Abstract

The development of deformities in farmed fish is largely the result of abiotic, biotic, and xenobiotic factors, information deficiencies in optimizing nutrition, and the genetic background to which the fish are exposed in their early life stages. In general, skeletal anomalies are considered to have significant adverse effects on animal welfare, biological performance of farmed fish, product quality, and production costs. In the data obtained by the meta-analysis method, the presence of negative effects on the formal structures of fish was found, regardless of the region, duration, stage, factor, stock density, and method used to detect deformation. In this regard, in the studies considered within the deformation region/type, 46% of deformities were found in the spine, 37% in the head, and 16% in the total skeleton. In turn, the results of the meta-analysis showed that the percentages of the apparent value were 35.82% in the spine, 33.12% in the skeleton, and 31.06% in the head. The deformation rate had an overall negative effect on the functional characteristics of the fish, regardless of the variables considered. In addition, all statistically significant individual response variables had a negative effect size. In the future, advanced statistical tools such as Bayesian meta-analysis, network meta-analysis, and meta-regression analysis can be used to explore more complex data structures. The rapid development of artificial intelligence techniques will increase the efficiency of data collection and the robustness of results for meta-analysis studies in aquaculture and other fields.

1. Introduction

Marine fish farming is now an industry that has become an important source of economic profit, employment, and livelihood in many countries. Due to the high economic value of the species produced and their suitability for intensive production, their market share is growing daily. In this context, marine fish farming accounts for 7.3 million tons of the 30.8 million tons of seafood produced worldwide [1]. With this intensive production, problems related to production have also arisen. The goal of good production is to maintain the characteristics of the morphologically natural forms of farmed fish in terms of quality and quantity. However, even with the best breeding techniques, some morphological disorders can occur in the farmed species. Permanent differences in shape (deformities), which are readily apparent to the consumer, usually occur in the early embryological and postlarval stages of production [2]. Production defects that occur at certain times in breeding facilities also cause undesirable deformations. Theoretically, developmental defects cause phenotypic differences in individuals who genetically exhibit the same traits under the same environmental conditions. As the increase in phenotypic changes negatively affects the feeding and swimming activities of the fish, the stress factor also comes into play [3]. However, it is not yet clear whether individuals with deformities in their embryonic and postembryonic stages exhibit the same characteristics throughout their lives. Although the causes of vertebral deformities in juvenile and adult fish are considered to be notochord deformities that occur during the larval period, the study of fish ontogeny development is understood to cause skeletal deformities in alternate biological processes. The development of deformities in reared fish is largely the result of abiotic, biotic, and xenobiotic influences, information deficits in diet optimization, and the genetic background to which fish are exposed in their early life stages [4]. Abiotic factors, such as photoperiod, light intensity, dissolved oxygen, and carbon dioxide, and high water current [511], such as nutritional (nutritional imbalance) deficiencies (lysine, tryptophan, phenylalanine, vitamin A (VA), C, D, E, K, n-3 highly unsaturated fatty acid, phospholipid, phosphorus, manganese, and zinc) swimbladder problems, stock densities, manipulations, parasites, bacterial, and viral infections, have an intense effect on deformations [6, 1219].

