Table 1.4.
LGA and high birth weight and long-term outcomes—type 1 and type 2 diabetes.
| Author, year, country | Study design | Cases | Outcomes (risk estimates) | Reference group (weight) | Comments/adjustments | Risk of bias | Directness | Precision |
|---|---|---|---|---|---|---|---|---|
| • Type 1 and type 2 diabetes • Systematic review/meta-analysis n = 6 | ||||||||
| Cardwell et al. (2010), UK (137) | • Type 1 diabetes • Meta-analysis • Cohort n = 5 case–control n = 20 • 30 populations |
12,087 | • Birth weight >4,000 g: • OR (cohort studies) 1.15 (95% CI 1.05–1.26) • OR (case–control studies) 1.05 (95% CI 0.95–1.17) • AOR (all studies) 1.11 (95% CI 1.03–1.20) |
3,000–3,500 g | All ages included in risk estimates not only children/adolescents <18 years | |||
| Harder et al. (2007), Germany (158) | • Type 2 diabetes • Meta-analysis • cohort n = 10 • case–control n = 3 |
6,901 | • Birth weight >4,000 g: • 1OR 1.27 (95% CI 1.01–1.59) • 2OR 1.36 (95% CI 1.07–1.73) |
• 1 ≤ 4,000 g • 22,500 g • 4,000 g |
No separate OR calculated for children/adolescents <18 years | |||
| Harder et al. (2009), Germany (18) | • Type 1 diabetes • Meta-analysis • cohort n = 2 • case–control n = 10 |
7,491 | • Birth weight >4,000 g: • OR. 1.17 (95% CI 1.09–1.26) • AOR 1.43 (95% CI 1.11–1.85) |
<4,000 g | Adjusted for confounders in seven of the included studies and wide difference in the number of confounders ranging from 2 to 14 | |||
| Knop et al. (2018), China (160) | • Type 2 diabetes • Systematic review, meta-analysis • 49 studies • Cohort n = 36 • Case–control n = 8 • Cross-sectional n = 5 • (for high birth weight 32 studies) |
43,549 | • Birth weight >4,500 g: • OR 1.19 (95% CI 1.04–1.36) |
4,000–4,500 g | Adult only (>18 years) | |||
| Whincup et al. (2008), UK (159) | Type 2 diabetes systematic review, meta-analysis | 6,260 | • Per 1,000-g increase: • OR 0.80 (95% CI 0.72–0.89) • Birth weight >4,000 g: • OR 1.35 (95% CI 0.67–2.72) |
<4,000 g | Adults | |||
| Zhao et al. (2018), China (161) | • Type 2 diabetes • Meta-analysis, • Cohort n = 16 • Case–control n = 5 |
22,341 | • Birth weight >4,000 g: • OR was calculated for all ages: • OR 1.11 (95% CI 1.00–1.24) |
2,500–4,000 g | Only 2 studies were limited to children/adolescents less than 18 years, both were case–control studies. No separate calculated OR for children/adolescents separately | |||
| Original articles | ||||||||
| • Type 1 diabetes • Original articles n =20 | ||||||||
| Bock et al. (1994), Denmark (144) | Case–control | 837 | • No statistical differences in mean birth weight between the cases and controls: • 3,381, SD 536 g vs. 3,351, SD 602 g |
• Exclusion criteria: mother with IDDM at the time of birth • No risk estimates |
Serious | Good | Fair | |
| Borras et al. (2011), Spain (145) | Case–control | 306 | • LGA >90 percentile • OR for diabetes 1.45 (95% CI 1.02–2.07) |
10–90th percentile | • No adjustment • 43 of originally 349 cases excluded due to missing data on birth weight |
Serious | Good | Fair |
| Cardwell et al. (2005), UK (138) | Cohort study | 991 | • Birth weight >4,000 g: • ARR 1.68 (95% CI 1.30–2.18) • Birth weight 3,500–3,999 g: • ARR 1.48 (95% CI 1.20–1.83) |
<3,000 g | • Adjusted for maternal age, birth order, year of birth, gestational age • Missing data 8% |
