Obesity class I and II and IOM 2009 gestational weight gain recommendations 5–9 kg. An audit on 10,000 term singleton deliveries

Abstract Objectives To compare several maternal-fetal morbidities comparing the Institute of Medicine IOM 2009 recommendations (IOMR: 5–9 kg in all obese women) between women with adequate gestational weight gain (GWG) and Inadequate (less than 5 kg), and excessive those gaining more than 9 kg among obese women class I (30–34.9 kg/m2) and class II (35–39.9 kg/m2). Study design South-Reunion University’s maternity (Reunion Island, Indian Ocean). 21-Year-observational cohort study (2001–2021). Epidemiological perinatal database with information on obstetrical and neonatal risk factors. Main outcome measures Cesarean sections, preeclampsia, means birthweight, rate of small (SGA) or large (LGA) for gestational age newborns and macrosomic babies (≥4 kg). Results Among the singleton term live births (37 weeks onward) we could define the pre-pregnancy body mass index and GWG in 85.9% of cases. The final study population focused on 10,296 obese women (7138 obesity class I − 30–34.9 kg/m2, 3158 obesity class II – 35–39.9 kg/m2). Concerning inadequate GWG (less than 5 kg), respectively for obese I and II, IOMR babies were heavier (plus 90 and 104 g, p < .001), were more prone to be LGA OR 1.61 and 1.69, p < .001, macrosomic OR 1.49 and 2.21, p < .0001, IOMR women had more cesarean sections OR 1.33, OR 1.45, p = .001, and for obese II a tendency for more term preeclampsia OR 1.83, p = .06. Conclusion This study demonstrates that for obese women these IOMR (5–9 kg) are mildly but significantly too high if we consider obesity class I and obviously too high for obesity class II (35–39.9 kg/m2).


Introduction
This study is complementary to a recent one where we analyzed gestational weight gain, gestational weight gain (GWG), in severe obese class III (! 40 kg/m 2 ) [1]. Obesity worldwide has become a major public health challenge in this twenty-first century with an apparent irresistible rise of this epidemic since the 1970s [2,3]. Nowadays, it is evaluated that obese people (!30 kg/m 2 ) represent some one billion inhabitants (out of seven) on this planet [2,3], and that more than 2.1 billion (30% of the global population) are either overweight or obese [4], and some projections evaluate that in 2030 almost half of the world population will be either overweight or obese [5]. Women (particularly comprising more than 50% cases of total obesity cases among the human population) are particularly affected and this problem has become a major challenge for obstetric care [6]. Knowing the optimal GWG (from conception to birth) among the annual 135 million human pregnancies may be considered to be one of the "Holy Grail" to achieve for maternity health care providers and for women themselves. Extensive literature exists on the subject with, in background, the current international cornerstone which are the 2009-IOM recommendations [7] based on the WHO-body mass index (BMI) classification standardized in 2000 [8]. Our purpose is to focus only on obese people class I and II (30-34.9 kg/m 2 and 35-39.9 kg/m 2 ) where the current recommendations are extremely simple: the recommended GWG for all obese women is 5-9 kg during pregnancy. However, a lot of controversies aroused on these "obese recommendations." Obese class I (30-34.9 kg/m 2 ) have different (and lower) morbidities than obesity class II (35-39.9 kg/m 2 ), not to speak of obesity class III (!40 kg/m 2 ). Second, for obese women, the major controversies concern the debate if, beginning at obesity class II, women should lose weight during their pregnancy [1,[9][10][11][12][13][14]. We already extensively discussed these controversies in other papers [1,15]. Additional to pre-pregnancy obesity, gestational weight gain during pregnancy is another rising attention for its strong influence on pregnancy outcomes. Concerning large for gestational age (LGA) newborns and particularly the macrosomic ones (>4 kg), these newborns are well-known to present a 10-fold higher risk of fearsome shoulder dystocia [16]. The limit of 4 kg is considered to be the point where significant morbidities at delivery may occur [17], moreover, in the case of associated GDM [18]. Besides some immediate morbidities, there are long-term effect on the future life of the individual [19][20][21].
In Reunion Island (a French overseas department in the Indian ocean) we have been facing an apparently ineluctable rise of obesity in our population for 21 years: our obese pregnant women (pre-pregnancy BMI) were 12% in 2004 and were 25% in 2021 (with a parallel rise of gestational diabetes, respectively 8% in 2004 to 20% in 2021). Obesity and pregnancy are therefore now becoming an important issue for us [22]. The aim of this audit is to compare the IOM 2009 recommendations (5-9 kg) for GWG in obese women with inadequate (<5 kg) and excessive (>9 kg) in our clinical/epidemiological experience.

