Exposure and Early Effect Biomarkers for Risk Assessment of Occupational Exposure to Formaldehyde: A Systematic Review

.


Introduction
Occupational exposure to formaldehyde (FA) and the related adverse effects for human health have been studied for many years and, nowadays, are still the subject of much research.
It is well known that exposure to FA is associated with a broad spectrum of negative outcomes for health, ranging from mild to severe [1].Specifically, acute exposure to FA can lead to irritation of the eyes, nose, throat, and skin, as well as symptoms like nasal congestion, sore throat, headaches, coughing, conjunctivitis, fatigue, skin rashes, shortness of breath, nausea, and nosebleeds [2].Besides this, FA is classified as a Group 1 human carcinogen [3], and chronic exposure to FA can lead to cancer both in humans 2 of 19 and animals [4].Despite this evidence, FA finds extensive use in the manufacturing of various products, including resins, adhesives, plywood binders, plastics, synthetic fibers, paints, and insulation foams.These materials serve as the fundamental components in the production of furniture, upholstery, carpets, curtains, and various other household items [5].Thus, FA is present ubiquitously in the environment and used in a great number of processes and activities in workplaces.The main sources of occupational exposure to FA are industrial production including resins, plastics, laminates, furniture, molding compounds, chemical manufacture, fertilizer, pesticides, paper, wood products, sanitizers, scientific supply, rubber, leather tanning, iron foundries, photographic film, textiles, and cosmetics.Other occupational settings in which occupational exposure to FA occurs are healthcare settings, especially for preserving tissue and specimens and for embalming procedures, agrifood scenarios, building, transportation, and fuel [6].
Given all the possibilities of occupational exposure to FA and the related adverse effects for human health, over the years, different mitigation strategies have been implemented in occupational environments to minimize exposure [7].However, these procedures are not standardized and not used in all the settings in which FA is used; thus, it is essential to evaluate this specific risk in the workplace and to monitor the exposed workers for their exposure and the related adverse effects, in particular those ones at an early and reversible stage.This topic is even more relevant considering that "promoting a safe and protected working environment for all workers" represents a target of the Sustainable Development Goal 8 of the UN Agenda 2030 for Sustainable Development, that is the goal to promote sustained, inclusive, and sustainable economic growth; full and productive employment; and decent work for all.At present, the gold standard for FA risk assessment and management in workplaces is the exposure measurement.In particular, occupational exposure to FA is actually assessed by measuring the levels of airborne FA.This measurement can be carried out through portable samplers/analyzers, equipped with photoacoustic spectroscopy detectors or electrochemical detectors, with a sensitivity of the order of a few µg/m 3 .More sensitive approaches are represented by active or passive environmental sampling with specific sorbent tubes containing 2,4-dinitrophenyhydrazine or 2-(hydroxymethyl) piperidine as derivatizer with a built-in ozone scrubber performed in fixed sites and/or by the use of personal samplers, then analyzed by high performance liquid chromatography mass spectrometry (HPLC-MS/MS) or gas chromatography mass spectrometry (GC-MS) [6].However, the airborne FA level is just an estimation of the true individual exposure.After the inhalation of FA, owing to its water solubility and reactivity, a great part of it is inactivated from mucus and peribronchial fluid and the remaining amount is absorbed in the body [8].Following its absorption, FA spontaneously reacts with glutathione (GSH) to generate hydroxy methyl glutathione (HMGSH).Subsequently, the enzyme formaldehyde dehydrogenase (FDH) oxidizes HMGSH into S-formylglutathione (FGSH).FGSH is then metabolized by S-formylglutathione hydrolase, resulting in the production of formate and the regeneration of reduced glutathione [9].Moreover, FA can also undergo oxidation facilitated by aldehyde dehydrogenase (ALDH), in coordination with cytochrome oxidase isoenzymes including and CYP2E1 [10].In this manner, the generated formate can be excreted in urine in the form of formic acid, interact with other biomolecules, or even be metabolized into carbon dioxide [11,12] Given the metabolism of FA, it should be very interesting to evaluate FA exposure by the use of indicators of internal dose and of early effects; however presently, official occupational health guidelines do not establish specific biomarkers for this purpose.
The aim of the present systematic review was to identify all the biomarkers used for assessing the exposure to FA in occupational environments and for evaluating the relative early adverse effects for human health.

