1. Introduction
Concerns regarding ecological and public health effects from the ubiquity of microplastics (MPs) have grown in the last decade. They have been detected in oceans [
1], surface water [
2], marine sediments [
3], marine organisms [
4], and generally throughout the marine environment [
5]. MPs have been proven to be a vector for major pollutants in the marine environment [
6]. This ability raises questions about the food safety of marine organisms. With the increase in the production and improper discharge of plastic waste, there is a growing need to understand the full range of MP pollution in the marine environment.
Thompson et al. [
7] first reported MPs in marine organisms and defined them as “plastic particles smaller than 5 mm found in marine ecosystems” [
7]. This research became a milestone in marine science in the following years as it addressed the issue of the accumulation of MP particles in the world’s oceans. Since 2004, the study of MPs has significantly increased, with the scientific community now focusing not only on their presence in the oceans but also more generally investigating the sources, transportation, and the impact of MPs on the environment as well as on human health.
MPs were first reported in marine organisms in the blubber of baleen whales [
8]. However, the ingestion of MP particles by marine benthic organisms was officially confirmed by the experimental study of [
9]. The study of MPs regarding fish and generally edible marine species expanded spectacularly as they are a very important food source for humans [
10].
Marine organisms are passively exposed to MP pollution, and actively as well, as they can easily ingest MPs mistaken for food and end up in the internal organs of the fish [
11]. MPs can also be consumed from marine organisms, by the consumption of already exposed organisms of lower trophic levels. Once ingested, they can accumulate in the gastrointestinal tract, leading to internal disruption [
12], feeding behavior disorders, as well as exposure to harmful chemicals adsorbed on the surface of MPs [
13]. These impacts are particularly concerning regarding edible marine species, which are part of the human diet, as they could be transferred through the food chain and finally affect human health.
Small pelagic and benthic fish, some of which are important commercial species with increased economic and social value, can be used as indicators for understanding the extent of environmental pollution and the potential health risks. These species include sardines and anchovies [
14], whose ecological role in the coastal ecosystems is very important as they transport energy from lower to higher trophic levels [
15]. As a result, studying MPs in edible marine organisms has become a primary topic in marine ecology, marine biology, and public health research.
The Mediterranean Sea is a semi-enclosed sea basin and one of the most polluted seas in the world, at least in relation to MP pollution [
16]. The high concentrations of MPs are probably due to their physical and oceanographic characteristics [
17]. The region’s highly populated coastal zone, intense marine traffic, and human activities such as fishing, shipping, industrial activity, and tourism contribute to making it particularly vulnerable [
18].
The Mediterranean Sea is known for its rich biodiversity [
19], and many species are considered to be highly important for both ecological and economic purposes. Many of the Mediterranean fish species are an important source of food, which demands an extensive investigation into the topic of MP pollution. The confirmed presence of MPs in this region poses a threat to marine species and their food chains, and consequently to human health [
20], making a better understanding of the extent of pollution and its impact on different water bodies a necessity.
To better understand the extent and variability of MP pollution in the Mediterranean Sea; this systematic review focuses on ecologically and economically important fish species. These species slightly differ in their ecology, feeding behavior and ecological niche, providing a range of exposure for these organisms to MPs, under different environmental conditions. These species are also important in local fisheries and in mediterranean human diet, which makes an ideal candidate for the assessment of the potential risks caused by MPs.
More specifically,
Engraulis encrasicolus (Linnaeus, 1758), a pelagic-neritic species [
21] of the coastal zone, occurs in depths up to 400 m and can survive in salinities ranging from 5 to 41 PSU [
22]. In general, but especially during the reproductive period, it can even enter lakes, lagoons, and river estuaries.
Sardina pilchardus (Walbaum, 1792), also a pelagic-neritic species, is found in coastal areas related to low salinity waters [
23], such as fresh, brackish, and marine waters [
21]. It has been recorded at depths from 10 to 100 m [
24], and its diet is based mainly on zooplankton and phytoplankton organisms [
25]. Reproduction usually takes place either in the open sea or close to the shore [
26,
27].
Boops boops (Linnaeus, 1758) is a benthic and semi-pelagic species of the Mediterranean, mainly found at depths from 0 to 350 m [
28]. The type of substrate this species inhabits varies, as it has been observed on sand, silt, and gravel substrates, as well as in seagrass meadows [
29]. Its diet is based on plankton organisms and small crustaceans [
30].
