Abstract

Objective. This systematic review and meta-analysis evaluates the relationship between gestational weight gain and the risk of small for gestational age in obese pregnant women. Methods. Studies were identified by searching the Web of Science, Embase, and PubMed databases up to June 30th, 2022. The meta-analysis was carried out to determine the risk of small for gestational age with gestational weight gain (GWG) below the 2009 Institute of Medicine (IOM) guidelines compared with within the guidelines in obese women. The Newcastle–Ottawa Scale was used to assess the methodological quality. The chi-squared test, Q test, and I2 test were used to evaluate statistical heterogeneity. Subgroup analyses were conducted, and publication bias was assessed by funnel plots and Egger’s test. Sensitivity analyses were performed for three groups of obese people (I: BMI 30–34.9 kg/m2, II: BMI 35–39.9 kg/m2, and III: BMI ≥ 40 kg/m2) to examine the association of obesity and SGA. Results. A total of 788 references were screened, and 29 studies (n = 1242420 obese women) were included in the systematic review. Obese women who gained weight below the IOM guideline had a higher risk of SGA than those who gained weight within the guideline (OR = 1.27, 95% CI = 1.16–1.38, Z = 5.36). Both weight loss (<0 kg) and inadequate weight (0–4.9 kg) during pregnancy in obese women are associated with an increased risk of SGA (OR = 1.50, 95% CI = 1.37–1.64, Z = 8.82) (OR = 1.18, 95% CI = 1.14–1.23, Z = 8.06). The same conclusions were also confirmed for the three obesity classes (I: OR = 1.38, 95% CI = 1.29–1.47; II: OR = 1.39, 95% CI = 1.30–1.49; and III: OR = 1.26, 95% CI = 1.16–1.37). Subgroup analysis by country showed that GWG below guidelines in obese women of the USA and Europe was associated with risk for SGA (USA (OR = 1.30, 95% CI = 1.15–1.46), Europe (OR = 1.24, 95% CI = 1.11–1.40)) and not in Asia (OR = 1.17, 95% CI = 0.91–1.50). Conclusion. Our findings indicated that obese pregnant women who had weight loss or inadequate weight (0–4.9 kg) according to the IOM guideline had increased risks for SGA. Moreover, we also evaluated that gestational weight loss (<0 kg) in these pregnancies was associated with an increased risk for SGA compared with inadequate weight (0–4.9 kg) in these pregnancies. Therefore, the clinical focus should assist obese women to achieve GWG within the IOM guidelines to decrease the risk for SGA.

1. Introduction

Obesity has increased dramatically around the world in these past decades, and it is a public health problem. Obesity in pregnancy is often associated with adverse outcomes such as pregnancy-induced hypertension, preeclampsia, gestational diabetes mellitus (GDM), cesarean section, macrosomia, and neonatal asphyxia [13]. The Institute of Medicine (IOM) guideline revised the gestational weight gain in 2009 [4] and recommended that obese women should gain between 5 and 9 kg to obtain the best maternal and perinatal outcomes. However, the revision of the 2009 guideline did not provide recommended GWG for different classes of obesity.

Given the known relationship between gestational weight gain (GWG) above recommended and adverse perinatal outcomes, along with the long-term maternal health effects of obesity, physicians and women alike are exploring the possible benefits of weight loss during pregnancy-about 8% of all pregnant women reported attempting to lose weight, with the highest prevalence (13%) reported in obese women [5]. Moreover, the prevalence of actual weight loss increases with increasing obesity class, reaching as high as 15% in obesity class III [6, 7]. In this context, both prepregnancy BMI and GWG have been associated with maternal and fetal complications. However, there was no agreement on whether inadequate weight (0–4.9 kg) or weight loss (<0 kg) in obese women can contribute to improving neonatal outcomes or on the correct GWG to be reached to reduce these complications.

Some groups and meta-analyses suggested inadequate weight (0–4.9 kg) or weight loss (<0 kg) in obese women was associated with increase of SGA and low birth weight [812]. SGA and low birth weight not only increased the neonatal morbidity and mortality, but also some other chronic diseases such as type 2 diabetes, cardiovascular disease, and mental problems in adulthood [1315]. However there is no agreement on inadequate weight (0–4.9 kg) or weight loss (<0 kg) in obese women. Some groups have suggested that appropriate management of inadequate weight (0–4.9 kg) or weight loss (<0 kg) can contribute to improving neonatal outcomes [6, 16, 17]. Therefore, the objective of this systematic review and meta-analysis was to assess the relationship between inadequate weight gain during pregnancy and the risk of SGA in obese women.

