Molecular prevalence of Coxiella burnetii in cheese samples: Systematic review and meta‐analysis

Abstract Background Cheese is a popular dairy product consumed worldwide, and it has been implicated as a source of Coxiella burnetii infections. Objectives The present study aimed to describe the molecular prevalence and source analysis of C. burnetii in cheese samples. Methods A systematic literature search was conducted using the Medline/PubMed, Science Direct, Web of Science, Scopus, and Google Scholar databases to identify studies reporting the molecular prevalence of C. burnetii in cheese samples. The pooled prevalence of C. burnetii in cheese samples was estimated using a random‐effects model. Results A meta‐analysis was conducted using the mean and standard deviation values obtained from 13 original studies. The overall molecular prevalence of C. burnetii in cheese was estimated to be 25.2% (95% confidence interval [CI]: 13.1%–39.7%). The I 2 value of 96.3% (CI95% 94.9–97.3) suggested high heterogeneity, with a τ 2 of 0.642 (CI95% −0.141 to 0.881), and an χ 2 statistic of 323.77 (p < 0.0001). Conclusions In conclusion, our meta‐analysis provides a thorough assessment of the molecular prevalence and source analysis of C. burnetii in cheese samples.


2019
).One scientific reason for using meta-analysis in food-related studies is that it allows researchers to gain a more precise estimate of the effect of a particular food or dietary pattern on health outcomes (Hamidiyan et al., 2018).By pooling data from multiple studies, metaanalysis can increase the statistical power of the analysis and reduce the impact of random variation in individual studies, thereby providing a more accurate estimate of the true effect size (Sutton et al., 2001).
Furthermore, meta-analysis can also help identify sources of heterogeneity or inconsistency across studies and can be used to explore potential sources of bias or confounding factors.Thus, meta-analysis can provide a more robust and comprehensive evaluation of the evidence base in food-related studies and can help guide future research and public health policy (Khaneghah et al., 2018;Khaneghah et al., 2019).
In this study, a systematic review and meta-analysis were conducted to determine the molecular prevalence and source analysis of C. burnetii in cheese samples.The findings of this study could provide important insights into the potential public health risks associated with the consumption of contaminated cheese and inform effective preventive measures to reduce the risk of transmission of C. burnetii to humans.

Literature search and study selection
The systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (Moher et al., 2009).A systematic literature search was conducted using the Medline/PubMed, Science Direct, Web of Science, Scopus, and Google Scholar databases to identify studies reporting the molecular prevalence of C. burnetii in cheese samples.The search was conducted using the following keywords: ('C.burnetii' or 'Q fever' or 'C.burnetii' or 'coxiellosis') and ('cheese') and ('molecular' or 'PCR' or 'real-time PCR' or 'qPCR' or 'conventional PCR' or 'nested PCR' or 'molecular diagnosis' or 'polymerase chain reaction').The search was limited to studies published in English between January 2000 and April 2023.

Inclusion criteria
To identify any additional relevant studies, the reference lists of the identified studies were searched by hand.Full texts of potentially eligible studies were reviewed to determine if they met the inclusion criteria.Studies were considered eligible for inclusion if they met the following criteria: (1) reported the molecular prevalence of C. burnetii in cheese samples, (2) used molecular methods to detect the presence of C. burnetii DNA in cheese samples, (3) provided sufficient data to calculate the prevalence estimate and its confidence interval (CI) and ( 4) were primary studies, not reviews.

