Abstract
Zinc (Zn) malnutrition is a prevalent micronutrient deficiency that can negatively impact cognitive and physical health, especially among women and children in developing countries. One effective strategy to address low Zn levels in staple cereals (e.g. wheat, rice and maize), which are essential for human health, is Zn biofortification. There are two main approaches to Zn biofortification: genetic biofortification, which involves using the genetics of the crops to increase their Zn content, and agronomic biofortification, which uses zinc-rich fertilizers to boost the Zn levels in the edible parts of the crops. This systematic review aims to evaluate the effectiveness of agronomic biofortification strategies in increasing grain zinc concentrations in major cereals, specifically wheat, rice, and maize, through a network meta-analysis. Cochrane Library, Google Scholar, Scopus, and agricultural journals were searched up to the year of 2017 to identify relevant field trials assessing the impact of Zn fertilisation on grain Zn concentration and yield. The search was updated in 2020. Eligible studies were those focusing on wheat, rice, and maize, field-based, addressing Zn fertilisation research question, where grain Zn concentration served as the primary outcome and grain yield as a secondary outcome. Data were extracted and assessed for quality of evidence in the included studies. Outcome data was limited to grain Zn concentration and grain yield. Pairwise and network meta-analysis for direct and indirect comparisons of Zn fertilisation methods were performed. This review summarises evidence from 44 independent field-based studies related to zinc (Zn) fertilisation, of which 27 focused on wheat, 13 on rice, and 4 on maize. These studies involved different application methods, including soil, foliar, and combined soil + foliar Zn fertilisation, compared against control groups with no Zn fertilisation. The design of the studies was randomised blocks, typically with three or four replicates. In total, 21 studies evaluated the effectiveness of soil, foliar, and combined soil + foliar Zn fertilisation methods against a control, while 15 studies focused solely on foliar application comparisons against control groups. The review highlights the prevalence of foliar Zn studies and the few instances of comparisons involving combined fertilization methods. The risk of bias in the included studies is mostly low. Soil + foliar Zn fertilisation significantly increases grain Zn concentration in wheat by 28.7 mg/kg [95%-CI: 22.77 to 34.53], while the increase is smaller for rice (6.8 mg/kg [95%-CI: 5.3 to 8.22]) and maize (7.9 mg/kg [95%-CI: 6.83 to 8.91]). Foliar Zn fertilisation also increases Zn concentration, reaching up to 18.0 mg/kg [95%-CI: 14.26 to 21.79] in wheat, 6.7 mg/kg [95%-CI: 3.25 to 10.04] in rice, and 7.6 mg/kg [95%-CI: 6.51 to 8.75] in maize, although it has minimal effects on grain yields. Conversely, soil Zn fertilisation results in a lower increase in Zn concentration, such as 4.7 mg/kg [95%-CI: 2.30 to 7.12] in wheat, with variable impacts on grain yields. Overall, considering the baseline of 16 mg/kg for rice and 25 mg/kg for wheat and maize, the effectiveness of Zn fertilisation strategies varies across different crops; the combination of soil and foliar Zn application significantly enhances grain Zn concentrations, especially in wheat with an increase of 114.8% over the baseline, while rice shows some statistically significant increase of 42.5% over the baseline, but maize does not show significant increases. Despite the biologically significant effects, there is low certainty of evidence, due mostly to high heterogeneity and unexplained inconsistency between studies. Agronomic strategies can increase wheat grain Zn concentrations, aiding in biofortification efforts to improve human Zn availability. The combination of soil and foliar applications is the most effective approach. Although increase in rice and maize grain Zn concentrations is statistically significant, it remains relatively small; however, even a slight increase (1-mg/kg) in rice could have an impact on human health due to its lower baseline Zn levels. Future research on the genetic variability of Zn concentration in rice could enhance farming practices and breeding programs, potentially incentivised by premium payments for high-Zn grain products.
Declarative Title
Combined soil and foliar fertilisation significantly increase grain zinc concentrations of wheat while offering smaller effects to rice and maize.
The Review in Brief
Combined soil and foliar fertilisation significantly enhance grain zinc concentration in wheat (28.7 mg/kg), rice (6.8 mg/kg), and maize (7.9 mg/kg) grain, with minimal or variable impacts on yield, when compared with individual soil or foliar application.
What Is This Review About?
Zinc deficiency is a prevalent issue that adversely affects cognitive and physical health, particularly among vulnerable groups such as women and children in developing countries. Therefore, it is critical to improve Zn levels in staple food crops e.g. cereals, due to their widespread global consumption and potential implications for public health. However, there are different strategies with different effectiveness to increase Zn levels in staple food crops. This review focuses on agronomic biofortification, which includes Zn fertilisation with a range of available methods (such as soil application, foliar spraying, or a combination of both) aimed at increasing the Zn content in staple foods. The outcomes evaluated were an increase of Zn concentration in the grain (primary outcome), and the overall grain yield of wheat, rice, and maize (secondary outcome) as affected by Zn fertilisation. Thus, this review considers the policy question pertaining to how effective agronomic biofortification strategies are in increasing Zn levels in staple crops with the potential for improving dietary Zn availability for populations suffering from Zn deficiency. It also explores the potential implications for public health initiatives aimed at addressing micronutrient deficiencies in developing countries, as well as the economic strategies for incentivising farmers to adopt agronomic biofortification practices. Overall, this review provides insights into effective agricultural interventions that could enhance Zn concentrations and improve health outcomes in at-risk populations.
What Is the Aim of This Review?
The aim of the review is to evaluate the effectiveness of agronomic biofortification strategies, specifically soil and foliar zinc fertilisation and the combination of both, in increasing grain Zn concentration of staple cereals—wheat, rice, and maize—while assessing their impact on grain yield and addressing the broader policy question of mitigating Zn deficiency among vulnerable populations in developing countries.
What Are the Main Findings of This Review?
This review summarises evidence from 44 independent studies conducted between 1999 and 2017, across Africa, Europe, Asia and America, focusing on the impact of Zn fertilisation on grain Zn concentration of the major cereals i.e. wheat, rice, and maize. These studies involved 105 crop genotypes grown on diverse soils with varying pH and Zn status, primarily in Asia (notably China, Turkey, Pakistan, and India). These independent studies investigated various Zn application methods, including soil and foliar treatments and a combination of both laid in randomised block designs with three or four replicates. Common wheat genotypes feature prominently, with specific cultivars appearing in multiple studies. All studies included control treatments with no Zn application and reported grain Zn concentrations, but only seven did not address grain yield, and none assessed Zn bioavailability in humans.
Combined soil and foliar Zn fertilisation significantly increases grain Zn concentration of wheat by 28.7 mg/kg, while rice and maize show smaller increases of 6.8 mg/kg and 7.9 mg/kg, respectively, along with slight yield improvements for wheat and rice but a decrease for maize. Foliar Zn application alone raised grain Zn levels in wheat by 18.0 mg/kg with minimal impact on yield across all crops, while soil Zn fertilisation yields modest increases in grain Zn with varying effects on crop yields. A network meta-analysis supports these findings, showing that combined fertilization is particularly effective for wheat. Although results are statistically significant for rice, no significant differences were observed for maize. Overall, the quality of evidence is low due to high study variability, but bias risk is low from an agronomic standpoint.