Another pelagic fish of the Mediterranean is
Sardinella aurita (Valenciennes, 1847). It has a short life duration and its diet is based on plankton organisms [
31]. This species is mainly found in the warm waters of the Mediterranean [
32]. Despite the similarities in its diet compared to other small pelagic fish,
S. aurita is hierarchically higher in the food chain [
33].
Lastly,
Trachurus mediterraneus (Steindachner, 1868) is a semi-pelagic, medium-sized carnivorous fish found throughout the Mediterranean Sea [
34], widely distributed in pelagic food chains [
35]. It usually inhabits shallow coastal areas [
36] and feeds mainly on crustaceans and fish larvae [
37].
These five fish species are all significant to human consumption, with a crucial role in the Mediterranean diet, well known for its health benefits. They are key species for the Mediterranean fisheries, providing economic and employment support for coastal communities. Thus, it is worth investigating other studies to better understand the effects of MPs in fish tissues.
A standard problem in reviewing studies is that the sample sizes vary significantly. What is the minimum sample size that needs to be analyzed in a study to detect MPs in fish, to ensure the results are reliable and accurate? Large sample sizes can waste human and material resources, but very small sample sizes can lead to inconclusive or misleading results. Establishing a minimum sample size should increase the reliability and efficiency of studies and reduce their cost [
38].
The aims of this systematic review were to (a) examine the time needed for the completion of a study from sampling to publishing an article for MPs in fish tissues, (b) investigate the difference in the concentrations of MPs consumed by the five commercial fish species mentioned above in different Mediterranean water bodies, (c) suggest the minimum number of replicated measurements needed in a survey to be able to detect differences between groups, and (d) evaluate whether the different methodologies used in every study could influence the reported MP concentrations.
2. Materials and Methods
2.1. Literature Search
The present systematic review, conducted under the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [
39] in October 2024, aimed to collect peer-reviewed research articles in three different databases: Web of Science, Scopus, and Science Direct. The search strategy was to find research articles that focus on the concentration of MPs ingested by five different fish species in eight different Mediterranean water bodies [
40] and in three different tissues.
More specifically, the species examined were E. encrasicolus, B. boops, S. pilchardus, S. aurita, and T. mediterraneus. Similarly, the Mediterranean regions under study were categorized into the following water bodies: the Alboran Sea, the Balearic Sea, the Ligurian Sea, the Tyrrhenian Sea, the Adriatic Sea, the Ionian Sea, the Aegean Sea, and the Levantine Sea. The tissues examined were the gastrointestinal tract compartment (GIT), the gills (GIL), and the muscle tissue (MU).
In all three search engines, the time period was set from 1970 (the minimum date of Web of Science) to October 1, 2024, when all records were retrieved. Any additional article was published within this time period. A total of 120 searches was performed, 40 on each search engine. The terms used in the search engines for E. encrasicolus were as follows:
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Alboran Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Balearic Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Ligurian Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Tyrrhenian Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Adriatic Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Ionian Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Aegean Sea”);
(“Microplastics” AND “Fish” AND “Engraulis encrasicolus” AND “Levantine Sea”).
Similar searches were carried out for the rest of the species under study.
2.2. Systematic Review Protocol
Using the commercial reference management software package EndNote 20, the systematic review was structured into four stages based on the PRISMA guidelines (
Figure 1). The first stage included the “Identification of records”, in which all the collected articles were imported into EndNote and duplicates were removed.
During the second stage, which was “Screening of abstracts”, all remaining articles from the first stage were screened regarding the content of their title and abstract. Therefore, articles whose screening indicated that they referred to different species or water bodies than the ones under study, as well as articles that were systematic reviews or meta-analyses, were removed. Articles whose content was not clear on any of the above were reviewed in the third stage. At this stage, information including the authors’ names, year of publication, type of publication, and DOI was collected.
In the third stage, “Eligibility criteria”, the remaining articles from the second stage were reevaluated in their full text coverage.
At this stage, the information collected included the following: (a) the species under study, (b) the water body under study, (c) the number of examined individuals, (d) the specific tissue under study, (e) the concentration of MPs in “particles per individual” units, (f) the standard deviation, (g) the inclusion or not of a chemical characterization technique, and (h) the aims of each study in relation to different categories.
