Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 https://doi.org/10.1186/s12884-021-03554-6 RESEARCH ARTICLE Open Access Which growth standards should be used to identify large- and small-for-gestational age infants of mothers with type 1 diabetes? A pre-specified analysis of the CONCEPTT trial Claire L. Meek1,2* , Rosa Corcoy3,4,5, Elizabeth Asztalos6, Laura C. Kusinski1,2, Esther López4,7, Denice S. Feig8,9, Helen R. Murphy2,10,11 and On behalf of the CONCEPTT collaborative group Abstract Background: Offspring of women with type 1 diabetes are at increased risk of fetal growth patterns which are associated with perinatal morbidity. Our aim was to compare rates of large- and small-for-gestational age (LGA; SGA) defined according to different criteria, using data from the Continuous Glucose Monitoring in Type 1 Diabetes Pregnancy Trial (CONCEPTT). Methods: This was a pre-specified analysis of CONCEPTT involving 225 pregnant women and liveborn infants from 31 international centres (ClinicalTrials.gov NCT01788527; registered 11/2/2013). Infants were weighed immediately at birth and GROW, INTERGROWTH and WHO centiles were calculated. Relative risk ratios, sensitivity and specificity were used to assess the different growth standards with respect to perinatal outcomes, including neonatal hypoglycaemia, hyperbilirubinaemia, respiratory distress, neonatal intensive care unit (NICU) admission and a composite neonatal outcome. Results: Accelerated fetal growth was common, with mean birthweight percentiles of 82.1, 85.7 and 63.9 and LGA rates of 62, 67 and 30% using GROW, INTERGROWTH and WHO standards respectively. Corresponding rates of SGA were 2.2, 1.3 and 8.9% respectively. LGA defined according to GROW centiles showed stronger associations with preterm delivery, neonatal hypoglycaemia, hyperbilirubinaemia and NICU admission. Infants born > 97.7th centile were at highest risk of complications. SGA defined according to INTERGROWTH centiles showed slightly stronger associations with perinatal outcomes. Conclusions: GROW and INTERGROWTH standards performed similarly and identified similar numbers of neonates with LGA and SGA. GROW-defined LGA and INTERGROWTH-defined SGA had slightly stronger associations with neonatal complications. WHO standards underestimated size in preterm infants and are less applicable for use in type 1 diabetes. (Continued on next page) * Correspondence: clm70@cam.ac.uk 1 Institute of Metabolic Science, University of Cambridge, Addenbrooke’s Hospital, Box 289, Cambridge CB2 0QQ, UK 2 Cambridge Universities NHS Foundation Trust, Cambridge, UK Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 2 of 10 (Continued from previous page) Trial registration: This trial is registered with ClinicalTrials.gov. number NCT01788527. Trial registered 11/2/2013. Keywords: Large-for-gestational-age, Small for gestational age, Macrosomia, Birth-weight, Diabetes, Pregnancy, CONCEPTT, Growth standards, INTERGROWTH, GROW Background women with T1D, LGA is five times more common than Birth weight is an important indicator of neonatal well- in the background maternity population [5]. Identifica- being [1, 2]. Infants who are small- or large-for- tion of these infants may improve outcomes by increas- gestational-age (SGA or LGA; birth weight < 10th or ing surveillance and targeting interventions to those at >90th percentile) experience higher risks of morbidity highest risk [15]. Furthermore, identification of a single and mortality [3, 4]. Recent population based data sug- growth standard with optimal performance in T1D preg- gests that despite improvements in care, infants of nancies would allow future standardisation of research women with type 1 diabetes (T1D) remain at high risk outcomes. of LGA (rates ~ 50%) [5]. LGA rates were also high in A further challenge in growth assessment in T1D the CONCEPTT international randomized controlled pregnancy regards the use of growth standards in pre- trial of the use of continuous glucose monitoring (CGM) term infants. GROW and INTERGROWTH approach in comparison with capillary blood glucose monitoring preterm growth differently. GROW centiles are based in pregnant women with T1D [6]. LGA rates were sig- upon the Hadlock formula for gestational age, suggesting nificantly reduced in infants of women who used CGM that the growth of preterm and term infants should be (53% vs 69% in home blood glucose monitoring group), exactly the same at any timepoint [16]. The INTE likely due to improved glycaemic control [7]. RGROWTH standards do not make this assumption and Currently there is controversy internationally about are based on the size of preterm infants at birth. Al- which growth standards to use to compare LGA rates in though this approach seems more scientifically justifi- different populations. Customised (Gestation Related able, as growth abnormalities may contribute to preterm Optimum Weight; GROW) centiles [8] were used in birth, the INTERGROWTH standards were based upon CONCEPTT for the comparison of birthweight across limited data from preterm (before 37 weeks) and very international sites and