Summary of Evidence and Findings: - Wheat benefits most from soil and foliar combined fertilisation, leading to a significant increase of 28.7 mg/kg [95%-CI: 22.77 to 34.53] in grain Zn concentration. - Rice and Maize show a lower increase in grain Zn concentration of 6.8 mg/kg [95%-CI: 5.3 to 8.22] and 7.9 mg/kg [95%-CI: 6.83 to 8.91] respectively under the same treatment. - Foliar Zn application increases Zn concentrations by 18.0 mg/kg [95%-CI: 14.26 to 21.79] for wheat, 6.7 mg/kg [95%-CI: 3.25 to 10.04] for rice, and 7.6 mg/kg [95%-CI: 6.51 to 8.75] for maize compared to the control treatments. - Soil Zn fertilisation resulted in a small increase in Zn concentration, particularly for wheat (4.7 mg/kg [95%-CI: 2.30 to 7.12]) and maize (4.9 mg/kg [95%-CI: 2.78 to 6.92]). - The yields across all crops show high variability, and the confidence intervals indicate uncertainty regarding the effect size; while wheat and rice grain yields show slight increase with Zn fertilization, maize shows a decrease. - Therefore, combined soil and foliar fertilization results in a greater increase in grain Zn concentration in wheat compared to foliar and soil fertilization. Despite low risk of bias from an agronomic perspective, the overall evidence quality was considered low due to high inconsistency among studies (I2 greater than 90%).
What do the Findings of This Review Mean?
- Policies could be implemented to incentivise farmers to adopt soil and foliar combined Zn fertilisation techniques, particularly for wheat production, to enhance grain Zn concentration and address micronutrient deficiencies. - Given the varying levels of Zn uptake and yield response across different crops, tailored fertilisation strategies should be implemented based on specific crop requirements and local soil conditions. - Supporting research initiatives that address the identified inconsistencies in response to Zn applications and factors influencing effectiveness could lead to more reliable guidelines for farmers. - Future studies should investigate the biological mechanisms behind crop-specific responses to Zn fertilisation, particularly focusing on soil pH, soil Zn status, and other agronomic factors. - Conducting long-term studies could provide more comprehensive data on the sustainability of Zn fertilisation practices and their impact on soil health (if any) and crop yields over time. - Reviews should encompass a broader range of geographical locations and environmental conditions to better understand how different factors influence Zn uptake across various crop species. - Emphasis on improving study designs to reduce inconsistency and enhance the reliability of findings is crucial for guiding agricultural practices and policies effectively.
How Up-To-Date Is This Review?
The review authors conducted the search for studies up to the year 2020. They are also aware of more recent publications (e.g., Bhardwaj et al., 2022; Singh et al., 2023) that were not included in this review. This new information will be considered for future updates to the review, should subsequent publications suggest changes in the direction or magnitude of the pooled effects or should sufficient studies likely to explain inconsistency and heterogeneity emerge. Unfortunately a substantial body of evidence is likely to be needed to address these deficiencies in the current evidence base.
Background
The Problem, Condition, or Issue
There are trends of declining mineral content in staple cereals, resulting either due to plant factors or changes in soil nutrient concentrations. For instance, zinc concentration in wheat has significantly decreased since mid-1960s, coinciding with the introduction of semi-dwarf, high-yielding cultivars, thus identified as significant plant factors contributing to the downward trend in grain mineral concentration (Fan et al., 2008) which can be related to zinc malnutrition in humans.
Zinc malnutrition poses a significant global health challenge, particularly prevalent in developing countries (Lowe et al., 2020), where it is a widespread micronutrient deficiency that adversely affects both cognitive and physical health, especially among vulnerable populations like women and children. The consequences of Zn deficiency are far-reaching, impacting immune function, growth, and neurological development (Alloway, 2009; Cakmak, Kalaycu, et al., 2010; Ohly et al., 2019; Velu et al., 2014; Zou et al., 2012). Addressing this deficiency is critical for improving overall public health, enhancing educational outcomes, and fostering economic development in regions where malnutrition is prevalent (Lassi et al., 2020; Staub et al., 2021; Zou et al., 2019). Hence, the context for exploring solutions to Zn malnutrition through biofortification has become increasingly urgent, as global awareness of the interplay between nutrient deficiencies and health outcomes grows (Ramadas et al., 2020). In this context, biofortification is seen as an effort to increase mineral concentrations back to levels before the mid-1960s agriculture systems, which may be a cost-effective and sustainable approach to reducing Zn deficiency, complementing fortification at processing stage and dietary diversification (Lowe et al., 2020; Tang et al., 2022).
The proposed intervention to combat Zn malnutrition involves genetic and agronomic strategies to enhance Zn content in food crops. Major food staples, such as rice, wheat, and maize, often contain inadequate levels of bioavailable Zn in addition to their recent loss of genetic diversity (Dapkekar et al., 2020; FAO & CIRAD, 2015; Zou et al., 2012), thus their byproducts also generally have low Zn concentration. As these crops serve as primary energy sources for large segments of the population, particularly in low-income countries, enhancing their Zn content is essential for combating malnutrition (Chattha et al., 2017; Khan et al., 2017; Olsen & Palmgren, 2014; Velu et al., 2014 ). With two thirds of the world agricultural land having marginal to severely low Zn availability, crops grown on such soils have inherently low Zn concentrations (Dapkekar et al., 2020; Gupta et al., 2016). A judicious use of Zn fertilisers (supported by appropriate methods of application, rate and timing) is likely to be the primary short-term strategy for correcting low soil availability and, in turn, increase Zn intake derived from these crops by humans globally (Alloway, 2009; Dapkekar et al., 2020; FAO & CIRAD, 2015; Tuyogon et al., 2016; Velu et al., 2014). Therefore, conducting a review in this topic is crucial as it addresses the need to tackle Zn malnutrition and its broad implications for public health and socioeconomic development. With an increasing global awareness of the link between nutrient deficiencies and health outcomes, this review can highlight effective strategies to enhance Zn bioavailability in food crops. This information is critical for policymakers, agricultural stakeholders, and health practitioners aiming to devise sustainable solutions that could alleviate Zn deficiency and improve overall health in vulnerable populations (Lassi et al., 2020; Ohly et al., 2019; Staub et al., 2021; Zou et al., 2019).
The Intervention
Zinc biofortification is a promising approach to address the inadequate Zn levels in the staple foods. There are two primary approaches to achieve this enhancement (Cakmak, 2008; Ramadas et al., 2020; Velu et al., 2014):
Genetic Biofortification
This approach focuses on breeding crop varieties that inherently have higher Zn concentrations. It can involve traditional breeding techniques or modern biotechnological methods. The goal is to develop new cultivars that not only produce high yields but also provide essential nutrients, including Zn. Genetic biofortification has the potential for long-term sustainability, as it can create crops that continually provide enhanced nutritional benefits (Borrill et al., 2014; Dapkekar et al., 2020; Garcia-Oliveira et al., 2018; Waters & Sankaran, 2011).
Agronomic Biofortification
This strategy focuses on the application of Zn-containing fertilizers during crop growth. By enriching the soil with Zn, farmers can enhance Zn concentration in the edible parts of their crops (Mao et al., 2014; Tuyogon et al., 2016). Agronomic biofortification offers a more immediate and flexible solution compared to genetic strategies, allowing for immediate improvements in crop nutrient profiles (Cakmak & Kutman, 2018; Liu et al., 2016). However, it relies on effective agronomic practices and adequate farmer education to ensure its success.