Once more, articles that referred to different species or water bodies from those investigated in the present study, as well as articles that were systematic reviews or meta-analyses, were removed from the whole dataset. In addition, after examining the full text, articles that did not include the necessary information for the systematic review were removed. In the fourth and final stage, the articles included in the qualitative synthesis were separated from those in the quantitative synthesis. In order to determine the eligibility for inclusion in the review, the screening of the records was performed by a single reviewer, without the use of any automation tools.
All included publications [
41,
42,
43,
44,
45,
46,
47,
48,
49,
50,
51,
52,
53,
54,
55,
56,
57,
58,
59,
60,
61,
62,
63,
64,
65,
66,
67,
68,
69,
70,
71,
72,
73,
74,
75,
76,
77,
78,
79,
80,
81,
82] and all related data were organized into a final database (
Table 1) and visualized using the ArcGIS Pro 3.0.3 program. The coordinates of the sampling locations were either extracted directly from the articles as “longitude” and “latitude”, or were digitalized through the maps included in each article under study.
The visualization of the sampling sites’ data in ArcGIS showed that they were not evenly distributed across different Mediterranean water bodies, which would create a problem in the subsequent statistical analyses. To address this, the eight water bodies were further distributed into six new water bodies (W, SW, CW, CE, NE, SE) to obtain a more even distribution (
Figure 2). The visualization also showed that in many studies, the reported sampling sites belonged to a different water body than the one mentioned. This apparent difference in the spatial distribution of the sampling sites could also pose a problem for the statistical analysis. In order to prevent these problems, the sampling sites were categorized based on the water bodies created.
Where “W” includes the Balearic Sea, “SW” the Alboran Sea, and “CW” the Tyrrhenian Sea and Ligurian Sea. The “CE” waterbody includes the Adriatic and Ionian Sea, “NE” the Aegean Sea, and “SE” the Levantine Sea.
2.3. Data Analysis
In order to study the concentration of MPs among the fish species and the Mediterranean water bodies, the collected data were used to extract qualitative and quantitative information. The concentration of MPs for each species and water body was recorded as units “Items per individual”. Additionally, for this systematic review, the standard deviation (SD) was calculated; therefore, all records with SE were converted to SD via the following equation [
83]:
where “SD” refers to the standard deviation, “SE” to the standard error, and “Sample size” to the size of the sample, i.e., the number of individuals under study.
Where an article contained concentration data for more than one species or more than one site, it was recorded as a stand-alone record for each species and site. The data analysis was divided into two separate parts. First, an analysis of the qualitative data was carried out, followed by the analysis of the quantitative data.
During the qualitative part, studies were categorized according to the data included in each one, presenting the concentration of MPs in the (i) studied tissues, (ii) studied fish species, and (iii) studied water bodies, and the percentage of occurrence in the papers pool was calculated. In addition, the aims of each study, the dates of sampling, and the dates of publication for each study were recorded. By subtracting the date of sampling from the date of publication, the time taken for each record, from the beginning to the end, was determined. The average time taken to publish the data was then calculated for each year of publication.
Regarding the quantitative data, the average and the standard deviation (SD) concentrations of MPs were calculated for all fish species within each Mediterranean water body, based on the reported values of the reviewed studies. A two-way design was used to assess whether there are statistically significant differences (
p-value < 0.05) between the recorded concentrations of MP particles across the studied species in different Mediterranean water bodies, where data permitted. PRIMER-E v.7 PERMANOVA+ software was employed to perform the non-parametric permutation multivariate analysis of variance (PERMANOVA) [
84,
85] using untransformed data based on Euclidean distances, with Type III sums of squares, permutation of residuals under a reduced model, and 999 permutations. The levels of the ‘Species’ factor were set at four (BOO, ENC, PIL, MED) while the levels of the ‘Water body’ factor were set at six (W, SW, CW, CE, NE, SE). A crossed design was used, where the factor ‘Species’ was chosen as a fixed factor and the ‘Water body’ as a random factor. Pairwise PERMANOVA tests were conducted on the statistically significant results (
p-value < 0.05) to identify the differences between all pairs of the studied factors, using PRIMER-E v.7 PERMANOVA+.