Data extraction and meta-analysis
A standardized form was used to extract data from eligible studies.
This form included the following information: author, year of publication, country, study area, period of study, origin of cheese, milk species, kind of production, country classification, DNA extraction, molecular approach, target gene, number of cheese samples tested, number of positive samples, prevalence and CI.
The meta-analysis included in this study used the C. burnetii determined by molecular methods in cheese samples as the dependent variable, which served as the effect size for the analysis.To perform the meta-analysis of proportions, the previous methodology was followed (Wang, 2018).Heterogeneity among studies was initially assessed using Cochran's Q (χ 2 ) test, which examines the null hypothesis of homogeneity.Subsequently, the extent of heterogeneity was quantified using Higgins' I 2 statistic (Borenstein et al., 2011).The level of heterogeneity was then measured to select the appropriate model for estimating the overall weighted C. burnetii prevalence.Given that the level of heterogeneity was high, a random-effects model was used to account for both within-study variance (sampling error) and between-studies variance (τ 2 ).
Potential origins of heterogeneity were explored through the examination of moderating factors.Five distinct moderators were subjected to evaluation: (1) distinction between developed and developing countries, (2) milk source (goat or sheep in comparison to cow), (3) target gene (the transposon-like repetitive region of the bacterial genome (IS1111) as opposed to others), (4) extraction method (commercial vs. non-commercial) and (5) methodological approach used in DNA extraction (mechanical vs. non-mechanical).A subgroup analysis was performed for the categorical moderators, which were analysed using a mixed-effects model.The statistical significance of the moderators was evaluated by an omnibus test (QM) within the mixed-effects model (Viechtbauer 2010).The R 2 index was used to explore the proportion of heterogeneity accounted for by each moderator.To further explore heterogeneity among the studies, meta-regression was utilized.All moderators and their interactions were entered into the initial model, and non-significant terms were then dropped stepwise (from lowest R 2 to highest R 2 ) (Li et al., 2020).The association among moderators (country classification, milk source and methodological approach employed) was assessed by the Pearson correlation coefficient (r).Results from the meta-analysis, along with the corresponding 95% CIs, were summarized using forest plots.Egger's test was used to test for the possibility of publication bias for studies with low or high effect sizes (Egger et al., 1997).All assessments were conducted using commercial software version 20.110;MedCalc Software Ltd).

RESULTS
An initial search yielded a total of 2082 articles, out of which 13 met the inclusion criteria.The study selection process is summarized studies utilized commercial extraction protocols, whereas two trials employed non-commercial methodologies.Nucleic acid isolation from food samples was accomplished using mechanical procedures in nine studies, whereas four studies abstained from incorporating mechanical procedures for nucleic acid isolation (Table 1).
The median size of eligible studies was 67 cheese samples, and out of the total 1107 cheese samples, 317 were diagnosed as positive by molecular techniques.The percentage of positive cheese samples across the studies ranged from 4.3% to 85%.The overall molecular prevalence of C. burnetii in cheese was estimated to be 25.2% (95% CI: 13.1%-39.7%).The I 2 value of 96.3% (CI 95% 94.9-97.3)suggested high heterogeneity, with a τ 2 of 0.642 (CI 95% −0.141 to 0.881), and an χ 2 statistic of 323.77 (p < 0.0001).The overall meta-analysis is shown in a forest plot (Figure 2).The meta-analysis showed no significant evidence of publication bias based on Egger's test (p = 0.715).
In the context of the subgroup analysis comparing the target gene, the observed prevalence was 28.2% (CI 95% 13.7-45.4),I 2 = 96.7%;Moreover, no significant correlation was evident between developing and developed countries (r = −0.255,p = 0.626).

DISCUSSION
The global consumption of cheese and cheese-related products is increasing, with a wide variety of cheese types consumed across different cultures and regions.Europe is the largest consumer of cheese, accounting for over 40% of the global cheese consumption.The consumption of cheese is deeply ingrained in European culture, with each country having its unique cheese-making traditions and varieties.The United States is the second-largest consumer of cheese globally, with an average annual consumption of over 37 pounds of cheese per person (Khanal et al., 2019).The consumption of raw milk cheese is common in some regions, particularly in Europe, where it is considered a traditional food.Raw milk cheese is made from unpasteurized milk, which has not undergone the process of heating to kill bacteria (Fusco et al., 2020).Several studies have investigated the prevalence of C. burnetii F I G U R E 2 Forest plot illustrating meta-analysis of Coxiella burnetii prevalence in cheese samples from thirteen studies conforming to inclusion criteria in the systematic review.Subgroup analysis of C. burnetii prevalence in cheese samples differentiating developed and developing countries.
2019; Basanisi et al., 2022).The global molecular prevalence of C. burnetii in milk samples has been estimated by meta-analysis (Rabaza et al., 2021).We carried out a comprehensive systematic review of the literature using keyword-based searches to determine the global molecular prevalence of C. burnetii in cheese samples.Data from cross-sectional studies that met our inclusion criteria were extracted and included in a meta-analysis.
Our systematic review and meta-analysis on the molecular prevalence of C. burnetii in cheese samples provide valuable insights into the global distribution and prevalence of this bacterium in cheese products.The estimated overall molecular prevalence of C. burnetii in cheese was found to be 25.2%, with a 95% CI of 13.1%-39.7%.These findings are consistent with previous studies that have reported the presence of C. burnetii in cheese samples from various regions worldwide (Basanisi et al., 2022;Eldin et al., 2013;Gyuranecz et al., 2012).
The high heterogeneity observed across the studies, as indicated by the I 2 value of 96.3%, highlights the need for caution in interpreting these results.The variation in prevalence estimates across different regions and types of cheese may be attributed to differences in sampling methods, sample sizes and the sensitivity and specificity of the molecular techniques used for diagnosis.Additionally, differences in the prevalence of C. burnetii in animal populations may also contribute to the observed heterogeneity (Pearson et al., 2014).Despite the high heterogeneity, the lack of significant evidence of publication bias based on Egger's test suggests that the results are reliable and representative of the available literature.
Frequently encountered in the isolation of nucleic acids from food samples are four distinct processes: (1) mechanical procedures, including homogenization and filtration; (2) enzymatic processes; (3) DNA extraction; and (4) DNA purification (Jana et al., 2020).Due to the uneven distribution of microorganisms within food matrices, homogenization plays a crucial role in enhancing the diagnostic efficacy (Jana et al., 2020).The aggregated positivity rate observed in studies that undertook DNA extraction without mechanical treatment was comparatively elevated in comparison to those studies that incorporated mechanical processing during the DNA extraction procedure.Enzymatic processes was administered in all studies incorporated within the investigation.Concerning DNA extraction methods, 84.6% of trials employed commercial extraction protocols, whereas two trials conducted in Iran and Japan employed non-commercial approaches.
Among the studies employing non-commercial methods for DNA extraction, a lower positivity rate is evident compared to the studies utilizing commercial kits.It is noteworthy that milk and dairy products may contain constituents such as fat and protein, which have the potential to impede PCR-based assays (Baptista et al., 2021;Jana et al., 2020).Notably, extraction methodologies based on spin columns have demonstrated effectiveness in mitigating inhibitory substances and facilitating the retrieval of abundant DNA from milk and dairy products (Baptista et al., 2021).
However, 84.6% (11/13) of the studies included in this research employed the diagnostic method targeting the IS1111 gene region.
However, in one study, the com1 gene region was utilized, whereas another study employed com1, htp-B, and icd gene regions.In studies involving the IS1111 gene region, the ratio of positive samples to the total number of cheese samples was 29.2%.However, in the two additional studies conducted in Iran and Japan, this rate dropped to 15.5%.The IS1111 gene region exhibits varying copy numbers depending on the strain, resulting in enhanced sensitivity (Rabaza et al., 2021;WOAH, 2018).The elevated positivity rate observed in studies focusing on the IS1111 gene region is believed to stem from its heightened sensitivity.
In the molecular investigations, diverse methodologies have been employed in prior studies, encompassing qPCR (n = 8), nPCR (n = 3), TdPCR (n = 1) and DdPCR (n = 1).These approaches yielded total positivity rates of 35.1%, 14.2%, 7.1% and 7.8%, respectively.The qPCR stands as a robust methodology in comparison to other PCR-based assays, as it facilitates the acquisition of quantitative data regarding the quantity of nucleic acid within the scrutinized specimen (Taylor et al., 2019).In the studies under consideration within this study, the diagnosis of C. burnetii involved the targeting of the IS1111 gene region using qPCR.However, owing to the copy number variability across strains, the IS1111 gene region cannot be precisely quantified through the employed targeted qPCR methodology (WOAH, 2018).Comparatively, nested PCR aimed at the com1 gene region exhibits lower sensitivity when contrasted with touchdown PCR that targets the IS1111 gene region (Kargar et al., 2015).Within the showcased studies, the prevalence of qPCR and the IS1111 gene region targeting is notable across various PCR techniques.The incorporation of these dual strategies in the diagnosis of C. burnetii is poised to deliver more accurate outcomes.
The present study provides important findings on the prevalence of C. burnetii in cheese samples across different countries and types of milk.Interestingly, the results show a significantly higher prevalence of C. burnetii in developed countries (33.0%) compared to developing countries (16.8%).This difference in prevalence may be attributed to differences in the level of awareness, hygiene practices and animal husbandry practices between developed and developing countries (Mura et al., 2014).The study also found a higher prevalence of C. burnetii in cheese samples obtained from cattle milk (26.4%) compared to those obtained from goat or sheep milk (20.5%).This is in contrast with previous studies that have reported a higher prevalence of C. burnetii in small ruminants than in cattle (Arricau-Bouvery et al., 2005;Guatteo et al., 2011).This is possibly because infected cattle can shed viable C. burnetii in milk for longer than one year (Enright et al., 1957).It should be noted that the prevalence rates reported in this study should be interpreted with caution due to the high degree of heterogeneity observed in the data (I 2 > 80%).This heterogeneity may be due to variations in the sampling methods, testing procedures and diagnostic criteria used in different studies.Therefore, future studies should aim to standardize these variables to obtain more reliable prevalence estimates.The findings of this study highlight the importance of implementing effective preventive measures to reduce the risk of transmission of C. burnetii to humans through contaminated cheese.These measures may include the use of pasteurized milk, strict hygiene measures during cheese production and consumer education on the potential risks of consuming unpasteurized cheese.However, our meta-analysis has some limitations that need to be taken into consideration when interpreting the results.First, the studies included in our analysis were conducted over a period of several years, and there may be variations in the prevalence of C. burnetii over time due to changes in animal husbandry practices, weather conditions and other factors.Second, the number of included studies in developed countries was higher than developing countries, which may limit the generalizability of the findings to other regions.Third, the molecular methods used to detect C.
burnetii in cheese samples varied among the included studies, which may have contributed to the observed heterogeneity.Finally, publication bias cannot be ruled out, as studies with negative results may not have been published.

CONCLUSIONS
Our meta-analysis presents a comprehensive estimation of the molecular prevalence and source analysis of C. burnetii in cheese samples.

F
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram illustrating the study selection process for the systematic review and meta-analysis of the molecular prevalence and source analysis of Coxiella burnetii in cheese samples.inFigure1, and the characteristics of the included studies are presented in Table1.These studies were conducted in eight different countries (Japan, Italy [3 studies], Poland, France, Iran [3 studies], Spain, Brazil [2 studies], Latvia) and were published between 2011 and 2022.A meta-analysis was conducted using the mean and standard deviation values obtained from 13 original studies.The sample sizes ranged from 28 to 178 cheese samples, collected over 15 years(2005 to 2020).The studies employed real-time PCR (n = 8), nested PCR (n = 3), touch-down PCR (n = 1) and droplet digital PCR (n = 1).The IS1111 was the gene most frequently targeted in these PCRs (n = 9), followed by com1 (n = 2), icd (n = 1), IS30A (n = 1) and htp-B (n = 1).Seven studies were conducted in developed countries and six in developing countries.Regarding DNA extraction methods, 11

F
Subgroup analysis of Coxiella burnetii prevalence in cheese samples based on milk source: cattle milk vs. goat or sheep milk.Subgroup analysis of C. burnetii prevalence in cheese samples comparing IS1111 gene detection with other genes.F I G U R E 4 Subgroup analysis of Coxiella burnetii prevalence in cheese samples commercial kit-based vs. non-commercial kit-based methods.Subgroup analysis of C. burnetii prevalence in cheese samples comparing mechanical procedures with non-mechanical procedures for sample processing.
The substantial heterogeneity observed underscores the necessity for standardized sampling and PCR protocols to ensure cross-study comparability.By doing so, identification of factors contributing to C. burnetii contamination in cheese becomes feasible, facilitating effective measures to prevent Q fever transmission to humans.Furthermore, our study reveals discernible prevalence patterns linked to distinct genetic markers, methodologies and sample processing procedures within the context of C. burnetii prevalence in cheese.The prominent prevalence observed for the IS1111 gene underscores its substantive presence, whereas other genes display comparatively lower prevalence rates.Methodological variations, particularly between commercial and non-commercial kit-based methods, yield disparate prevalence outcomes.Variations in sample processing procedures similarly yield significant differences in prevalence rates between mechanical and non-mechanical techniques.Our findings highlight the absence of substantial correlations among milk type, study origin and development status with C. burnetii prevalence.This underscores the complex and multifaceted character of these relationships.This study further underscores the imperative of improving hygiene and animal husbandry practices in both developed and developing nations to effectively lower the potential for C. burnetii contamination in cheese products.