Common agronomic practices to correct nutrient deficiencies include soil and foliar application of fertilisers. Recent interest and research have concentrated on using these strategies to increase Zn concentration in the edible parts of crops, particularly the major cereals i.e. wheat, maize and rice (Cakmak, 2008). By improving Zn availability for plant uptake, remobilisation and storage in edible plant parts, these strategies aim to alleviate Zn malnutrition in humans and enhance dietary Zn availability (Alloway, 2009; Dapkekar et al., 2020; Velu et al., 2014). The economic viability of these farming practices is also crucial, especially as investments are made in nutraceutical preparations and the development of functional foods to address Zn deficiencies in the human diet (Basu et al., 2007). For instance, Zn is utilised in nutraceuticals and dietary supplements to improve the health of patients with chronic liver disease and hepatic encephalopathy (Almoselhy, 2023; Puri et al., 2022).
Soil fertilisation strategy typically involves spreading granular Zn fertiliser, usually zinc sulphate (ZnSO4.7H2O), before sowing or planting. The Zn is taken up by the roots and then translocated to the vegetative tissues from which it is transferred into developing grain. Foliar strategy involves spraying zinc solution (ZnSO4.7H2O) on the leaves either once or multiple times during the growing season. The absorbed Zn is subsequently translocated preferentially to the actively growing vegetative tissues (to be remobilised later in the growing season) or directly to the developing grains if applied during post-anthesis growth (Cakmak, Kalaycu, et al., 2010; Dapkekar et al., 2020; Li et al., 2015; Phattarakul et al., 2012; Wang, Li, et al., 2015; Zhang, Deng, et al., 2012; Zhao et al., 2014; Zou et al., 2012).
How the Intervention Might Work
Zinc biofortification is an approach linking agricultural practices with public health outcomes. By enhancing the nutritional quality of staple crops, this strategy has the potential for improving dietary Zn availability and lead to better health for populations suffering from Zn malnutrition. For decades, the use of Zn fertilisers in agriculture has been justified economically for enhancing availability for plant uptake with the potential to increase Zn concentration and crop yields (Dapkekar et al., 2020; Manzeke et al., 2012, 2014). When Zn fertilisers are applied, there is an increase in Zn availability to the plant roots and leaves, allowing for greater uptake and potential accumulation in the grain. This effect is particularly pronounced in some cereal cultivars bred for biofortification, as these cultivars have a genetic predisposition to accumulate higher levels of the nutrient (Wang et al., 2015). However, the effectiveness of Zn fertilisation is influenced by several factors, including Zn solubility/availability in the soil, different forms of Zn fertilisers and timing of fertilisation. Furthermore, although soil Zn applications tend to produce approximately ten times greater effects on grain Zn content compared to overall yield, foliar Zn applications are commonly employed in soils with severe Zn deficiency (Cakmak, Kalaycu, et al., 2010; Zhao et al., 2014; Wang, Li, et al., 2015).
Zinc concentration in cereal grain can be increased through several mechanisms: uptake from the soil, absorption through plant tissue or by remobilisation from vegetative tissues as the plant matures. However, as stated previously, crop genetics, soil pH and soil available Zn concentration can influence the uptake and subsequent remobilisation of Zn from the senescing vegetative tissues to the developing grains (Dapkekar et al., 2020; Olsen & Palmgren, 2014; Palmgren et al., 2008; Waters & Sankaran, 2011; Zhang et al., 2013a, 2013b). In particular, high-pH (alkaline, calcareous) soils have exceptionally low Zn availability (Rengel & Graham, 1995; Zhao et al., 2014). In these conditions, both soil and foliar Zn fertilisation become acutely important to ensure adequate Zn supply (Cakmak, Pfeiffer, et al., 2010; Phattarakul et al., 2012; Li et al., 2016; Ram et al., 2016; Dapkekar et al., 2020).
The effectiveness of agricultural interventions, such as the application of Zn fertilizers, are informed by agronomic studies that investigate the relationship between soil properties, crop yield, and nutrient density (Alloway, 2009; Dapkekar et al., 2020; Farooq et al., 2018; Olsen & Palmgren, 2014). The biofortification process entails several effects that influence health outcomes through direct and indirect pathways: development and dissemination of crop varieties with higher Zn concentrations, application of Zn fertilizers (granular and foliar) to enhance the zinc content in crops. These pathways lead to increased Zn concentration in staple crops (e.g., wheat, maize, rice), improved agricultural productivity and profitability for farmers due to enhanced crop yield, increased dietary Zn intake in the population, and prevention or reduction of Zn deficiency and related health issues (e.g., stunted growth, impaired immune function).
Therefore, the application of Zn-rich fertilisers enhances Zn availability in the soil, promoting better uptake by plants. Studies have shown positive correlations between soil Zn levels and crop Zn concentrations (Cakmak, Kalaycu, et al., 2010; Dapkekar et al., 2020; Shivay et al., 2016). Although evidence remains limited and inconclusive in certain cases—primarily due to challenges in accurately measuring body Zn uptake (Lowe et al., 2020)—consuming Zn biofortified foods holds significant potential to enhance the overall Zn status of populations, leading to better health outcomes (Borrill et al., 2014). The introduction of biofortified crops can provide farmers with diversified income streams and improve livelihoods, potentially leading to better health-seeking behaviors and greater investment in health and nutrition (Ramadas et al., 2020; Velu et al., 2014). The use of Zn fertiliser has the potential for improving Zn status of crops up to fourfold from the current baseline, and thus, for producing crops with Zn concentrations that meet human requirements for dietary intakes. Additionally, improving Zn concentration in crops via agronomic strategies (i.e. soil and foliar fertilisation) lowers the risk of malnutrition and chronic diseases, especially in children and pregnant women, due to inadequate Zn intake. Most importantly, it limits the need for improving Zn nutrition via supplementation or via commercial food fortification (Alloway, 2009; Velu et al., 2014; Wang, Li, et al., 2015) for treatment and prevention of various diseases and health problems.
Why it Is Important to do This Review
The Harvest Plus Zinc programme has successfully implemented biofortification through agronomic strategies. This programme aims to enhance the nutritional and agronomic qualities of several key crops, including wheat, rice, maize, sweet potato, common bean, and cassava. Research by Waddington et al. (2012) suggests that Zn supplementation, with or without riboflavin, may support the growth of children under six months. Additionally, Staub et al. (2021) reported that Zn supplementation could facilitate weight gain, promote linear growth, and decrease mortality rates among pre-term infants. Tang et al. (2022) modelled food fortification in Malawi to inform national nutritional strategies by balancing efforts to improve micronutrient density and ensure adequate dietary intake, and found that additional micronutrient interventions are needed to meet dietary needs of vulnerable groups.
To our knowledge, there is currently no systematic review or network meta-analysis that specifically assesses the impact of soil and foliar application on grain Zn concentration and yield of major cereals. The increasing recognition of the role of nutrition in the health and development of countries by government and internation organizations underscores the significance of this review. This review could provide essential evidence to advocate for the integration of biofortification into national agricultural strategies, contribute to global discussion on tackling malnutrition, and support ongoing investment in biofortification research.
The policy implications of this review are notable in several key areas: (i) it could help address Zn deficiencies by providing robust evidence to guide nutrition policies and enhance dietary guidelines; (ii) it may enhance food security by offering actionable insights on the effectiveness of biofortification strategies, directing research and funding priorities for countries aiming for long-term agricultural policies aligned with Sustainable Development Goal 2 (Zero Hunger); (iii) it could promote sustainable agricultural practices that improve crop nutrition with minimal changes in agricultural inputs; and (iv) it may inform public health initiatives focused on combating micronutrient deficiencies, elucidating how Zn biofortification aligns with broader sustainability goals.