2. Materials and Methods

2.1. Data Source and Search Strategy

This review was registered in PROSPERO with the number CRD42022345753. We comprehensively searched the Web of Science, PubMed, and Embase to identify related articles published before June 30th, 2022, using keywords and MeSH headings for Pregnant Women, Pregnancy, Obesity, Gestational Weight Gain, Weight Gain, Infant, and Small for Gestational Age (Table S1). No language restriction was imposed. Reference lists were also assessed to acquire additional relevant articles. All relevant terms, including free-text terms and MeSH terms, were used in the literature search. All reference lists of the relevant reviews were hand-searched for additional relevant trials.

2.2. Eligibility Criteria

Studies were selected if they examined outcomes in women with BMI defined as obesity (BMI > 30 kg/m2, I: BMI 30–34.9 kg/m2, II: BMI 35–39.9 kg/m2, and III: BMI ≥ 40 kg/m2) assessed by self-reported or objective measurement before pregnancy, during pregnancy, or postpartum). Studies were included if the following criteria were met: (1) Population of singleton pregnancies. (2) The primary outcome assessed was the SGA (defined as birth weight less than the 10th percentile of birth weight for sex and gestational age). (3) Women with BMI defined as obesity (BMI > 30 kg/m2, I: BMI 30–34.9 kg/m2, II: BMI 35–39.9 kg/m2, and III: BMI ≥ 40 kg/m2) assessed by self-reported or objective measurement before pregnancy, during pregnancy, or postpartum) [18]. (4) Obese women who gained weight below the recommendation of the 2009 guideline including less than 0 kg (weight loss) and 0–4.9 kg (inadequate weight) compared the gaining with the guidelines (5–9 kg) [4].

Studies were excluded if they assessed a population that is not representative (diabetes women and women with second pregnancy), if the combined effects between obesity and weight gain in obese women were not examined, and if they were duplicate or secondary publications, opinion articles, reviews, guidelines, posters, conference papers, case reports, nonhuman studies, non-English articles, and without enough data.

2.3. Data Extraction and Quality Assessment

Two investigators (LJ and LBY) independently searched, selected, and extracted publications from the literature. Inconsistent data were discussed by the two investigators to reach consensus or evaluated by a third senior investigator (GKX). To assess the methodological quality of included studies, we used a modified version of the Newcastle–Ottawa Quality Scale. Two researchers (CW and LBY) independently evaluated the study quality and assigned the quality grades. Discrepancies were resolved by consensus of them and another researcher (GKX). The Newcastle–Ottawa Scale is composed of three categories: “Selection,” “Comparability,” and “Outcome.” Our modified Newcastle–Ottawa Scale excluded one item (“demonstration that outcome of interest was not present at the start of study”) of the “Selection” category since the lack of relevance for our meta-analysis. The elimination of the item left a maximum of three points for the “Selection” category. As our outcomes required follow-up until the end of pregnancy, another item, namely, “was follow-up long enough for an outcome to occur” under the “Outcome” category was excluded. A maximum of two points were awarded for this column. The two “most important confounding factors” of the “Comparability” criteria were selected on the basis of a prior knowledge of their association with GWG and each outcome. This modified Newcastle–Ottawa scale [19] ultimately conferred up to six points. Due to the shortage of validation studies that provided a cutoff score for rating low-quality studies, an arbitrary cutoff of four or fewer was used to categorize a study as “low quality.”

For the outcome (SGA), the points for confounding were allocated as follows: one point was allocated for controlling for parity, and an additional point for age, smoking, or diabetes mellitus (DM). We designated the lowest score for the outcome (SGA) without controlling all the items. The final comparability score was the minimum score that a study received for all the outcomes. “0” means no point awarded; “1” means one point awarded [9].

Two reviewers (CW and LBY) independently extracted the following data from full-text articles: name of the first author, year of publication, country of study, time span of the study (years), study setting, study design, characteristics of participants (including the population, source, and categories of BMI), confounding factors, and adjusted OR (95% CI). The results were verified again by another independent reviewer (CW) (Table 1).

2.4. Statistical Analysis

The multivariate-adjusted odds ratio (OR) and corresponding 95% CI reported in the studies were used to produce forest plots in our meta-analysis. Data were statistically analyzed using RevMan 5.3 software, and measured values were quantified using the weighted mean difference (WMD) with a 95% confidence interval (CI). Heterogeneity among different studies was quantified using I2. Heterogeneity was deemed statistically significant and insignificant when I2 values were >50% or ≤50%, respectively, and analyzed using random and fixed effects models, respectively. The reasons for heterogeneity were explored using subgroup analyses. Publication bias was quantified using funnel plots and Egger tests. Values with were considered statistically significant, suggesting that publication bias was not excluded. Sensitivity analysis for outcome was performed for three groups of obese people (I: BMI 30–34.9 kg/m2, II: BMI 35–39.9 kg/m2, and III: BMI ≥ 40 kg/m2) to examine the effects of obesity and SGA.

