Nothing Special   »   [go: up one dir, main page]
Next Article in Journal
Iodine Deficiency Exacerbates Thyroidal and Neurological Effects of Developmental Perchlorate Exposure in the Neonatal and Adult Rat
Previous Article in Journal
Microplastics Exposure Aggravates Synovitis and Pyroptosis in SLE by Activating NF-κB and NRF2/KEAP1 Signaling
Previous Article in Special Issue
Micro/Nanoplastic Exposure on Placental Health and Adverse Pregnancy Risks: Novel Assessment System Based upon Targeted Risk Assessment Environmental Chemicals Strategy
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Toxics
Volume 12
Issue 12
10.3390/toxics12120841
toxics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Is Polyhexamethylene Guanidine and Oligo(2-(2-Ethoxy) Ethoxyethyl Guanidium Chloride Exposure Related to Gestational Diabetes?

by
Hyowon Choi
1,
Nam-Yun Kim
2,
Nalai Kim
2 and
Yeon-Soon Ahn
1,3,*
1
Department of Prevention Medicine, Yonsei University Wonju College of Medicine, Wonju 26426, Republic of Korea
2
Department of Obstetric and Gynecology, Yonsei University Wonju College of Medicine, Wonju 26426, Republic of Korea
3
Genomic Cohort Institute, Yonsei University Wonju College of Medicine, Wonju 26426, Republic of Korea
*
Author to whom correspondence should be addressed.
Toxics 2024, 12(12), 841; https://doi.org/10.3390/toxics12120841
Submission received: 23 September 2024 / Revised: 10 November 2024 / Accepted: 14 November 2024 / Published: 22 November 2024
(This article belongs to the Special Issue Identification of Toxic Effects of Emerging Chemicals throughout the Reproductive Process)
Browse Figures
Figure 1
<p>Study flowcharts. KEITI, Korean Environmental Industry and Technology Institute; NHIS, National Health Insurance Services; DM, diabetes mellitus; GDM, gestational diabetes mellitus.</p> "> Figure 2
<p>Direct acyclic graph of this study. PHMG/PGH, polyhexamethylene guanidine/oligo(2-(2-ethoxy)ethoxyethyl guanidium chloride; BMI, body mass index; DM, diabetes mellitus; GDM, gestational diabetes mellitus.</p> "> Figure 3
<p>Results of mediation analysis (adjusted model). ACME, average causal mediation effect; ADE, average direct effect.</p> ">
Versions Notes

Abstract

:
This study aims to investigate the association between exposure to toxic indoor chemicals, specifically polyhexamethylene guanidine (PHMG) and oligo(2-(2-ethoxy) ethoxyethyl guanidinium) chloride (PGH), used in humidifier disinfectants, and gestational diabetes mellitus (GDM). We confirmed pregnancy from 2003 to 2017 and identified GDM by linking a cohort of claimants who reported exposure to PHMG/PGH with National Health Insurance Service data. The GDM incidence was calculated, and PHMG/PGH exposure characteristics—exposure status, the humidifier’s distance/location, and exposure duration/hours—were investigated. Logistic regression and mediation analysis were applied using asthma, frequently treated with steroids, as a mediator. Among 521 pregnancies, 38 were identified as GDM, with 2.4% before exposure and 8.9% after exposure. Pregnancies after exposure had a higher odds ratio (OR) for GDM (OR 2.968, 95% CI: 1.004–12.725). A trend of increased GDM risk was observed with longer exposure duration/hours. Additionally, pregnancies after exposure demonstrated total and direct effects on GDM (β = 0.0435, p = 0.036, β = 0.0432, p = 0.030) independent of the indirect effects by asthma. The incidence of GDM was higher after PHMG/PGH exposure compared to before. PHMG/PGH exposure was associated with GDM, independent of asthma. Further research is warranted to confirm these findings in exposed cohorts and to explore the underlying mechanisms.

1. Introduction

Gestational diabetes mellitus (GDM), the most common pregnancy complication [1], leads to several maternal complications, including an increased risk of developing type 2 diabetes and cardiovascular diseases. Additionally, there may be adverse effects on the child’s health [2]. It typically results from β-cell dysfunction against a background of chronic insulin resistance during pregnancy [2]. Chemical exposures, including endocrine disruptors, fine particulate matter, and organic pollutants, are recognized for damaging pancreatic β-cells and contributing to the development of GDM [3,4]. Humidifier disinfectants (HDs), a class of indoor toxicants, have been associated with severe lung diseases, including HD-associated lung injury and interstitial lung disease (ILD) [5]. HDs were widely used in Korea, particularly among children and pregnant women. According to Chang et al., among 1144 pregnant women who responded to questionnaires, 28.2% of subjects reported using a humidifier [6]. Despite the widespread use of HDs among pregnant women, human epidemiological studies examining the impact of HD exposure on pregnancy complications, including GDM, are lacking.
Recent toxicologic studies suggest that HDs can have systemic effects beyond the lungs, including metabolic disorders and reproductive toxicities [7,8,9]. This mechanism may also be associated with the development of GDM. Also, asthma, a respiratory condition strongly associated with HD exposure [10], is itself a recognized risk factor for GDM due to chronic inflammation or hormonal changes [11]. In addition, asthma is often treated with steroids [12], which are well-known to induce glucose intolerance [13] and further increase the risk of GDM [14]. Given the high prevalence of asthma among individuals exposed to HDs, the relationship between asthma and GDM in this population warrants further investigation.
Most HD brands contained polyhexamethylene guanidine/oligo(2-(2-ethoxy)ethoxyethyl guanidium chloride(PHMG/PGH) and a mixture of chloromethylisothiazolinone (CMIT) and methylisothiazolinone (MIT) [15]. In in vivo studies with a maximum observation period of 7 days, guanidine-type compounds, such as PHMG and PGH, demonstrated a retention rate of over 66% in the body when inhaled [16]. In contrast, CMIT/MIT compounds showed a retention rate of 22% in the body when inhaled [17]. Furthermore, in vivo studies have reported that PHMG/PGH contributes to pregnancy-induced toxicities [18]. Given these differences, our study focuses on guanidine-type compounds, particularly PHMG and PGH.
In this study, we aimed to (1) estimate the incidence of GDM by PHMG/PGH exposure status; (2) investigate the association between GDM and various exposure characteristics, including pregnancy affected by PHMG/PGH exposure, cumulative exposure duration/hour, and humidifier location/distance; and (3) explore the effect of PHMG/PGH exposure on GDM through both direct pathways and indirect pathways mediated by asthma using mediation analysis based on a causal diagram.