Search Strategy
The systematic review was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [13].PRISMA Checklist is reported as Supplementary Materials.Additionally, the review protocol was registered in PROSPERO with the reference number CRD42023416960.
The review question focuses on the occupational exposure to FA in adult workers of all genders, to identify biomarkers of exposure used for evaluating FA exposure and/or early negative effects.
The studies selection procedure used the "PICOS" methodology (P stands for patient, population or problem; I for intervention; C for control group or comparison; O for outcome; S for study design) to generate the search query, and adherence to the following eligibility criteria was required: the population was made up of adult workers of all genders who were occupationally exposed to FA; intervention was not applicable; the outcome was the identification of all the biomarkers of exposure and/or effects used for assessing FA risk of exposure in the workplace; study design included observational, semi-experimental, and experimental studies.
We queried three electronic databases to search for articles: PubMed, Scopus, and Web of Science.

Inclusion/Exclusion Criteria
Articles were deemed eligible if they included data from observational studies, semiexperimental, and experimental studies, on adult workers who were occupationally exposed to FA, regardless of gender or age.We only included items published in English from the beginning of each database until 18 September 2023.
Studies including data about general population were excluded.We also excluded studies on biomarkers in workers not exposed to FA and with controls who had different socio-demographic characteristics than the exposed group, as well as controls who were not workers.Other types of studies, such as reviews, meta-analysis, case studies, qualitative investigations, book chapters, editorials, commentary studies, and so on, were not considered.
The titles and abstracts obtained from the three databases were imported into the reference management software Zotero (version 6.0.27), which was used for the initial assessment of relevance.Subsequently, the next phase involved a title and abstract screening, where potentially suitable studies were independently reviewed by five authors (A.A., A.D.G., S.Z., E.M., and V.C.).Following this, the full texts of these studies were independently examined by the same five authors, and a subsequent discussion took place regarding their potential inclusion in the review.Any disagreements that arose were resolved through consensus among the authors.All the steps were supervised by two other investigators (C.P. and M.V.).
The collected data were organized into a table that presented bibliographic details (including author, year of publication, origin country), sample size, age, and gender of participants.The table also included information about employment characteristics, biomarkers of exposure and/or early effects investigated, confounding and interfering variables considered, and the key findings of the selected studies.

Study Quality and Evaluation
We conducted the quality assessment using the Newcastle-Ottawa Quality Assessment Scale, modified for cohort and case-control studies, which enabled us to determine the overall rating.Specifically, selection, comparability, and outcome were the three evaluation categories.An overall quality rating was assigned to each eligible article according to the number of criteria met, as follows: For cross-sectional studies in the three evaluation categories of selection, comparability, and outcome: good quality (all criteria met, low risk of bias); fair quality (1 criterion not met or 2 criteria unclear, moderate risk of bias); poor quality (2 or more criteria not met, high risk of bias).For case-control and cohort studies: good quality (3 or 4 criteria in selection domain, 1 or 2 criteria in comparability domain, and 2 or 3 criteria in outcome domain); fair quality (2 criteria in selection domain, 1 or 2 criteria in comparability domain, and 2 or 3 criteria in outcome domain); poor quality (0 or 1 criterion in selection domain, 0 criteria in comparability domain, or 0 or 1 criteria in outcome domain).
Each study was individually scored by five authors (A.A., A.D.G., S.Z., E.M., and V.C.), and any discrepancies were resolved through consensus among all the authors.The ultimate rating for each article was calculated as the average of the five authors' scores.

Article Selection
Figure 1 shows the steps of the article selection process used for the systematic review following the PRISMA guidelines [13].Each study was individually scored by five authors (A.A., A.D.G., S.Z., E.M., and V.C.), and any discrepancies were resolved through consensus among all the authors.The ultimate rating for each article was calculated as the average of the five authors' scores.

Article Selection
Figure 1 shows the steps of the article selection process used for the systematic review following the PRISMA guidelines [13].On a total of 1524 articles found in all searched databases (416 from PubMed, 706 from Scopus, and 402 from Web of Science); after duplicate deletion, 991 records were screened for inclusion; of the remaining studies, 891 were deleted after analyzing the title and abstract.Then, the full texts of 100 articles were assessed for eligibility and evaluated considering the inclusion and exclusion criteria.After the evaluation, 48 articles were excluded on the basis of the exclusion criteria and specifically for the following reasons: 19 articles because they were not specific to FA, 15 articles only considered environmental monitoring, 9 did not consider original data, 3 did not assess an occupationally exposed population, 1 because it was in a different language from English, and 1 because it was not found.Finally, 52 articles met the inclusion criteria, and were included in the analysis.

