Systematic Review and Meta-Analysis of Melatonin Quantification in Wine

Abstract

The identification of melatonin in grapes has led to the publication of numerous studies on melatonin in wines, and the aim of this study was to perform a systematic review and meta-analysis of published data on melatonin concentrations in wines. In this context, international databases such as Scopus, Web of Science and PubMed were searched for relevant articles (437) up to 29 March 2025. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used. A total of 15 studies from eight countries, involving various wine types and analytical methods, were included in the meta-analysis. Considerable analytical variation was observed across studies, and high-performance liquid chromatography (HPLC) coupled with either mass spectrometry (MS) or fluorescence (FL) detection was shown to be the most accurate and sensitive method for quantifying melatonin. The highest concentrations were found in Spanish red Tempranillo wine, Romanian white Noah wine, and Romanian rosé Lidia wine. Red wines, particularly those produced from Cabernet Sauvignon (CS) grapes, were the most frequently studied. The results of this work provide a clearer picture of melatonin levels in wine. Further research is needed to explore the implications of melatonin content in wine for human health and the wine industry.

1. Introduction

Melatonin is an essential indoleamine hormone produced in the pineal gland involved in modulating several physiological processes [1]. For example, it suppresses prooxidant enzymes, stimulates the activity of antioxidant enzymes, reduces radical formation and metal-induced toxicity due to improving mitochondrial function [2], and can be a powerful direct free radical scavenger [3], twice more effective than vitamin E [4].

It was considered a characteristic compound of vertebrates [5], but in recent decades, it has been found in a variety of medicinal and food plants [6]. Melatonin is present in drinks, including wine [7,8,9], as one of its dietary components [10], even if its acceptable daily intake is very difficult to estimate. Because melatonin can cross physiological barriers [11], the intake of drinks containing melatonin increases its concentration in plasma and the number of melatonin-derived metabolites [12]. After the identification of melatonin in grapes [13], studies were carried out to quantify it in wines [14]. Melatonin is naturally present in wines because it is contained in wine grapes. Also, in some stages of the winemaking process, melatonin is biosynthesized by yeasts from its precursors, tryptophan and serotonin [15,16,17], mainly during alcoholic fermentation [18]. The final amount can be affected by environmental conditions during vine cultivation [19].

The presence of the melatonin in wines offers an exciting perspective related to moderate red wine consumption associated with potential health benefits, strengthening the premise that health benefits of the Mediterranean diet are due in part to wine [20]. The implication of melatonin as a potential health benefit is firstly due to its endogenous antioxidant capacity. Likewise, its additive and synergistic effects with other natural antioxidants in the winemaking process, such as resveratrol [21], can lead to the enhancement of polyphenolic compounds [22]. However, studies are still in the early stages [23], and it is not fully understood how melatonin changes the profile of phenolic compounds in wines.

The low amounts of melatonin in wines necessitate complex analytical techniques for their quantification. Many proper methods have been developed, such as the voltammetry of immobilized particles methodology [24]; the capillary electrochromatography (CEC) method [25]; the enzyme-linked immunosorbent assay (ELISA) method [26]; and the methods of HPLC with different detectors, diode array detection (DAD) [27], FL [28,29,30], and tandem MS [31,32]. HPLC is most affordable and suitable for assaying melatonin in wines since it enables unambiguous identification, it has a large capacity for discrimination and performs one of the most precise analyses. Because wine represents a complex matrix and melatonin concentrations are in the ng mL⁻¹ range, these methods require the pretreatment of sample. Several extraction and concentration methods were evaluated, such as solid-phase extraction (SPE) [14,29,33], dispersive liquid–liquid micro-extraction (DLLME) [27,30], and microextraction by packed sorbent (MEPS) [28].