2. Materials and Methods

2.1. Study Populations

From 2011 to 2020, a total of 7056 claimants registered with the Korean Environmental Industry and Technology Institute (KEITI) for compensation related to damages caused by HDs. The data for these claimants have been systematically documented and stored within KEITI’s comprehensive portal for HD damage support [19]. We selected subjects who reported exposure to PHMG/PGH. By linking this database with the Korean National Health Insurance Service (NHIS) claims data, we tracked the medical utilization of these subjects from 2002 to 2021. Indexes of deliveries were identified using the International Classification of Diseases, 10th Revision (ICD-10), codes O8*, in combination with inpatient records and delivery procedure codes. The conception date for each index delivery was estimated by subtracting 266 days (38 weeks) from the date of delivery, and this duration was defined as the pregnancy episode [20,21]. We selected pregnancy episodes with conception dates between January 2002 and December 2016, which correspond to delivery dates between January 2003 and December 2017, reflecting the five years following the cessation of HD sales.
Episodes were excluded if they had twin or preterm deliveries (26 episodes), if they had been diagnosed with diabetes mellitus (DM) prior to conception (4 episodes), if they had incomplete exposure surveys (12 episodes), or if the date of the index delivery was more than five years after exposure cessation (50 episodes), as severe pregnancy outcomes were observed to manifest within a latency period after HD exposure, with a decreasing trend after 2017 [22]. Ultimately, 521 episodes and 374 subjects were included in the analysis (Figure 1).

2.2. Outcomes (GDM) and Mediator (Asthma)

In Korea, all pregnant women between 24 and 28 weeks of pregnancy are screened for gestational diabetes mellitus (GDM). From 2003 to 2012, the two-step method was primarily used for GDM diagnosis. However, following the results of the international Hyperglycemia and Adverse Pregnancy Outcome [23], both the one-step and the two-step methods have been utilized since 2013 [24]. As data on the oral glucose tolerance test are not available in the NHIS claims data, GDM was defined as having been diagnosed with the ICD-10 codes O244 and O249, recorded at least twice during pregnancy episodes.
Asthma was defined as having at least two occurrences based on the ICD-10 codes J45 and J46. It was considered if diagnosed before 28 weeks of pregnancy or prior to the GDM diagnosis.

2.3. Assessment of Exposure Characteristic

Exposure data related to PHMG/PGH were obtained from KEITI’s comprehensive portal for HD damage support. The data were collected through a self-administered questionnaire followed by a 1:1 interview with victims who reported damage [19]. Subjects’ exposure was verified using HD receipts and photos of HD purchases.
We classified pregnancy episodes based on PHMG/PGH exposure status as either “after-exposure episodes” if PHMG/PGH exposure began before GDM diagnosis in GDM cases or before 28 weeks of pregnancy in control episodes. For “non-exposure episodes”, we included episodes without any reported PHMG/PGH exposure during the relevant time period.
To identify which exposure characteristics were associated with GDM, we investigated the distance from the humidifier and the humidifier’s location. To estimate the exposure duration, we calculated the cumulative exposure duration. For the GDM, we determined the cumulative exposure duration before the GDM diagnosis. For the control, we determined the cumulative exposure duration before 28 weeks of pregnancy. And, we computed the cumulative exposure time (hours) by multiplying the years of use (year), annual use (months/year), monthly use (weeks/month), daily use (days/week), and hourly use (hours/day). Also, we computed the indoor air concentrations, suggested by Ryu et al. [25], by multiplying the used HD product concentration by the daily average usage amount and dividing by the volume of the space where it was used. For cases where measuring the space volume was not feasible, we substituted with the mean volume from other measured spaces in this study. We then categorized these three variables—cumulative exposure duration, cumulative exposure hours, and concentration—into terciles for analysis.

2.4. Covariates and Study Modeling

Potential confounders were selected using a directed acyclic graph (DAG) to identify variables that could influence the relationship between the main exposure and the outcome, thus reducing bias in the study [26] (Figure 2, Table S1). Known risk factors for GDM, including maternal age at delivery, parity, educational level, smoking status, and urbanization, were considered as confounders [27,28]. The year of conception (≤2013, >2013) was also included due to its correlation with the change in the diagnosis method for GDM [24]. Given the seasonality of humidifier use (prevalent between fall and spring), the season of delivery was categorized into four groups: winter (December to February), spring (March to May), summer (June to August), and fall (September to November).
Educational level was dichotomized into “higher than high school” or not. Smoking status was categorized as either a history of smoking (ex and current smokers) or never smokers. Rural areas were defined based on the Korean residential zoning classification, with regions classified as “eup”, “myeon”, or “ri” considered rural.