To examine the effects of different analytical protocols on MP concentration variability, two separate PERMANOVA analyses were conducted, based on the Euclidean distances, with Type III sums of squares, permutation of residuals under a reduced model, and 999 permutations. At first, “Filter pore size” was defined as a fixed factor with three levels (0.2 μm to <1 μm, 1 μm to ≤2 μm and >2 μm) to examine whether it could contribute to differences in the reported MP concentrations across studies. For this analysis, only the filter pore size used in relation to the final MP extraction step reported was considered. However, many of the studies did not provide sufficient data, and it was not possible to determine the filter pore size used. It should also be noted that the filter pore does not define the detection limit, since the sensitivity of the analysis methods is usually lower than the filter pore size used. As a result, the available data were insufficient to draw meaningful conclusions regarding the effect of this factor.
Secondly, in order to investigate if the chemicals used during the sample digestion step could cause statistically significant variability (p < 0.05), the “Chemicals” was defined as a fixed factor with four levels based on the specific chemical used (KOH, H2O2, Both, None). Pairwise tests were also performed on the statistically significant results (p < 0.05) in order to identify differences between different factor levels.
In order to estimate and propose a recommendation for the minimum number of replicated measurements (samples) needed in a survey, a “Power Analysis” estimation was performed. The methodology was based on the results of the PERMANOVA analysis comparing regions and species. More specifically, the aim was to calculate the number of samples needed for a survey to be able to detect the effect size differences found, based on the pairwise comparisons that showed statistically significant differences (
p-value < 0.05). Therefore, in the cases where significant differences were found between regions, the effect sizes were calculated using Cohen’s d [
86], on an online effect-size calculator [
87], and categorized into small (d = 0.2), medium (d = 0.5), and large (d ≥ 0.8) [
88]. The analysis was performed using G*Power 3.1 software [
89], with the selections “F test, ANOVA: Fixed effects, main effects and interactions” and type of analysis “A priori: Compute required sample size” given a power and effect size. The settings inputted in G*Power used for analysis were as follows: (i) Type a error probability α = 0.05, (ii) (1 − Type b) error probability (power = 0.8), (iii) degree of freedom calculated as: df = (water bodies − 1) × (species − 1) − 1 = (6 − 1) × (4 − 1) − 1 = 14, (iv) number of groups calculated as all the possible combinations of species × regions (groups = 4, regions = 6, i.e., 4 × 6 = 24), and (v) the effect size “f” provided by the PERMANOVA results. Finally, the total sample size (i.e., number of fish measurements) obtained from the above analysis was divided by the number of groups to determine the number of replicated samples required per group.
4. Discussion
This systematic review provides useful information about the distribution of MPs in the GIT of five commercial fish species across different Mediterranean water bodies. Through this study, a holistic overview of the widespread distribution of MPs in the Mediterranean Sea was provided. From further analysis and visualization of the data, it is worth mentioning that the western Mediterranean has been studied more extensively than the eastern basin, which leaves critical knowledge gaps for understanding MP pollution throughout the Mediterranean Sea. Therefore, this systematic review highlights the need for further analysis of the eastern Mediterranean region, at least regarding the five species under study.
4.1. Distribution of MP Research in Fish
The fact that only two studies of
S. aurita were found indicates a serious lack of data regarding this species. Therefore, the findings of the present study imply that there is a need for further investigation of this species, as it is a commercial species of importance and a source of food mostly for the North African Mediterranean countries [
93].
Regarding studies on different organs and tissues, this systematic review identified only two studies involving the gills, and no studies involving the muscle tissue, despite the fact that the latter is the one directly consumed by humans. The reviewed studies do not provide evidence for food safety risks from MPs, as none of them investigated MU tissue. However, future research focusing on the MP contamination of this tissue could be crucial for evaluating the potential implications for human consumption.
Regarding the aims of the studies, the majority of those focused on the quantification of MPs, implying that during this early stage of MP research, there was a priority towards identifying simple distribution patterns rather than qualitative differences or involved processes. Concerning the different aims of the studies examined, the focus of the aims on the assessment, characterization, and identification of possible sources of MPs may reflect concerns over their presence in the marine environment and the potential ecological risks and health implications (as shown in
Figure 5). It should be noted that the included studies of the systematic review referred only to the GI of fish and not to the MU. As a result, no implications can be made regarding potential risks for human consumption or direct health effects.
However, the fact that there are fewer studies focusing on the development of bioindicators, the feeding habits of the marine organisms, and food safety issues highlights a gap in the understanding of the long-term impacts of MPs. The increased number of studies included in the “Other” category demonstrates the multidimensional nature of MPs, as well as the new and understudied research areas. These findings reflect the need for more research into the less studied aims to understand the impacts of MPs across the overall research range.