diagnosis of LGA. GROW centiles preterm babies (before 34 weeks), which introduces un- are customised to maternal and neonatal factors includ- certainty [17]. More studies in preterm infants are ing maternal ethnicity, height, weight, parity, neonatal needed to identify which growth standard performs best sex and gestational age [7, 9]. GROW provides country- in this group [18]. specific customised centiles, enabling international com- Our aim was therefore to assess the incidence of LGA parisons between populations. Advocates suggest that and SGA using different definitions and growth stan- customised centiles reduce over-investigation of normal dards in T1D pregnancy, and to assess which standard fetuses and can more accurately predict fetuses at in- was able to identify infants at highest risk of perinatal creased risk of stillbirth and perinatal mortality [9, 10]. complications. Other growth standards based on data from the INTE RGROWTH-21st study (20,486 infants across eight geo- graphical areas) and the World Health Organisation Methods (WHO) Multicentre Growth Reference Study (8500 in- The recruitment, rationale and methodology of the fants across six geographical areas) [11, 12] are used CONCEPTT trial are described in detail elsewhere [6] internationally. Both standards assume < 3.5% of the (ClinicalTrials.gov NCT01788527; registered 11/2/2013; variability in growth is due to differences in ethnicity see Appendix S1 for CONCEPTT collaborative group). and population when circumstances are optimal (e.g In brief, women with T1D were recruited before or dur- healthy, well-nourished mothers) [13]. INTE ing pregnancy and randomized to real-time continuous RGROWTH-21st standards focus on fetal growth and glucose monitoring or capillary glucose monitoring neonatal size at birth, while the WHO charts assess alone. Women in the capillary glucose monitoring group weight-for-age at 0–60 months [12]. In addition, some also had short periods of masked continuous glucose studies in diabetes pregnancy report standard deviation monitoring, to allow comparison of glycaemic control (SD) based birthweight categorisation. between groups. Women were followed-up until delivery Population studies comparing growth standards have with collection of information about birth outcomes. focused on the identification of SGA in general mater- Local policies in the study sites were used to determine nity populations [9, 10, 14]. However, in infants of the optimal timing and method of delivery. Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 3 of 10 Pre-specified neonatal outcomes for the CONCEPTT Statistical analysis study included miscarriage, stillbirth, neonatal death, Continuous data were described as mean (SD) and cat- birth injury, shoulder dystocia, preterm delivery, neo- egorical data as n (%) as appropriate. Data regarding natal hypoglycaemia requiring intravenous dextrose, birthweight were analysed as percentiles. An unadjusted hyperbilirubinaemia, respiratory distress syndrome, neo- log-binomial regression model was used to assess associ- natal intensive care unit admission requiring a duration ations between different growth standards, at different of at least 24 h, total length of hospital stay, birthweight, thresholds, and perinatal outcomes. Results are pre- macrosomia (birthweight ≥4 kg), LGA (>90th centile) sented as relative risk ratios (RR) and 95% confidence in- and SGA (< 10th centile) based on customised centiles. tervals. As other comparable studies in the literature Definitions for CONCEPTT outcomes are given in Ap- provide odds ratios, we have included a table in the sup- pendix S2 and were standardised across all the CON- plementary material with odds ratios and 95% confi- CEPTT sites. This data was collected using participant’s dence intervals, calculated using unadjusted logistic medical records. For the CONCEPTT trial, the compos- regression (Table S1). The threshold for statistical sig- ite neonatal endpoint incorporated pregnancy loss (mis- nificance was p< 0.05. We considered that the best per- carriage, stillbirth, or neonatal death), birth injury, forming growth standard to be that which was neonatal hypoglycaemia, hyperbilirubinaemia, respiratory significantly associated with the most suboptimal peri- distress syndrome, or neonatal intensive care admis- natal outcomes. sion> 24 h. This pre-specified secondary analysis includes data from pregnant and pre-pregnant recruits who be- Results came pregnant during the 6-month pre-pregnancy trial Two hundred twenty-five women and infants were in- and who gave birth to a liveborn infant. Therefore, in cluded in this analysis, including 200 from the pregnancy the current study, the composite neonatal endpoint in- arm and 25 from the pre-pregnancy arm who became corporated birth injury, neonatal hypoglycaemia, hyper- pregnant during the trial. Baseline characteristics and bilirubinaemia, respiratory distress syndrome, or pregnancy outcomes are detailed in Table 1. Most neonatal intensive care admission> 24 h but not preg- women were over 30 years old (mean age 31.4 years), nancy loss. overweight (mean BMI 25.8 kg/m2), of European or Mediterranean ethnicity (86.2%) and approximately half Calculation of birth weight centiles used insulin pump therapy (48.9%). They had T1D of Gestational age at delivery was based upon ultrasound 16.5 years’ duration with suboptimal glucose levels measurements in early pregnancy (approximately 12 (HbA1c 6.9%; 51.8 mmol/mol) in early pregnancy as de- weeks). Maternal height and weight required for the fined by the trial eligibility criteria, which required GROW calculation was measured by trained staff at the HbA1c >=6.5% (48 mmol/mol). Their infants were born baseline study visit. As the CONCEPTT study recruited at 37.0 weeks of gestation, predominantly by Caesarean women in early pregnancy, this was considered broadly section (68.9%). similar to pre-pregnancy weight. Large-for-gestational-age (LGA) rates varied (Table 2. GROW centiles were calculated using version 8 (2017) GROW: 62.2%; INTERGROWTH 66.7%; WHO 29.8%) of the GROW calculator using data about maternal self- and there were also differences in the mean and median reported ethnicity, parity, height, weight, gestational age birthweight centile (Table 1. Mean centiles GROW: 82.1; at birth and neonatal sex [8]. INTERGROWTH centiles INTERGROWTH 85.7; WHO 63.9 centiles; Median cen- were calculated using the windows app (available at tiles GROW: 95.2; INTERGROWTH 95.0; WHO 73.6 https://intergrowth21.tghn.org/intergrowth-21st- centiles). Other measures of birth weight are shown in applications/; accessed 31/03/2019) using information Table 2. Other common perinatal complications in- about infant sex, weight and age (0–60 months). WHO cluded neonatal hypoglycaemia (25.3%), hyperbilirubi- centiles were calculated using data about infant sex, naemia (27.6%), respiratory distress (8.4%) which all weight and age (0 months) using the igrowup package contributed to frequent NICU admissions > 24 h (36.9%). for Stata (available at http://www.who.int/childgrowth/ Birth injury and shoulder dystocia were uncommon, (1/ software/en/; accessed 31/03/2019). To demonstrate dif- 225 (0.4%)) and only occurred in one infant who was ferences and similarities between centile-based methods considered LGA by all criteria. LGA according to and SD-based methods in the assessment of birth GROW, INTERGROWTH and WHO criteria was asso- weight, we also included SD-based definitions for LGA ciated with increased risks of perinatal complications (+ 1 and 2 SDs) using methods which were commonly (Table 2). While each growth standard was associated used in the literature. Calculation of SGA using defini- with some complications, no growth standard identified tions <5th and < 2.5th percentile resulted in too few in- all complications studied. Increased birthweight accord- fants to permit meaningful analysis. ing to GROW displayed more significant associations Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 4 of 10 Table 1 Maternal Infant Characteristics. BMI: body mass index; GROW: gestation related optimum weight; NICU: neonatal intensive care unit; WHO: World Health Organisation. Composite outcome: birth injury, neonatal hypoglycaemia, hyperbilirubinaemia, respiratory distress syndrome, or neonatal intensive care admission. Diabetes complications defined as any retinopathy, neuropathy or nephropathy Mean (SD) or n (%) n=225 MATERNAL CHARACTERISTICS Maternal age, years 31.4 (4.5) BMI at enrolment, kg/m2 25.8 (4.6) Ethnicity European/ Mediterranean origin 194 (86.2) Primiparous 89 (39.6) Duration of diabetes, years 16.5 (7.7) Diabetes complications 58 (25.8) Hypertension pre-pregnancy 15 (6.7) HbA1c at randomisation mmol/mol 51.8 (6.6) HbA1c at randomisation % 6.9 (0.6) Smoking 20 (8.9) Insulin pump 110 (48.9) INFANT CHARACTERISTICS Sex (% male) 115 (51.1) Gestational age at delivery 37.0 (1.6) BIRTHWEIGHT MEASURES Birthweight g 3583.6 (705) Macrosomia >=4 kg 59 (26.2) Median (range) GROW centile 82.1 (25.9) 95.2 (0.1–100.0) INTERGROWTH centile 85.7 (20.8) 95.0 (3.9–100.0) WHO centile 63.9 (32.0) 73.6 (0.0–100.0) OBSTETRIC AND PERINATAL OUTCOMES Caesarean section 155 (68.9) Preterm delivery 89 (39.6) Neonatal hypoglycaemia requiring intravenous dextrose 57 (25.3) NICU admission 83 (36.9) Hyperbilirubinaemia 62 (27.6) Respiratory distress 19 (8.4) Composite neonatal outcome 107 (47.6) with perinatal outcomes than INTERGROWTH (i.e. Overall GROW and INTERGROWTH standards per- with neonatal hypoglycemia, hyperbilirubinemia, NICU formed similarly and identified similar numbers with admission and the composite outcome). LGA (slightly higher for INTERGROWTH) and SGA Most perinatal complications demonstrated a U-shaped (slightly higher for GROW). These standards performed relationship with the birth centile (Fig. 1). Neonatal consistently regardless of sex, ethnicity and timing of de- hypoglycaemia was most frequent in infants born ex- livery (see supporting Tables S2, S3, S4). The positive tremely large for gestational age (ELGA; > 97.7th centile). and negative predictive values of the neonatal outcomes In smaller infants (<25th centile), GROW and INTE and their associated sensitivity and specificity varied de- RGROWTH were associated with hyperbilirubinaemia, pending on outcome (Table 3). The WHO standards do while WHO centiles at < 10th and/or <25th centile not take gestational age at birth into account and there- thresholds were associated with multiple outcomes, in- fore underestimated size in preterm infants. This re- cluding NICU admission, hyperbilirubinaemia, respiratory sulted in a linear association between WHO centile and distress and the composite outcome (Table 2; Fig. 1). preterm delivery (Fig. 1). Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 5 of 10 Table 2 Association of adverse neonatal outcomes with low and high birthweights after GROW, INTERGROWTH and WHO criteria. Relative risk ratios and 95% confidence intervals are reported in comparison to all other pregnancies. Neonatal hypoglycaemia included only infants who required IV dextrose. GROW: gestation related optimum weight; NICU: neonatal intensive care unit; WHO: world health organisation. Composite outcome: birth injury, neonatal hypoglycaemia, hyperbilirubinaemia, respiratory distress syndrome, or neonatal intensive care admission. * p< 0.05; **p< 0.01; ***p< 0.001 N (%) Neonatal NICU Admission Hyper- Respiratory Composite hypoglycaemia RR (95% CI) bilirubinaemia Distress Outcome RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI) Birthweight > 84.1st centile (mean + 1 sd) GROW > 84.1st centile 146/225 1.66 (0.97 to 2.84) 1.41 (0.95 to 1.56 (0.94 to 2.56) 2.03 (0.70 to 5.91) 1.46 (1.05 to 2.02)* (64.9%) 2.10) INTERGROWTH > 84.1st 167/225 1.45 (0.81 to 2.61) 1.17 (0.77 to 1.19 (0.71 to 1.99) 1.30 (0.45 to 3.77) 1.20 (0.85 to 1.69) centile (74.2%) 1.77) WHO > 84.1st centile 82/225 1.36 (0.87 to 2.13) 0.99 (0.69 to 0.77 (0.48 to 1.23) 0.80 (0.32 to 2.04) 1.00 (0.75 to 1.33) (36.4%) 1.41) Birthweight >90th centile (mean + 1.28 sd) GROW >90th centile 140/225 1.86 (1.09 to 3.20)* 1.49 (1.01 to 1.61 (0.99 to 2.62) 2.28 (0.78 to 6.63) 1.56 (1.13 to (62.2%) 2.21)* 2.16)** INTERGROWTH >90th 150/225 1.69 (0.97 to 2.94) 1.48 (0.98 to 1.44 (0.87 to 2.36) 1.88 (0.64 to 5.45) 1.41 (1.01 to 1.96)* centile (66.7%) 2.22) WHO >90th centile 67/225 1.48 (0.94 to 2.33) 1.02 (0.70 to 0.82 (0.50 to 1.34) 0.63 (0.22 to 1.82) 1.05 (0.78 to 1.41) (29.8%) 1.47) Birthweight > 97.7th centile (mean + 2 sd) GROW > 97.7th centile 95/225 2.35 (1.47 to 3.75)*** 1.62 (1.15 to 1.56 (1.02 to 2.38)* 1.88 (0.79 to 4.50) 1.50 (1.14 to (42.2%) 2.28)** 1.97)** INTERGROWTH > 97.7th 92/225 2.30 (1.45 to 3.65)*** 1.48 (1.06 to 1.19 (0.78 to 1.82) 1.61 (0.68 to 3.80) 1.42 (1.08 to 1.86)* centile (40.9%) 2.08)* WHO > 97.7th centile 28/225 2.29 (1.45 to 3.61)*** 1.55 (1.05 to 0.90 (0.45 to 1.77) 1.32 (0.41 to 4.24) 1.33 (0.95 to 1.86) (12.4%) 2.30)* Birthweight <10th centile (<1.28 sd below mean) GROW <10th centile 5/225 (2.2%) 0.79 (0.13 to 4.60) 1.65 (0.79 to 2.24 (1.06 to 4.73)* 2.44 (0.40 to 1.27 (0.61 to 2.63) 3.45) 14.91) INTERGROWTH <10th 3/225 (1.3%) Insufficient events 1.83 (0.81 to 2.47 (1.08 to 5.65)* 4.11 (0.78 to 1.41 (0.63 to 3.18) centile 4.14) 21.63) WHO <10th centile 20/225 (8.9%) 1.21 (0.59 to 2.45) 1.90 (1.31 to 1.74 (1.02 to 2.98)* 3.66 (1.47 to 1.67 (1.24 to 2.77)*** 9.11)** 2.24)*** Birthweight <25th centile (< 0.675 sd below mean) GROW <25th centile 13/225 (5.8%) 0.59 (0.16 to 2.17) 1.05 (0.51 to 1.43 (0.69 to 2.95) 1.92 (0.50 to 7.43) 0.80 (0.40 to 1.61) 2.13) INTERGROWTH <25th 7/225 (3.1%) 0.56 (0.09 to 3.46) 1.58 (0.81 to 2.15 (1.09 to 4.23)* 3.66 (1.04 to 1.21 (0.63 to 2.33) centile 3.07) 12.88)* WHO <25th centile 37/225 1.22 (0.70 to 2.12) 1.83 (1.31 to 1.77 (1.13 to 2.76)* 2.96 (1.25 to 1.22 (0.70 to 2.12) (16.4%) 2.57)*** 7.02)* Discussion admission. In smaller infants, SGA defined according to This study demonstrates that GROW and INTE INTERGROWTH criteria showed slightly stronger asso- RGROWTH growth standards perform comparably in ciations with outcomes. However, using any thresholds type 1 diabetes pregnancies, giving similar median birth- studied, weight related measures alone were not strong weight centiles and comparable rates of LGA and SGA predictors of suboptimal perinatal outcomes (Table 3). neonates. Furthermore, both GROW and INTE WHO standards do not incorporate gestational age at RGROWTH defined LGA identified neonates at in- delivery, and thus fail to adequately describe size at birth creased risk of complications. LGA defined according to in preterm infants. For term infants, the WHO criteria GROW (>90th and/or > 97.7th centile) showed stronger gave a true birthweight centile, but for preterm infants, associations with preterm delivery, neonatal the WHO criteria gave a low centile, which reflected hypoglycaemia, hyperbilirubinaemia and NICU their prematurity, not their comparative size at birth. Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 6 of 10 Fig. 1 Rates (%) of caesarean delivery, preterm delivery, neonatal hypoglycaemia, hyperbilirubinaemia, respiratory distress, NICU admission and the composite neonatal outcome according to birth centile category based on GROW, INTERGROWTH and WHO standards. Numbers in each category are given at the bottom right of this figure This measure of prematurity means that the WHO cri- customised centiles believe that incorporating maternal teria were still able to predict outcomes, which demon- variables results in a more accurate representation of strates the importance of preterm delivery in relation to size at birth [9, 19]. Conversely, proponents of the INTE multiple neonatal complications. However, the inability RGROWTH-21st standards claim that variables such as to reliably attribute a birthweight centile is a substantial ethnicity make little difference to size at birth, in well- limitation in type 1 diabetes pregnancies, where rates of nourished populations with access to adequate antenatal preterm delivery are as high as 40% [5]. care [13]. A limitation of the CONCEPTT trial is that A central aspect to the controversy about GROW and while it international, 86% of women recruited were of INTERGROWTH centiles involves the perceived im- European / Mediterranean origin, which reduced the op- portance of maternal factors to the growth of the infant, portunity to look at growth standard performance in dif- and ethnicity in particular. Proponents of GROW ferent ethnicities. A further issue is that women who Table 3 Association of adverse neonatal outcomes with their positive and negative predictive values for the INTERGROWTH and WHO criteria. GROW: gestation related optimum weight; NICU: neonatal intensive care unit; WHO: world health organisation. Composite outcome: birth injury, neonatal hypoglycaemia, hyperbilirubinaemia, respiratory distress syndrome, or neonatal intensive care admission; sen: sensitivity; spe: specificity; ppv: positive predictive value; npv: negative predictive value Caesarean delivery Preterm delivery Neonatal NICU Admission Hyperbilirubinaemia Respiratory Composite hypoglycaemia Distress Outcome Meek et al. BMC Pregnancy and Childbirth sen spe ppv npv sen spe ppv npv sen spe ppv npv sen spe ppv npv sen spe ppv npv sen spe ppv npv sen spe ppv npv Birthweight > 84.1st centile (mean + 1 sd) GROW > 84.1st centile 71 49 75 43 66 36 40 62 75 39 29 82 72 39 41 71 74 39 32 80 79 36 10 95 73 42 53 63 INTERGROWTH > 84.1st centile 81 40 75 48 75 26 40 62 81 28 28 81 77 27 38 67 77 27 29 76 79 26 9 93 78 29 50 59 (2021) 21:96 WHO > 84.1st centile 43 77 80 38 22 54 24 52 44 66 30 78 36 63 37 63 31 61 23 70 32 63 7 91 36 64 48 52 Birthweight >90th centile (mean + 1.28 sd) GROW >90th centile 68 51 76 42 66 40 42 65 75 42 31 84 71 43 42 72 73 42 32 80 79 39 11 95 72 47 55 65 INTERGROWTH >90th centile 72 46 75 43 66 33 39 60 77 37 29 83 75 38 41 72 74 36 31 79 79 34 10 95 74 40 53 63 WHO >90th centile 36 84 84 37 17 62 22 53 39 73 33 78 30 70 37 63 26 69 24 71 21 69 6 91 31 71 49 53 Birthweight > 97.7th centile (mean + 2 sd) GROW > 97.7th centile 50 76 82 41 51 63 47 66 63 65 38 84 54 65 47 71 53 62 35 78 58 59 12 94 52 67 59 61 INTERGROWTH > 97.7th centile 48 76 82 40 43 60 41 62 61 66 38 83 51 65 46 69 45 61 30 74 53 60 11 93 50 67 58 59 WHO > 97.7th centile 14 90 75 32 9 85 29 59 25 92 50 78 18 91 54 65 11 87 25 72 16 88 11 92 16 91 61 54 GROW <10th centile 3 99 80 31 3 99 60 61 2 98 20 75 4 99 60 64 5 99 60 73 5 98 20 92 3 98 60 53 INTERGROWTH <10th centile 2 100 100 32 2 99 67 61 0 98 0 74 2 99 67 64 3 99 67 73 5 99 33 92 2 99 67 53 WHO <10th centile 10 94 80 32 18 97 80 64 11 92 30 75 16 95 65 66 15 93 45 74 26 93 25 93 14 96 75 55 GROW <25th centile 7 97 85 32 8 96 54 61 4 93 15 74 6 94 38 63 8 95 38 73 11 95 15 92 5 93 38 52 INTERGROWTH <25th centile 3 97 71 31 4 98 57 61 2 96 14 74 5 98 57 64 6 98 57 73 11 98 29 92 4 97 57 53 WHO <25th centile 18 87 76 32 31 93 76 68 19 85 30 76 27 89 59 68 26 87 43 76 37 85 19 94 24 91 70 57 Page 7 of 10 Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 8 of 10 choose to participate in studies are often affluent, well- Despite maternal diabetes being a risk factor of peri- nourished and educated, and may not represent mothers natal morbidity, there has been relatively little assess- with different socioeconomic circumstances. ment of different growth standards in this population. A major focus for growth standards has been on the Kase and colleagues reported that customised centiles identification of infants who are SGA with a view to re- identified more infants as SGA/LGA compared to popu- ducing stillbirth rates [9]. Although we have identified lation centiles in diabetes pregnancies [25]. Narchi and that infants <25th centile displayed a trend to be at high- Skinner had similar findings but concluded there was no est risk of multiple complications, very few infants fell evidence of a difference in mortality or morbidity be- into this category which made detailed assessment of tween the infants identified by customised vs population SGA outcomes challenging. Although SGA is uncom- growth standards [26]. The current study adds to the lit- mon in type 1 diabetes pregnancy, it is plausible that in- erature by highlighting differences between the common fants born < 10th centile do not represent all those with growth standards in a generalisable population of preg- growth