 202, no. 6, pp. 654.e1–654.e6, 2010.
8 Journal of Diabetes Research

[3] C. A. Crowther, J. E. Hiller, J. R. Moss, A. J. McPhee, W. S. Jeffries, [17] J. B. O’Sullivan, S. S. Gellis, R. V. Dandrow, and B. O. Ten-
and J. S. Robinson, “Effect of treatment of gestational diabetes ney, “The potential diabetic and her treatment in preg-
mellitus on pregnancy outcomes,” The New England Journal of nancy,” Obstetrics and Gynecology, vol. 27, no. 5, pp. 683–
Medicine, vol. 352, no. 24, pp. 2477–2486, 2005. 689, 1966.
[4] M. N. Landon, C. Y. Spong, E. Thom et al., “A multicenter, [18] D. R. Coustan and J. Imarah, “Prophylactic insulin treat-
randomized trial of treatment for mild gestational diabetes,” ment of gestational diabetes reduces the incidence of macro-
The New England Journal of Medicine, vol. 361, no. 14, somia, operative delivery, and birth trauma,” American
pp. 1339–1348, 2009. Journal of Obstetrics and Gynecology, vol. 150, no. 7,
[5] A. L. Hui, G. Sevenhuysen, D. Harvey, and E. Salamon, “Stress pp. 836–842, 1984.
and anxiety in women with gestational diabetes during dietary [19] D. J. Thompson, K. B. Porter, D. J. Gunnells, P. C. Wagner,
management,” The Diabetes Educator, vol. 40, no. 5, pp. 668– and J. A. Spinnato, “Prophylactic insulin in the management
677, 2014. of gestational diabetes,” Obstetrics and Gynecology, vol. 75,
[6] U. M. Schaefer-Graf and H. Kleinwechte, “Diagnosis and no. 6, pp. 960–964, 1990.
new approaches in the therapy of gestational diabetes melli- [20] L. Jovanovic and C. M. Peterson, “Management of the
tus,” Current Diabetes Reviews, vol. 2, no. 3, pp. 343–352, pregnant, insulin-dependent diabetic woman,” Diabetes Care,
2006. vol. 3, no. 1, pp. 63–68, 1980.
[7] H. Yamashita, I. Yasuhi, Y. Kugishima et al., “Factors associ- [21] N. Langer and O. Langer, “Emotional adjustment to diagnosis
ated with patients with gestational diabetes in Japan being at and intensified treatment of gestational diabetes,” Obstetrics
increased risk of requiring intensive care,” International Jour- and Gynecology, vol. 84, no. 3, pp. 329–334, 1994.
nal of Gynecology & Obstetrics, vol. 140, no. 2, pp. 170–174, [22] Z. Nachum, I. Ben-Shlomo, E. Weiner, and E. Shalev, “Twice
2018. daily versus four times daily insulin dose regimens for diabetes
[8] I. Yasuhi, H. Yamashita, K. Maeda et al., “High-intensity in pregnancy: randomised controlled trial,” BMJ, vol. 319,
breastfeeding improves insulin sensitivity during early post‐ no. 7219, pp. 1223–1227, 1999.
partum period in obese women with gestational diabetes,” [23] A. McElduff and R. G. Moses, “Insulin therapy in pregnancy,”
Diabetes/Metabolism Research and Reviews, vol. 35, no. 4, arti- Endocrinology and Metabolism Clinics of North America,
cle e3127, 2019. vol. 41, no. 1, pp. 161–173, 2012.
[9] H. Minakami, Y. Hiramatsu, M. Koresawa et al., “Guidelines [24] M. G. Carlson and P. J. Campbell, “Intensive insulin therapy
for obstetrical practice in Japan: Japan Society of Obstetrics and weight gain in IDDM,” Diabetes, vol. 42, no. 12,
and Gynecology (JSOG) and Japan Association of Obstetri- pp. 1700–1707, 1993.
cians and Gynecologists (JAOG) 2011 edition,” The Journal [25] D. Russell-Jones and R. Khan, “Insulin-associated weight gain
of Obstetrics and Gynaecology Research, vol. 37, no. 9, in diabetes – causes, effects and coping strategies,” Diabetes,
pp. 1174–1197, 2011. Obesity & Metabolism, vol. 9, no. 6, pp. 799–812, 2007.
[10] M. B. Landon, P. M. Catalano, and S. G. Gabbe, “Diabetes mel- [26] D. L. Russell-Jones, “Insulin detemir and basal insulin ther-
litus complicating pregnancy,” in Obstetrics: Normal and Prob- apy,” Endocrinology and Metabolism Clinics of North America,