Evidence from this review would support the inclusion of Zn-biofortified cereal crops in national dietary guidelines and food composition databases to improve population Zn intake through food-based approaches. It also would encourage policies integrating agronomic biofortification into national nutrition and agriculture strategies. From a sustainability perspective, evidence from this review would promote efficient nutrient use, improved soil health, and reduced environmental losses, aligning with climate-smart and resource-efficient farming principles. Overall, this evidence would support a nutrition-sensitive, “farm-to-fork” approach that simultaneously advances human health and sustainable agricultural production.
Major cereals such as wheat, maize, and rice are typically low in grain Zn, which can lead to Zn deficiency and malnutrition in populations reliant on cereal-based diets. Various strategies—such as diet diversification, food fortification, nutrient supplementation, and agronomic biofortification—have been proposed to combat low-Zn grain and associated deficiencies. Among these, biofortification stands out as a cost-effective and accessible means of producing Zn-enriched crops that enhance nutritional security (Bhardwaj et al., 2022; Bouis & Saltzman, 2017; Singh et al., 2023).
While Zn supplementation through nutraceuticals has significantly addressed Zn malnutrition, empirical evidence suggests that agronomic Zn biofortification could be a viable strategy to improve soil Zn availability and alleviate related human malnutrition (Cakmak, 2008; Yadav et al., 2020). Therefore, it is crucial to identify fertilisation methods that not only increase grain Zn concentration and yield but also enhance Zn bioavailability in human digestive system.
Several studies have explored the impact of soil and foliar Zn fertilization on Zn concentration and bioavailability in grains of major cereals (Cakmak, Kalaycu, et al., 2010; Li et al., 2015; Phattarakul et al., 2012; Wang, Li, et al., 2015; Zhang, Deng, et al., 2012; Zou et al., 2012). Findings indicate that foliar fertilisation may be generally more effective than soil application in increasing grain Zn concentration, with evidence showing an inverse relationship between grain Zn levels and yield (Farooq et al., 2018; Phattarakul et al., 2012; Ram et al., 2016). However, a significant portion of Zn in wheat and rice grains comes from remobilisation from senescing leaves and continuous uptake from soil, suggesting that both soil and foliar fertilisation are effective for Zn biofortification in these crops (Phattarakul et al., 2012; Shivay et al., 2016; Waters & Sankaran, 2011). Cakmak et al. (2010) and Farooq et al. (2018) further recommended a combined approach of soil and foliar fertilisation for increasing grain Zn concentrations and yields in wheat and rice.
Caulfield and Black (2004) reviewed the effect of Zn deficiency on human healt in relation to mortality risk of diarrhoea, pneumonia and malaria among children in developing countries but did not assess biofortification strategies to mitigate the prevalence of inadequate Zn intakes. Ashong et al. (2012) reviewed the advantages and disadvantages of rice biofortification with Zn and other minerals concerning micronutrient status and health outcomes. Their findings imply that industrial Zn fortification during rice processing could help mitigate Zn malnutrition. However, their estimates varied widely, indicating that while some interventions may be highly effective, others exhibit negligible impact. Brnić et al. (2016) compared two methods of increasing Zn content in rice (growing rice in nutrient Zn-enriched solution and industrial biofortification) to determine which approach more effectively enhances Zn availability to consumers. However, the results were inconclusive regrading which method is more effective, primarily because of high imprecision. Khan et al. (2017) found that flour made from agronomically biofortified wheat can positively influence Zn status, suggesting that agronomic biofortification of wheat is a viable approach to tackling Zn deficiency. Signorell et al. (2023) also noted that while consuming Zn biofortified whole wheat flour improved overall Zn status, it did not significantly affect plasma Zn levels or deficiency in children aged 4 to 12 years.
Despite substantial primary research on Zn biofortification through soil and foliar applications, conducting a systematic review with comprehensive data collection, critical assessment, and statistical synthesis via network meta-analysis would enhance understanding in this area. An existing synthesis by Joy et al. (2015) offered insights into the cost-effectiveness of Zn-enriched fertilisers for alleviating dietary Zn deficiency in sub-Saharan Africa but lacked the methodological rigor typical of systematic reviews, as it did not follow a guiding protocol, define search and inclusion criteria, or conduct a critical appraisal. Additionally, it did not investigate effect modifiers or publication bias and did not adhere to the standards of the Methodological Expectations of Campbell Collaboration Intervention Reviews (MECCIR). As of now, no network meta-analysis has been conducted in agricultural research, despite its application in various other fields.
Objectives
To assess the effectiveness of agronomic biofortification strategies in increasing grain Zn concentration and grain yield of wheat, rice and maize, using conventional and network meta-analysis to rank Zn fertilisation methods.
Methods
This review followed the methods described in the pre-registered study protocol (Domingos et al., 2017).
Criteria for Considering Studies for This Review
Types of Studies
Characteristics of Included Studies
Types of Participants
The major cereals i.e. wheat, rice and maize were the target crops because, globally, they represent a significant component of the human diet.
Types of Interventions
Types of Outcome Measures
Primary outcomes: Zn concentration in the grain.
Secondary outcomes: grain yield.
Search Methods for Identification of Studies
Briefly describe the anticipated search strategy.
Electronic Searches
The following electronic databases were searched: Cochrane Central register of Controlled Trials (CENTRAL), Google Scholar, Scopus (Elsevier B.V.), Web of Science, and Science Direct. The search also included leading Agricultural journals such as Plant and Soil, Journal of Agricultural and Food Chemistry, Field Crops Research, Journal of Cereal Science, Trends in Plant Science, New Zealand Journal of Agricultural Research, Communications in Soil Science and Plant Nutrition, Journal of Plant Nutrition, and Journal of Agricultural Science. The search also included the agricultural trials registry https://ccafs.cgiar.org/resources/tools/agtrials formerly https://www.agtrials.org/. These journals and website were selected for their known strength in covering the agricultural and plant nutrition literature. Results of the search were limited to the period 1990 – 2017. The search was then updated in 2020.
Data Collection and Analysis
Description of Methods Used in Primary Research
Included studies are anticipated to employ Zn fertilisation methods for agronomic zinc biofortification which include soil application, foliar spraying, soil + foliar application and seed coating. Soil application involves adding Zn fertilisers like zinc sulfate or zinc oxide directly to the soil to enhance Zn availability for crops. Foliar spraying applies zinc solutions directly onto plant leaves, facilitating rapid Zn absorption, especially in Zn-deficient soils. Seed coating involves treating seeds with zinc-containing compounds before planting to improve Zn uptake during early growth stages. These methods aim to increase grain Zn concentration, thereby addressing Zn deficiency in human diets through biofortified crops.
Selection of Studies
The search was filtered as shown (Figure 1) and outlined below: • Title and abstract: in addition to full title, abstracts of the studies were read to minimise the risk of error. Flow-Chart of Data Search
Titles and abstracts of the articles retrieved were screened independently by two review authors to assess study eligibility as determined by the inclusion criteria. • Full text: full text of eligible studies was read and assessed for relevance.
Full copies of all eligible papers were retrieved. When a title or abstract could not be rejected with certainty, the full text of the article was obtained for further evaluation. If full articles could not be obtained, authors were contacted to obtain further details of the study. Disagreements at any stage of the eligibility assessment process were resolved through discussion and consultation with a third author, where necessary. Details of excluded studies at stage two are listed (Table S5).
Data Extraction and Management
Two review authors extracted data using MS-Excel spreadsheets. The following information was extracted from eligible studies: • study location (country, experimental site and year). • study characteristics (crop species and genotype, initial soil pH and Zn status, form, application rate and timing of Zn fertiliser). • other management practices (NPK fertilisation, use of animal manures, irrigation, pest/disease control).