4.2. Time Between Sampling and Publication
The intense research into MPs in 2020 and 2022 indicates the increasing interest in the topic. The rapid decrease in 2021, with only one recorded publication, is probably related to the coronavirus pandemic (COVID-19), which caused global disruption and dysfunction, affecting research activity in universities and research centers. For the years 2016 to 2018, publications maintained a stable and relatively low level of interest in the topic. This may be due to the fact that the study of MPs in marine organisms is a rather recent topic, whose research is still ongoing.
The increase in the required time, from sampling to publication, observed until 2019, reflects the growing complexity of the different research methods applied. The peak in 2022 is probably related to publications whose sampling was conducted at an earlier time. From 2022 to 2024, there is a decrease in scientific publications, which may reflect the negative effects of the pandemic on research activity. However, there are more publications after 2022 compared to the period 2016–2018. The decrease in the time needed in 2023–2024 may reflect methodological improvements and lower levels of uncertainty in the scientific protocols used, although this cannot be directly confirmed from the data of this systematic review. The absence of data for the year 2021 underlines the need for both complete and accurate data recording in publications, and highlights the need for continuous improvements of methods for MP investigation.
This provides information on the time needed for a scientific work to be completed regarding the concentration of MPs in commercially important fish species. The review also raises concerns about possible delays in environmental risk assessment and public awareness. In some cases, samples were published over a decade after collection, reflecting how extended preparation times can reduce the relevance and immediacy of the results. Environmental conditions can change significantly throughout this time, reducing the applicability of these results to current conditions. By understanding the time needed for the completion of a study, from sampling to publication, it can be useful for the development of new policies, as well as for the update of the current marine monitoring programs, which are based on up-to-date data. Over time, there has been an improvement in the chemical characterization techniques used to study MPs, indicating the need for more comprehensive and reliable methods. Between 2016 and 2018, the use of physical characterization methods in MP studies was more common, which may be due to the limited development of chemical characterization methods. From 2019 to date, the significant increase in the use of chemical characterization techniques indicates their importance in terms of polymer identification.
The increasing use of chemical characterization techniques for MPs is evident from 2022 and beyond, as studies that rely exclusively on physical characterization decrease. Therefore, this trend highlights a methodological development aiming at improving the reliability and accuracy of the results, and at a more detailed identification of MPs, including the determination of their exact composition. Chemical characterization techniques are useful as they minimize the number of falsely identified MP particles, and allow the polymer identification, which is necessary to understand the sources of MP pollution.
Furthermore, the increase in the use of chemical characterization techniques during the same period as the increase in the time required to complete a study likely reflects methodological complexity. These techniques provide lower detection limits and more accurate polymer identification, which improves the reliability of the reported data. The adoption of quality assurance practices such as contamination controls and recovery tests can further contribute to the enhancement of the robustness and comparability of the existing literature, apart from the concentrations of the MPs found.
4.3. Comparison Between Fish Species and Water Bodies
The interaction between the factors ‘Species’ and ‘Waterbodies’, as derived from PERMANOVA, was not found to be statistically significant (p-value > 0.05). On the contrary, the fact that there are statistically significant differences between the different water bodies (p-value < 0.05) implies that this factor is responsible for the variability observed in the MP concentrations, although it cannot be excluded that methodological differences (e.g., variation in the detection size limits) may also contribute to this finding. Regarding the factor ‘Species’, the results indicated that it does not seem to play a significant role in the concentration of MPs. The statistically significant differences found between the pairs of water bodies (W-CW, W-SE, and SW-SE) (p-value < 0.05), as obtained from the pairwise tests, are probably related to the environmental factors of the geographical areas involved, possibly including water circulation but also anthropogenic aspects such as the discharge and management of plastic wastes.
For other pairs of water bodies, no significant differences were found, at least under the levels of sampling effort reported. The multiple pressures that a coastal ecosystem is exposed to can affect the amount of microplastics ingested by its organisms. This can be partly related to heterogeneity. As [
61] demonstrated, the anthropization factor has an important role in microplastic ingestion. Therefore, the comparison of the MP content of the different water bodies is more difficult as the heterogeneity of methodological factors, especially the lower detection size limitations, increases. Further research on the geographic distribution of MP concentrations in fish is likely to reveal the environmental mechanisms involved in the uptake and retention of MPs.