2.5. Statistical Analysis

Data analysis was conducted using R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria). Descriptive statistics (mean ± standard deviation for continuous variables and proportions for categorical variables) were calculated for all episode characteristics. The incidence of GDM (per 100 pregnancies) was estimated and stratified by PHMG/PGH exposure status. Incidence cases were identified as GDM diagnoses, and the incidence rate was calculated by dividing the number of incident cases by the target episodes. We calculated the incidence of GDM by performing age standardization using the age structure (≥30, or <30) of pregnancy episodes in the non-exposure.
Comparative analyses of sociodemographic factors, PHMG/PGH exposure status, exposure characteristics, and comorbidities between GDM and control episodes were conducted using t-tests and chi-square tests. Fisher’s exact test was used for variables with a sample size of less than 10. Logistic regression analysis was performed to examine the relationship between PHMG/PGH exposure status/characteristics and GDM.
In the mediation analysis, PHMG/PGH exposure status was considered the exposure, asthma the mediator, and GDM the outcome. All analyses were conducted using both crude and adjusted models. The crude model included only the exposure and the outcome, while the adjusted model included the age at delivery, urbanization, educational level, and smoking status as covariates, based on the DAG. In logistic regression, we adjusted the mediator and the selected confounders. The mediation analysis was performed using the R package “mediation” (version 4.4.7) [29] by employing bootstrapping with 1000 simulations to perform the analysis. Statistical significance was determined at p < 0.05 for all analyses.
This study used retrospective cohort data, and there may be unmeasurable variables that may affect causal assumptions. Therefore, we used the same R package to perform the mediation analysis. In the sensitivity analysis, we calculated the correlation (Rho) between the mediator and the outcome variables, as well as the product of the R-squared values between the mediator and the outcome variables (R2M × R2Y). We specified a range of values for Rho and R2M × R2Y, and for each combination of values, we estimated the average causal mediation effect (ACME) using 1000 simulations. Furthermore, due to the limited number of GDM cases, we conducted an additional sensitivity analysis with 1:4 propensity score matching based on selected confounders, followed by conditional logistic regression.

3. Results

Table 1 presents the characteristics of the study participants. The participants had a mean birth year of 1976.3 ± 4.4 years, with a total exposure duration of 39.1 ± 32.8 months. Among the participants, 12.3% resided in rural areas, and only 2.4% reported a history of smoking. Additionally, 14.7% had a pre-pregnancy diagnosis of asthma.
The incidence of GDM was 7.3 per 100 pregnancies overall. In non-exposure pregnancy episodes, the incidence of GDM was 2.4 per 100 pregnancies. In pregnancy episodes after PHMG/PGH exposure, the incidence was 8.9 per 100 pregnancies. Based on age standardization of the data using the age structure of women with deliveries in non-exposure, the incidence of GDM was 2.4 per 100 pregnancies in non-exposure cases and 8.2 per 100 pregnancies among after-exposure cases (Table 2).
The characteristics of the study episodes were as follows: the mean age was 31.2 ± 3.8 years, the mean delivery year was 2008.2 ± 3.3, the cumulative total exposure duration was 18.7 ± 23.5 months, the mean cumulative exposure time was 7135.6 ± 11,226.1 h, and the mean concentration of PHMG/PGH was 1.3 ± 2.7 μg/m3. A total 4.2% of episodes occurred after 2013, 12.1% of episodes occurred in rural areas, 2.3% of episodes were in ever-smokers, and 28.8% of episodes involved participants with education levels at or below high school. In total, 73.5% of participants reported the humidifier being located “less than 1 m” away, and 73.1% reported the humidifier being positioned “close to the nose and mouth”. The proportion of asthma was 10.6%. A total of 38 episodes (7.3%) were classified as GDM cases. Compared to the control group, significant differences were observed in the age, delivery year, year of conception, exposure status, cumulative exposure hour distribution, and tercile of the concentration of PHMG/PGH (p = 0.019, p < 0.001, p = 0.001, p = 0.024, p = 0044, and p = 0.012, respectively; Table 3).
Table 4 presents the logistic regression results. After-exposure episodes showed increased odds ratios in both crude and adjusted models (crude OR: 4.062, 95% CI: 1.431–17.062; adjusted OR: 2.698, 95% CI: 1.004–12.725). Cumulative exposure duration and cumulative exposure hours were not individually statistically significant. However, the p for the trend for cumulative exposure duration was significant (adjusted p = 0.009), while the trend for cumulative exposure hours approached significance (adjusted p = 0.059). Also, the second tercile value of concentration showed increased odd ratios in both crude and adjusted models (crude OR: 3.6, 95% CI: 1.535~9.118; adjusted OR: 3.059, 95% CI: 1.26~7.956)
In the mediation analysis, the crude model showed that the average direct effect (ADE) of exposure status on GDM was statistically significant (β = 0.0650, 95% CI: 0.0276–0.10), as was the total effect (β = 0.0654, 95% CI: 0.0292–0.10). However, the ACME was not significant. In the adjusted model, both the ADE and the total effect remained statistically significant (β = 0.0432, 95% CI: 0.0050–0.08; β = 0.0435, 95% CI: 0.0057–0.08), but the ACME was still not significant (Table 4, Figure 3).
A sensitivity analysis confirmed that the mediation effect of exposure on GDM was insufficient (Table S2). A sensitivity analysis confirmed that the results of the propensity score matching showed no statistically significant values (Tables S3 and S4). However, the direction of the OR values was consistent with the main analysis, and the p for the trend for cumulative exposure duration was significant at 0.017.