restriction. nant women with type 1 diabetes. We had no women In this study, standard-deviation-based criteria for the with optimal glucose control (defined as HbA1c< 48 diagnosis of LGA have been assessed. Although a birth mmol/mol) in early pregnancy as these women were ex- weight z score > 1 is considered consistent with LGA, cluded from the CONCEPTT trial. However, population this definition is different to standard centile-based defi- based studies confirm that only 15% of women with nitions (>90th centile) [20]. Different approaches to the T1D achieve a first trimester HbA1c level < 6.5% (48 LGA diagnosis contribute to difficulty in comparing mmol/mol [5]) and thus the data presented here is rep- populations internationally [21–23]. resentative for the vast majority of women. Future stud- A fundamental aim of antenatal care in T1D pregnan- ies should evaluate growth standards and definitions of cies involves careful control of maternal glycaemia to LGA and SGA in larger cohorts of women with diabetes, normalise fetal growth. This study shows that growth – including among women with target HbA1c levels. Bet- related measures alone are not strong predictors of sub- ter understanding of the causes and early identification optimal perinatal outcomes. For example, a birthweight of growth restriction in diabetes pregnancy should be a >90th centile on GROW and INTERGROWTH criteria research priority. could identify neonatal hypoglycaemia, NICU admission and respiratory distress with 71–77% sensitivity, but the Conclusions specificity for these outcomes was around 32–42%. WHO growth standards do not incorporate gestational These data are consistent with other work which high- age at birth and therefore are unsuitable for use in type lights the challenges of accurate prediction of neonatal 1 diabetes pregnancy, where preterm delivery is com- outcomes in T1D pregnancy. For example, Yamamoto monplace. However, GROW and INTERGROWTH and colleagues demonstrated that LGA was the only sig- standards are both suitable. Infants born > 97.7th centile nificant predictor for neonatal hypoglycaemia on ad- had the highest risks of suboptimal outcomes. LGA de- justed logistic regression analysis (odds ratio 2.51, 95% fined by GROW and SGA defined by INTERGROWTH CI 1.10–5.70) [24]. However, 36% of infants with neo- showed strongest associations with neonatal outcomes. natal hypoglycaemia were appropriate for gestational Abbreviations age, resulting in similar levels of sensitivity and specifi- LGA: Large for gestational age; SGA: Small for gestational age; city seen in the current report. CONCEPTT: Continuous Glucose Monitoring in Type 1 Diabetes Pregnancy As only a small proportion of the general maternity Trial; NICU: Neonatal intensive care unit; T1D: Type 1 diabetes; CGM: Continuous glucose monitoring; GROW: Gestation Related Optimum population has type 1 diabetes, CONCEPTT represents Weight; WHO: World Health Organisation; SD: Standard deviation; BMI: Body one of the larger randomised trials with detailed data on mass index perinatal outcomes, making it useful to assess fetal Acknowledgements growth, and infant birthweight outcomes. Customised The authors would like to thank all the women with type 1 diabetes who (GROW) centiles were reported for CONCEPTT, but participated in the CONCEPTT trial. We also acknowledge the invaluable the effect of the intervention was also seen using INTE support from the 31 clinical care teams and the CONCEPTT Steering Committee: Denice S Feig, Helen R Murphy, Elisabeth Asztalos, Jon F R RGROWTH standards. Although accelerated fetal Barrett, Rosa Corcoy, Alberto de Leiva, Lois E Donovan, J Moshe Hod, Lois growth is common in T1D pregnancies, the rates of Jovanovic, Erin Keely, Craig Kollman, Ruth McManus, Kellie E Murphy, Katrina LGA in the CONCEPTT infants were higher than ex- Ruedy and George Tomlinson. pected (66% in CONCEPTT compared to ~ 50% in a UK Authors’ contributions population using similar methodology [5]). The reasons CLM designed the study, analysed and interpreted the data, wrote and for this are unclear, particularly as the CONCEPTT revised the manuscript. HRM identified the study question, contributed to data analysis, reviewed and revised the manuscript and contributed to the population had better glycaemic control compared to discussion. DSF, EL and RC contributed to study design, data collection, data the UK clinical population [5]. collation and reviewed and revised the final manuscript. EA and LCK Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 9 of 10 contributed to the discussion and reviewed the final manuscript. All authors Foundation Trust, Cambridge, UK. 3Servei d’Endocrinologia i Nutrició, Hospital gave approval of the final version of the manuscript prior to publication. de la Santa Creu i Sant Pau, Barcelona, Spain. 4Institut de Recerca, Hospital de la Santa Creu i Sant Pau, CIBER-BBN, Barcelona, Spain. 5Department de Funding Medicina, Universitat Autònoma de Barcelona, Barcelona, Spain. 6Sunnybrook The trial is funded by Juvenile Diabetes Research Foundation (JDRF) grants Health Sciences Centre, University of Toronto, Toronto, Canada. 