Material and methods
From 1st January 2001 to 31st December 2021, the hospital records of all women delivered at the maternity of the University South Reunion Island (ap. 4,000 births per year now, 4300 the preceding decade) were abstracted in standardized fashion. All data were entered into an epidemiological perinatal database which contained information on obstetrical risk factors, description of deliveries and neonatal outcomes. As participants in the French national health care system, all pregnant women in Reunion Island have their prenatal visits, biological and ultrasound examinations, and anthropological characteristics recorded in their maternity booklet. In our term pregnancies, the average of prenatal visits was 9.2 ± 2, and 4.2 ± 1.7 ultrasounds per pregnancy. Weight gain was defined by the difference between the weight measured at the delivery room minus the weight written down in the maternity booklet by the health worker (midwife, physician) at the first prenatal visit during the first trimester. In a few cases (less than 10%), the prepregnancy weight was defined by the personal recollection of the woman.
In the general analysis, there were three criteria of exclusion: Preterm births (<37 weeks), multiple births and fetal deaths (in utero fatal deaths, stillbirths and medical termination of pregnancy).
Small for gestational age (SGA, 10th percentile) and Large for gestational age babies (LGA, 90th percentile) have been defined by a local curve for male and female infants established from our perinatal data base [22].
Preeclampsia was defined according to the World Health Organization recommendations [23] and the International Society for the study of Hypertension in Pregnancy as the new onset of hypertension (BP !140 mmHg systolic or !90 mm Hg diastolic) at or after 20 weeks' gestation and substantial proteinuria (>0.3 g/24 h).
Reunion Island is a French department in the Southern Indian Ocean. The peculiarity of this tropical region lays in the multiethnic origin of inhabitants [Africa and intermixed population (50%), Europe (27%), India (20%) and China (3%)]. Compared to Europe and mainland France, there is a younger reproductive population (the mean age at primiparity is of 25 years). Finally, accessibility to maternity services is easy, and high-quality care is provided free of charge by the French healthcare system.

Statistical analysis
Data are presented as numbers and proportions (%) for categorical variables and as mean and standard deviation (SD) for continuous ones. Comparisons between groups were performed by using v 2 -test; odds ratio (OR) with 95% confidence interval (CI) was also calculated. Paired t-test was used for parametric and the Mann-Whitney U test for non-parametric continuous variables. p-Values <.05 were considered statistically significant. Epidemiological data have been recorded and analyzed with the software EPI-INFO 7.1.5 (2008, CDC Atlanta, OMS), EPIDATA 3.0 and EPIDATA Analysis V2.2.2.183. Denmark.

Ethical approval
This study analyzing data anonymously was exempt from approval of the institutional review board (Comit e de Protection des Personnes Sud-Ouest et Outre Mer III) and according to French legislation written consent.