4. Discussion

This study demonstrated a higher incidence of GDM in pregnancy episodes after PHMG/PGH exposure compared to non-exposure episodes, indicating a potential association between PHMG/PGH exposure and GDM. Specifically, increasing cumulative exposure duration and cumulative exposure hours showed a trend toward increasing GDM risk. A moderate exposure concentration also demonstrated a significant association with GDM. Additionally, PHMG/PGH exposure demonstrated a positive total effect and β value for GDM, independent of asthma mediation.
We conducted this analysis using NHIS claim data, which cover 97% of the Korean population and include detailed information about the age, sex, residential regions, medical services, and diagnoses classified through ICD-10 codes [30]. The NHIS claims data, consistently used since 2002, likely captured most GDM cases in our analysis. However, the administrative nature of the NHIS claim data may limit the accuracy of diagnoses. To enhance the accuracy of GDM diagnoses in this study, unlike in other studies [31,32], we additionally confirmed at least two occasions.
A study based on data from the Health Insurance Review and Assessment Service reported that GDM occurred in 7.5% of all pregnancies in Korea between 2009 and 2011 [32]. Similarly, in our study, the incidence was 7.294 per 100 pregnancies, indicating that the analyzed claimants were likely representative of the general population and not subject to selection bias. Nonetheless, when stratified by non-exposure and after-exposure status, we observed a higher incidence of GDM in the after-exposure group, suggesting a potential link between PHMG/PGH exposure and GDM. Furthermore, when the after-exposure group was divided into exposure overlapping pregnancy episodes and pregnancy episodes after exposure had ended, age-standardized data revealed increasing trends in GDM incidence of 6.0% in the exposure overlapping group and 17.1% in the pregnancy episodes after exposure end group (Table S5). Notably, only the pregnancy episodes after exposure ended occurred after 2013, when Korea began using the one-step diagnosis method for GDM alongside the two-step method [24]. According to Chang et al., the incidence of GDM increased from 12.0% in 2013 to 14.9% in 2016, which aligns with the trends observed in our study [33]. Notably, no cases of DM progression were observed among individuals diagnosed with GDM in this study.
In our study, only one episode of ILD was identified, and therefore it was not included as a mediator. Despite the well-established association between ILD and PHMG/PGH exposure, the low incidence observed here may be due to the high risk of adverse pregnancy outcomes—such as miscarriage or preterm birth—when ILD is diagnosed during pregnancy. This could lead to early termination of pregnancies with higher exposure levels, resulting in a potential underrepresentation of high-level PHMG/PGH exposure and ILD cases among full-term pregnant women. Consequently, this may partially explain why the effects of high-concentration exposure are not fully captured in our findings [10].
Our findings also suggest that PHMG/PGH exposure exerts a direct effect on GDM development independent of asthma associated with steroid use and asthma itself. The primary mechanisms driving GDM development involve β-cell dysfunction and chronic insulin resistance during pregnancy [2]. While some chemicals are known to damage or disrupt β-cell function, there is no in vivo study to date directly linking PHMG/PGH to β-cell impairment [4]. However, Ahn et al. reported that women and children exposed to PHMG 5 to 7 years prior had lower serum levels of cholesterol-carrying proteins (APOA4, APOB, APOC1-3, APOE, APOF, APOJ, APOM, and SAA4) and HDL compared to control groups [34]. Low plasma levels of these apolipoproteins and HDL are known to impair lipid metabolism, contributing to insulin resistance [35].
One of the strengths of our study is the comparison of GDM incidence before and after PHMG/PGH exposure, which allowed us to demonstrate the impact of PHMG/PGH on GDM development. Additionally, this is the first study to demonstrate a dose–response relationship and to highlight the direct effect of PHMG/PGH exposure on GDM without asthma mediation. Importantly, we utilized a DAG to guide our analysis, thus ensuring that we avoided over-adjustment, which further strengthens the validity of our findings.
Despite the insights provided by this study, several limitations should be considered. First, we lacked access to data on certain potential confounders, such as pre-pregnancy BMI and family history, which may influence outcomes. However, we confirmed through a DAG that these variables are unlikely to act as confounders in this study. Second, the retrospective nature of the PHMG/PGH exposure data could introduce information bias, although this bias is likely non-differential as it affects both the GDM and control groups equally. Third, our analysis was based on data from individuals who submitted damage claims, which may introduce bias. To mitigate the potential for selection bias from external comparisons, we focused on internal comparisons in this study. Lastly, as this is a retrospective cohort study, it can only establish associations between PHMG/PGH exposure and GDM, not causation. Future experimental studies examining pathophysiology or research with individuals with known exposure histories are necessary to clarify underlying mechanisms.

5. Conclusions

In conclusion, this study is the first to demonstrate an association between cumulative PHMG/PGH exposure and GDM, showing that PHMG/PGH directly affects GDM risk without asthma mediation. These findings underscore the need for further research to address potential confounding factors and to conduct mechanistic studies in both animals and humans to further elucidate the GDM risk associated with PHMG/PGH exposure.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/toxics12120841/s1, Table S1: Code of daggity; Table S2: Sensitivity analysis of mediation analysis; Table S3: Characteristics of study episode after propensity score matching; Table S4: Conditional logistic regression analysis after PSM; Table S5: Crude and age standardization incidence of GDM, further stratified after exposure.

Author Contributions

Conceptualization, H.C.; Methodology, H.C.; Validation, Y.-S.A.; Formal Analysis, H.C.; Writing—Original Draft Preparation, H.C.; Writing—Review and Editing, N.-Y.K., N.K. and Y.-S.A.; Visualization, H.C.; Funding Acquisition, Y.-S.A. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Wonju Severance Christian Hospital (CR323316).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The datasets presented in this article are not readily available due to the privacy of the participants. Requests to access the datasets should be directed to Y.-S.A.