The protruding abdominal areas, relatively small head structure, and scale deformities observed in fish in the first years of production took on other shapes as the production level increased and the number of deformed individuals increased. In addition, some chemicals are known to cause deformations in fish larvae [2024]. Contaminants mixed with water such as dioxins, polychlorinated biphenyls, and tributyltin can negatively affect the mineralization process and cause the weakening of the skeletal structure [2529]. Inevitably, the genetic background [3034] can be shown as the main causes of these deformations. The protruding abdominal regions, relatively small head structure, and scale deformities previously observed in fish took other forms as production levels increased and the number of deformed individuals increased. In addition, inappropriate net structures caused the formation of irregular scratches on the scales and skin of the fish, which became apparent as excessive mucus production and desquamation during the marketing phase [7, 35, 36]. In addition, problems with poor coloration (the color of the fish deviates from its natural color) have increased. Subsequently, the increase in production capacity and survivability over the years has led to various deformation structures and related production problems. In this context, skeletal deformities are defined as one of the most important biological problems in contemporary production. Skeletal deformities are generally observed in the axial skeleton, the jaw and operculum, the fin carrier, and the complex structures of the tail bone [7, 17, 3739]. The most common types of skeletal deformations of the axial skeletal structure under aquaculture conditions are lordosis, kyphosis, and scoliosis [15, 17, 4042]. These deformation types are intensively studied in red sea bream (Pagrus major), gilthead sea bream (Sparus aurata), European sea bass (Dicentrarchus labrax), Japanese flounder (Paralichthys olivaceus), Atlantic salmon (Salmo salar), rainbow trout (Oncohrynchus mykiss), Senegal sole (Solea senegalensis), Atlantic cod (Gudas morhua), Yellowtail (Seriola sp.), and grouper (Epinephelus sp.) farmed in different regions of the world [37, 4249]. Sometimes very different types of deformation can be detected at the same time in the same individual. These types of deformation can be detected by staining the bone cartilage, observation under the microscope, X-rays, and by the naked eye. Skeletal development is important for external morphology and functional mobility during juvenile rearing, and skeletal deformities increase production costs [50, 51]. Movement difficulties, poor growth, high mortality, and diseases in deformed fish cause great economic losses [36, 5154]. It has been noted that it is difficult to obtain official data on the incidence of deformities from commercial hatcheries, but generally, 7%–20% of deformed individuals occur, and this rate may even increase to 45%–100% from time to time [37, 39, 4345, 47, 5560]. In the first stages of production, fish are stunned in high salinity to separate individuals without a swim bladder. Additionally, deformed fish are separated on illuminated tables before being sent to net cages and/or earthen ponds. However, despite all these practices, deformed individuals are encountered in the later stages of production. Since deformed fish are not preferred by consumers, market opportunities, and prices for these products are lower. In addition, these individuals must be separated from other individuals at the hatchery, which increases production costs. It is estimated that European aquaculture suffers annual losses of 50 million euros due to deformities [51]. This situation also puts a lot of pressure on the sustainable development of the aquaculture industry [51, 53, 54, 61, 62]. For this reason, researchers tend to find the reason for these negative characteristics that occur in production with the new techniques that they constantly use.

In a meta-analysis, the results of multiple studies on specific topics are collected independently, combined, and analyzed using statistical methods. Thanks to this method, it can also be defined as a measure that expresses the extent and direction of the relationship between variables. A meta-analysis is a useful approach for extracting and integrating information from different studies. It recognizes the need to integrate and synthesize experimental observations using a quantitative approach [6366]. For this purpose, a meta-analysis was conducted, which is a systematic literature review, and used. The goal of the study conducted in this context is to analyze the results of research on deformities in marine fishes using the method of meta-analysis, to evaluate the cause–effect relationships of the problems uncovered so far, and to develop different approaches for new research.

2. Materials and Methods

In the study, the data on the deformation rate of control and experimental groups in all scientific articles on marine fish larvae farming were searched. Thus, it was investigated whether the deformation rate affects the deformation region/type and the factors that cause the deformation. For this purpose, a meta-analysis was conducted, which is a systematic literature review, and used. In the data analysis, the appropriate methods for the data type were selected from the possible combinations of research results. In this study, subgroup analysis for deformity region/type and causes was performed following the guidelines of empirically based published studies for researchers [67]. First, the characteristics of these studies were coded. Then, effect sizes were calculated using a common scale. Finally, the moderation effects of the studies on the outcome measure were examined.

In addition to the effect of the deformation rate on the deformation area in the hypotheses established for the study, the continent where the study was conducted was evaluated in terms of working time, fish production stage, deformation factor, stocking density, and method of determining deformation. A literature review was then planned to determine the validity of these hypotheses [68, 69]. To define the problem under study, the fish species studied, deformation factor, working time (day), fish production stage, number of fish used for analysis (n), stocking density (individual/l), deformation analysis method, statistical analysis method, significance value (P), deformation type, and deformation rate (%; control and experimental) were analyzed by meta-analysis method.