Means (across sites and genotypes), sample size and standard deviation of the interventions were also extracted for comparison of the effects of soil, foliar and soil + foliar Zn fertilisation on grain Zn concentration and grain yield. Means were extracted on a study-country basis. In the case of a study carried out in multiple years, using various rates and/or sources of zinc were used, we calculated and used the arithmetic average values. The same approach was used to calculate the arithmetic average for multiple site locations and genotypes used as determined by each study. Where measure of dispersion (i.e., standard error, standard deviation, confidence intervals) was not reported, corresponding authors were contacted via email whether they would make that information available. Otherwise, standard deviation of the outcome variable was imputed using a bootstrapping procedure based on runif in the R statistical software environment. Study location (country), initial soil pH and Zn status were used as potential effect modifiers to investigate between-study heterogeneity and inconsistency.
Assessment of Risk of Bias in Included Studies
Eligible studies were assessed in terms of clarity of aims, clarity and appropriateness of methodology, objectivity of outcome measurements, use of controls, and clarity of findings, which were considered and reported in the descriptive analysis. Value judgements categorised as “high”, “low” or “unclear” were supported by a transparent rationale for the judgement. Potential sources of bias were scored using a system indicating high, moderate and low risk of bias, as appropriate. The following categories of bias were assessed (Waddington et al., 2012): confounding and sample selection bias; reporting biases; and other sources of bias. This involved consideration of within-study risk of bias (study limitations), directness of evidence, and heterogeneity and precision of effect estimates. Sources of bias for statistical modelling studies included factors relating to model specification (e.g. source of model coefficients) and methods of inference (e.g. use of systematic sensitivity analysis). No study was excluded based on the critical assessment tool, but the findings were considered during the evidence synthesis. The critical appraisal (Shea et al., 2017) informed the overall strength of the evidence.
Measures of Treatment Effect
Only continuous data were expected, that is, mean concentration of Zn in grain in mg/kg and grain yield in t/ha. Therefore, the effectiveness of the Zn fertilisation methods for increasing grain Zn concentrations was compared by computing mean differences (MDs) of the interventions with their 95% confidence intervals (CIs). As for effectiveness, an MD greater than 1.0 favours the intervention (as compared to the comparator/control). A negative MD favours the comparator. The effect size was subsequently interpreted in relation to a minimum increase of 8 mg/kg from the baseline levels—16 mg/kg for rice and 25 mg/kg for wheat and maize—as established by the Harvest Plus project (Harvest Plus, 2024). These thresholds correspond to a 50% increase in Zn concentration for rice and a 68% increase for wheat and maize.
Direct and Indirect Comparisons of Treatment Effects
Firstly, a pairwise meta-analysis of evidence from studies comparing the same interventions on the same crop species was carried out using a random-effects model. A random-effects model was used only for wheat and rice (Tables S1, S3; Figs. S1-S6), whereas a fixed-effect model was used for maize due to the low number of studies in that species (Table S2; Figs. S7-S9). The analysis primarily concentrated on agronomic outcomes, specifically grain Zn concentration and grain yield. While potential improvements in dietary Zn intake and reductions in Zn deficiency were acknowledged, these aspects were discussed only marginally, as they fall outside the scope of this review.
List of the Excluded Studies
Summary of Network Meta-Analysis for the Effects of Zn Fertilisation Method on Grain Zn Concentration of Wheat
Summary of Network Meta-Analysis for the Effects of Zn Fertilisation Method on Grain Zn Concentration of Maize
Transitivity Across Treatment Comparisons
It is well established that phytate levels and varietal differences in Zn bioavailability can significantly impact nutritional outcomes. However, the included studies did not report data on phytate content or Zn bioavailability. As a result, potential nutritional effect modifiers could not be evaluated.
Two main transitivity assumptions were assessed: • the Zn fertiliser used to compare different methods of application is similar when it appears in different studies; and • the design of a given soil versus foliar Zn fertilisation comparison study is similar to those studies comparing soil to soil + foliar Zn fertilisation.
To this aim, the following potential effect modifiers were taken into consideration: initial soil pH, Zn status and study location.
Criteria for Determination of Independent Findings
To ensure the independence of results in the included studies, a thorough assessment of their design and data sources was conducted. This involved checking for overlapping populations or datasets that could indicate dependence, as well as examining whether multiple studies originate from the same authors or countries and experimental sites, which may suggest correlated results. Additionally, the study timelines were reviewed to determine whether multiple studies stem from the same cohort or data collection period, and identify duplicated data to prevent double counting.
To address potential dependencies, overlapping data were adjusted, either by excluding duplicate reports or averaging them appropriately. Multilevel (hierarchical) models were employed to account for data correlations, while sensitivity analyses were used to evaluate how the inclusion or exclusion of dependent studies influences overall findings.
Dealing With Missing Data
Where measure of dispersion (e.g. standard error, standard deviation, confidence intervals) was not reported, primary study authors were contacted to obtain the data, where possible. Failing this, standard deviation of the outcome variable was imputed using a bootstrapping procedure based on runif in the R statistical software environment.
Assessment of Heterogeneity
Pairwise meta-regressions were conducted to further explore the sources of heterogeneity where appropriate. Between-study heterogeneity was quantified using the I2 statistic. Potential sources of heterogeneity such as genotype variation within crop species, time and number of foliar Zn applications and rate of fertilisation were not investigated. Between-study heterogeneity was also quantified using the Tau2 statistic. Given that the interpretation of this variance is not straightforward, summary effects of the network meta-analysis were presented together with their predictive intervals to facilitate understanding of the results in light of the heterogeneity magnitude.
Assessment of Reporting Biases
We had anticipated utilising funnel plots and regression tests of funnel plot asymmetry to make judgements about whether reporting biases were strongly suspected or not detected. Unfortunately, low numbers of studies and extreme heterogeneity precluded the use of these methods.
Data Synthesis
Summary of Network Meta-Analysis for the Effects of Zn Fertilisation Method on Grain Zn Concentration of Rice
Subgroup Analysis and Investigation of Heterogeneity
Sensitivity Analysis
The study design interaction approach was used as the data provided multi-arm trials with estimates deriving from potentially heterogenous sources. Therefore, direct and indirect evidence were contrasted, and the difference was considered as the inconsistency factor. However, inconsistency within the network itself (despite the closed path of the network) was not explored because the models did not converge with the high number of covariates included in the exploration of heterogeneity. Instead, pairwise meta-regressions using the Akaike Information Criterion (AIC) and/or Deviance Information Criterion (DIC) were performed to avoid overfitting.
Summary of Findings and Assessment of the Certainty of the Evidence
The main findings of the review were set out in summary of findings (SoF) tables incorporating primary outcomes only (grain Zn concentration and grain yield), to explain the significance of findings. The outcome for each comparison was listed with estimates of relative effects contributing data for those outcomes.
Results
Description of Studies
Results of the Search
The literature search returned 628 records. After the exclusion of duplicates, 571 titles were screened, and 210 potentially eligible records were identified. Abstracts were then screened, and 90 potentially eligible records were identified. After full-text review, 66 records were excluded, and the remaining 24 records corresponding to 44 independent studies that fulfilled the inclusion criteria were used in the qualitative and quantitative analyses (Figure 1).
Included Studies
The 44 independent studies were conducted across four continents, 55 site locations and used 105 crop cultivars grown on eight types of soil with pH 4.8-8.8 and Zn status varying from 0.1-6.5 mg/kg of soil. The 44 studies were published between 1999 and 2017, and most studies were carried out in Asia i.e. China (13), Turkey (7), Pakistan (6) and India (4) between 1993 and 2013. There were 27 studies on wheat, 13 on rice and four focused on maize (Table 1).