Results
There were 76,189 term births (37 weeks onward) at the South Reunion maternity during the 21-year period (1 January 2001-31 December 2021). Multiple pregnancies (N ¼ 653 pregnancies and 1305 newborns), fetal deaths (antepartum term fetal deaths N ¼ 140, five per-partum still births at term and four medical termination of pregnancy) were excluded from further analysis. The live-birth term population consisted of 74,704 singleton pregnancies. Term pregnancies (37-42 weeks gestation) where we could define the maternal pre-pregnancy body mass index and the Gestational Weight Gain comprised 65,446 patients (85.9% of the total). The final study population focused on 10,296 obese women (7138 obesity class I À 30-34.9 kg/m 2 , 3158 obesity class II -35-39.9 kg/m 2 ) Comparisons excessive > 9 kg vs adequate "normals" (5-9 kg) (Tables 1 and 2) In excessive GWG mean birthweight were significantly heavier than "normals" (adequate): þ79 g obese I, þ115 g obese II, p < .001. Respectively for obese I and II there were more LGA OR 1.39 and 1.49, p < .001; more macrosomic newborns (!4 kg) OR 1.52 l p ¼ .0005; more cesarean sections (c-sections) OR 1.24, 1.20 (p values .002 and .07). While the incidence of SGA was between 7.5 to 8% in adequate GWG, it was of 7.7% amongst obese I (NS) and 4.6% amongst obese II, OR 0.58, p ¼ .003. Incidence of preeclampsia was higher in excessive vs normals (class I: 2.4% vs 1.1%, OR 2.23, p ¼ .0005, 3.8% vs 2.7%, OR 1.46, p ¼ .13 in obese class II).
Comparisons inadequate < 5 kg vs adequate "normals" (5-9 kg) (Tables 1 and 2) Conversely mean birthweight were significantly heavier in obese normals (adequate 5-9 kg) vs inadequate (<5 kg): þ90 g obese I, þ104 g obese II, p < .0001. Comparing inadequate vs adequate GWGs, there were less LGA, OR 0.62 p ¼ .001, less macrosomic newborns (!4 kg), OR 0.67 p ¼ .02, and also less c-sections, 0.75 p ¼ .001. Incidence of maternal preeclampsia was similar (1.0 and 1.1%) in the inadequate as compared to normals for the obese class I women (Table 1), but there was a tendency to have a lower rate of preeclampsia in the inadequate group among obese II: 1.5% vs 2.7% normals (OR 0.55, p ¼ .06, Table 2). However, the incidence of SGA was higher in inadequate GWG (<5 kg) than in normals 10% vs 7.5 to 8% (obese I and II respectively OR 1.35, p ¼ .009 and OR 1.32, p ¼ .08). Figures 1 and 2 visualize incidences of different maternal fetal risk factors (cesarean sections, term preeclampsia, SGA and LGA newborns, macrosomic babies !4 kg), for obese class I, 30-34.9 kg/m 2 (Figure 1), class II, 35-39.9 kg/m 2 (Figure 2), and obese class III (over 40 kg/m 2 ). In blue are the scales of IOM 2009 recommendations (5-9 kg) (IOMR). It is of note that in the two figures, the crossing points for the incidence 10% of SGA and LGA babies in the function of maternal GWG are outside of the IOM 2009 recommendations on the left, corresponding also with less incidence of maternal preeclampsia, rate of LGA and macrosomic newborns and rate of cesarean sections.

Principal findings
First of all, it is epidemiologically interesting in this cohort to notice that only 29% of our class I and II have fitted with the IOM2009 recommendations, IOMR (5-9 kg. This audit validates the IOM2009 recommendations: gaining more than 9 kg in obese women ("excessive" weight gain") is harmful for the maternalfetal studied morbidities (birthweights, LGA, macrosomic newborns, c-section rates and preeclampsia), see Tables 1 and 2. In our cohort 48% of class I and 39% class II were in this excessive group.
The aim of this study was however to test also the recommendations 5-9 kg. Looking at details in our real experience, Table 1 and 2, we found that inadequate GWG (<5 kg) may have important advantages as compared with the IOMR for both obese class I and II: Table 1, obese class I, lighter newborns (minus 92 g), less LGA, OR 0.62 p ¼ .001, less macrosomic (!4 kg), OR 0.67 p ¼ .02, less c-sections, OR 0.75 p ¼ .001. Table 2, obese class II less LGA babies, OR 0.59 p ¼ .0002, less macrosomic (!4 kg), OR 0.45 p ¼ .0008, less c-sections, OR 0.69 p ¼ .001, but also a tendency to have less preeclampsia, OR 0.55 p ¼ .06. All these positive results were at the expense of a higher rate of SGA babies 10% in inadequate GWG vs 7.5% in normals p ¼ .009 in obesity class II. However, it is of note that the 10% rate of SGA in inadequate GWG is the physiological by-definition rate in a neonatal population.

