BACKGROUND To unravel true links between diet and health, it is important that dietary exposure is accurately measured. Currently, mainly self-reporting methods (e.g. food frequency questionnaires and 24-h recalls) are used to assess food intake in epidemiological studies. However, these traditional instruments are subjective measures and contain well-known biases. Especially, estimating the intake of the group of confectionary products, such as products containing cocoa and liquorice, remains a challenge. The use biomarkers of food intake (BFIs) may provide a more objective measurement. However, an overview of current candidate biomarkers and their validity is missing for both cocoa- and liquorice-containing foods. OBJECTIVE The purpose of the current study was to (1) identify currently described candidate BFIs for cocoa (products) and liquorice, (2) to evaluate the validity of these identified candidate BFIs and (3) to address further validation and/or identification work to be done. METHODS This systematic review was based on a comprehensive literature search of three databases (PubMed, Scopus and ISI web of Science), to identify candidate BFIs. Via a second search step in the Human Metabolome Database (HMDB), the Food Database (FooDB) and Phenol-Explorer, the specificity of the candidate BFIs was evaluated, followed by an evaluation of the validity of the specific candidate BFIs, via pre-defined criteria. RESULTS In total, 37 papers were included for cocoa and 8 papers for liquorice. For cocoa, 164 unique candidate BFIs were obtained, and for liquorice, four were identified in total. Despite the high number of identified BFIs for cocoa, none of the metabolites was specific. Therefore, the validity of these compounds was not further examined. For liquorice intake, 18-glycyrrhetinic acid (18-GA) was found to have the highest assumed validity. CONCLUSIONS For cocoa, specific BFIs were missing, mainly because the individual BFIs were also found in foods having a similar composition, such as tea (polyphenols) or coffee (caffeine). However, a combination of individual BFIs might lead to discriminating profiles between cocoa (products) and foods with a similar composition. Therefore, studies directly comparing the consumption of cocoa to these similar products are needed, enabling efforts to find a unique profile per product. For liquorice, we identified 18-GA as a promising BFI; however, important information on its validity is missing; thus, more research is necessary. Our findings indicate a need for more studies to determine acceptable BFIs for both cocoa and liquorice.

中文翻译：

---

可可和甘草（产品）食物摄入的生物标志物：系统评价。

背景为了揭示饮食与健康之间的真正联系，准确测量饮食暴露是很重要的。目前，主要使用自我报告方法（例如食物频率问卷和 24 小时回忆）来评估流行病学研究中的食物摄入量。然而，这些传统工具是主观的测量方法，并且包含众所周知的偏见。特别是，估计糖果产品组的摄入量，例如含有可可和甘草的产品，仍然是一个挑战。使用食物摄入生物标志物 (BFI) 可以提供更客观的测量。然而，目前对含可可和甘草食品的候选生物标志物及其有效性的概述都缺失。目标 本研究的目的是 (1) 确定目前描述的可可（产品）和甘草候选 BFI，(2) 评估这些已识别的候选 BFI 的有效性，以及 (3) 解决有待完成的进一步验证和/或识别工作。方法 本系统评价基于对三个数据库（PubMed、Scopus 和 ISI web of Science）的全面文献检索，以确定候选 BFI。通过在人类代谢组数据库 (HMDB)、食品数据库 (FooDB) 和 Phenol-Explorer 中的第二个搜索步骤，评估候选 BFI 的特异性，然后评估特定候选 BFI 的有效性，通过预定义的标准。结果 总共包括 37 篇关于可可的论文和 8 篇关于甘草的论文。对于可可，获得了 164 个独特的候选 BFI，对于甘草，总共确定了四个。尽管已确定的可可 BFI 数量众多，但没有一种代谢物具有特异性。所以，这些化合物的有效性没有进一步检查。对于甘草摄入量，发现 18-甘草次酸 (18-GA) 具有最高的假定有效性。结论 对于可可，特定的 BFI 缺失，主要是因为在具有相似成分的食物中也发现了单个 BFI，例如茶（多酚）或咖啡（咖啡因）。然而，单个 BFI 的组合可能会导致可可（产品）和具有相似成分的食物之间的差异化特征。因此，需要进行直接比较可可消费与这些类似产品的研究，以便努力找到每种产品的独特概况。对于甘草，我们将 18-GA 确定为有前景的 BFI；但是，缺少有关其有效性的重要信息；因此，需要更多的研究。