Wheat studies consisted of common, bread and durum species with 92 different genotypes/cultivars of which Faisalabad-2008, Jinmai47, Bezostaya1 (bread wheat) and Kunduru1149 (durum wheat) were used in at least three different studies. Studies on rice included 13 different genotypes/cultivars of which CNT1, PR120 and GLY No.6 were used in at least two studies. Work on maize included two genotypes/cultivars, with Zhengdan958 used in at least two studies.
All 44 studies assessed the effectiveness of soil, foliar and/or soil + foliar Zn fertilisation for increasing grain Zn concentration. Soil application consisted of 10, 23, 25, 40, and/or 50 kg/ha of ZnSO4.7H2O as a single application to the soil surface but incorporated (15-20 cm depth) into the soil before sowing/planting. Foliar applications consisted of 0.2%, 0.3%, 0.4% and 0.5% w/v of ZnSO4.7H2O at an application rate of 500-1000 L/ha, with 1-3 applications at stem elongation, tillering, booting, flowering, milk, anthesis, panicle initiation and/or heading growth stages. Co-interventions were administered in four studies (Li et al., 2015; Manzeke et al., 2014; Ram et al., 2016; Zhang, Deng et al., 2012), which included insecticides and fungicides, urea and superphosphate, phosphorus PKS blend, ammonium nitrate, cattle manure, woodland litter and urea. For the organic amendments, only cattle manure, and possibly woodland litter contained Zn which may have influenced the response to Zn fertiliser. Basal fertilisation treatments included application of ammonium nitrate, ammonium sulfate, urea, superphosphate and triple superphosphate applied before sowing/planting and/or before flowering based on the common crop management practices for each study.
In the analysis of the 44 independent studies on Zn fertilisation, all included a control treatment that involved no Zn fertilisation. Among these, one study by Manzeke et al. (2014) featured an absolute control group, which did not receive any form of fertiliser or amendment. A set of 20 studies examined the effects of various treatments, including soil, foliar, and combined soil and foliar Zn fertilisation, comparing against the control treatment. This group consisted of 11 studies focusing on wheat, eight on rice, and one on maize. Additionally, 15 studies specifically looked at the impact of foliar Zn fertilisation compared to the control treatment, with 10 studies on wheat and five on rice. There were five studies, two on wheat and three on maize, that evaluated the effects of soil Zn fertilisation against the control group. Lastly, four studies, comprising three on wheat and one on rice, investigated the effects of both soil and foliar or combined soil and foliar Zn fertilisation relative to the control (Figure 2). Network Plots for Direct Comparisons Between Eligible Studies. Numbers on the Lines are the Number of Studies Contributing Data for Each Comparison, Highlighted by the Thickness of the Line
Regarding the outcomes, all 44 studies reported grain Zn concentrations, whereas seven studies (Ajiboye et al., 2015; Boonchuay et al., 2013; Cakmak, 2008; Cakmak, Kalayci, et al., 2010; Imran et al., 2015; Li et al., 2015; Manzeke et al., 2014) did not report the corresponding effects on grain yield. Five studies reported grain phytate or phytic acid concentration (Barrameda-Medina et al., 2017; Gomez-Coronado et al., 2016; Imran et al., 2015; Li et al., 2015; Wei et al., 2012) but none of the studies measured Zn bioavailability in the human digestive tract, except Cakmak et al. (2010) which tested and reported the role of both soil and foliar Zn fertilisation in Zn concentration of whole grain fractions. Only two studies reported Zn concentration in different grain fractions (Ajiboye et al., 2015; Cakmak, Kalayci, et al., 2010)
For the effect on grain Zn concentration, data were available for the following direct comparisons (Figure 2): soil fertilisation versus control (27 studies: 15 for wheat, 8 for rice and 4 for maize), foliar fertilisation versus control (36 studies: 22 for wheat, 13 for rice and 1 for maize), soil + foliar versus control (22 studies: 13 for wheat, 8 for rice and 1 for maize), foliar versus soil fertilisation (21 studies: 12 for wheat, 8 for rice and 1for maize), soil + foliar versus foliar fertilisation (20 studies: 11 for wheat, 8 for rice and 1 for maize) and soil + foliar versus soil fertilisation (21 studies: 12 for wheat, 8 for rice and 1 for maize). For the effect on grain yield, data were available for the following direct comparisons: soil fertilisation versus control (24 studies: 14 for wheat, 7 for rice and 3 for maize), foliar fertilisation versus control (32 studies: 20 for wheat, 11 for rice and 1 for maize), soil + foliar fertilisation versus control (19 studies: 11 for wheat, 7 for rice and 1 for maize), foliar versus soil fertilisation (19 studies: 11 for wheat, 7 for rice and 1 for maize), soil + foliar versus foliar fertilisation (18 studies: 10 for wheat, 7 for rice and 1 for maize) and soil + foliar versus soil fertilisation (19 studies: 11 for wheat, 7 for rice and 1 for maize).
The included studies were relatively similar in terms of weather conditions but differed in terms of crop species/genotype, soil pH and soil Zn concentration. This heterogeneity was taken into consideration when performing meta-analyses and interpreting the findings.
The studies Zou et al. (2012), Ram et al. (2016), and Phattarakul et al. (2012) stood out as particularly influential because their multi-country coverage and multi-site designs tended to yield large within-study weights and broad geographic representativeness. These studies combined soil and foliar Zn applications and encompassed crops such as wheat and rice, amplifying their impact on the pooled estimates for “any Zn application” and for crop-specific outcomes. Their cross-country scope (spanning Asia, Africa, Europe, and the Americas in various experimental designs) and the inclusion of multiple years or sites contribute substantially to cross-context generalisability but may also introduce substantial between-study heterogeneity. As such, their influence was formally evaluated not only through overall effect sizes but also through how their inclusion shapes heterogeneity and subgroup. However, diagnostic checks showed that the direction of the pooled effect did not shift by removing either study except the magnitude of the effect. Likewise, although a weighted meta-analysis was used, studies from China contributed more data points or weight to the overall effect size estimate than the rest of the countries, the direction of the pooled effect did not shift by removing these studies except the magnitude of the effect.
Excluded Studies
Returned records/studies were excluded after abstract screening and full-text evaluation. The two main reasons for exclusion were lack of a full description of Zn fertilisation strategy and/or limited description of the experiment set-up/methodology (Table 2).
Risk of Bias in Included Studies
Overall, the included studies were judged to have a low risk of bias as the quality was considered high from an agricultural research perspective and in causal terms. Nonetheless, the risk of selection and reporting biases was unclear because some studies did not report the methods and findings adequately. Furthermore, although the overall bias is low, the lack of data on Zn bioavailability and human nutritional outcomes represents a significant limitation that weakens the overall strength of the evidence. Additional problems with lack of reporting and small numbers of replicates mean that baseline comparisons cannot be assumed to be equivalent with regard to confounders. Formal assessment of this issue is non-discriminatory in agri-evidence but contributes to considerable uncertainty.