7Servei de #17–2011-533, and grants under the JDRF Canadian Clinical Trial Network, a Pediatria, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. 8Mount Sinai public-private partnership including JDRF and FedDev Ontario and sup- Hospital, Department of Medicine, University of Toronto, Toronto, Canada. 9 ported by JDRF #80–2010-585. The funders provided solely the finance re- Lunenfeld-Tanenbaum Research Institute, Toronto, Canada. 10Norwich quired for this study and had no role in the trial design, data collection, data Medical School, University of East Anglia, Norwich, UK. 11Department of analysis or data interpretation. Medtronic supplied the CGM sensors and Women and Children’s Health, King’s College London, London, UK. CGM systems at reduced cost. They also had no role in trial design, collec- tion, handling, analysis or interpretation of data. Received: 13 May 2020 Accepted: 12 January 2021 HRM conducts independent research supported by the National Institute for Health Research (Career Development Fellowship, CDF-2013-06-035), and is supported by Tommy’s charity. CLM is supported by the Diabetes UK Harry Keen Intermediate Clinical Fellowship (DUK-HKF 17/0005712) and the EFSD- References Novo Nordisk Foundation Future Leader’s Award (NNF19SA058974). 1. Weaver LT. In the balance: weighing babies and the birth of the infant welfare clinic. Bull Hist Med. 2010;84(1):30–57 [published Online First: 2010/ Availability of data and materials 07/17]. The datasets used and/or analysed during the current study are available 2. Clarke JD. Observations on some causes of excess mortality of males above from the corresponding author on reasonable request. that of females. Philos Trans R Soc Lond. 1786;76:349–64. 3. Ray JG, Park AL, Fell DB. Mortality in Infants Affected by Preterm Birth and Ethics approval and consent to participate Severe Small-for-Gestational Age Birth Weight. Pediatrics. 2017;140(6). The CONCEPTT study was approved by the Research Ethics Committee East https://doi.org/10.1542/peds.2017-1881 [published Online First: 2017/11/10]. of England (Essex) under reference 12/EE/0310 in 2012. All participants 4. Doty MS, Chen HY, Sibai BM, et al. Maternal and Neonatal Morbidity provided written informed consent. Further ethical approval was not Associated With Early Term Delivery of Large-for-Gestational-Age But required for this analysis. Nonmacrosomic Neonates. Obstet Gynecol. 2019;133(6):1160–6. https://doi. org/10.1097/aog.0000000000003285 [published Online First: 2019/05/29]. Consent for publication 5. Murphy HR, Bell R, Cartwright C, et al. Improved pregnancy outcomes in Not applicable. women with type 1 and type 2 diabetes but substantial clinic-to-clinic variations: a prospective nationwide study. Diabetologia. 2017;60(9):1668–77. Competing interests https://doi.org/10.1007/s00125-017-4314-3 [published Online First: 2017/06/ Denice Feig has received honoraria for speaking engagements from 10]. Medtronic and has been on an. 6. Feig DS, Donovan LE, Corcoy R, et al. Continuous glucose monitoring in Helen Murphy has received honoraria for speaking engagements from pregnant women with type 1 diabetes (CONCEPTT): a multicentre Medtronic, Roche, Novo Nordisk, Eli-Lilly and is a member of the Medtronic international randomised controlled trial. Lancet. 2017;390(10110):2347–59. European Advisory Board. https://doi.org/10.1016/S0140-6736(17)32400-5 [published Online First: CLM is the guarantor of this work and, as such, has had full access to all the 2017/09/15]. data in the study and takes responsibility for the integrity of the data and 7. Murphy HR. Continuous glucose monitoring targets in type 1 diabetes the accuracy of the data analysis. pregnancy: every 5% time in range matters. Diabetologia. 2019;62(7):1123–8. https://doi.org/10.1007/s00125-019-4904-3 [published Online First: 2019/06/ Author details 05]. 1 Institute of Metabolic Science, University of Cambridge, Addenbrooke’s 8. Gardosi JFA. Gestation network. Customised weight centile calculator. Hospital, Box 289, Cambridge CB2 0QQ, UK. 2Cambridge Universities NHS GROW v8.1 UK. http://www.gestation.net/birthweight_centiles/centile_ Foundation Trust, Cambridge, UK. 3Servei d’Endocrinologia i Nutrició, Hospital object.htm. de la Santa Creu i Sant Pau, Barcelona, Spain. 4Institut de Recerca, Hospital 9. Francis A, Hugh O, Gardosi J. Customized vs INTERGROWTH-21(st) standards de la Santa Creu i Sant Pau, CIBER-BBN, Barcelona, Spain. 5Department de for the assessment of birthweight and stillbirth risk at term. Am J Obstet Medicina, Universitat Autònoma de Barcelona, Barcelona, Spain. 6Sunnybrook Gynecol. 2018;218(2s):S692–s99. https://doi.org/10.1016/j.ajog.2017.12.013 Health Sciences Centre, University of Toronto, Toronto, Canada. 7Servei de [published Online First: 2018/02/10]. Pediatria, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. 8Mount Sinai 10. Gardosi J, Clausson B, Francis A. The value of customised centiles in Hospital, Department of Medicine, University of Toronto, Toronto, Canada. assessing perinatal mortality risk associated with parity and maternal size. 9 Lunenfeld-Tanenbaum Research Institute, Toronto, Canada. 