Clinical implications
The interesting lesson of this study is the two figures analyzing GWG by an increment of 5 kg/m 2 with different maternal-fetal risk factors: we notice immediately that the crossing points where we have 10% of LGA and 10% SGA newborns (the very definition of a normal Gaussian neonatal population, what we already proposed to coin "maternal foetal corpulence symbiosis, MFCS" [15]) between the 2 classes of obesities are always outside the IOMR scale on the left. For obesity class I, the MFCS point is rather between a GWG of 0 to 6 kg, for obesity class II rather losing weight during pregnancy À1 to À5 kg. Considering the MFCS point as a goal to achieve to have an optimal GWG during a pregnancy going to term may be of interest. We have recently described that late onset preeclampsia (34 weeks onward) was linearly associated with rising BMI among overweight/obese women [24]. Focusing then on a Gaussian neonatal population ((physiological, 10% SGA and 10% LGA) has probably also some influence on other maternal-neonatal morbidities [25,26]. This unexpected phenomenon surprised us in 2018 [15]: each category of maternal pre-pregnancy BMI, prepregnancy BMI (ppBMI), have a specific SGA-LGA crossing point in function of their GWG. Looking at the first study in 2018, made on a total population, and not only obese women, this drift continues for overweight (25-29.9 kg/m 2 ), normal (20-24.9 kg/m 2 ) and lean women (<20 kg/m 2 ), on the right this time. This global and constant drift is so regular that, indeed, it is, after verification, a linear curve [15]). This linearity has absolutely to be verified elsewhere, especially among team facing a high population of severe obese women !40 kg/m 2 . This linearity has just been confirmed for the optimal gestational weight gain in twin pregnancies among a cohort of 200,000 US pregnancies [27,28] For possible consequences of a new paradigm (achieving to have normal Gaussian curve in neonatal birthweights), we have also tested our population for term preeclampsia [25] and other morbidities (c-sections, macrosomes, LGA, etc … ) [26] showing that for obesity class II losing weight during pregnancy is advantageous for everybody (mother, fetus and social health care expenses) [1].
We did not analyze gestational diabetes mellitus as in a recent publication we found in this same population an inverse relationship between optimal GWG and the risk of gestational diabetes: Lowering the GWG seems to heighten the risk of GDM, and excessive weight gain to lower it [26]. This phenomenon has also been described by preceding authors [28][29][30][31], and in fact, it might be an "optical" or a "perspective" bias due to our retrospective data. Li et al. [29] proposed an explanation that may be the good one: because the diagnosis of GDM occurs primarily at 26-28 weeks of gestation, treatment with diet and/or insulin plus increased physical activity may affect subsequent weight gain resulting in decreased weight gain in late pregnancy. This is emphasized in a recent paper in the United Kingdom (UPBEAT study): obese women with a positive OGTT at 27 weeks, and afterward a strong follow-up until delivery present lower weight gain than obesity considered as non-diabetic [31,32]. There is a need for future prospective studies where the optimal GWG being defined since the beginning of pregnancy, those women following the new recommendations (physiological neonatal population, linear association with pre-pregnancy BMI) would probably reach the 26-28th week period with different OTT's results.
Fetal sex is recognized to impact the intrauterine response to an adverse environment [33][34][35], such as inadequate or excessive gestational weight gain, one of a major outcome in our mind in this study was SGA (small for gestational age) scrupulously defined per week gestation (37-42 weeks in this study) and importantly by the sex of the infant.
The strength of our study is the capturing of several perinatal outcomes in a population of the area (ap. 360,000 inhabitants, and 5000 births per year. With 4,300 births per year, the university maternity represents 85% of all births in the south of the island, all receiving level 3, European standard of care. The data in this large cohort are homogeneous as they were collected in a single center (no intercenter variability) and not based on national birth registers but directly from medical records (avoiding inadequate codes).
The weakness is the retrospective nature of this study, in a relatively small population but we sincerely hope that our observations will trigger proper prospective trials and induce clinicians as a systematic verification in their own population if there is also a linear law concerning the association between 10% of SGA/LGA in term newborns and maternal GWG (whatever the initial pre-pregnancy BMI). this study only foucse on some of maternal-fetal outcomes which might not estimate the overall health risk brought by IOM recommendation.

Conclusions
This study demonstrates that for obesity Class I and II IOM 2009 recommendations of a GWG of 5-9 kg are too high and must be urgently revised. Since 2018 [15], some 80 million obese women have delivered on this planet without any change in the international recommendations, while we have certainly an achievable pathway to actively counterbalance the morbid effects of high BMIs i.e. term preeclampsia [25], LGA babies, macrosomic newborns, cesarean sections, etc. [26]. At least, it should exist a strong discussion (and/or refutation), and first of all a simple verification: Yes or No is there in other countries a constant and linear drift concerning the 10th crossing points of neonatal SGA/LGA ("Maternal Foetal Corpulence Symbiosis," MFCS) in different maternal categories of pre-pregnancy BMI associated with GWG? A confirmation would allow to recommend gestational weight gains with a prerequisite rationale as it has been shown recently for twin pregnancies in the USA [27,28]. Achieving to have in obese women a Gaussian normal neonatal population (10% of SGA and LGA). Similar research in larger populations is strongly recommended as the stake is immense.

Disclosure statement
The author attests that no conflict of interest exists regarding this work.

Funding
No specific funding besides the French Government funding for the South-Reunion perinatal database.

Condensation
International recommendations (IOM 2009): 5-9 kg of gestational weight gain in obese women class I and II are too high.