Although melatonin can modify the phenolic profile and antioxidant activity via phenylpropanoid metabolism, only one systematic review has been reported on the melatonin content of foods in the Mediterranean diet, such as tomatoes, olive oil, red wine, beer, nuts, and vegetables [34]. It highlighted that the consumption of these foods could have an impact on antioxidant status and that components of food matrix may play a role in the effect of melatonin intake. Taking into consideration the potential health benefits of melatonin with regard to regular moderate wine consumption, several reviews have focused on the presence of melatonin in wine [7,20,35], but to date, no systematic review and meta-analysis regarding melatonin quantification in wine has been performed. Hence, the aim of this study was to conduct a systematic review and meta-analysis of published data on melatonin concentrations in wines from different countries, varieties of wine, and analytical methods. This work seeks to consolidate existing findings and provide a clearer picture of melatonin levels in wine.

2. Materials and Methods

This systematic review was conducted in accordance with the PRISMA guidelines [36], and the PRISMA statement checklist was used to evaluate eligible reports, covering various sections of the review process. It is not registered in any international systematic review registries, as it is not a systematic review of studies involving humans as participants or a systematic review of studies involving animals in research that is directly relevant to human health.

2.1. Search Strategy

We conducted the present systematic review by searching papers (articles, conference papers, book chapter, reviews, and so on) on different amounts and quantitative analyses of melatonin in wine from inception to 29 March 2025. International databases, including Scopus, Web of Science, and PubMed, were used to retrieve the relevant articles. The first screening was performed using Article title, Abstract, Keywords (Scopus) or All Fields (Web of Science and Pub Med), without filters. The systematic search was carried out using the following terms in all databases: “melatonin” and “wine” and Boolean operators (AND).

2.2. Inclusion/Exclusion Criteria and Data Extraction

Our primary purpose was to review every single paper for new evidence related to melatonin concentration in wine. The results of the searches were filtered by the English language criterium and document type. Because information published in one original article can also be found in a different document, e.g., review, conference paper, etc., it is crucial to meticulously de-duplicate the data. To avoid introducing duplicate information into the meta-analyses, the exclusion criteria were review, book chapter, conference paper, short survey, note, letter, and article commentary. The title and abstract of papers were evaluated and categorized with the terms “retain” or “exclude”, and relevant documents were downloaded by one investigator (G.L.R.). Articles eligible for assessment were those involving measurements of melatonin concentrations in wine samples. The references lists of the collected articles were also screened manually to find additional articles. Then all retained full texts were read by two investigators (S.A.V.E. and C.A.) to assess each article’s suitability for inclusion based on the pre-specified inclusion and exclusion criteria. We excluded all articles that did not provide the average melatonin concentration of the studied wine in a numerical manner. Afterward, the data from each study, including country, method of detection, sample pretreatment, variety of wine (red, white/sparkling, rosé or dessert), type of grapes (e.g., Malbec, Feteasca Neagra (FN), CS), the concentration of melatonin in wines and the standard deviation error, were extracted independently using a data collection form by all investigators. Any disagreements were resolved through discussion between investigators.

2.3. Statistical Analyses

Meta-analyses were performed on all 15 the studies that reported melatonin concentrations in wines, excluding studies in which concentrations of melatonin isomers were presented [37,38,39].

An online statistical tool, METAANALYSISONLINE, supported by ELIXIR Hungary (https://metaanalysisonline.com/, (accessed on 7 April 2025)), was used to perform meta-analyses and to generate forest plot layouts using Cochrane’s RevMan (https://revman.cochrane.org/info). As a forest plot is the most efficient tool for presenting heterogeneity and pooled results, we chose to generate only forest plots instead of funnel plots, and included Egger’s test and omitted potential publication error for readers. Meta-analyses were calculated and expressed as means of melatonin concentrations with standard deviations and the number of wine samples.

Studies were weighted to reflect their relative contribution to the overall summary effect, and the method used was inverse variance weighting. A random effects model was utilized to determine the most appropriate model for meta-analysis and panel data analysis, taking the variability between different studies into consideration. To estimate the between-study heterogeneity, the DerSimonian and Laird method was used.

Heterogeneity in meta-analysis refers to the variations in different study outcomes. The classical measure of heterogeneity is Cochran’s Q test.

The p-value associated with heterogeneity tests indicates the probability of observing the given level of heterogeneity if the true effect is the same across all studies. A low p-value (typically less than 0.05) suggests statistically significant heterogeneity, meaning there is evidence of variability in the true effects across studies beyond what would be expected by chance. High p-values suggest homogeneity or a lack of heterogeneity.