In the data collection, a literature search strategy was first established for the studies to be included in the meta-analysis of the factors affecting the development of deformations. For this purpose, the Web of Science (WOS), Google Scholar, and Scopus databases were searched, covering the years 1997–2022 (June). The search term was ((“deformities” AND “marine fish larvae”), (“abnormalities” AND “marine fish larvae”), (“malformation” AND “marine fish larvae”), (“osteological abnormalities” AND “marine fish larvae”), and (“skeletal deformation” AND “marine fish larvae”)).

As a result of the search, 1,700 references were obtained. The search strategy was changed due to a large number of literatures, the fact that reviewing individual abstracts can lead to errors, and the number of literatures that did not meet the acceptance criteria, we had previously established in the Google Scholar database. It was decided to use the WOS database because it was believed that randomized controlled trials, one of the acceptance criteria, could be obtained at a higher rate than in the Google Scholar database. In addition, in order to reduce the number of literatures to some extent and to examine the lower number of relevant literatures in more detail, the terms “deformities,” “marine fish larvae,” and “aquaculture” were set as keywords.

In accepting the literature on the topic, English-language full-text articles were first identified. In addition, articles that were evaluated first asked about early developmental stages (larval stage, weaning, pregrowth, and juvenile fish), cause of deformity (feeding, production model, and other factors), type of deformity (head, spine, and skeleton), and rate of deformity. Articles that did not meet the above criteria and those that did not provide statistical results were not included in the study.

In the coding phase of the studies, after the literature was transferred to the Mendeley program, the title and abstract were first reviewed and evaluated. After the preliminary evaluation, the full text of all the literature to be analyzed was obtained. The literature to be included in the analysis was coded by full-text evaluations according to the inclusion and exclusion criteria. The number of literatures to be included in the search and review is shown in the PRISMA flowchart [69]. The PRISMA flowchart for the literatures was included in the analysis, inspired by the work of Cozer et al. [70] (Figure 1).

In coding the studies, the studies included in the analysis were grouped under the heading of diet, production system, and other factors to show the general cause of deformation. These studies are listed in Table 1.

2.1. Data Analysis

In the effect size meta-analysis, it was necessary to calculate the measures of the outcome variables (effect sizes) for each of the studies before numerically combining each research article included in the study. In calculating the effect size, the values of “OR” were used as the basis for evaluating the overall effect size in the analyzes conducted for binary data. The limit of statistical significance in the evaluation of the overall effect meant that the risk of an OR > 1 outcome was increased [101103]. The minus (−) sign at the beginning of the effect values represents the control group; the plus (+) sign indicates that the effect was positive for the experimental group. When the effect size was zero or close to zero, it indicated that there was no result for or against the control and experimental groups. A composite effect size of 0.80 and above was considered a significant effect; values between 0.50 and 0.79 were considered moderate, and values below 0.50 were considered no effect [104]. If the study had a homogeneous distribution, the fixed effects model was used; if the distribution was not homogeneous, the random effects model was used [105].

To adjust or calibrate the results of studies meeting the criteria to a common scale, effect sizes were calculated using odds ratios and Hedge’s g values. Statistical analyses were performed using the metaessentials workbook Version 1.4, which is licensed under Creative Commons. Data were interpreted using the user’s guide under the Attribution-Non-Commercial-ShareAlike 4.0 International License [106] and developed after [107]. Subgroup and moderator analyses were performed when heterogeneity (I2) was high [108]. This allowed the researchers to further investigate the role of the different variables.

After calculating the deformation rate and effect size values of each trial in the control and experimental groups, the I2 statistic was reviewed for heterogeneity analysis. I2 is the heterogeneity ratio of the total change in the observed effect. Heterogeneity is related to the percentage of variance disclosure of the available studies. Heterogeneity increases as the percentage of disclosure increases [66]. When assessing heterogeneity, a heterogeneity rate (I2) of less than 25% is considered absent, 25%–50% is considered low, 51%–75% is considered moderate, and more than 75% is considered high [103]. This test tests the null hypothesis that all items rate the same effect. This analysis is used to determine if there is a statistically significant variance [109].