3. Results

3.1. Literature Search

The process of study identification and inclusion, and the reasons for exclusion are presented in Figure 1. A total of 788 studies were identified by the search. Following the removal of duplicates, 684 titles and abstracts were screened. Eighty-three studies were selected for full-text review, and 29 studies [8, 16, 17, 2045], involving 1242420 obese pregnancies, met our eligibility criteria and were included in the systematic review. The kappa coefficient of agreement for included studies between the reviewers (LJ and LBY) was 0.99.

3.2. Study Characteristics

We got 29 articles [8, 16, 17, 2045] including prospective [17, 29] and retrospective cohort studies [8, 16, 2028, 3045], built into our meta-analysis. The included studies reported on at least 1242420 obese pregnant women. At least 175247(14.11%) obese women were weight loss (<0 kg) and inadequate weight (0–4.9 kg) during the pregnancy compared with the recommendations of IOM guidelines. Twenty studies were American [8, 2225, 2731, 33, 34, 36, 37, 4045], one was Japanese [32], one was from Lebanon [26], and seven were in Europe [16, 17, 20, 21, 35, 38, 39]. All but fifteen studies [8, 20, 21, 23, 24, 27, 28, 31, 33, 3740, 45] investigated outcomes according to three groups of obese people. In addition, twenty-six studies [8, 16, 17, 2022, 2438, 4042, 44, 45] also investigated outcomes for overall obesity except for three studies [23, 39, 43]. Eleven studies [8, 21, 22, 24, 28, 3134, 40, 45] investigated outcomes for obese women who weight loss (<0 kg), and reported outcomes for obese women who inadequate weight (0–4.9 kg). Table 1 provides detailed information.

3.3. Quality Score

One study scored two points, three scored three points, six scored four points, twelve scored five points, and the others scored six points. The articles scoring below or equal to four points were regarded as “low quality” and would be subsequently involved in the sensitivity analysis (Table 2).

3.4. Outcomes
3.4.1. Primary Outcomes

Obese women who gained weight below the guideline recommendations had a higher risk of SGA than women who gained weight within the guidelines (OR = 1.27, 95% CI = 1.16–1.38, Z = 5.36, ; 26 studies). Heterogeneity as defined by the I2 statistics was high (I2 = 59%, ); therefore, a random effect model was used for the analysis (Figure 2). In obese women, weight loss (<0 kg) and inadequate weight (0–4.9 kg) during pregnancy were associated with an increased risk of SGA. (OR = 1.50, 95% CI = 1.37–1.64, Z = 8.82, ; 11 studies) (OR = 1.18, 95% CI = 1.14–1.23, Z = 8.06, ; 11 studies), as shown in Figures 3 and 4.

Data from each class group showed the differences between the obese women who gained weight below the guidelines and those who gained weight within. Class I (OR = 1.38, 95% CI = 1.29–1.47, Z = 9.94, ; 14 studies); Class II (OR = 1.39, 95% CI = 1.30–1.49, Z = 9.08, ; 15 studies) Class III (OR = 1.26, 95% CI = 1.16–1.37, Z = 5.53, ; 14 studies) (Figure 5)

The same results were identified by Class I, Class II, and Class III of obese women who weight loss (<0 kg) and inadequate weight (0–4.9 kg) during pregnancy (Figure 6). In addition, we found that gestational weight loss (<0 kg) was associated with an increased risk for SGA (OR = 1.49, 95% CI = 1.33–1.66, Z = 6.85, ; 5 studies), as compared with inadequate weight (0–4.9 kg) (Figures 7 and 8).

3.4.2. Subgroup Analysis

The results of subgroup analysis by countries showed that GWG below guidelines in obese women were associated with risk for SGA: USA (OR = 1.30, 95% CI = 1.15–1.46, Z = 4.27, ; 18 studies) and Europe (OR = 1.24, 95% CI = 1.11–1.40, Z = 3.68, ; 6 studies), with no statistically significant result for Asia. (OR = 1.17; 95% CI = 0.91–1.50; Z = 1.26, ; 2 studies) (Figure 9).