The I², also known as Cochran’s Q statistic, is a statistical measure that quantifies the degree of heterogeneity or inconsistency between the results of different studies. I² was calculated for each meta-analysis. If I² is more than 50%, then the heterogeneity is considerable, and if I² is lower than 50%, heterogeneity is insignificant.

The random effects model with the inverse variance method was used to obtain the summarized raw means (MRAW) and the confidence interval, a measure of dispersion. The number of eligible studies was limited, and it may be difficult to assign reasonable values to the bias parameters and priors for quantitative bias analysis methods. The review authors were required to search all possible sources for study reports and results. Due to the fact that only 15 non-randomized records were included in the meta-analyses, it was difficult to accurately evaluate whether the results were influenced by publication bias.

3. Results and Discussion

3.1. Selection of Studies

A flow chart of the literature search process is detailed in Figure 1. A total of 437 initial articles were identified through database searching, i.e., Pub Med (93), Scopus (159), and Web of Science (185). Duplicates were removed, leaving a set of 225 records that were reviewed, and 28 records were found to be eligible. Finally, 10 records were excluded for not meeting the selection criteria, and 18 records were included in review [8,9,14,17,25,26,27,28,29,30,32,37,38,39,40,41,42,43].

The details of these studies are reported in

Table 1. Studies on the concentration of melatonin and its isomers from diverse wines.

Method	Sampling Method	Country	Red Wine	Conc. ng mL−1	White Wine	Conc. ng mL−1	Rose/Dessert Wine	Conc. ng mL−1	Ref.
HPLC-FL	SPE	Italy	Sangiovese	0.5	Trebbiano	0.4	-		[14]
HPLC-FL	MEPS	Italy	-		Albana	0.6	-		[28]
HPLC-FL	DLLME	Portugal	Syrah	4.29	Multivarietal	0.63	-		[8]
			Alicante Bouschet	1.92	Moscatel Graúdo	3.93
			Touriga Nacional	2.71	-
			Castelão	4.48
			CS	3.06
			Touriga Franca	1.05
			Syrah	2.61
			CS	2.85
			Touriga Nacional	2.82
			Castelão	7.44
			Aragonez	4.27
			Trincadeira	1.90
HPLC-FL	DLLME	Romania	FN	0.74	-		-		[30,41]
			CS	1.09
			FN	0.84
			CS	1.36
HPLC-MS/MS	only filtered	Italy	Groppello	6.3	-		-		[42]
			Merlot	4.4
HPLC-MS/MS	only filtered	Argentine	Malbec	7.47	-		-		[9]
				5.88
				5.80
				6.15
				4.67
				4.15
				4.81
				6.56
				6.11
				9.67
				9.46
				9.68
				7.81
				7.33
				10.49
				8.34
				12.53
				12.51
				4.55
				7.80
				12.39
				10.89
				7.62
				5.93
				3.30
				3.56
		Brazil	Tempranillo, Tannat, Alicante, CS, Egiodola Syrah, Petit Syrah	12.08	-		-
				4.97
				19.89
				7.40
				4.15
				8.06
				8.06
				3.51
				2.74
UPLC-TSQ Access Max-MS	only filtered	Romania	-		Aligoté and Fetească albă	0.22	-		[40]
						0.65
					Sauvignon Blanc	7.81
						0.37
HPLC-MS/MS	concentration	Spain	CS	14.2	-				[26]
			Petit Verdot	5.1
			Prieto Picudo	49
			Syrah	86.5
			Tempranillo	129.5
HPLC-MS/MS	SPE	Italy	Groppello	0.35	Chaudelune	0.18	Recioto di Soave	0.14	[43]
			Melag	0.62
			Nebbiolo	0.14			Santelmo	0.18
			Terre di Rubinoro	0.17			Passito di Pantelleria	0.31
			Syrah	0.23			Marsala	0.11
			Placido Rizzotto	0.05			Moscato di Pantelleria	0.29
			La Segreta	0.31
UHPLC/ESI-QTRAP	SPE	Italy	Nebbiolo	0.06	-				[29]
				0.06
				0.06
				0.04
				0.05
				0.05
				0.06
				0.04
HPLC-QTOF-MS	SPE	China	Marselan	5.84	-				[17]
				6.02
HPLC-MS/MS	SPE	Romania	CS	17.7	Sparkling	19.6	Lidia	30.8	[32]
			Merlot and Pinot Noir	23.0	Riesling	11.8	CS	8.4
			Merlot	18.0
			CS and Merlot	11.6	Riesling	17.0	Lidia	82.6
			Merlot	2.0			Lidia	32.6
			Babeasca neagra	3.8	Noah	35.4	Riesling and Chasselas	2.0
			Seibel 1	10
			Isabelle and Babeasca neagra	11.6	Riesling and Feteasca	1.0	CS	5.8
			Seibel 1	1
			Othello	66.6	CS	12.5	CS	1.4
			CS	4.2
CEC	LE	Argentine	Malbec	0.24	Chardonnay	0.16	-		[25]
			CS	0.32
ELISA	SPE	Spain	CS	0.23					[26]
			Jaen Tinto	0.16
			Merlot	0.21
			Palomino Negro	0.28
			Petit Verdot	0.22
			Prieto Picudo	0.19
			Syrah	0.22
			Tempranillo	0.14
HPLC-DAD	US-DLLME	Argentine	Malbec	3.02	-				[27]
			Malbec	3.70
			Malbec	4.95
			Malbec	3.54
		Brazil	CS	2.23
			Tannat	3.54
			Alicante	4.56
HPLC-MS/MS	US	Argentine	Malbec	3.3
			Malbec	4.1
			Malbec	4.8
			Malbec	3.6
		Brazil	CS	2.74
			Tannat	3.51
			Alicante	4.7
HPLC-MS/MS	concentration	Spain	CS	74.13 *	Palomino Fino	390.82 *	-		[39]
			Merlot	241.22 *
			Tempranillo	77.72 *
			Tintilla de Rota	322.68 *
			Merlot	245.46 *
			Syrah	423.01 *
			Tempranillo	306.86 *
HPLC-MS/MS	-	Turkey	-	170.7 *	-		-		[37]
HPLC-MS/MS	SPE	Argentine	Tannat	151.74 *	-		-		[38]
			Merlot	211.28 *
			Malbec	145.26 *
			CS	185.09 *
			Malbec	60.16 *