Publication bias above a certain level affects the average effect size to be calculated and makes it higher than it should be [110]. For this reason, publication bias is determined by calculating Kendall’s tau coefficient from the “funnel plot” graph, and another statistic of Begg and Mazumdar’s rank correlations. If there is no publication bias, this coefficient should be close to 1 and the double-tailed p-value should not make a significant difference, i.e., the p-value will be greater than 0.05 [103]. In addition, the results were interpreted using Orwin’s number and Egger’s regression analysis to reset the statistically significant effect of the deformation rate in the meta-analysis.

A meta-analysis was performed between the deformation rates using the studies that compared the percent change between the deformation region/type and cause between the control and experimental groups, which are included in Table 1. The I2 statistic was used for heterogeneity analysis. When this value was above 50% and the p-value of the heterogeneity test (Q) was significant (), the analysis was performed with the random effects model. These variables were working time, stock density, deformation detection analysis methods, deformation region/type (head, spine, and skeleton), deformation factors (feeding, production model, and other factors), and stage (larval stage, weaning, on growing, and juvenile). Publication bias was visualized with a funnel plot and quantitatively defined with the Begg–Mazumdar test, which is preferable to Egger regression because it considers smaller samples [111, 112]. All tests were performed at a 95% confidence level (CI), and all p-values less than 0.05 were considered significant.

3. Results

The following are the outcome issues:(1)Location of deformation zone/type (head, spine, and skeleton).(2)Deformation factors (feeding, production model, and other factors).(3)It consists of the publications considered for the deformation stage (larval stage, weaning, on growing, and juvenile).

In the studies examined, it was found that the continent, working time, phase, factor, stocking density, and variables considered by the deformation detection method of the publication that determines the deformation were studied and the data belonging to these common factors were analyzed.

The data on the deformation rates obtained from the results of the factors considered to cause deformation in the study are shown in Table 1.

As shown in Figure 2, the effect size of each study in the meta-analysis was represented with a square symbol. The total effect size, reflecting the sum of the individual studies, was shown in black at the bottom. The width of the circles was also indicated in proportion to the weight of each paper. The result of the meta-analysis calculating the difference between the means was −0.82 (−1.48, −0.45). Thus, the relationship between the deformation rate and the control and experimental groups was significant () (Table 2).

According to the hypothesis that the deformation rate has a significant relationship with the deformation location of the control and experimental groups, the I2 value was 83.95%. The Q-value of the heterogeneity test was 379.96 (), the random effects model was used, and the results were presented with the forest plot diagram in Figure 2 and the meta-analysis in Table 2.

As shown in Figure 3, the effect size of each study in the meta-analysis was represented by the round red symbol. The overall effect size reflects the sum of the individual studies, as defined by the green symbol at the bottom. The width of the circles was also indicated in proportion to the weight of the individual papers. In the meta-analysis, performed by calculating the difference between the means, 0.44 (0.17, 1.15) was found.

These results were in favor of the control group. Consequently, the relationship between the deformation rate and the control and experimental groups was significant () (Table 3).

To determine the influence of the deformation rate in the publications included in the study on the deformation region/type, a p-value of less than 0.05 was determined. The Q-value (379.96) was more significant than the result of the heterogeneity test, and the statistical I2 value was calculated as 83.95. As a result of the individual studies included in the analysis, it was found that the studies examined by deformation region/type had a heterogeneous structure in the applications of the meta-analysis. Therefore, the distribution of effect sizes was evaluated as a result of the random effects model calculations. The results of the meta-analysis of 62 studies that examined the effect of region/type of deformation and were included in the study are shown in Figure 2 with the forest plot. In the analysis performed according to the random effects model, the effect size was statistically significant with a value of 0.44 (0.08–2.35; ).

To this end, a subgroup analysis was performed to examine the effects of deformation rate on deformation region/type. The results of the subgroup analysis are shown in Figure 3 and Table 3. In the study, the deformation region/type was divided into three areas: head, spine, and skeleton. All subgroups were evaluated according to the random effects model.