10Norwich Bjog. 2009;116(10):1356–63. https://doi.org/10.1111/j.1471-0528.2009.02245.x Medical School, University of East Anglia, Norwich, UK. 11Department of [published Online First: 2009/06/23]. Women and Children’s Health, King’s College London, London, UK. 11. Villar J, Cheikh Ismail L, Victora CG, et al. International standards for newborn weight, length, and head circumference by gestational age and Received: 13 May 2020 Accepted: 12 January 2021 sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project. Lancet. 2014;384(9946):857–68. https://doi.org/10.1016/s0140-6736(14)60932- 6 [published Online First: 2014/09/12]. 12. WHO Multicentre Growth Reference Study Group. WHO Child Growth Supplementary Information Standards based on length/height, weight and age. Acta Paediatr Suppl. The online version contains supplementary material available at https://doi. 2006;450:76–85 [published Online First: 2006/07/05]. https://www.who.int/ org/10.1186/s12884-021-03554-6. childgrowth/standards/Acta_95_S450.pdf?ua=1. 13. Papageorghiou AT, Kennedy SH, Salomon LJ, et al. The INTERGROWTH-21(st) fetal growth standards: toward the global integration of pregnancy and Additional file 1. pediatric care. Am J Obstet Gynecol. 2018;218(2s):S630–s40. https://doi.org/ 10.1016/j.ajog.2018.01.011 [published Online First: 2018/02/10]. Author details 14. Anderson NH, Sadler LC, Stewart AW, et al. Maternal and pathological The online version contains supplementary material available at https://doi. pregnancy characteristics in customised birthweight centiles and org/10.1186/s12884-021-03554-6. identification of at-risk small-for-gestational-age infants: a retrospective 1 Institute of Metabolic Science, University of Cambridge, Addenbrooke’s cohort study. BJOG. 2012;119(7):848–56. https://doi.org/10.1111/j.1471-0528. Hospital, Box 289, Cambridge CB2 0QQ, UK. 2Cambridge Universities NHS 2012.03313.x [published Online First: 2012/04/04]. Meek et al. BMC Pregnancy and Childbirth (2021) 21:96 Page 10 of 10 15. Mackin ST, Nelson SM, Kerssens JJ, et al. Diabetes and pregnancy: national trends over a 15 year period. Diabetologia. 2018;61(5):1081–8. https://doi. org/10.1007/s00125-017-4529-3. 16. Hadlock FP, Harrist RB, Martinez-Poyer J. In utero analysis of fetal growth: a sonographic weight standard. Radiology. 1991;181(1):129–33. https://doi. org/10.1148/radiology.181.1.1887021. 17. Villar J, Giuliani F, Fenton TR, Ohuma EO, Ismail LC, Kennedy SH; INTERGROWTH-21st Consortium. INTERGROWTH-21st very preterm size at birth reference charts. Lancet. 2016;387(10021):844–5. 18. Stefanelli S, Groom KM. The accuracy of ultrasound-estimated fetal weight in extremely preterm infants: a comparison of small for gestational age and appropriate for gestational age. Aust N Z J Obstet Gynaecol. 2014;54(2):126– 31. https://doi.org/10.1111/ajo.12198 [published Online First: 2014/02/28]. 19. Anderson NH, Sadler LC, McKinlay CJD, et al. INTERGROWTH-21st vs customized birthweight standards for identification of perinatal mortality and morbidity. Am J Obstet Gynecol. 2016;214(4):509.e1–7. https://doi.org/ 10.1016/j.ajog.2015.10.931 [published Online First: 2015/11/08]. 20. Blickstein I, Salihu HM, Keith LG, et al. The association between small-for- gestational age triplet pregnancies and neonatal mortality: a novel approach to growth assessment in multiple gestations. Pediatr Res. 2006; 59(4 Pt 1):565–9. https://doi.org/10.1203/01.pdr.0000202757.47761.76 [published Online First: 2006/03/22]. 21. Persson M, Norman M, Hanson U. Obstetric and perinatal outcomes in type 1 diabetic pregnancies: A large, population-based study. Diabetes Care. 2009;32(11):2005–9. https://doi.org/10.2337/dc09-0656 [published Online First: 2009/08/14]. 22. Persson M, Pasupathy D, Hanson U, et al. Birth size distribution in 3,705 infants born to mothers with type 1 diabetes: a population-based study. Diabetes Care. 2011;34(5):1145–9. https://doi.org/10.2337/dc10-2406 [published Online First: 2011/03/25]. 23. Murphy HR, Rayman G, Lewis K, et al. Effectiveness of continuous glucose monitoring in pregnant women with diabetes: randomised clinical trial. BMJ. 2008;337:a1680 [published Online First: 2008/09/25]. 24. Yamamoto JM, Kallas-Koeman MM, Butalia S, et al. Large-for-gestational-age (LGA) neonate predicts a 2.5-fold increased odds of neonatal hypoglycaemia in women with type 1 diabetes. Diabetes Metab Res Rev. 2017;33(1). https:// doi.org/10.1002/dmrr.2824 [published Online First: 2016/05/18]. 25. Kase BA, Cormier CM, Costantine MM, et al. Population standards of birth weight underestimate fetal growth abnormalities in diabetic pregnancies. Am J Perinatol. 2012;29(2):147–52. https://doi.org/10.1055/s-0031-1295656 [published Online First: 2011/11/23]. 26. Narchi H, Skinner A. Infants of diabetic mothers with abnormal fetal growth missed by standard growth charts. J Obstet Gynaecol. 2009;29(7):609–13. https:// doi.org/10.1080/01443610903100625 [published Online First: 2009/09/17]. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.