Acknowledgments

We would like to express our gratitude to the National Institute of Environmental Research (NIER), South Korea, for their invaluable support in facilitating this research. All individuals or institutions included in this section have provided their consent to be acknowledged.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. American Diabetes Association. 2. Classification and diagnosis of diabetes: Standards of medical care in diabetes—2018. Diabetes Care 2018, 41, S13–S27. [Google Scholar] [CrossRef] [PubMed]
  2. Plows, J.F.; Stanley, J.L.; Baker, P.N.; Reynolds, C.M.; Vickers, M.H. The Pathophysiology of Gestational Diabetes Mellitus. Int. J. Mol. Sci. 2018, 19, 3342. [Google Scholar] [CrossRef] [PubMed]
  3. Eberle, C.; Stichling, S. Environmental health influences in pregnancy and risk of gestational diabetes mellitus: A systematic review. BMC Public Health 2022, 22, 1572. [Google Scholar] [CrossRef] [PubMed]
  4. Varshavsky, J.; Smith, A.; Wang, A.; Hom, E.; Izano, M.; Huang, H.; Padula, A.; Woodruff, T.J. Heightened susceptibility: A review of how pregnancy and chemical exposures influence maternal health. Reprod. Toxicol. 2020, 92, 14–56. [Google Scholar] [CrossRef] [PubMed]
  5. Lee, M.-S.; Kim, H.J. Epidemiologic research on lung damage caused by humidifier disinfectants. Epidemiol. Health 2016, 38, e2016031. [Google Scholar] [CrossRef]
  6. Chang, M.H.; Park, H.; Ha, M.; Kim, Y.; Hong, Y.C.; Ha, E.H. Characteristics of Humidifier Use in Korean Pregnant Women: The Mothers and Children’s Environmental Health (MOCEH) Study. Environ. Health Toxicol. 2012, 27, e2012003. [Google Scholar] [CrossRef]
  7. Song, J.-H.; Ahn, J.; Park, M.Y.; Park, J.; Lee, Y.M.; Myong, J.-P.; Koo, J.-W.; Lee, J. Health Effects Associated with Humidifier Disinfectant Use: A Systematic Review for Exploration. J. Korean Med. Sci. 2022, 37, e257. [Google Scholar] [CrossRef]
  8. Kim, Y.; Choi, J. Early life exposure of a biocide, CMIT/MIT causes metabolic toxicity via the O-GlcNAc transferase pathway in the nematode C. elegans. Toxicol. Appl. Pharmacol. 2019, 376, 1–8. [Google Scholar] [CrossRef]
  9. Lee, J.; Jeong, J.S.; Kim, S.Y.; Im, W.J.; Shin, Y.J.; Lee, K.; Choi, S.J.; Heo, Y.; Jeong, E.J.; Nam, S.Y.; et al. Reproductive and developmental toxicity screening of polyhexamethylene guanidine phosphate by oral gavage in rats. Regul. Toxicol. Pharmacol. 2019, 108, 104440. [Google Scholar] [CrossRef]
  10. Ministry of Environment. Special Law for Relieving Damage from Humidifier Disinfectants. Sejong: Ministry of Environment. Available online: https://www.me.go.kr/home/web/board/read.do?menuId=286&boardId=807010&boardMasterId=1 (accessed on 21 August 2024).
  11. Jensen, M.E.; Barrett, H.L.; Peek, M.J.; Gibson, P.G.; Murphy, V.E. Maternal asthma and gestational diabetes mellitus: Exploration of potential associations. Obstet. Med. 2021, 14, 12–18. [Google Scholar] [CrossRef]
  12. Rajvanshi, N.; Kumar, P.; Goyal, J.P. Global Initiative for Asthma Guidelines 2024: An Update. Indian Pediatr. 2024, 61, 781–786. [Google Scholar] [CrossRef] [PubMed]
  13. Van Raalte, D.; Diamant, M. Steroid diabetes: From mechanism to treatment. Neth. J. Med. 2014, 72, 62–72. [Google Scholar] [PubMed]
  14. Yildirim, Y.; Tinar, S.; Oner, R.S.; Kaya, B.; Toz, E. Gestational diabetes mellitus in patients receiving long-term corticosteroid therapy during pregnancy. J. Perinat. Med. 2006, 34, 280–284. [Google Scholar] [CrossRef] [PubMed]
  15. Park, D.-U.; Ryu, S.-H.; Lim, H.-K.; Kim, S.-K.; Choi, Y.-Y.; Ahn, J.-J.; Lee, E.; Hong, S.-B.; Do, K.-H.; Cho, J.-l.; et al. Types of household humidifier disinfectant and associated risk of lung injury (HDLI) in South Korea. Sci. Total Environ. 2017, 596–597, 53–60. [Google Scholar] [CrossRef]
  16. Mushtaq, S.; Park, J.E.; Shim, H.E.; Lee, C.H.; Shin, H.S.; Lee, S.-Y.; Jeon, J. Study on biological distribution of polyhexamethylene guanidine (PHMG), a toxic household chemical, using radiolabeling and molecular imaging tools. Environ. Eng. Res. 2022, 27, 210393. [Google Scholar] [CrossRef]