lem Pregnancies, M. B. Landon, H. L. Galan, and E. R. M. vol. 36, Supplement 1, pp. 7–13, 2007.
Jauniaux, Eds., pp. 871–907, Elsevier, Philadelphia, 8th edition, [27] J. A. Rowan, W. M. Hague, W. Gao, M. R. Battin, and M. P.
2021. Moore, “Metformin versus insulin for the treatment of gesta-
[11] K. Tonyushkina and J. H. Nichols, “Glucose meters: a review of tional diabetes,” The New England Journal of Medicine,
technical challenges to obtaining accurate results,” Journal of vol. 358, no. 19, pp. 2003–2015, 2008.
Diabetes Science and Technology, vol. 3, no. 4, pp. 971–980, [28] J. Balani, S. L. Hyer, D. A. Rodin, and H. Shehata, “Pregnancy
2009. outcomes in women with gestational diabetes treated with
[12] K. Itabashi, F. Miura, R. Uehara, and Y. Nakamura, “New Jap- metformin or insulin: a case–control study,” Diabetic Medi-
anese neonatal anthropometric charts for gestational age at cine, vol. 26, no. 8, pp. 798–802, 2009.
birth,” Pediatrics International, vol. 56, no. 5, pp. 702–708, [29] S. Ikenoue, K. Miyakoshi, Y. Kasuga, D. Ochiai, T. Matsumoto,
2014. and M. Tanaka, “Impaired fetal growth in mothers with inad-
[13] A. K. Blum, “Insulin use in pregnancy: an update,” Diabetes equate gestational weight gain: a retrospective study in Japanese
Spectrum: A Publication of the American Diabetes Association, uncomplicated pregnancy,” The Journal of Maternal-Fetal &
vol. 29, no. 2, pp. 92–97, 2016. Neonatal Medicine, vol. 33, no. 13, pp. 2227–2231, 2020.
[14] L. A. Barbour, C. Scifres, A. M. Valent et al., “A cautionary [30] T. Pertot, L. Molyneaux, K. Tan, G. P. Ross, D. K. Yue, and
response to SMFM statement: pharmacological treatment of J. Wong, “Can common clinical parameters be used to iden-
gestational diabetes,” American Journal of Obstetrics and tify patients who will need insulin treatment in gestational
Gynecology, vol. 219, no. 4, pp. 367.e1–367.e7, 2018. diabetes mellitus?,” Diabetes Care, vol. 34, no. 10, pp. 2214–
[15] M. Subriabre, R. Villalobos-Labra, L. Silva, F. Pardo, and 2216, 2011.
L. Sobrevia, “Insulin therapy and foetoplacental endothelial [31] V. W. Wong and B. Jalaludin, “Gestational diabetes mellitus:
dysfunction in gestational diabetes mellitus,” in Frontier in who requires insulin therapy?,” The Australian & New Zeal-
Clinical Drug Research-Dibetes and Obesity (FCDR-DO), and Journal of Obstetrics & Gynaecology, vol. 51, no. 5,
Atta-ur-Rahman, Ed., vol. 5, pp. 66–86, Bentham Science Pub- pp. 432–436, 2011.
lishers, Sharjah, United Arab Emirates, 2019. [32] S. H. Koning, K. A. Scheuneman, H. L. Lutgers et al., “Risk
[16] J. B. O’Sullivan, C. M. Mahan, D. Charles, and R. V. Dandrow, stratification for healthcare planning in women with gesta-
“Medical treatment of the gestational diabetic,” Obstetrics and tional diabetes mellitus,” The Netherlands Journal of Medicine,
Gynecology, vol. 43, no. 6, pp. 817–821, 1974. vol. 74, no. 6, pp. 262–269, 2016.
Journal of Diabetes Research 9

[33] T. L. Hernandez, J. E. Friedman, R. E. Van Pelt, and L. A. Bar-

bour, “Patterns of glycemia in normal pregnancy should the
current therapeutic targets be challenged?,” Diabetes Care,
vol. 34, no. 7, pp. 1660–1668, 2011.
[34] R. C. W. Ma and J. C. N. Chan, “Type 2 diabetes in East Asians:
similarities and differences with populations in Europe and the
United States,” Annals of the New York Academy of Sciences,
vol. 1281, no. 1, pp. 64–91, 2013.
[35] M. Takahara, H. Kaneto, N. Katakami, T. Matsuoka,
M. Matsuhisa, and I. Shimomura, “Impaired suppression of
endogenous glucose production in lean Japanese patients with
type 2 diabetes mellitus,” Diabetes Research and Clinical Prac-
tice, vol. 93, no. 1, pp. e1–e2, 2011.