When the results of the meta-analysis of the deformation zone/deformation type are examined in detail, the risk difference in the direction of the deformation zone/deformation type in an analysis of heterogeneity among them based on the head region (Q = 154.40, ), the risk difference in the direction of the deformation region /type is 0.21 (0.08, 0.57) in favor of the experimental group. When analyzed among themselves based on spine region (Q = 147.27, ), the risk difference toward deformation region/type is 0.36 (0.17, 0.76) in favor of the experimental group when analyzed among themselves based on skeletal structure (Q = 59.01, ) based on the skeletal structure. The risk difference for per type was found to be 1.10 (0.42, 2.91) in favor of the results of the funnel scatterplot, which also serves as a visual summary of the meta-analysis data set and highlights the possibility of publication bias, are shown in Figure 4. A large proportion of the 62 studies included in this study are very close to the combined effect size and in the upper ranges. According to Figure 4, it can be said that there is no picture of publication bias. Since Kendal’s tau coefficient is −0.16 and , no publication bias was observed in the studies included in the meta-analysis according to the calculated values.

Examination of the funnel plot indicates the possibility of publication bias. To confirm this finding, the Begg–Mazumdar rank correlation was used and this value was used to determine whether or not the number of studies included in the study was safe. The average effect of the included publications on study time formed a uniform and symmetrical funnel. The Begg–Mazumber rank correlation yielded a coefficient of 0.14 () for the value of Kendal’s tau (Figure 5). Therefore, the number of studies considered appeared to be valid for the overall effect size determined in the meta-analysis. In accordance with this finding, the articles of the authors who drew common conclusions in the studies on the deformation rate were used.

The average impact of the publications on the studied continent, phase, factor, stock density, and deformation method formed a uniform and symmetrical funnel. The Begg–Mazumdar rank correlation yielded a value of 0.17 () for Kendall’s tau (Figure 6). Therefore, the number of studies considered seemed to be valid for the overall effect size obtained in the meta-analysis. According to this result, the articles of the authors who had drawn common conclusions in the studies on the deformation rate were used. The results of the moderators’ effect size analysis are presented in Table 4.

Regression plots and lines can be interpreted according to the steepness of the line. At run time, the regression lines with stock density were not steep, but almost straight. This meant that changing the moderator did not affect the effect size. High p-values supported this result for run time () and stock density (). Therefore, the publications on deformation showed that the deformation rate in the different subgroups was not affected by the studied continent, phase, factor, and moderators of the deformation detection method. A steep sloping line was observed for the factor and deformation detection method, and it was interpreted that when these studies are published next year, a sharp change in effect size will be observed. However, data analysis showed that factor () and method of deformation detection (0.7839) had no effect-on-effect size. When the phase was examined, a very steep upward line was seen, which was interpreted as a strong change in the effect size of the studies in other studies to be conducted. However, data analysis showed that the factor () had no effect on the effect size.

4. Discussion and Conclusion

Shape deformities can be caused by inadequate knowledge of larval development and a lack of management and production techniques. Deformities that occur in larval production systems follow a complex structure. According to the obtained data, it was found that there were negative effects on the formal structure of fish regardless of the region, duration, stage, factor, stocking density, and method used to detect deformities. Although environmental conditions, supported by advanced application techniques, have an essential place in solving the problem, the high rate of deformation, which occurs frequently despite all developments, forces production.

Although the studies investigated were carried out on different or the same species using different or similar factors in different parts of the world, similar deformations were found in different or the same parts of the larval body [71, 85, 89]. In this analysis, we clearly understood that both abiotic, biotic, xenobiotic nutritional, and production systems have caused many different or similar deformations in many countries from Australia to Asia, Europe to Africa, and even in the Americas (Table 1).