Synthesis of Results
Primary Outcome: Grain Zn Concentration
Network Meta-Analysis
The data from the 44 studies allowed both direct and indirect comparisons of the three strategies for Zn fertilisation aimed at increasing grain Zn concentrations of wheat, rice, and maize. Inconsistencies within the network could not be evaluated due to the multitude of covariates considered in assessing heterogeneity. Nonetheless, the transitivity assumption was met. Thus, the network meta-analysis indicates that the combination of soil and foliar Zn fertilisation significantly increases grain Zn concentrations with observed increases of 23.9 mg/kg [95%-CI: 19.41 to 28.42] for wheat, 7.3 mg/kg [95%-CI: 5.08 to 9.59] for rice, and 8.8 mg/kg [95%-CI: 7.80 to 9.74] for maize. Foliar Zn fertilisation alone increases significantly grain Zn concentrations by 18.0 mg/kg [95%-CI: 14.30 to 21.60] for wheat, 6.6 mg/kg [95%-CI: 4.75 to 8.48] for rice, and 8.5 mg/kg [95%-CI: 7.47 to 9.59] for maize. In contrast, considering the confidence intervals of the pooled effects, soil Zn fertilisation does not significantly increase grain Zn concentrations, with increases of only 3.1 mg/kg [95%-CI: −1.15 to 7.38] for wheat, 2.2 mg/kg [95%-CI: −0.04 to 4.47] for rice, and 4.4 mg/kg [95%-CI: 3.75 to 5.12] for maize (Figures S1-S9).
Wheat
Compared to control treatment, soil Zn fertilisation increases grain Zn concentration of wheat by 4.7 mg/kg [95%-CI: 2.30 to 7.12], foliar Zn fertilisation increases by 18 mg/kg [95%-CI: 14.26 to 21.79] and the combination of soil and foliar Zn fertilisation by 28.7 mg/kg [95%-CI: 18.40 to 32.36]. However, when outlier studies (Cakmak, Kalaycu, et al., 2010; Gomez-Coronado et al., 2016; Phattarakul et al., 2012) are removed from analysis, the increase of grain Zn concentration is 16.7 mg/kg by foliar Zn fertilisation and 22.9 mg/kg by the combination of soil and foliar Zn fertilisation. Regarding the relative effectiveness between the interventions, the combination of soil and foliar Zn fertilisation increases grain Zn concentration 21.0 mg/kg [95%-CI: 15.02 to 26.88] more than soil Zn fertilisation and 4.8 mg/kg [95%-CI: 0.95 to 8.54] than foliar Zn fertilisation. The level of certainty of these findings is moderate due to high inconsistency (I2 > 80%,), with indication that study location, soil pH and Zn concentration do not modify the effect size of the interventions significantly (Table 3).
Rice
The effect of the interventions is small but significant. Compared to control, soil Zn fertilisation increases grain Zn concentrations of rice by 1.6 mg/kg [95%-CI: 0.48 to 2.78], foliar Zn fertilisation increases by 6.7 mg/kg [95%-CI: 3.25 to 10.04] and the combination of soil and foliar Zn fertilisation increases by 6.8 mg/kg [95%-CI: 5.33 to 8.22]. Combined soil and foliar Zn fertilisation increases 5.1 mg/kg [95%-CI: 3.88 to 6.32] more than soil Zn fertilisation and 1.2 mg/kg [95%-CI: 0.76 to 1.73] than foliar Zn fertilisation. The level of certainty of these findings is moderate due to high inconsistency (I2 > 80%,), with indication that study location, soil pH and Zn concentration do not modify the effect size of the interventions significantly (Table 4).
Maize
Grain Zn concentrations of maize increase 4.9 mg/kg [95%-CI: 2.78 to 6.92] by soil Zn fertilisation, and 9.0 mg/kg [95%-CI: 5.24 to 12.80] by foliar Zn fertilisation. The level of confidence for other comparisons is very low because the evidence is derived from a single stud, with mean differences between 0.2 and 7.9 mg/kg The level of certainty of these findings is very low due to reliance on single studies for most results (Table 5).
In terms of effectiveness, the network meta-analysis shows that Zn fertilisation methods differed substantially with respect to increasing grain Zn concentrations in wheat, and thus the methods are ranked as: soil + foliar >foliar >soil. Zinc fertilisation methods do not differ substantially with respect to increasing grain Zn concentrations in rice.
Secondary Outcome: Grain Yield
Summary of Network Meta-Analysis for the Effects of Zn Fertilisation Method on Grain Yield of Wheat, Rice and Maize
Discussion
Summary of Main Results
Soil Zn fertilisation generally increases Zn concentrations of wheat, rice, and maize, with an average of 4.7 mg/kg for wheat, 4.9 mg/kg for maize, and 1.6 mg/kg for rice, which correspond to 18.8% relative increase in Zn concentration for wheat, 19.6% for maize and 10% for rice from the baseline concentration. Eevidence from studies conducted in Portugal, Turkey, and Thailand indicate instances where soil Zn fertilisation leads to a decrease in grain Zn concentrations. On the other side, foliar Zn fertilisation increases grain Zn concentrations of wheat, rice, and maize, resulting in increase of 18 mg/kg for wheat, 6.7 mg/kg for rice, and 9.0 mg/kg for maize, which correspond to an increase of 72.0% for wheat, 41.9% for rice and 36.0% for maize from the baseline concentrations. However, when outlier studies are excluded from the analysis, the increase is slightly lower, highlighting that the effectiveness of foliar Zn fertilisation can significantly differ depending on the country (study location). Furthermore, combined soil and foliar Zn fertilisation significantly increases grain Zn concentrations of the major cereals, with wheat showing the highest increase of 28.7 mg/kg on average, followed by rice at 6.8 mg/kg, which correspond to an increase of 114.8% for wheat and 42.5% for rice, respectively. Maize shows a smaller increase of 7.9 mg/kg, though this evidence is derived from a single study and may not represent a consistent trend regarding maize research.
Foliar Zn fertilisation generally results in a greater increase in grain Zn concentrations compared to soil fertilisation, while the combination of both soil and foliar fertilisation method results in the highest increase (21.0 mg/kg = 84.0% in wheat and 5.1 mg/kg = 31.9% in rice) compared to soil Zn fertilisation alone. In this regard, the most effective Zn fertilisation strategy in increasing grain Zn concentrations of the major cereals is the combination of both soil and foliar Zn fertilisation, followed by foliar fertilisation alone, with soil Zn fertilisation causing the least effect. None of the Zn fertilisation methods significantly affects grain yield of wheat, rice, and maize. Therefore, using foliar and the combination of soil and foliar Zn fertilisation can substantially increase Zn concentrations in cereal grains, potentially improving their nutritional value without negatively impacting crop yields. Thus, effective Zn fertilisation could be a valuable strategy for improving the quality and nutrition of the cereal grains.
This review establishes a significant connection between Zn fertilisation methods and the increase in grain Zn concentrations, particularly for wheat, achieving up to threefold increases compared to control treatments. While rice and maize show only marginal increases of approximately 1.0 mg/kg, this change is more noteworthy for rice due to its lower baseline Zn concentration. Thus, these findings underscore the efficacy of Zn fertilisation in enhancing grain quality of wheat, which is crucial for improving nutritional outcomes.
Among the various Zn fertilisation methods, the combination of soil and foliar application emerges as the most effective approach for increasing grain Zn concentration of wheat. Although the difference in effectiveness compared to foliar application alone is small but statistically significant, the highest effect from the combination of soil and foliar approach suggests it may serve as the best practice for achieving optimal Zn enrichment in wheat grains. However, we cannot generalise these findings across all three crops comprehensively due to the limited data available for maize, thus restricting the robustness of the conclusions only to wheat and rice. Nonetheless, this review reinforces the potential of Zn fertilisation strategies to increase grain Zn concentrations of wheat, thereby addressing malnutrition in populations highly dependent on wheat-based diets. Notably, the lack of significant grain yield response to Zn fertilisation implies that the existing soil conditions may be adequate for crop growth, but not optimal for promoting sufficient Zn uptake necessary for increased grain quality with respect to human nutrition. The review also highlights an important distinction between achieving good crop yields and meeting the specific nutritional needs linked to micronutrient content.