  17. Song, M.-K.; Eun Park, J.; Ryu, S.-H.; Baek, Y.-W.; Kim, Y.-H.; Im Kim, D.; Yoon, S.-H.; Shin, H.; Jeon, J.; Lee, K. Biodistribution and respiratory toxicity of chloromethylisothiazolinone/methylisothiazolinone following intranasal and intratracheal administration. Environ. Int. 2022, 170, 107643. [Google Scholar] [CrossRef]
  18. Lee, J.; Choi, S.-J.; Jeong, J.-S.; Kim, S.Y.; Lee, S.-H.; Yang, M.J.; Lee, S.-J.; Shin, Y.-J.; Lee, K.; Jeong, E.J. A humidifier disinfectant biocide, polyhexamethylene guanidine phosphate, inhalation exposure during pregnancy induced toxicities in rats. J. Hazard. Mater. 2021, 404, 124007. [Google Scholar] [CrossRef]
  19. Korea Environmental Industry & Technology Institute. Korea Environmental Industry & Technology Institute Humidifier Disinfectant Damage Support Site. Available online: https://healthrelief.or.kr/home/main.do (accessed on 20 August 2024).
  20. Kim, H.O.; Lee, K.; Lee, S.M.; Seo, G.H. Association between pregnancy outcomes and radioactive iodine treatment after thyroidectomy among women with thyroid cancer. JAMA Intern. Med. 2020, 180, 54–61. [Google Scholar] [CrossRef]
  21. Park, K.Y.; Kwon, H.J.; Wie, J.H.; Lee, H.H.; Cho, S.B.; Kim, B.J.; Bae, J.M. Pregnancy outcomes in patients with vitiligo: A nationwide population-based cohort study from Korea. J. Am. Acad. Dermatol. 2018, 79, 836–842. [Google Scholar] [CrossRef]
  22. Kim, J.Y.; Lee, J.H.; Park, S.Y.; Joo, Y.S.; Byun, G.H.; Lee, N.K.; Heo, S.Y.; Kim, H.S.; Song, S.H.; Choi, J.Y.; et al. A Study on Health Damage of Humidifier Disinfectants Using National Health Insurance Big Data; 2 National Institute of Environmental Research: Incheon, Republic of Korea, 2020. (In Korean) [Google Scholar]
  23. Metzger, B.E.; Lowe, L.P.; Dyer, A.R.; Trimble, E.R.; Chaovarindr, U.; Coustan, D.R.; Hadden, D.R.; McCance, D.R.; Hod, M.; McIntyre, H.D.; et al. Hyperglycemia and adverse pregnancy outcomes. N. Engl. J. Med. 2008, 358, 1991–2002. [Google Scholar] [CrossRef]
  24. Hur, K.Y. New Diagnostic Criteria for Gestational Diabetes Mellitus and Pregnancy Outcomes in Korea. Diabetes Metab. J. 2019, 43, 763–765. [Google Scholar] [CrossRef] [PubMed]
  25. Ryu, H.; Choi, Y.-H.; Kim, E.; Park, J.; Lee, S.; Yoon, J.; Jo, E.-K.; Choe, Y.; Heo, J.; Yang, W. Misclassification and characterization of exposure to humidifier disinfectants using a questionnaire. BMC Public Health 2021, 21, 1458. [Google Scholar] [CrossRef] [PubMed]
  26. Textor, J.; Van der Zander, B.; Gilthorpe, M.S.; Liśkiewicz, M.; Ellison, G.T. Robust causal inference using directed acyclic graphs: The R package ‘dagitty’. Int. J. Epidemiol. 2016, 45, 1887–1894. [Google Scholar] [CrossRef] [PubMed]
  27. Giannakou, K.; Evangelou, E.; Yiallouros, P.; Christophi, C.A.; Middleton, N.; Papatheodorou, E.; Papatheodorou, S.I. Risk factors for gestational diabetes: An umbrella review of meta-analyses of observational studies. PLoS ONE 2019, 14, e0215372. [Google Scholar] [CrossRef] [PubMed]
  28. Zhang, C.; Rawal, S.; Chong, Y.S. Risk factors for gestational diabetes: Is prevention possible? Diabetologia 2016, 59, 1385–1390. [Google Scholar] [CrossRef]
  29. Tingley, D.; Yamamoto, T.; Hirose, K.; Keele, L.; Imai, K. Mediation: R package for causal mediation analysis. J. Stat. Softw. 2014, 59, 1–38. [Google Scholar] [CrossRef]
  30. Shin, D.W.; Cho, B.; Guallar, E. Korean national health insurance database. JAMA Intern. Med. 2016, 176, 138. [Google Scholar] [CrossRef]
  31. Sun, J.; Kim, G.R.; Lee, S.J.; Kim, H.C. Gestational diabetes mellitus and the role of intercurrent type 2 diabetes on long-term risk of cardiovascular events. Sci. Rep. 2021, 11, 21140. [Google Scholar] [CrossRef]
  32. Koo, B.K.; Lee, J.H.; Kim, J.; Jang, E.J.; Lee, C.-H. Prevalence of gestational diabetes mellitus in Korea: A national health insurance database study. PLoS ONE 2016, 11, e0153107. [Google Scholar] [CrossRef]
  33. Jung, C.-H.; Jung, S.-H.; Choi, D.; Kim, B.-Y.; Kim, C.-H.; Mok, J.-O. Gestational diabetes in Korea: Temporal trends in prevalence, treatment, and short-term consequences from a national health insurance claims database between 2012 and 2016. Diabetes Res. Clin. Pract. 2021, 171, 108586. [Google Scholar] [CrossRef]