Moderate | Good | Good |
| Goldacre (2017), UK (139) | Cohort study | 2,969 | • Birth weight 4,000–5,499 g: • AHR 1.12 (95% CI 0.99–1.27) • Birth weight 3,500–3,999 g: • AHR 1.11 (95% CI 1.02–1.22) |
3,000–3,499 g | Adjusted for infant sex, gestational age, maternal type 1 diabetes, maternal obesity, deprivation quintile, and caesarean section | Moderate | Good | Good |
| Haynes et al. (2007), Australia (146) | Cohort | 840 | • Birth weight ≥4,000 g: • IRR 1.19 (95% CI 0.95–1.49) • Birth weight 3,500–3,999 g: • IRR 1.09 (95% CI 0.92–1.28) |
3,000–3,499 g | Adjusted for maternal age, gestational age, birth order, and year of birth | Moderate | Good | Good |
| Levins et al. (2007), UK (140) | Cohort | 518 | • Estimated rate of diabetes (<15 years) in birth weight categories: • 3,500–3,999: Rate 1.55 (95% CI 1.28–1.86) • ≥4,000: Rate 1.65 (95% CI 1.17–2.26) |
No ref group | Adjusted for year of birth, Rates only per 1,000 individuals presented. No difference between birth categories | Serious | Good | Fair |
| Jones et al. (1999), UK (147) | Case–control study | 315 | • Birth weight 3,500–3,900 g: • ARR 1.00 (95% CI 0.74–1.36) • Birth weight ≥4,000 g: • ARR 1.15 (95% CI 0.76–1.75) |
3,000–3,499 g | Adjusted for maternal age, parity, birth weight for gestational age, gestational age and year of birth. Data included in Ievins (1997) and more restricted data material | Moderate | Good | Fair |
| Khashan et al. (2015), Sweden (141) | Cohort study | 13,944 | • Birth weight 4,000–5,500 g: • ARR 1.01 (95% CI 0.96–1.05) • LGA (+2 SD above mean) vs. AGA • RR 1.14 (95% CI 1.04–1.24) |
3,000–3,999 g | Adjusted for offspring age as a time-dependent variable, year of birth, maternal age, education, BMI, country of origin, pre-gestational diabetes, gestational diabetes and infant sex | Low | Good | Good |
| Kuchlbauer et al. (2014), Germany (142) | Cohort study | 1,117 | No risk estimate available. cases with type 1 diabetes had higher birth weight measured as SDS (0.15 vs. 0.03) than the newborn in the control SDS (z-scores) are calculated from birth weights based on population reference values | No adjustment. No risk estimates | Critical | Good | Fair | |
| Lawler-Heavner et al. (1994), USA (148) | Case–control study | 221 | • Birth weight 3,500–3,999 g: • AOR 0.9 (95% CI 0.5–1.7) • Birth weight ≥4,000 g: • AOR 1.0 (95% CI 0.4–2.5) |
<3,000 g | Adjusted for sex, age and birth in Colorado | Serious | Good | Fair |
| McKinney et al. (1999), UK (149) | Case–control study | 196 | • Birth weight ≥3,500 g: • OR 1.01 (95% CI 0.68–1.51) |
2,500–3,000 g | Uncertain whether the results are adjusted or not | Serious | Good | Fair |
| Metcalfe and Baum (1992), UK (150) | Case–control study | 952 | • Results given according to proportions in three categories of birth weight: • <2,500: insulin-dependent diabetes mellitus (IDDM) 65 (7%), Office of Population Censuses and Surveys (OPCS) 32,779 (6%) • 2,500–3,999: IDDM 783(82%), OPCS 509707 (86%) • ≥4,000: IDDM 104 (11%), OPCS 46012 (8%) |
No adjustments. No risk estimates. No conclusions drawn | Serious | Good | Fair | |
| Patterson et al. (1994), UK (151) | Case–control study | 529 | • Birth weight ≥4,000 g; • OR 1.14 (95% CI 0.75–1.74) |