In a closer look at the water bodies that are statistically significant (
p-value < 0.05), the Balearic Sea is located in the western part of the Mediterranean Sea. It is an open area with strong water circulation coming from the Gibraltar Strait. The coastal zone of the Balearic Sea is moderately urbanized [
94]; therefore, the exposure of the area to MP pollution through the terrestrial environment is expected to be of medium intensity.
The Tyrrhenian Sea, which on the northern part is connected to the Ligurian Sea, is located in the central part of the Mediterranean Sea. It is an enclosed area with low water circulation. This waterbody has a high exposure to MP pollution, as it is located close to areas of high urbanization [
94], such as Rome, and areas of high industrial and maritime activity, such as Naples and the wider area of Palermo. Consequently, this waterbody (CW) is expected to be contaminated with higher concentrations of MPs than the Balearic (W) waterbody.
The Levantine Sea covers the southeastern part of the Mediterranean Sea (SE waterbody) and, despite being a region characterized by high urbanization [
94], there is limited research on MP pollution [
68]. This region is exposed to intense anthropogenic pressures resulting from harbors, intense navigation and fishing activities, the untreated urban and industrial wastewater treatment discharges, and the unregulated dumping sites along the coast [
68]. This probably explains the statistically significant difference in concentrations between the water bodies of the Balearic and the Levantine Sea (
p-value < 0.05).
The Alboran Sea covers the southwestern part of the Mediterranean. It is characterized by intense water circulation as the waters are renewed by the Atlantic Ocean through the Gibraltar Strait. The coastal area is characterized by moderate urbanization and neighbors’ industrial areas, such as Spain and the northwestern part of Africa. However, concentrations of MPs are low, probably because the incoming waters through the Atlantic Ocean have the capacity to transport pollutants away from this area. Thus, the statistically significant difference between the SW and SE water bodies (p-value < 0.05) can be attributed to the different water circulation between the two regions.
4.4. Influence of Different Methodologies on Reported MP Concentrations
Regarding the use of filters with different pore sizes, the results showed that there were no statistically significant differences (p > 0.05), and based on the available data of the studies, the decision over the filter pore size does not appear to affect the results of MP concentrations detected. However, this finding should be carefully evaluated, since many studies did not clearly report the filter pore size used in every step of the process, or even in general. It should be mentioned that in most cases, the detection limits of the applied methods were higher than the filter pore size used. Therefore, the filter pore size cannot be used as an indicator of the detection limit, and this is considered a limitation of the present analysis. The limited information regarding the protocol details of each study, such as in this case, the lack of data for the filters used in all studies, may cause difficulties or wrong estimations in meta-analysis and systematic reviews. This underlines the need for protocol harmonization in future studies.
On the other hand, the statistically significant differences (p < 0.05) found between the studies using different chemicals underline the impact of chemical selection on the quantification of MP in the fish tissues. The highest MP concentrations were observed in studies that used chemicals, compared to those that did not. However, this result may be potentially misleading, as studies using chemical digestion are usually designed to target smaller and numerous particle sizes that are probably the actual factor responsible for the higher MP concentrations observed. This implies that the use of chemicals could increase the efficiency of MP extraction through a more efficient organic material digestion. However, the absence of statistically significant differences between both chemical categories (KOH, H2O2, or Both) suggests that while the use of chemicals generally improves the MP detection, the type and number of chemicals used is of minor importance.
These results indicate that the chemical digestion methodology used in each study might lead to potential variability in reported MP concentrations. According to
Table S2, it was found that 86.3% of studies did not perform a recovery analysis, 65.3% of studies did not report the size of the filter used, 48.4% of studies did not report the minimum sample size (or the detection limit), 37.1% did not perform a chemical characterization technique, and 10.5% did not perform a physical characterization technique. The findings of this study are once again in agreement with the scientific literature regarding the lack of methodological harmonization in MP studies and support the necessity for a standardized protocol. The adoption of harmonized protocols would not only improve the comparability between studies but also enhance the reliability of systematic reviews and meta-analyses, as well as risk assessments related to MP contamination in marine edible organisms.