* melatonin isomer.

. The melatonin quantification techniques and sampling methods, the country of research, the type of wine, the melatonin concentration determined, and the references were extracted from articles.

3.2. The Concentration of Melatonin in Wines from Different Countries

The published studies were performed on wines from eight different countries, namely Italy (n = 5), Romania (n = 4), Argentina (n = 4), Brazil (n = 2), Spain (n = 2), Portugal (n = 1), China (n = 1), and Turkey (n = 1). Among these countries, the highest concentrations of melatonin were found in Spanish red Tempranillo wine (129.5 ng mL⁻¹), Romanian white Noah wine (35.4 ng mL⁻¹), and Romanian rosé Lidia wine (82.6 ng mL⁻¹).

For countries with more than one study published, a meta-analysis of the results was performed to obtain MRAW and confidence intervals for each country.

Figure 2, a forest plot layout, created using Cochrane’s RevMan [44], shows the meta-analysis results of five studies from Italy, in which a total of 26 wine samples were analyzed. Based on the analysis performed using random effects model with an inverse variance method, the MRAW was 0.53, with a 95% confidence interval of [0.36; 0.7]. Significant heterogeneity was detected (p < 0.01), suggesting inconsistent effects in magnitude and/or direction. The I² value indicates that 94% of the variability among studies arises from heterogeneity rather than random chance.

The results obtained from the Romanian studies, in which 36 wine samples were analyzed, are presented in Figure 3. The MRAW calculated was 1.19, with a 95% confidence interval of [0.46; 1.92], and significant heterogeneity was detected (p < 0.01; I² = 83%).

In Figure 4, the meta-analysis results of 37 Argentinean sample wines are shown. An MRAW of 3.84, with a 95% confidence interval of [0.26; 7.43], was obtained, and the heterogeneity was considerable (p < 0.01; I² = 99%). Figure 5 displays the results obtained from two Brazilian studies, in which 15 samples wines were analyzed. The MRAW was 5.36, with a 95% confidence interval of [1.18; 9.54]. The values of the degrees of heterogeneity were p = 0.02 and I² = 82%.

The presence of a 0.53–1.19 MRAW range in studies from European countries and a 3.84–5.36 MRAW range in South America studies can be explained by the fact that the endogenous and exogenous factors may influence the biosynthesis and accumulation of melatonin in grapevines. The climate and environmental conditions during vine cultivation and agricultural practices influence the final melatonin concentrations in wines [45].

The variations in melatonin amounts in different wines from the same country could be attributed to the changes that occurred during the aging process of the wines. The stability of the active principles is influenced by the age of the wines [32], and not all eligible studies included in the meta-analysis provided the age of the wines.

Another possible factor that leads to the considerable heterogeneity is the fact that winemaking practices differ significantly across the globe, influenced by historical traditions, climate, and grape varieties. In Europe, the focus is on terroir and traditional methods with strict regulations, while New World regions like South America embrace experimentation and technology. For example, Saccharomyces cerevisiae yeast is the primary microorganism responsible for alcoholic fermentation in wine production. However, phenotypic differences within the same species reveal metabolic differences that lead to the diversity of compounds in wines [46], subsequently also leading to variations in melatonin concentration.

3.3. The Methodology of Melatonin Quantification in Wines

The ranked order of methods used for quantification in studies included in the review is as follows: HPLC-MS/MS (60%) > HPLC-FL (25%) > HPLC-DAD (5%) > CEC (5%) > ELISA (5%). Different extraction and cleanup procedures were used in the sample preparation step, and the overall rank order is as follows: SPE (35%) > DLLME (25%) > just filtered (20%) > concentration (10%) > MEPS (5%) > liquid extraction (LE) (5%).

Two major challenges are generally present in melatonin quantification: basal melatonin levels and the desired sensitivity. HPLC is the most commonly used method under different conditions with an isocratic reversed mobile phase and several types of detectors. The DAD detector is useful in different matrices, but in the analysis of wine samples, it was not frequently used due to detection/quantification limits and the concentration level of this analyte. The detection limit (LoD) and the quantification limit (LoQ) of the HPLC-DAD technique are in the µg mL⁻¹ domain. For Brazilian and Argentinian wine sample analysis, prior to determination, an ultrasound-dispersive liquid–liquid microextraction (US-DLLME) method for the preconcentration of melatonin was developed, optimized, and validated [27].

HPLC-MS/MS is a powerful and robust analytical technique for precise and sensitive measurement, even at low concentrations with a shorter run time. Due to its capability of selectively detecting and quantifying active principles in complex matrices, it is often used in melatonin research in wine samples. Of all the developed methods for the analysis of melatonin in wines, HPLC-MS/MS methods are the most widely used. The application of HPLC-MS/MS in melatonin quantification has enabled tremendous quantitative capabilities in the MRM mode with high sensitivity, specificity, and selectivity [47]. One of the many advantages of HPLC-MS/MS is its ability to determine melatonin isomers, which show cytoprotective and antioxidant activity depending on the changing position of two side chains in the indole ring [48]. Although melatonin and its identified isomer had the same fragment ions, melatonin’s most stable abundant fragment is 174. The most abundant fragment for the isomer found in wine is 216, and the ratios of these ions were different for melatonin and melatonin isomers. The relative abundance of the fragment ion 216 was higher than 174 for melatonin isomers [37,38,39]. Another advantage is that the wine samples can be analyzed without the involvement of sample pretreatments, which can only be filtered [9,40,42].

HPLC-FL is a simple, fast and sensitive detection technique. In addition to the most common HPLC-MS methods, considering melatonin has strong native fluorescence, a HPLC-FL method was applied for analysis of food with simple matrix [49]. The LoD and LoQ of this technique are in pg mL⁻¹ domain and are suitable for the determination of melatonin in wines [8,14,28,30,41].

Compared with HPLC, several unique characteristics of CEC, such as minimal reagent consumption, high resolving power, low cost, and rapid analyses, make it particularly attractive for the analysis of compounds in wine [50]. As a result, the high electro-chromatography robustness, high resolution, and high sensitivity, accompanied by the straightforwardness and reproducibility of the extraction procedure, results in a useful tool for determination of melatonin in wine samples [25]. Today, techniques used to detect and quantify melatonin include immunological assays, such as ELISA. The results obtained using this method are not considered accurate, since the various compounds present in grapes can cross-react with antibodies and enzymes, overestimating the real values of melatonin [51]. The main drawbacks of the ELISA test are a lack of linearity, false positive and false negative results, and the relatively high price of the kit. In an attempt to determine melatonin in wines using this test, a 25.0% false positive rate and a 28.6% false negative rate were obtained. While the ELISA kit could not be validated for accurate melatonin quantification in wines [26], it could still be useful for screening if validated for matrix effects.

Differences in the methodology employed for analyzing melatonin in wine can indeed lead to variations in results. Variances in sample preparation/extraction techniques, quantification techniques, and detection methods may influence the reported levels of melatonin in different studies.

Sample pretreatment, purification, separation, and concentration usually require more than 60% of the analysis time and mainly determine the accuracy of the analysis. It is difficult to use appropriate sampling methods to achieve satisfactory recovery. A variety of pretreatment methods have been used for purifying and extracting melatonin in wine samples. Their characteristics and superiorities are worth considering in order to achieve a level of recovery that is appropriate for analytical methods. LE is a general extraction technology known for its simple operation and ease-of-use. Among the different pretreatment methods for melatonin, SPE has replaced LE due to its advantages of extraction efficiency, timesaving capability, low levels of consumption, user-friendly control, and so on. To further achieve miniaturization and high-throughput development requirements, numerous technologies like microextraction technologies such as MEPS and DLLME were also used. These methods have excellent extraction efficiency that can achieve a-hundred-times-higher enrichment and are quicker and less expensive than standard SPE [52].

A variety of pretreatment and analytical methods used for identifying melatonin in wines are summarized in

Table 2. Methods that have been used in published research for melatonin quantification.

Sample Pretreatment	Cartridge/Sorbent	Elution/Extraction Solvent	Analytical Method	LoD and LoQ ng mL−1	References
SPE	C8 cartridges	methanol	HPLC-MS/MS	0.12 and 0.42	[38]
SPE	Bond Elut C18	methanol	HPLC-MS/MS	0.6	[32]
SPE	Strata X-Polymeric Reversed Phase	methanol	HPLC-MS/MS	2.3 and 18	[29]
SPE	ProElut C18	methanol	HPLC-MS/MS	-	[17]
SPE	HLB Oasys	methanol	HPLC-MS/MS	-	[43]
SPE	C18	methanol	ELISA	-	[26]
SPE	BondElut C18	methanol	HPLC-FL	0.03 and 0.01	[14]
MEPS	C8	methanol	HPLC-FL	0.05 and 0.02	[28]
DLLME	-	chloroform	HPLC-FL	0.01 and 0.05	[41]
DLLME		chloroform	HPLC-FL	0.01 and 0.05	[30]
DLLME	-	chloroform	HPLC-FL	0.07 and 0.24	[8]
US-DLLME	-	dichloromethane	HPLC-DAD	230 and 700	[27]
only filtered	-	-	HPLC-MS/MS	0.059 and 0.12	[40]
only filtered	-	-	HPLC-MS/MS	-	[9]
only filtered	-	-	HPLC-MS/MS	-	[42]
LE	-	methanol	CEC	0.01 and 0.03	[25]
concentration	3:1 (v:v)	methanol	HPLC-MS/MS	-	[39]
-	-	-	HPLC-MS/MS	0.033 and 0.112	[37]

.

The results of meta-analyses of studies, in which the HPLC-MS or HPLC-FL methods were used to quantify melatonin in wines, are reported in

Table 3. The meta-analyses of melatonin concentrations in wines quantified using different HPLC methods (ng mL⁻¹).

HPLC-MS
HPLC-FL

.

The meta-analysis was performed on the results obtained from a total of 100 wine samples quantified with HPLC-MS across nine studies. The highest total samples used in one study was 35 [9]. The MRAW was 4.3, with a 95% confidence interval of [2.59; 6]. It should be noted that the lowest mean level of melatonin concentration in wines was 0.05 ng mL⁻¹ [29], confirming once again the capabilities of this technique. The highest mean level of melatonin concentration was 56.86 ng mL⁻¹ [26]. Significant heterogeneity was detected (p < 0.01 and I² = 100%).

In five studies, for the analysis of 25 wine samples, the analytical method used was HPLC-FL. The smallest total number of samples used in a study was one [28]. The MRAW obtained was 1.21, with a 95% confidence interval of [0.64; 1.78] and the heterogeneity was considerable (p < 0.01; I² = 95%).

The significant heterogeneities detected were not due to the analysis techniques but were influenced by grape type, winemaking practice, and country. For example, in the study carried out by Carneiro [27], the determination of melatonin in wine samples was performed using HPLC-DAD after an DLLME ultrasound procedure and, also, with a comparative method, i.e., HPLC-MS/MS. The t-test showed that there was no significant difference between the results found using the HPLC-DAD method and the HPLC-MS/MS method.

These meta-analyses results indicate that SPE and DLLME are highly efficient sample pretreatment and sensitive detectors, and like MS or FL, are considered suitable for the accurate detection of melatonin in wines.

3.4. The Concentration of Melatonin According to the Type of Wine

The presence of melatonin in wines also depends on the variety of wine, i.e., red, white, rosé, or dessert, and the rank order of type, based on number of the conducted studies, is as follows: red wines (66.6%) > white/sparkling wines (26.6%) > rosé/dessert (6.6%).

Figure 6 shows the meta-analysis results for studies on red wines. It can be seen that the wine type most used to identify melatonin was red wine. A total of 115 wine samples were analyzed, and the largest number of samples used in a single study was 35 [9]. In only one study was a single sample analyzed [14]. The MRAW was 2.83 with a 95% confidence interval of [1.87; 3.79]. The highest concentration mean, 22 ng mL⁻¹, was obtained in 2011 [26]. In 2019, the lowest melatonin concentrations were found in the 0.05–0.06 ng mL⁻¹ range [29]. The values of the degree of heterogeneity were p < 0.01 and I² = 100%.

These results confirm, once again, that the final amounts of melatonin from red wines are dependent on the microorganisms involved in the winemaking process, the grape variety, geographic regions of vineyards, and the aging of wines [32].

Figure 7 presented the results obtained from the meta-analysis of 16 white wine samples. It should be noted that white wines are relatively underexplored in melatonin studies (six studies), and there were two studies in which only one sample was analyzed [14,28]. The widest confidence interval obtained was [7.13; 25.30] [32], and the MRAW was 2.94, with a 95% confidence interval of [−0.49; 6.36]. The p-value obtained was <0.01, and the I² was 80%.

Figure 8 shows the results of studies on rose/dessert wines, in which a total of 12 samples were analyzed. Only two studies on the melatonin quantification in this type of wine have been published [32,43]. The mean melatonin concentration determined was of the order of ng mL⁻¹. The highest concentration, 82.6 ng mL⁻¹, was quantified in wine obtained from Lidia grapes [32], a grape variety very popular in Eastern Europe, which produces a semi-sweet wine with a pleasant strawberry aroma. The MRAW was 9.16, with a 95% confidence interval of [−12.95; 31.27]. Significant heterogeneity was detected (p = 0.04 and I² = 77%).

As previously highlighted, the most studied wine from the point of view of melatonin content was red wine, probably also for the possible synergistic effects with polyphenolic compounds, especially anthocyanins, which lead to the increase of the antioxidant activity of these wines [23]. The most of the red wine samples analyzed in the published studies were obtained from Malbec (29.6%), CS (13.6%), Nebbiolo (7.2%), and Syrah (6.4%) grapes. CS, the wine produced from the most widely cultivated grape variety in the world, is found in studies published in both Europe and South America, while Nebbiolo and Syrah wines were used in research from Mediterranean countries (Spain, Italy, Portugal) [8,26,29,39,43]. It was observed that wines specific to a certain area were also studied, such as Malbec wine, a red wine grape that grows mostly in Argentina [9,27], or FN, an old grape variety that grows mainly in Romania [30,41]. Melatonin quantification is also observed in a relatively new wine variety, Marselan [17]. White and rosé wines have been analyzed less than red wines and almost only in Europe [32,39,40,43], with one exception—a single study from Argentina [25].

The red wines predominantly used in the published studies are Malbec and CS and the results of their meta-analyses are presented in

Table 4. The meta-analyses of melatonin concentration results in Malbec and Cabernet Sauvignon wines (ng mL⁻¹).

Malbec
Cabernet Sauvignon

. Based on the analysis performed on 35 red wines-Malbec, the MRAW was 5.67, with a 95% confidence interval of [2.1; 9.23]. The highest mean concentration was 7.52 ng mL⁻¹, and similar concentrations were determined in 34 samples from two different studies with HPLC methods, with the only exception being a sample quantified using the CEC technique [25]. The values of the degree of heterogeneity were p < 0.01 and I² = 97%.

In a meta-analysis performed on 14 CS samples from seven diverse studies, the MRAW is 2.14 with a 95% confidence interval of [1.15; 3.14]. A similarity was observed in the results obtained, and the mean concentration was in the order 1–2 ng mL⁻¹, except for a 7.21 value that was obtained through analysis with two different techniques [26]. When, during the vinification process, wines were produced from grapes grown in vineyards for family consumption and not for sale, the yeasts used in the fermentation stage were natural ambient yeasts from grapes or from the air and not all parameters were able to be controlled, and thus the mean concentration was 11.17 ng mL⁻¹ [32]. The heterogeneity was considerable (p < 0.01 and I² = 97%).

Foods containing melatonin are becoming increasingly popular and are beginning to be considered promising nutraceuticals as it has been demonstrated that melatonin consumed through food products is absorbed, enters the circulation, and could have physiological effects through receptor-mediated or non-receptor-mediated processes. Melatonin supplementation has been shown to have positive effects on different pathologies, such as cancer [53], neurodegenerative diseases [54], and sleep problems [55]. Melatonin is a molecule that can provide a multitude of health benefits, but its importance is undervalued due to its difficulty in determination and the problems of estimating its total intake from a regular diet [56]. The exact concentrations of melatonin in different foods and the effects on health through the intake of foods rich in melatonin are not known. For this reason, it is necessary to evaluate the melatonin content in wines in order to estimate its total intake in a regular diet and to better explore its potential impact on health.

4. Conclusions

This systematic review and meta-analysis compiled and evaluated data on melatonin concentrations in wine published before 29 March 2025. To our knowledge, this is the first study to systematically quantify melatonin content in wine using meta-analytic methods. A total of 18 studies from eight countries, involving various wine types and analytical methods, were included in the analysis. Our results suggest that the average melatonin concentration in wine is around 8.09 ng/mL. The highest concentrations were found in Spanish red Tempranillo wine (129.5 ng/mL), Romanian white Noah wine (35.4 ng/mL), and Romanian rosé Lidia wine (82.6 ng/mL). Red wines, particularly those produced from Malbec and Cabernet Sauvignon grapes, were the most frequently studied. Considerable analytical variation was observed across studies, with solid-phase extraction and dispersive liquid–liquid micro-extraction being identified as the most effective pretreatment techniques. High-performance liquid chromatography coupled with either mass spectrometry or fluorescence detection was shown to be the most accurate and sensitive method for quantifying melatonin. However, our review also identified important limitations. These include significant heterogeneity in analytical methods, inconsistent reporting of wine age and origin, and the small sample sizes of many studies. Furthermore, environmental factors such as climate, agricultural practices, and vinification processes appear to influence melatonin levels, making direct comparisons difficult. Despite these limitations, our findings provide a valuable baseline for future studies and suggest that moderate wine consumption may contribute to dietary melatonin intake. Melatonin consumed through food products has positive effects on different pathologies. Further standardized research is needed to explore the implications of melatonin content in wine for human health and the wine industry.