This review reveals a gap in understanding the bioavailability of Zn post-processing as limited data hindered adequate estimation of how Zn fertilisation influences grain Zn concentration after processing, which is also a critical factor for nutritional adequacy. The concern is that a substantial quantity of grain Zn could be lost during milling and polishing processes, particularly for wheat and rice. Most grain Zn is found in the outer layers of grain, which are often removed during processing, potentially compromising the nutritional benefits gained through fertilisation.
Interestingly, this review shows that there is a negligible impact of Zn fertilisation on grain yield across wheat, rice, and maize, regardless of the specific fertilisation method employed. This lack of yield improvement, even in soils classified as low-Zn (below 1.0 mg Zn/kg), defies expectations that such soil condition would result in a direct yield response to Zn intervention. Moreover, 31 of the 44 studies reported pre-intervention soil Zn below the threshold (1.0 mg Zn/kg), suggesting that while Zn fertilisation can enhance grain Zn concentration, its role in influencing overall grain yield presents a more complex scenario that warrants further investigation.
Overall Completeness and Applicability of Evidence
The external validity of the findings of this review can be assessed on several levels, including: (1) This review examines wheat, rice, and maize, which are essential staple crops in many countries and play a vital role in ensuring global food security. The studies included in this review cover a wide range of geographical regions and various local farming practices. The review considers different soil types, climate conditions, and agricultural practices that can greatly affect how well crops absorb Zn and how effective fertilisation Zn strategies are. Since only a limited number of studies have been done in areas with distinct climatic or soil characteristics, questions regarding whether the findings can be applied to other contexts are minimised. (2) The review compares various common methods for applying Zn, including a combination of soil and foliar fertilisation together with foliar and soil fertilisation alone. This comparison helps provide a thorough understanding of different intervention strategies for increasing grain Zn concentrations and grain yield. Hence, all potential fertilisation approaches suitable for various agricultural contexts were explored. (3) While the review does not present findings on the specific varieties of wheat, rice, and maize, or the farming practices used in the 44 independent studies, the findings still fulfil the overall goal of the review. Nonetheless, the findings may have limited applicability to maize crops because few studies met the inclusion criteria, and the included studies focused on a small selection of maize cultivars that do not adequately represent the diversity seen in global agriculture. (4) The outcomes measured were primarily focused on grain Zn concentration and grain yield. While these are critical indicators, the review does not discuss other possible benefits or interactions, such as soil health improvements (if any), cost-benefit analysis of different Zn application methods, or potential impacts on human nutrition beyond Zn concentration in the grain. Including these aspects would provide a more holistic view of the effectiveness of the intervention and broader implications for different settings. (5) Variability in point estimates and changes in effect directions that were not linked to known modifiers like soil pH imply that additional research may be necessary to confirm the findings across various soil conditions.
Quality of the Evidence
The findings suggest that the review generally met its objectives regarding the evaluation of agronomic Zn biofortification strategies to increase grain Zn concentration and grain yield. However, the overall evidence quality was rated as low due to high inconsistency among studies, evidenced by an I2 greater than 90%. This inconsistency suggests that findings might not be universally applicable or reproducible, raising questions about generalizability. Additional uncertainty arises regarding publication and reporting biases. None of the included studies were pre-registered and reporting was generally too poor to allow assessment of baseline comparability. Open science and adoption of reporting standards are long overdue for agri-evidence and other environmental fields.
Potential Biases in the Review Process
Analysis 1.1. Meta-Analysis for Direct Comparisons for Grain Zn Concentrations of Wheat
Analysis 1.1. Meta-Analysis for Direct Comparisons for Grain Zn Concentrations of Rice
Analysis 1.1. Meta-Analysis for Direct Comparisons for Grain Zn Concentrations of Maize
Therefore, conducting a more exhaustive literature search that includes grey literature, and unpublished studies could help mitigate the risk of publication bias and provide a fuller picture of the evidence on Zn fertilization. Future reviews could benefit from standardized methodologies in included studies to reduce variability and improve comparability. This could involve prescriptive guidelines for study design when studying Zn fertilization effects; Utilising advanced statistical techniques to investigate heterogeneity and explore the influence of specific effect modifiers could provide additional insights that clarify why certain outcomes differ between studies. Including a wider range of agronomic and environmental outcomes could provide a more holistic understanding of the implications of Zn fertilization practices beyond just grain Zn concentration and yield.
Authors’ Conclusions
Implications for Practice and Policy
Implications for Research
Supporting research initiatives aimed at addressing inconsistencies in Zn application responses, while exploring the biological mechanisms underlying crop-specific responses to Zn—such as soil pH and Zn status—is essential. Long-term studies are necessary to assess the sustainability of Zn fertilisation practices and their effects on soil health and crop yields over time. Furthermore, conducting reviews across diverse geographical locations and environmental conditions will enhance understanding of the factors influencing Zn uptake in various crops. Lastly, improving study designs is vital to reduce variability and ensure reliable findings that can effectively inform agricultural practices and policies.
Soil + foliar and foliar Zn fertilisation increases grain Zn concentration in wheat but not in rice in terms of absolute amount of increase. This is indicative of lower efficiency for Zn uptake and remobilisation following foliar Zn fertilisation in rice. Therefore, there is the need for: • characterising wild relatives or testing more rice genotypes for Zn uptake and remobilisation. • estimation of grain Zn bioavailability in the human digestive system. • further analysis of sufficient empirical evidence of fertilisation strategies to identify efficiencies and weaknesses particularly with respect to rice and maize.
Broader implications for agri and environmental evidence relate to the immaturity of these fields as evidence-informed disciplines. Adoption of open science practices-particularly pre-registration would greatly improve the utility of the evidence. Reporting guidelines and adoption of core-common outcomes could also improve standards making research more useful and useable.
Supplemental Material
Supplemental Material - Agronomic Biofortification Strategies to Increase Grain Zinc Concentrations of Wheat, Rice and Maize: A Systematic Review and Network-Meta-Analysis
Supplemental Material for Agronomic Biofortification Strategies to Increase Grain Zinc Concentrations of Wheat, Rice and Maize: A Systematic Review and Network-Meta-Analysis by Israel F. N. Domingos, Marcin Baranski, Zed Rengel, Paul Bilsborrow, Gavin Stewart in Campbell Systematic Reviews
Footnotes
Acknowledgements
The authors thank INAGBE (Instituto Nacional de Gestão de Bolsas de Estudos – Angola).
Author Contributions
Funding
The authors received no financial support for the research, authorship, and/or publication of this article. But received scholarship for the PhD program.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Zed Rengel has conducted primary research and literature reviews on zinc biofortification and is Editor-in-Chief of Crop & Pasture Science Journal.
Gavin Stewart is an Associate Editor of Peer J, Research synthesis methods, and multiple Cochrane and Campbell entities.
Paul Bilsborrow has conducted primary research and reviews on agronomic management of cereal crops and has been a Senior Editor of the European Journal of Agricultural Science.
Israel F. N. Domingos received financial support from the National Institute for the Management of Scholarships of the Ministry of Higher Education, Science, Technology, and Innovation of the Republic of Angola. However, declares that there are no personal or financial conflicts of interest that may have influenced the conduct of the work reported in this review or the interpretation of the results generated from it.
Plans for Updating This Review
First author (Israel F. N. Domingos) will be responsible for updating the review if new evidence is available every five years.
Differences Between Protocol and Review
There are no differences between protocol and review.
Data and Analytic Code
Data and analytic code will be made available upon request to primary review author.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