Production of portioned fish in marine fish farming is a long and difficult process. The deformities that occur when fish reach the portion stage are observed in the first days of the larval stage, when skeletal tissue is weak or not yet differentiated, and can be seen by the naked eye in the early stages of production. The proportion of fish with abnormalities differs not only between other facilities but also between different batches in the same incubator and even among larvae from the same breeder. In addition, extreme skeletal abnormalities can occur throughout the life cycle of fish, but their development usually begins with slight abnormalities in internal elements [51]. Even when egg quality criteria are met, the negative factors that occur in the early stages of larval life can affect the larva in a short period of time. The signs of deformation can be seen at the blastomere stage of embryonic development of the egg. In this context, production techniques and environmental conditions for the management of the rootstock should be carefully monitored [113, 114]. At the beginning of the larval period, the most critical problem in the fish produced was swim bladder inflation, which caused significant problems, especially in the production of sea bream and sea bass. Although the rate of the noninflated swim bladder problem averages about 10%, this situation sometimes reaches up to 50%. The entire tank can be lost, especially due to the thickness of the oil film layer formed on the tank surface and inadequate cleaning [6, 115, 116]. Remaining larvae experience deformities of the pectoral fins, notochord/vertebral axis, and prehaemal vertebrae. Not only the noninflating of the swim bladder but also the hypertrophic (overinflated) swim bladder causes severe losses, notochord, and spinal deformities [117]. Deformities of the larval notochord that occur in the early life stages have a risk to the entire culture process.

Spinal deformities in the subsequent period are the main problem, especially in the breeding of marine fish. Different names denote disorders of the spine in different forms, and naming is done according to the type of deformity, its severity, and frequency. Lordosis is one of the most common types of deformity in farmed species, and it is commonly seen in scoliosis and kyphosis, which are other major deformities of the spine. These deformities can sometimes occur simultaneously [4]. Although it has been associated with the inflation of the swim bladder in some species, it has been associated with various causes such as feeding, water flow, and genetics in other species [6, 52, 59, 118, 119]. In addition, there are some deformities ((a) dislocation, fusion, shortening, deformation, or absence of the centers [9, 120]; (b) dislocation, compression, deformation, absence, or additional formation of the haemal and neural arches, and apophysis [16]; (c) dislocation, shortening, deformation, absence, or separation of the ribs [44, 118, 121]; and (d) platyspondyly, vertebral ankylosis, and platyspondyly [122]) that have been described. Not only the spine or related systems but also the fins are deformed. These deformities can often originate from breeding techniques, tank structure, and stocking density. The disorders, which are intensely observed under the name saddleback syndrome, are not only found in many species but also have different structures in themselves. However, it has been noted that these deformations can also be formed in the early stages of larvae, although it is not known whether they are caused by fish behavior or by nature [7]. Although these deformities affect the welfare of the animals, they do not have a decisive effect on market conditions because the main structure of the fish has not deteriorated.

Another condition that is common in fish and differs in classification is head deformities. Although they cannot be fully explained, genetic disorders, sudden temperature changes during the incubation period of eggs, parasites, environmental conditions, and the use of feeds with insufficient vitamin content, in particular, are thought to cause the development of such deformities [7, 51, 76, 96, 99]. In general, these morphological disturbances in the cephalic region adversely affect the respiratory metabolism and feed intake of the fish.

The techniques used to detect the above deformations have now reached the standard. Depending on the life stage of the fish, different techniques are used to determine the intensity of deformation patterns and different degrees of accuracy. Deformations in fish can be detected by techniques such as external observation, X-ray, palpation, staining, synchrotron microcomputed tomography, computed tomography, histology, histopathology, histochemistry, and immunohistochemistry. These techniques may have some advantages and disadvantages depending on their application and/or larval age. More detailed applications are used in scientific studies, especially for larval stages. However, it has been shown that commercial companies confirm the quality standard by taking X-rays of a certain number of fish to register their quality at the sales stage, after sorting out the deformed fish at the appropriate stage.