3.4.3. Sensitivity Analyses and Publication Bias

Sensitivity analysis was used to evaluate the stability of the results. The sensitivity analysis indicated, compared with the original pooled OR, excluding the ten studies assessed as “low quality” also resulted in a similar (OR = 1.31; 95% CI = 1.25–1.38; Z = 10.37, ).

No more evidence of publication bias showed in the funnel plots for the overall obesity, weight loss, and inadequate weight group (Figure 10).

4. Discussion

Our meta-analysis demonstrates that obese women who gained weight below the guideline recommendations had more risks of SGA than those of gained weight within the guidelines. These data covered not only the population of overall obese women but all three classes of obesity of pregnant women. These results were similar to prior systematic reviews [911, 46]. However, these studies did not account for the differing socioeconomic, lifestyle, and racial backgrounds of patients. The repercussions of weight loss in obese gravida may varied based on race and socioeconomic classes, so studying these topics in diverse patient populations was important. Moreover, in our study, we also evaluated gestational weight loss in obese pregnant women was associated with an increases risk for SGA, compared with inadequate weight (0–4.9 kg). Of the 29 articles we selected, only five provided detailed information on the number of pregnancies at risk of weight loss (<0 kg) and inadequate weight (0–4.9 kg) for SGA. Therefore, our study added a subgroup analysis of race and found that obese women who gain weight below the guideline in the United States and Europe were associated with a higher risk for SGA, but not in Asia because the USA and Europe had the greatest prevalence of overweight and obesity [47, 48]. Asia women were more likely to be underweight than those in the USA and Europe [32].

Obesity during pregnancy is associated with a myriad of adverse outcomes such as preeclampsia, labour induction, postpartum haemorrhage, cesarean delivery, and preterm birth [49, 50]. Therefore, more obese women in the USA and Europe attempted to lose weight during pregnancy [51]. Our study also analyzed that not only weight loss (<0 kg) but also inadequate weight (0–4.9 kg) in obese pregnant women were associated with an increased risk of SGA. Our findings in these meta-analyses were also in line with the findings of a previous meta-analysis [10, 11]. Moreover, our results were identified by Class I, Class II, and Class III of obese women. The mechanism of weight gain within the guideline range during pregnancy contributes to SGA may be that the lack of maternal nutrition can lead to the placental vascular development change and barrier thickness increases, thus resulting in reduced glucose, amino acid, and lipid transport, as well as chronic hypoxia, which ultimately affected the fetus normal growth and development process.

4.1. Strength and Limitations

The strength of this systematic review included the comprehensiveness of the search strategies in three databases. We performed a careful quality assessment using a modified Newcastle–Ottawa scale. Sensitivity analyses corroborated the robustness of our findings and argued in favour of their validity. Importantly, we addressed the evidence for each obesity class. All included studies were adjusted for multiple important confounders, and all but ten studies were of high quality.

However, this meta-analysis has several limitations. First, it lacked studies from developing countries. The studies that met our inclusion criteria originated predominantly from United States. Hence, more research is needed from diverse populations to be able to generalize our findings. Second, some variables may also have influenced findings, such as maternal ethnicity, behavioral factors (diet, physical activity, and smoking), socioeconomic status, and women with prenatal complications, although some studies did adjust for these variables or excluded women with preexisting complications from their analysis. Third, the limits are related to the precision of self-reported GWG, the no possibility to obtain data on the dietary advice and dietary compliance of the women and on the long-term outcomes of neonates.

5. Conclusions

Our findings indicated that obese pregnant women who weight loss (<0 kg) and inadequate weight (0–4.9 kg) below the IOM guideline had increased risks for SGA. Therefore, the clinical focus should intensify efforts to assist obese women to achieve GWG within the IOM guidelines to decrease the risk for SGA. We also found that gestational weight loss in these pregnancies was associated with an increased risk for SGA compared with weight inadequate. These findings underline the importance considering the IOM guidelines in terms of gestational weight gain taking into consideration the different classifications of obese women.

Data Availability

The data described in this article can be freely and openly accessed from the original published articles in the database.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

Wen Chen, Beiyi Li, Kexin Gan, Jing Liu, Yajing Yang, and Xiuqin Lv contributed substantially to the concept and design of the study, performed data collection or analysis, and interpreted the data. The authors revised important substantive content. Wen Chen, Jing Liu, and Huijuan Ma have read and approved the final version of the manuscript.

Acknowledgments

This work was supported by the Medical Science Research Project of Hebei Provincial Health Commission of China (Grant number 20220802).

Supplementary Materials

Table S1: summary of the medical search strategy for this article. (Supplementary Materials)