  34. Ahn, H.-S.; Lee, S.-Y.; Kang, M.-J.; Hong, S.B.; Song, J.W.; Do, K.H.; Yeom, J.; Yu, J.; Oh, Y.; Hong, J.Y.; et al. Polyhexamethylene guanidine aerosol causes irreversible changes in blood proteins that associated with the severity of lung injury. J. Hazard. Mater. 2024, 478, 135359. [Google Scholar] [CrossRef] [PubMed]
  35. Baldi, S.; Bonnet, F.; Laville, M.; Morgantini, C.; Monti, L.; Hojlund, K.; Ferrannini, E.; Natali, A. Influence of apolipoproteins on the association between lipids and insulin sensitivity: A cross-sectional analysis of the RISC Study. Diabetes Care 2013, 36, 4125–4131. [Google Scholar] [CrossRef] [PubMed]
Toxics 12 00841 g001
Figure 1. Study flowcharts. KEITI, Korean Environmental Industry and Technology Institute; NHIS, National Health Insurance Services; DM, diabetes mellitus; GDM, gestational diabetes mellitus.
Figure 1. Study flowcharts. KEITI, Korean Environmental Industry and Technology Institute; NHIS, National Health Insurance Services; DM, diabetes mellitus; GDM, gestational diabetes mellitus.
Toxics 12 00841 g001
Toxics 12 00841 g002
Figure 2. Direct acyclic graph of this study. PHMG/PGH, polyhexamethylene guanidine/oligo(2-(2-ethoxy)ethoxyethyl guanidium chloride; BMI, body mass index; DM, diabetes mellitus; GDM, gestational diabetes mellitus.
Figure 2. Direct acyclic graph of this study. PHMG/PGH, polyhexamethylene guanidine/oligo(2-(2-ethoxy)ethoxyethyl guanidium chloride; BMI, body mass index; DM, diabetes mellitus; GDM, gestational diabetes mellitus.
Toxics 12 00841 g002
Toxics 12 00841 g003
Figure 3. Results of mediation analysis (adjusted model). ACME, average causal mediation effect; ADE, average direct effect.
Figure 3. Results of mediation analysis (adjusted model). ACME, average causal mediation effect; ADE, average direct effect.
Toxics 12 00841 g003
Table 1. Characteristics of the study participants.
Table 1. Characteristics of the study participants.
CharacteristicsMean ± SD or n (%)
(N = 374)
Birth year1976.3 ± 4.4
Urbanization
    - Urban area328 (87.7%)
    - Rural area46 (12.3%)
Smoking state
    - Never smoker365 (97.6%)
    - Ever smoker9 (2.4%)
Educational level
    - >High school272 (72.7%)
    - ≤High school102 (27.3%)
Total exposure duration39.1 ± 32.8
Parity
    −1245 (65.5%)
    −2111 (29.7%)
    −318 (4.8%)
Asthma diagnosis
before GDM diagnosis/G 28 weeks
    - Y55 (14.7%)
    - N319 (85.3%)
GDM, gestational diabetes mellitus.
Table 2. Crude and age standardization incidence of GDM.
Table 2. Crude and age standardization incidence of GDM.
AgeNon-ExposureAfter Exposure
Total Pregnancies, EpisodesCrude Incidence, %Standardized Incidence, %Total Pregnancies, EpisodesCrude Incidence, %Standardized Incidence, %
Total1272.36%2.36%3948.9%8.2%
<30 years731.37%1.37%1107.3%7.3%
≥30 years543.70%3.70%2849.5%9.5%
GDM, gestational diabetes mellitus; after exposure began before GDM diagnosis in GDM cases or before 28 weeks of pregnancy in control episodes; non-exposure, not affected by PHMG/PGH exposure.
Table 3. Characteristics of study episodes.
Table 3. Characteristics of study episodes.
N(%) or Mean ± SDGDMControlpTotal cases
(N = 38)(N = 483) (N = 521)
Age32.5 ± 3.431.0 ± 3.80.01931.2 ± 3.8
Delivery year2010.7 ± 3.22008.0 ± 3.2<0.0012008.2 ± 3.3
Year of conception 0.001
<201332 (84.2%)467 (96.7%) 499 (95.8%)
≥20136 (15.8%)16 (3.3%) 22 (4.2%)
Delivery seasons 0.837
        Winter (Dec~Feb)7 (18.4%)110 (22.8%) 117 (22.5%)
        Fall (Sep~Nov)9 (23.7%)118 (24.4%) 127 (24.4%)
        Spring (Mar~May)13 (34.2%)134 (27.7%) 147 (28.2%)
        Summer (Jun~Aug)9 (23.7%)121 (25.1%) 130 (25.0%)
Parity 0.554
        −115 (39.5%)230 (47.6%) 245 (47.0%)
        −218 (47.4%)204 (42.2%) 222 (42.6%)
        −35 (13.2%)49 (10.1%) 54 (10.4%)
Urbanization 1
    - Urban area33 (86.8%)425 (88.0%) 458 (87.9%)
    - Rural area5 (13.2%)58 (12.0%) 63 (12.1%)
Smoking state 1
    - Never smoker37 (97.4%)472 (97.7%) 509 (97.7%)
    - Ever smoker1 (2.6%)11 (2.3%) 12 (2.3%)
Educational level 0.835
    - >High school26 (68.4%)345 (71.4%) 371 (71.2%)
    - ≤High school12 (31.6%)138 (28.6%) 150 (28.8%)
Exposure 0.024