Main Characteristics of the Included Studies
The included articles were grouped based on the studied occupational scenarios, as follows: healthcare and research (Table 1), industrial (Table 2), and other settings or miscellanea of settings (Table 3).These groups were chosen on the basis of the results of a recent systematic review [6], which categorized those exposed to FA according to the activities carried out in the work environment.The included articles studied different categories of workers who were grouped as miscellanea of settings.Statistically significant higher frequency of micronuclei in the exposed group, both in peripheral blood lymphocytes and in epithelial buccal cells in the exposed group compared to control.Moderate positive correlation between years of exposure and frequency of micronuclei in peripheral blood lymphocytes and in epithelial cells

Good
In total, 21 articles out of a total of 52 included in the present review were focused on healthcare and research settings, 23 in industrial scenarios, and 8 in other settings.
Several factors extrinsic to occupational exposure can affect the FA exposure levels and the related frequency of observed early biological effects; hence, with the exception of one study [26], all the included studies considered some potential confounding and/or interfering factors on biomarkers measurements, such as age, sex, body mass index, current cigarette smoking status and alcohol consumption, recent infections, current use of medication, and duration of exposure.For this reason, in order to adjust the results related to the biomarkers of exposure or effects, these variables were included in models if they were significant at p < 0.05 or if there was evidence of confounding (e.g., greater than a 15% change in the regression coefficient).In the multiple linear regression analysis, adjustments were made for possible influence of such variables and selected biomarkers were subsequently tested using the adjusted multiple linear regression.

Discussion
The present systematic review was focused on the identification of biomarkers used for assessing the exposure to FA in occupational environments and the relative early negative effects for human health.
The first relevant result is related to the use of specific biomarkers of occupational exposure to FA.Indeed, just about one-sixth of the articles included have monitored an exposure biomarker using urinary concentrations of formic acid [20,35,47,48,61] or unmodified FA [26] or FA human serum albumin conjugate [28,50], and the results showed a statistically significant increase in the levels of these substances in the exposed group compared to control in only half of the studies [20,26,28,48].In particular, the urinary concentration of formic acid was found significantly higher in FA-exposure with respect to the control by Costa et al. [20] and by Peteffi et al. [11,48], but not by the other included studies.These contrasting results can be due to the influence of other individual or external factors beyond FA-exposure on the excretion of formic acid in the urine.For example, food composition can determine inter-and intra-individual daily fluctuations of urinary formic acid concentrations [64].Indeed, it has been reported that an excessive intake of proteins and carbohydrates seems to increase the formic acid excretion in urine [65].Besides this, age presents a positive correlation with the urinary levels of formic acid [20].Furthermore, it has been demonstrated that a non-negligible level of formic acid can be produced by endogenous FA [8].All these influences on the concentrations of formic acid make this substance a poor biological index for the human biomonitoring of exposure to FA.As regards urinary unmodified FA, Motta et al. [26] evidenced a potential FA weekly accumulation, but the same authors highlighted that the use of urinary FA as a biomarker of exposure poses severe limitations because of its very short half-life [26].Regarding FA human serum albumin conjugate, Ref. [28] recovered a statistically significant increase in this biomarker; in contrast, the results found by Regazzoni et al. [50] did not confirm the association between FA exposure and the formation of the adduct.This difference can be due to several factors, such as the levels and the duration of exposure, the methods used for measuring the adducts, the exposure to other substances as well as differences in the genetic polymorphism for specifying metabolizing enzymes.
Another important finding is linked to the biomarkers of early adverse effects for human health.Except for three studies [26,50,61] evaluating only biomarkers of exposure, the other included articles monitored also one or more biomarkers of effects in workers exposed to FA.The most common studied effect was the cytogenetic one by micronucleus assay in peripheral blood lymphocytes and/or epithelial buccal cells followed by the structural chromosomal aberrations and sister chromatide exchanges.Among the other studied early effects there were oxidative stress, DNA damage, and DNA protein crosslink.The results of almost all the included articles agree in demonstrating a statistically significant increase in early effect biomarkers in FA-exposed workers with respect to controls.These results confirm that occupational FA exposure is associated with a large number of early adverse effects and these risks should be carefully evaluated and managed.Besides this, these findings evidence that there are many effects related to FA exposure that are detectable early; consequently, these can be a useful tool for assessing the risks linked to this occupational exposure and for predicting the possibility of diseases, including cancer.A research agenda in this field is represented by the choice of which early effect indicators to use.In fact, there are some critical issues that need to be considered.
All the included studies take into account several confounding and/or interfering variables on the exposure to FA or on the levels of biomarkers, such as age, sex, body mass index, current cigarette smoking status and alcohol consumption, recent infections, current use of medication, and duration of exposure.As regard to sex and gender differences, a recent review highlighted that in the last few decades, given the increasing number of women workers, the difference between male and female populations in terms of occupational health has become evident and the effect of risk factors and work-related exposures is intrinsically different in male and female workers.However, further research in this field is needed in order to study in depth the extent of sex and gender differences in the context of occupational health [66].Besides this, for the interpretation of biomarkers of exposure and effects, the issue of confounding factors such as age, sex, and gender differences; genetic make-up; and other exogenous confounding factors including lifestyle habits such as smoking habit, alcohol consumption, and others should have been addressed and discussed in previous research, and they should be carefully taken into account in future studies [67].
In addition to the aforementioned interfering and confounding factors that can influence the levels of biological indices of effects, it must be considered that the indicators studied by the articles included in the present review are not only determined by exogenous FA, but also by endogenous FA and, therefore, it is necessary to find appropriate ways to exclude this quota produced within the organism.Furthermore, the identified indicators of early effect are not specific to the exposure to FA, but they can also form following exposure to other toxic and/or carcinogenic substances in workplaces.Several studies demonstrated that DNA damage and other early adverse effects can be determined by the occupational exposure to a large number of inorganic and organic compounds, such as cadmium [68], cytostatic/antineoplastic drugs [69], benzene and other polycyclic aromatic hydrocarbons [70], organochlorides [71], and others.
Regarding the setting groups, we found no differences in the results of the studies involving healthcare and research or industrial or other scenarios.This result is in line with the findings of a previous review [6], reporting that airborne FA was found at concentrations higher than outdoors in almost all the studied scenarios/activities.
The results of this systematic review add some scientific evidence on the use of the biomarkers for evaluating the exposure and the related early effects related to occupational exposure to FA.This evidence can support appropriate choices in the risk assessment and management process, necessary for helping to achieve the target of "promoting a safe and protected working environment for all workers" in the Sustainable Development Goal 8 of the UN Agenda 2030 for Sustainable Development.
This systematic review has some limitations.First of all, only studies published in the English language were considered, excluding articles published in other languages a priori.Besides this, a formal meta-analysis was not carried out because the results of the articles included in the review were different in term of populations, number of participants (from 16 to 399), study period (ranging from 3 weeks to 13 years), working context, study design, methodological approach for estimating exposure and effects, confounding and/or interfering factors considered, and kinds of biomarkers of exposure or effects investigated.Thus, publication bias and statistical heterogeneity were not assessed.This choice is supported by Cochrane, who recently declared that meta-analysis should be considered only when the studies are adequately homogeneous for participants, interventions, and outcomes [72].However, in our knowledge, this is the first systematic review of the scientific literature that gives a complete picture of biomarkers of exposure and early effects used in the assessment of FA exposure in occupational settings.

