Espinosa Alan, Mendoza Kenny, Laviada-Molina Hugo, Rangel-Méndez Jorge Aarón, Molina-Segui Fernanda, Sun Qi, Tobias Deirdre K, Willett Walter C, Mattei Josiemer
Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston MA, USA.
Department of Human Nutrition and Metabolic Research, Health Sciences School, Universidad Marista de Merida, Mexico.
Adv Nutr. 2024 Dec;15(12):100292. doi: 10.1016/j.advnut.2024.100292. Epub 2024 Sep 13.
The effect of nonnutritive sweeteners (NNSs) on the body mass index [BMI (in kg/m)] of children and adolescents remains unclear despite rising consumption. Detailed systematic evaluations are warranted. We aimed to summarize evidence on NNS consumption and BMI sex- and age-specific absolute changes (kg/m) in pediatric populations, by NNS type, study design, duration, analysis type, conflicts of interest (COI), geographical region, age, sex, and baseline BMI. We searched randomized controlled trials (RCTs) and prospective cohort studies in children (2-9 y), adolescents (10-24 y), and young adults (20-24 y). Pooled estimates derived from random-effects meta-analysis for BMI changes, and the evidence quality was evaluated overall and by subgroup. From 2789 results, we included 4 RCTs [n = 1372; mean follow-up = 42.6 wk (standard deviation = 18.4); 2 (50%) with COI], and 8 prospective cohort studies [n = 35,340; median follow-up 2.5 y (interquartile range = 1.7-6.3), 2 (25%) with COI]. No identified studies evaluated NNS in food, NNS beverages compared with water, or participants aged 20-24 y. Random allocation to NNS beverages (25-2400 mg/d, from beverages) showed less BMI gain [mean difference = -0.114 kg/m (95% confidence interval [CI]: -0.207, -0.021); I = 87.02%] compared with sugar-sweetened beverages (SSBs). Stratified estimates resulted in less BMI gain in adolescents, participants with baseline obesity, consumers of mixed NNS, longer trials, and trials without COI. Pooled estimates from prospective cohorts showed a nonsignificant association between NNS beverages and BMI gain [0.05 kg/m (95% CI: -0.03, 0.13); I = 75.06%; per daily 355 mL serving]. Stratified estimates remained consistent. Removing studies with COI attenuated estimates. Evidence had low to moderate quality. In summary, pooled results from RCTs comparing NNS beverages compared with SSBs showed less BMI gain in adolescents with obesity. Meta-analyses of long-term cohort studies did not display a significant association between NNS beverages and BMI changes. This trial was registered at PROSPERO as CRD42022352284.
尽管非营养性甜味剂（NNSs）的消费量在不断增加，但其对儿童和青少年体重指数[BMI（kg/m）]的影响仍不明确。有必要进行详细的系统评估。我们旨在按NNS类型、研究设计、持续时间、分析类型、利益冲突（COI）、地理区域、年龄、性别和基线BMI，总结儿科人群中NNS消费与BMI性别和年龄特异性绝对变化（kg/m）的证据。我们检索了针对儿童（2 - 9岁）、青少年（10 - 24岁）和青年成年人（20 - 24岁）的随机对照试验（RCTs）和前瞻性队列研究。通过随机效应荟萃分析得出BMI变化的合并估计值，并对总体和各亚组的证据质量进行评估。从2789个结果中，我们纳入了4项RCTs[n = 1372；平均随访时间 = 42.6周（标准差 = 18.4）；2项（50%）存在COI]，以及8项前瞻性队列研究[n = 35,340；中位随访时间2.5年（四分位间距 = 1.7 - 6.3），2项（25%）存在COI]。未发现有研究评估食品中的NNS、NNS饮料与水的比较，或20 - 24岁参与者。随机分配至NNS饮料组（每日25 - 2400毫克，来自饮料）与含糖饮料（SSBs）相比，BMI增加较少[平均差值 = -0.114 kg/m（95%置信区间[CI]：-0.207，-0.021）；I² = 87.02%]。分层估计显示，青少年、基线肥胖参与者、混合NNS消费者、试验时间较长以及无COI的试验中BMI增加较少。前瞻性队列的合并估计显示，NNS饮料与BMI增加之间无显著关联[0.05 kg/m（95% CI：-0.03，0.13）；I² = 75.06%；每份355毫升每日摄入量]。分层估计结果保持一致。剔除存在COI的研究后，估计值减弱。证据质量为低到中等。总之，比较NNS饮料与SSBs的RCTs合并结果显示，肥胖青少年的BMI增加较少。长期队列研究的荟萃分析未显示NNS饮料与BMI变化之间存在显著关联。本试验已在国际前瞻性系统评价注册库（PROSPERO）注册，注册号为CRD42022352284。
