Table 1 Characteristics of the experimental studies reviewed

Walking alone

 Bo et al. 2014 [32]

Italy

2 × 2 factorial design, baseline at 24–26 weeks gestation, follow up evaluation at 38th week or before delivery.

Four different participant groups:

-D: Dietary recommendations only

-B: Behavioral dietary recommendations only

-E: Exercise only

-BE: Behavioral dietary recommendations, and exercise.

(Groups E + BE and B + BE were combined for analysis)

200

Supervision: No

Frequency: Every day

Intensity: 12–14 on the Borg Rating of Perceived Exertion scale

Type: Walking

Duration: 20 min/ day or 140 min/week

Behavioral dietary recommendation

Primary outcome: Fasting blood glucose

Secondary outcomes: Postprandial glucose, Hemoglobin A1c (HbA1c), Insulin, Homeostasis-Model-Assessment-Insulin Resistance (HOMAIR)

Adjusted difference in postprandial glucose (mg/dl) for exercise versus control was − 11.1(95 % CI: -16.1, -0.2; p < 0.001).

Adjusted difference in HbA1c (%) for exercise versus control was − 0.3 (95 % CI: −0.4, − 0.2); p < 0.001).

No significant difference in Fasting glucose, Log-fasting insulin, and Log-HOMA-IR.

Walking combined with other Aerobic and/or Resistance Exercises

 Avery et al. 1997 [34]

USA

RCT, enrolled ≤ 34 weeks gestation, follow-up measurements 4-weeks later

29

Supervision: Partly

Frequency: 3–4 times/week till the end of pregnancy

Intensity: 70 % of calculated maximal heart rate

Type: Cycle ergometer for supervised, either walking or cycle ergometer for unsupervised (most participants chose walking for unsupervised)

Duration: 30 min (this included a 5 min warm-up and 5 min cool-down before/after a 20 min session)

Dietary therapy and maintaining current physical activity level.

Hemoglobin A1c,

fasting and postprandial blood glucose.

No significant difference in hemoglobin A1c, fasting blood glucose and post prandial glucose.

 Halse et al. 2014 [33]

Australia

RCT, enrolled 26–30 weeks gestation, follow-up measurements 6-weeks later

40

Supervision: Partly, though all exercise occurring at home

Frequency: 3 supervised and 2 unsupervised per week for approximately 6 weeks

Intensity: 5 min warm-up at 9–11 RPE (Borg scale) followed by phases of continuous moderate-intensity (12–14 RPE) and interval bouts higher intensity (15–15 RPE) mixed with lower intensity (9–11 RPE). Ends with 5–10 min cool-down at 9–11 RPE

Type: Upright stationary cycle ergometer for supervised. Not specified for unsupervised sessions, depends on the participant’s choice (walking and cycling were the most commonly selected).

Duration: Range from 25 to 45 min according to participant ability and progression.

Conventional GDM treatment.

Fasting blood glucose,

Postprandial blood glucose, glycosylated hemoglobin

and insulin levels

Mean postprandial glucose was significantly lower in exercise group compared with control (P = 0.046)

No significant difference in daily fasting glucose, HbA1C, fasting glucose and insulin levels after Oral Glucose Tolerance Test.

 Sklempe Kokic et al. 2018 [38]

Croatia

RCT, enrolled ≤ 30 weeks gestation, follow up data on glucose levels collected monthly /bi-monthly till end of pregnancy (38–40 weeks; data abstracted following childbirth)

42

Supervision: Partly

Frequency: 2 times/ week of supervised session and unsupervised walking daily for a minimum of 6 weeks

Intensity: 13–14 on the Borg Rating of Perceived Exertion scale for aerobic and resistance exercises parts

Type: Unsupervised walking plus supervised session which includes aerobic (on treadmill), resistance, pelvic floor and stretching, relaxation

Duration: 30 min for unsupervised walking, 50–55 min of the supervised session

Standard Antenatal care for GDM

Fasting and postprandial glucose levels at the end of pregnancy

Mean postprandial glucose was lower for exercise group (4.66 ± 0.46 mmol/L) vs. control group (5.30 ± 0.47 mmol/L); (p < 0.001)

No significant differences in fasting blood glucose.

Other Aerobic Exercises

 Bung et al. 1991 [39]

USA

RCT, enrolled 27–32 weeks gestation, blood glucose determined weekly thereafter.

41

Supervision: Yes

Frequency: 3 times/week

Intensity: 50 % VO2 max

Type: Recumbent bicycle

Duration: 45 min

Insulin therapy

Mean weekly fasting plasma glucose

No statistical differences between exercise and Insulin subjects were observed in blood glucose measurements.

 Jovanovic-Peterson et al. 1989 [37]USA

RCT,

Gestational age at enrollment unknown, weekly follow up for 6 weeks

19

Supervision: Yes

Frequency: 3 times/week

Intensity: Moderate intensity

Type: Arm ergometer training

Duration: 20 min for 6 weeks

Diet alone

Fasting glucose levels, post prandial glucose levels, glycosylated hemoglobin

Significant difference in fasting and glycated Hemoglobin.

Resistance Training

 deBarros et al. 2010 [36]

Brazil

RCT, enrolled sedentary women at 24–34 weeks gestation, followed to end of pregnancy

64

Supervision: Partly

Frequency: 3 times/week

Intensity: Perceived exertion scale “somewhat heavy”

Description: Resistance band to target main muscle groups (i.e., stations for chest, back, biceps, triceps, deltoid, quadriceps, thigh, and calf muscles); performed 15 repetitions at each station, with a minimum resting period of 30 s and a maximum of 1 min in between stations. In weeks 1 and 2, underwent 2 circuit series, then 3 circuit series from week 3 to end of pregnancy

Usual prenatal care

Self-monitored capillary glucose values and need for insulin

Exercise group had significantly higher percentage of weeks with 80 % of capillary glucose measurements within target range compared with controls (63 % vs. 41 %).

No significant difference in mean glucose levels.

Significant decrease in the number of patients who required insulin in exercise group (21.9 %) compared with control group (56.3 %).

 Brankston et al. 2004 [35]

Canada

RCT, enrolled 26–32 weeks gestation, followed to end of pregnancy

32

Supervision: Partly

Frequency: 3 times/week

Intensity: felt “somewhat hard”

Description: Anchored rubber tubing to complete 8 exercises [i.e., plies (squats with outward facing knees), military press, knee extension, hamstring curl, bench press, lateral pull down, seated row, and triceps press] with < 1 min rest in between. Weeks 1 and 2 included 2 sets of 15 repetitions for each exercise; 3 sets of 15 repetitions for each in week 3; 3 sets of 20 repetitions for each from week 4 to end of pregnancy

Standard Diabetic Diet

Primary outcome: insulin requirement

Secondary outcome: Latency to insulin treatment and amount of insulin

Significant difference in amount of insulin required (units/kg) (diet alone, 0.48 ± 0.3 versus diet and exercise, 0.22 ± 0.2)

Significant difference in latency (weeks) to insulin requirement (diet alone, 1.11 ± 0.8 versus diet and exercise, 3.71 ± 3.1)