Conclusions
The results of the included articles evidenced that the use of biomarkers of exposure for assessing occupational exposure to FA is under debate.Further studies are needed to find suitable biological indicators of FA exposure.
Almost all the studies show a statistically significant increase in early effect biomarkers (particularly cytogenetic assays) in FA-exposed workers with respect to controls demonstrating their usefulness for biomonitoring studies of exposed workers.FA exposure in working contexts should be eliminated or reduced as much as possible thanks to the use of individual and collective protective equipment and mitigation strategies.Besides this, biomarkers of early effect can be used for evaluating the health hazards to human health in a very early and reversible phase.In particular, micronucleus assays on buccal cells are interesting and promising for their sensitivity and also because of their non-invasiveness, which makes them easily usable and acceptable to workers.
Additional efforts must be made to eliminate the effect due to simultaneous exposure to other toxic and carcinogenic substances and the influence of confounding and interfering factors.

Figure 1 .
Figure 1.Flow diagram of study selection.

Figure 1 .
Figure 1.Flow diagram of study selection.
, low risk of bias); fair quality (1 criterion not met or 2 criteria unclear, moderate risk of bias); poor quality (2 or more criteria not met, high risk of bias).For case-control and cohort studies: good quality (3 or 4 criteria in selection domain, 1 or 2 criteria in comparability domain, and 2 or 3 criteria in outcome domain); fair quality (2 criteria in selection domain, 1 or 2 criteria in comparability domain, and 2 or 3 criteria in outcome domain); poor quality (0 or 1 criterion in selection domain, 0 criteria in comparability domain, or 0 or 1 criteria in outcome domain). met

Table 1 .
Main characteristics of studies (n = 21) involving healthcare and research settings included in the systematic review.

Country Working Context Sample Size Age (Mean Value ± SD and/or Range) Gender (%) Biomarkers of Internal Dose and/or Early Effect Confounding and Interfering Factors Main Results Quality Evaluation According to NOS Scale
38 ± 9 years; 20 males (33%) and 40 females (67%) Biomarkers of effect: micronucleus frequencies, sister chromatid exchanges, comet tail length in peripheral lymphocytes Age, gender, smoking habits, years of employment Statistically significant increases in micronucleus frequency, sister chromatid exchanges, and comet tail length in the exposed group compared to controls

Table 2 .
Main characteristics of studies (n = 23) involving industrial settings included in the systematic review.

Table 3 .
Main characteristics of studies (n = 8) involving other settings or miscellanea of settings included in the systematic review.