So far, various factors have been shown to play a role in the development of skeletal abnormalities in different fish species under culture conditions. In general, skeletal abnormalities are thought to have significant adverse effects on animal welfare, biological performance of cultured fish, product quality, and production costs. Many of these explained factors indicate that, even if possible contaminants and pathogens in rearing conditions are controlled, improper feeding and/or starvation, adverse abiotic conditions, and genetic factors are the most likely causes of skeletal abnormalities in reared fish [51]. If nutritional requirements of preferred food are not met or more elements are supplied than necessary, especially during the early life development, this has a strong effect on skeletal deformities [123]. Proteins and amino acids, lipids and fatty acids (essential fatty acids, phospholipids, and oxidized lipids), and vitamins (water-soluble vitamins, fat-soluble vitamins, and minerals (macrominerals, trace elements) form the basis and content of feeds, and their effects on larvae are both it has been explained in detail in the form of a general assessment and the species in the review [51, 124]. On the other hand, the environmental factors most frequently cited as possible causes of skeletal abnormalities in reared fish are broodstock condition, egg quality, stocking density, rapid growth conditions, handling stress, hydrodynamic turbulence/water turbulence/water supply, rearing methods, light regime, mechanical factors, oil films on the water surface, O2/CO2 content, pH, physical trauma/mechanical stress, pathogens, parasites, toxins, radiation, salinity fluctuations, substrate condition (especially for flatfish), tank characteristics (volume, shape, color, and material), temperature fluctuations, antibiotics, and xenobiotics [13, 52, 59, 125127]. In addition, genetic selection programs aimed at achieving rapid growth rates have been introduced over time; studies showing the inheritance of skeletal abnormalities, genetic drift, or gene mutations in deformed fish, their effects on phenotype, inbreeding, and selective breeding polyploidy were performed and interpreted [4, 34, 128134].

The final success of the production is possible primarily by raising quality offspring. The most important precaution in this regard is to avoid sudden changes, keeping the physicochemical structure of the surrounding water under constant control at every stage of production and avoiding the factors that cause stress. In addition, it is important to obtain the broodstocks to be used in production from the natural environment, if possible, or to replace the population of broodstocks used in production with broodstocks obtained from nature at a certain rate (around 30%) every year. Feeding of broodstock with high-quality feed throughout the year or at least from the stage of egg formation will inevitably has a positive effect on the quality of the eggs. Fish eggs should be obtained in an environment free of pesticides, metal ions, hypochlorite, and other pollutants. Although it is not done most of the time, the bacterial load should be reduced by disinfecting the live eggs after the dead eggs are separated. Also, attention should be paid to egg stock density by hatching. In the larval period, it is important to prepare the environment of the tank according to the needs of the larvae (avoiding the formation of air bubbles, especially on the tank walls), to keep the stock density of the larvae optimal, and to avoid sudden changes in parameters and mechanical shocks. In particular, the influence of the light factor on the larval stage should not be forgotten. During this time, stress, especially from a sudden power outage, has a very strong effect on swimbladder hypertrophy. For this reason, the generator systems must be suddenly activated. In addition, sudden changes in salinity have a very strong effect on swimbladder hypertrophy during larval production in low salinity. Therefore, the sustainability of freshwater resources is important. Removal of the oil layer accumulated on the tank surface is inevitable for the formation of a functioning swimbladder. Surface cleaners should be placed according to the tank surface area and their cleaning should be done uninterruptedly. The flow rate applied to the tanks should be increased in direct proportion to the age of the larva, and the larva’s swimming speed and resistance should be taken into consideration when making flow rate calculations. Automatic flow meters should be used in tanks. If the survival rate around day 20 in the tanks does not reach the desired level of quality larvae, the tanks should be quenched by chlorination. Chlorination is essential to maintain the natural population. Live food sources should be fortified with essential fatty acids and vitamins and refrigerated during the day, so they do not lose their nutritional value. In addition, the content of artificial feeds should be defined separately for each species, and the nutritional composition of the feeds should be determined for different developmental stages of the same species. In addition, before transferring the fish to net cages and/or earthen ponds for rearing, they should be examined on light tables for shape deformation, and deformed individuals should be separated and destroyed. In addition, it would be beneficial to follow the current literature and integrate all kinds of scientific data with the production system, to record the current production data system in a healthy way and to use these data in subsequent productions.

Despite these evaluations, which are considered a success for quality production, it is inevitable to encounter different problems at different stages of production, even when trying to meet all conditions. In a process where the survival rate under cultural conditions is so high compared to the natural environment, there will continue to be deformed individuals. Changes in deformation types and rates will inevitably affect ecosystem structure and function in the long run, as well as production, individuals escaping, or being released into the natural environment. The combination of degraded population structure in the natural environment with environmental impacts (especially microplastics and pollutants) poses a major threat to the ecological balance. There is no guarantee that this will not impact ecosystems and human communities.

This study examined 62 national and international studies on deformation rates between 1997 and 2022. The studies included in the study were identified by scanning Google Scholar, WOS, Scopus, Science Direct, and ULAKBİM data. In total, 100% of the national and international studies included in the study are covered within foreign articles. In this study, the meta-analysis method was used to analyze the studies on the topic of deformation “because it allows the results of many independent studies on a given topic to be combined and the findings obtained to be analyzed using statistical techniques.” Unlike other screening methods, the meta-analysis method is based on numerical data and statistical techniques and provides a comprehensive and systematic summary of the literature on the studied concept [135].

Our results showed that the deformation rate had an overall negative effect on the functional properties of fish, regardless of the variables considered. The effect size appeared statistically significant and negative when analyzed as a whole, including all heterogeneous data. The number and content of the publications studied were determined to include the source of deformation. In addition, all statistically significant individual response variables had a negative effect size. Also in our study, the effects of deformation on region/type were supported by high p-values for maturity and stock density under the experimental conditions analyzed as a function of conditions in the variables considered. Therefore, the deformation publications showed that the deformation rate among the different subgroups was not affected by the continent, phase, factor, and moderators of the deformation detection method studied. In addition, it showed that the factor and the deformation detection method did not affect the effect size, and it was found that there was a strong change depending on the phase.

In this study, when the results were evaluated based on the meta-analysis method, it was found that the deformation rate was in the spine, head, and skeleton. In this regard, in the studies that were treated within the deformity region/type, 46% of the deformities were found in the spine, 37% in the head, and 16% in the skeleton. According to the meta-analysis results, the apparent value weight percentages were 35.82% in the spine, 33.12% in the skeleton, and 31.06% in the head.

Consequently, meta-analyzes may process results from multiple trials in aquaculture studies. However, results may differ between experimental conditions due to experimental conditions, parametric properties, and functional variability. Although meta-analysis is used in many fields, aquaculture is still developing. Meta-analysis of aquaculture studies also presents many challenges, including high heterogeneity, data sources, biased information, and nonautomated content of literature data. However, the findings from our global meta-analysis can be considered as a key indicator of deformities limiting fish production. In the future, advanced statistical tools such as Bayesian meta-analysis, network meta-analysis, and meta-regression analysis can be used to explore more complex data structures. The rapid development of artificial intelligence techniques will increase the efficiency of data collection and the robustness of results for meta-analysis studies in aquaculture and other fields.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Disclosure

An earlier version of this work was published as a preprint by this team on Research Square: https://www.researchsquare.com/article/rs-3202114/v1 [136].

Conflicts of Interest

The authors have no conflicts of interest to declare that are relevant to the content of this article.

Authors’ Contributions

All authors contributed to the study’s conception and design. Hülya Eminçe Saygı contributed to conceptualization and writing—original draft. Hülya Eminçe Saygı, Hatice Tekoğul, Muammer Kürşat Fırat, Cüneyt Suzer, Osman Özden, and Fatih Güleç contributed to investigation. Hülya Eminçe Saygı and Muammer Kürşat Fırat contributed to methodology. Hülya Eminçe Saygı, Muammer Kürşat Fırat, Hatice Tekoğul, and Deniz Çoban contributed to supervision. Muammer Kürşat Fırat contributed to validation. Müge Aliye Hekimoğlu and Şahin Saka contributed to writing—review and editing. Şahin Saka contributed to visualization. Cüneyt Suzer, Osman Özden, Hatice Tekoğul, and Fatih Güleç contributed to resources. Deniz Çoban contributed to formal analysis. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Each member of the author’s team made a significant contribution to this study. All the authors are aware of and agree with the content of the manuscript.

Acknowledgments

The authors would like to thank Research Square for including the presented study in the preview for publication in the international literature.