4.5. Minimum Recommended Sample Size
The power analysis conducted in this study provides important information for future research programs in the field of MP concentration in marine organisms. Based on the findings, the number of samples required for a survey to produce statistically valid results on the concentration of MPs in different fish species, as derived from the two-way ANOVA: fixed effect and interactions, are a minimum of six samples.
However, in the present review, the data varied with respect to the number of samples. More specifically, the minimum number of samples in a published study was found to be two, while the highest number was 180. This means that the results of studies with fewer than six samples may be of limited accuracy.
Through the use of this methodology, statistical sufficiency can be ensured in the results of a study. The application of this method is recommended in large-scale surveys, where there may be a large number of samples or study areas, to reduce the time and cost of the study and maintain the reliability of the data.
It is worth noting that the recommended sample size of this systematic review does not conflict with the existing literature, such as the 50 individuals proposed by OSPAR [
95]. The difference lies in the study design, as OSPAR’s guidelines are based on surveys with fewer study areas (N = 3) and species (N = 1), while the power analysis presented in this study showed that the required number of individuals decreases as the number of species and study areas increases. However, the use of 50 individuals for MP detection for every species and every study area is often not applicable. According to the findings of this review (
Table 1), only 28.2% of the published studies included more than 50 individuals, which highlights the difficulty of using this protocol universally, since this number of fish may not be caught, and even if it does, the process will substantially increase time, cost, and labor required.
By establishing a recommended minimum sample size, the development of a more methodologically consistent and statistically accurate MP research can be supported. With the inclusion of this analysis, the present study not only synthesizes current findings, but also contributes to the improvement of future study designs for MP environmental research.
Nevertheless, it should be emphasized that the derived results presented are exploratory and have important limitations, since the dataset used for the effect sizes was based on heterogeneous studies with variation in methods, detection limits, and incomplete report of quality assurance and quality control.
5. Conclusions
The present study demonstrates the extensive presence of MPs in four of the five species studied in different Mediterranean water bodies and provides a comprehensive overview of the extent of the studies in the Mediterranean region. In particular, this systematic review highlights the need for further investigation of MP concentrations in pelagic fish from the eastern part of the Mediterranean, which appears to be understudied compared to the central and western Mediterranean Sea.
The scientific community has recognized the need to harmonize methodologies in the field of MP research. Harmonized approaches provide accuracy in the assessment of MP ingestion and improve comparability between studies, both in sub-regional and regional scales [
50]. The ability to compare different studies is essential to provide a reliable estimation of MP levels in the marine environment across wider geographical areas. In addition, the lack of information regarding the protocol used, as revealed by a large number of studies—for example, the pore size of the used filters, or the lower size detection limit for visual identification—can lead to the limited comparability of data. Protocol standardization would be highly beneficial, but is still difficult to achieve. For these reasons, this study suggests that harmonization of protocols should be prioritized in order to enhance the reliability and applicability of monitoring efforts. Thus, harmonization should be pursued by using key methodological characteristics, such as lower size detection limits, recovery analysis, filter pore size, and the chemicals used.
The fact that there was not a sufficient number of studies for S. aurita (two studies only), and S. aurita was therefore excluded from the statistical analyses, underlines the need for further research on major commercial fish species in the Mediterranean.
Missing data over MPs were also identified for the different tissue types. Most of the studies concern the gastrointestinal tract of fish, very few studies concern the MP concentration in gills, and there are no studies of MP in muscle tissue. Therefore, our results highlight the need for multi-tissue studies, particularly for MU tissue, which is directly consumed by humans either in raw or cooked form.
Statistical analyses showed that the different water bodies are a significant factor affecting the concentration of MPs in fish, but the difference among fish species is not significant. The eastward increase in concentrations of MPs may be due to the environmental characteristics of local seas, water circulation patterns, differences in management of plastic debris wastes in coastal areas, as well as in the catchment basins of major rivers, population densities, or a combination of some or all of these. More focused studies could shed light on the dominant factors affecting this unequal distribution.
Concerning the minimum number of samples required for a survey in order to obtain reliable results, the present systematic review has shown that this number is at least six. We expect that the adoption of this minimum number will help ensure the reliability of the results of a study while reducing the required time and cost.
Overall, the findings from this study can provide useful information for the investigation of MPs in commercial Mediterranean fish species. Finally, it highlights the need for further research in understudied areas, species, and processes, so the impact of the extent of MP pollution in the Mediterranean marine area can be better understood.