2,500–3,999 g | No adjustments | Serious | Good | Fair |
| Rosenbauer et al. (2008), Germany (152) | • Case–control • Nationwide hospital-based surveillance (ESPED) |
• 760 • 719 cases in birthweight analysis |
• Birth weight ≥4,000 g: • AOR 1.28 (95% CI 0.94–1.73) |
3,000–3,999 g | Probably adjusted for familiar type 1 diabetes, social status, maternal age, number of siblings and change of residency | Moderate | Good | Fair |
| Stene et al. (2001), Norway (143) | Cohort study | 1,824 | • 3,500–3,999 g: RR 2.11 (95% CI 1.24–3.58) • 4,000–4,499 g: RR 2.38 (95% CI 1.39–4,06) • ≥4,500 g: RR 2.21 (95% CI 1.24–3.94) |
<2,000 g | Adjusted for sex, maternal age, plurality, birth weight, gestational age, caesarean section, pre-eclampsia, year of birth | Low | Good | Fair |
| Stene and Joner (2004), Norway (153) | Case–control study | 545 | • 3,500–3,999 g: AOR 0.94 (95% CI 0.44–1.99) • ≥4,000 g: AOR 1.01 (95% CI 0.46–2.29) |
<2,500 g | Adjusted for sex, maternal age, plurality, birth weight, gestational age, caesarean section, pre-eclampsia, duration of breast feeding, maternal education, atopic eczema, allergic rhino-conjunctivitis and asthma | Low | Good | Fair |
| Tai et al. (1998), Taiwan (154) | Case–control | 117 | • Birth weight ≥4,000 g: • AOR 0.97 (95% CI 0.39–2.45) |
<3,000 g | Adjusted for age, sex | Critical | Poor | Poor |
| Wadsworth et al. (1997), UK (155) | Case–control | • 281 • 218 cases included in the analysis |
• No significant association with birth weight analyzed as a continuous variable • Unadjusted OR per kg increase in birth weight 0.94 (95% CI 0.65–1.35) |
Unadjusted | Serious | Good | Poor | |
| Waernbaum et al. (2019), Sweden (156) | Case–control study | 14,949 | AOR 1.08 (95% CI 1.06–1.10) | Birth weight z-score category with the interval 0–1 as reference | Adjusted for urinary tract infection, PROM, maternal age, PTB, maternal BMI | Low | Good | Good |
| Wei et al. (2006), Taiwan (157) | Case–control study | 277 | ≥4,000 g: AOR 1.01 (95% CI 0.46–2.29) | <2,600 g | Adjusted for age, sex, socioeconomy, family history of diabetes„ delivery order, breast feeding, BMI, and GDM | Moderate | Fair | Fair |
| Type 2 diabetes | ||||||||
| Hu et al. (2020), China (163) | Cohort | 48,118 | ≥4,000 g: AOR 1.20 (95% CI 1.07–1.34) | 2,500–3,499 g | Adjustments: age, gender, smoking, drinking, education, physical activity, diet habits, systolic blood pressure, dyslipidemia, BMI | Moderate | Fair | Good |
| Zhu et al. (2013), China (164) | Cross-sectional survey | • 903 children with overweight • 2 with type 2 diabetes • 6 with impaired fasting glucose • 16 with impaired glucose tolerance • 2 with impaired fasting glucose + impaired glucose intolerance |
• Birth weight ≥4,000 g: • AOR 1.92 (95% CI 1.06–3.49) • Subgroup of girls analyzed separately: • AOR 4.38 (95% CI 1.21–15.85) |
2,500–3,999 g | Adjusted for age, gender, parental education. Only few children with type 2 diabetes or impaired fasting glucose | Moderate | Fair | Fair |
LGA, large-for-gestational-age; AGA, appropriate-for-gestational-age; HOMA-IR, homeostasis model assessment-insulin resistance; MS, metabolic syndrome; GDM, gestational diabetes mellitus; LBW, low birth weight; HBW, high birth weight; NBW, normal birth weight.