    - Non-exposure3 (7.9%)124 (25.7%) 127 (24.4%)
    - After exposure35 (92.1%)359 (74.3%) 394 (75.6%)
Cumulative exposure duration, months
(before diagnosis of GDM or G28 weeks)		26.1 ± 25.918.2 ± 23.20.04418.7 ± 23.5
Cumulative exposure duration
(before diagnosis of GDM or G28 weeks) (tercile)		 0.175
    - T1 (≥0, <4)10 (26.3%)192 (39.8%) 202 (38.8%)
    - T2 (≥4, <20)11 (28.9%)138 (28.6%) 149 (28.6%)
    - T3 (≥20, <115)17 (44.7%)153 (31.7%) 170 (32.6%)
Cumulative exposure hours, hours
(before diagnosis of GDM or G28 weeks)		11,687.1 ± 12,757.86777.5 ± 11,032.00.0097135.6 ± 11,226.1
Cumulative exposure hours
(before diagnosis of GDM or G28 weeks) (tercile)		 0.059
    - T1 (≥0, <1008)7 (18.4%)170 (35.2%) 177 (34.0%)
    - T2 (≥1008, <6720)13 (34.2%)161 (33.3%) 174 (33.4%)
    - T3 (≥6720, <122,304)18 (47.4%)152 (31.5%) 170 (32.6%)
Concentration (μg/m3)2.8 ± 5.51.2 ± 2.30.0881.3 ± 2.7
Concentration (μg/m3) (tercile) 0.012
    - T1(≥0, <0.001)8 (21.1%)207 (42.9%) 215 (41.3%)
    - T2 (≥0.001, <1.22)16 (42.1%)115 (23.8%) 131 (25.1%)
    - T3 (≥1.22, <35.38)14 (36.8%)161 (33.3%) 175 (33.6%)
Distance from humidifier 0.401
    - ≤1 m25 (65.8%)355 (73.5%) 380 (72.9%)
    - >1 m13 (34.2%)128 (26.5%) 141 (27.1%)
Location of humidifier 0.665
    - Close to nose or mouth26 (68.4%)353 (73.1%) 379 (72.7%)
    - Other location12 (31.6%)130 (26.9%) 142 (27.3%)
Asthma diagnosis before GDM diagnosis/G28 weeks 0.414
    - Y6 (15.8%)49 (10.1%) 55 (10.6%)
    - N32 (84.2%)434 (89.9%) 466 (89.4%)
GDM, gestational diabetes mellitus; after exposure began before GDM diagnosis in GDM cases or before 28 weeks of pregnancy in control episodes; non-exposure, not affected by PHMG/PGH exposure.
Table 4. Logistic regression and mediation analysis by PHMG/PGH exposure characteristics.
Table 4. Logistic regression and mediation analysis by PHMG/PGH exposure characteristics.
VariablesCrude, OR (95% CI)Adjusted, OR (95% CI)
Exposure state, after exposure, ref. non-exposure4.03 (1.419~16.926)2.968 (1.004~12.725)
Cumulative exposure duration
(before diagnosis of GDM) (tercile), ref. T1
    - T21.53 (0.628~3.77)1.245 (0.5~3.123)
    - T32.133 (0.965~4.961)1.579 (0.681~3.808)
p for trendp = 0.024p = 0.056
Cumulative exposure hours
(before diagnosis of GDM) (tercile), ref. T1
    - T21.961 (0.783~5.333)1.68 (0.65~4.686)
    - T32.876 (1.217~7.57)2.3 (0.929~6.271)
p for trendp = 0.009p = 0.017
Concentration (μg/m3) (tercile), ref. T1
    - T23.6 (1.535~9.118)3.059 (1.26~7.956)
    - T32.25 (0.94~5.755)2.041 (0.844~5.264)
p for trendp = 0.663p = 0.681
Distance from humidifier, ref. ≤ 1 m
    - >1 m1.442 (0.697~2.859)1.46 (0.689~2.98)
Location of humidifier, ref. close to nose or mouth
    - Other places1.253 (0.594~2.506)1.198 (0.555~2.46)
Mediated effects of affected state Crude, β (95%CI)Adjusted, β (95%CI)
    ACME0.0004 (−0.0029~0.00)0.0004 (−0.0031~0.00)
    ADE0.0650 (0.0276~0.10)0.0432 (0.0050~0.08)
    Total effects0.0654 (0.0292~0.10)0.0435 (0.0057~0.08)
ACME, average causal mediating effect; ADE, average direct effect.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Choi, H.; Kim, N.-Y.; Kim, N.; Ahn, Y.-S. Is Polyhexamethylene Guanidine and Oligo(2-(2-Ethoxy) Ethoxyethyl Guanidium Chloride Exposure Related to Gestational Diabetes? Toxics 2024, 12, 841. https://doi.org/10.3390/toxics12120841

AMA Style

Choi H, Kim N-Y, Kim N, Ahn Y-S. Is Polyhexamethylene Guanidine and Oligo(2-(2-Ethoxy) Ethoxyethyl Guanidium Chloride Exposure Related to Gestational Diabetes? Toxics. 2024; 12(12):841. https://doi.org/10.3390/toxics12120841

Chicago/Turabian Style

Choi, Hyowon, Nam-Yun Kim, Nalai Kim, and Yeon-Soon Ahn. 2024. "Is Polyhexamethylene Guanidine and Oligo(2-(2-Ethoxy) Ethoxyethyl Guanidium Chloride Exposure Related to Gestational Diabetes?" Toxics 12, no. 12: 841. https://doi.org/10.3390/toxics12120841

APA Style

Choi, H., Kim, N.-Y., Kim, N., & Ahn, Y.-S. (2024). Is Polyhexamethylene Guanidine and Oligo(2-(2-Ethoxy) Ethoxyethyl Guanidium Chloride Exposure Related to Gestational Diabetes? Toxics, 12(12), 841. https://doi.org/10.3390/toxics12120841

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop