Abstract
Background: Volatile organic compounds (VOCs) have been detected in feminine hygiene products (FHPs), especially in tampons and sanitary pads. However, little is known about whether menstrual products can contribute to VOC exposure in women. Our objectives were to: (1) examine the variations of urinary VOC concentrations during menstrual cycles; (2) evaluate the relationships between the use of menstrual products and urinary VOC concentrations; and (3) link urinary VOC concentrations to those measured in menstrual products.
Methods: We measured urinary concentrations of 98 target VOCs in 25 reproductive-aged women with 100 repeated measures collected between October 2018 and February 2019. First-morning-void urine samples were collected four times for each woman during one menstrual cycle. Urinary VOC concentrations were measured using gas chromatography–mass spectroscopy.
Results: Of 98 target VOCs measured in the urine samples, 36 VOCs were detected. We did not see statistically significant variations in VOC concentrations across the menstrual cycle. After multivariable adjustment, tampon users had significantly higher concentrations of 2-butanone (β = 1.58 log ng/g, 95% confidence interval [CI]: 0.16–3.00, p = 0.03) and methyl isobutyl ketone (β = 0.63 log ng/g, 95% CI: 0.03–1.22, p = 0.04), compared with pad users. Higher n-nonane, benzene, and toluene estimated from menstrual products were associated with higher urinary concentrations in women.
Conclusion: The use of FHPs during menses might be a potential source of VOCs. A larger cohort study is warranted to confirm our results and evaluate clinical implications.
Keywords: feminine hygiene products, volatile organic compounds, endocrine-disrupting chemicals, women
Introduction
Volatile organic compounds (VOCs) include a wide variety of chemicals in many household products, including personal care products, paints, adhesives, gasoline, and building materials.1 In addition to their widespread use, VOCs are common environmental contaminants, found in indoor and ambient air, soils, and ground and drinking water.2 As a result, human exposure can occur via inhalation, ingestion, and dermal contact. In the National Health and Nutrition Examination Survey (NHANES), urinary concentrations of several VOCs were found to increase between 2005 and 2014, despite a decline in ambient air levels, which suggests that ambient VOCs may not be the primary source of VOC exposure in the general population.3
VOC exposure has been associated with developmental, reproductive, neurologic, immunologic, and carcinogenic effects in animal models and humans.4 Several VOCs are listed on the Agency for Toxic Substances and Disease Registry (ATSDR)'s Substance Priority List due to their known or suspected toxicity and the potential threat to human health.5 Given their toxicity, it is essential to identify VOC sources to target exposure reduction strategies.
Chemical exposure to VOCs through feminine hygiene products (FHPs) is a public health concern. In August 2014, the consumer group Women's Voices for Earth found menstrual pads emitted specific VOCs, including styrene, chloromethane, and chloroform.6 Park et al. measured the amounts of three VOCs (i.e., methylene chloride, toluene, and xylene) in the air contained in the packages of sanitary pads and diapers, finding VOC concentrations similar to that of indoor air.7 Kim et al. evaluated 74 targeted VOCs in the sanitary pads retailed in Korea.8 Our research team also found toxic VOCs, for example, benzene, n-heptane, and 1,4-dioxane, in FHPs sold in the U.S. market, including washes, tampons, menstrual pads, wipes, sprays, powders, and moisturizers.9 Using the NHANES 2001–2004 data, we further found a significant and positive association between the frequency of vaginal douching and whole blood concentrations of 1,4-dichlorobenzene in women 20–49 years of age.10
FHPs are widely used. In the U.S., ∼$3 billion per year is spent on these products.11 FHPs intended for use on vaginal and vulvar tissues, which are highly permeable, are marketed and sold with little or no data identifying their ingredients or potential hazards. However, it remains unknown whether tampon or pad use is related to VOC exposure.
Therefore, we conducted a pilot study with a cohort of 25 women (11 White, 6 Black, and 7 Asian) to identify possible associations between the use of menstrual products and longitudinal changes in urinary concentrations of VOCs during the menstrual cycles. We hypothesized that VOC concentrations would be higher when tampons or pads were used during menstruation and then decrease at the end of period and go back to the background level. We also compared urinary VOC concentrations among tampon users to pad users. VOC concentrations would be different in women using different types of products. In a secondary analysis, we examined relationships between VOCs measured in menstrual products and urinary VOC concentrations. Higher VOC levels in tampons or pads were hypothesized to be linked to higher urinary VOCs in women.
Materials and Methods
Study population
To conduct this pilot study, we recruited 25 female participants affiliated with the University of Michigan School of Public Health (SPH) in Ann Arbor, MI, in September 2018. SPH students, staff, and faculty received an email about the project and inviting them to participate in the study. The email also included a short survey about eligibility. Eligibility criteria included age between 20 and 49 years, have at least one menstrual period in the past 3 months, variations of menstrual cycle lengths within 7 days, the average menstrual cycle length of 21–40 days, and self-identified with one of the designated racial/ethnic groups (White, Black, Asian). Women were excluded if they were currently pregnant, breastfeeding, or diagnosed by physicians with a vaginal infection, uterine fibroids, polycystic ovarian syndrome, or endometriosis in the past 12 months.
To evaluate racial differences, we aimed to recruit three self-identified racial groups: 10 White women, 10 Black women, and 10 Asians. Because some participants refused to participate in the study or lost to follow-up, the final study sample included 11 White, 6 Black, and 8 Asians. All participants provided written informed consent approved by the Institutional Review Board at the University of Michigan.
Participants were provided with free menstrual products they preferred to use. To better capture potential VOC exposure through the use of menstrual products, first-morning-void urine samples were collected at four time points, that is, 7 days before menstruation, 3 days after the first day of menstruation or the end of heavy bleeding, 7 days after the first day of menstruation or the end of menstrual period, and 7 days after a period ends (Fig. 1). Women who were uncertain about the onset of menstruation were instructed to use ovulation kits to predict the menstrual cycle's timing.
FIG. 1.
Open in a new tab
At each of the four sample collection time points, the participants received a $15 gift card, resulting in a total of $60 in compensation. Women were instructed to return the sample at the date of collection, and they were provided with cooler bags and ice packs for sample delivery. Women unable to return the sample on the collection date were instructed to keep the urine sample in the freezer and return it at their earliest convenience. After completing the processing of samples, sample aliquots were stored in a −80°C freezer. One woman failed to collect her first urine sample before the start of menstruation, resulting in 99 urine samples for the analysis.
VOC measurements
Given that urine samples were collected soon after exposure, we measured parent compounds using headspace sampling and gas chromatography–mass spectrometry (GC-MS) analyses, similar to that described previously for the measurements of VOCs in FHPs.9 A 10 mL aliquot of each urine sample was transferred to a 40-mL glass vial, which was immediately sealed using a Teflon septum and screw cap. After 10 minutes equilibrium at temperature of 40°C (close to normal human body temperature), 1000 mL of N2 was purged to the bottom of the vial using a needle over a 30-minute period. Flow exiting the vial passed through a 10-cm-long stainless steel adsorbent sampling tube (Scientific Instrument Services, Inc., Ringoes, NJ) equipped with a needle inlet that also pierced the septum. The sampling tubes were packed with 250 mg anhydrous sodium sulfate (Fisher Scientific, Fair Lawn, NJ) to remove water vapor, and 160 mg of 60/80 mesh Tenax-GR (Scientific Instrument Services, Inc., Palmer, MA). The temperature was kept at 40°C.
For the analysis, tubes were injected with the internal standards, loaded into a short-path automated thermal desorption system (Scientific Instrument Services, Inc.) coupled to a GC-MS (Model 6890/5973; Agilent Technologies, Santa Clara). Chromatographic separation was performed using a DB-VRX capillary column (60 m × 0.25 mm, 1.4 μm film thickness) with an optimized temperature program. Peak areas were extracted by a ChemStation macro, adjusted for internal standards, and transferred electronically to a spreadsheet. In parallel with sample analyses, quality control procedures were conducted, including the use of standard reference materials, linearity, and drift checks with each sample batch; spike recovery tests performed periodically during analyses; daily analyses of a calibration/quality assurance (QA) sample (a freshly loaded adsorbent tube containing 10 ng of target compounds); calculation of method detection limit for individual target compounds; and analyses of blanks with each sample batch. All solvents and other materials contacting samples were proved clean as confirmed using blanks. VOC calibrations used authentic standards. Supplementary Table S1 shows limits of detection (LODs) and percent above LODs for 98 target VOCs, which included 12 alkanes, 6 aldehydes, 19 aromatics, 7 esters, 40 halohydrocarbons, 4 ketones, 2 terpenes, and 8 others. Values below LOD were replaced by LOD/. Urinary specific gravity, to help account for urine dilution, were assessed using a refractometer (Master Refractometer; ATAGO Co., Inc., Japan).
Covariates
Information on age, employment or registration status, education, smoking status, household smoke exposure, and use of FHPs were collected at enrollment through a self-administered questionnaire. Education was categorized into some college or associate degree, bachelor's degree, master's degree, or higher. Women were either students or staff at the time of enrollment. Smoking status was classified into never smoker, former smoker, or current smoker. Passive smoke exposure was defined as whether or not living with someone who smoked cigarettes. We also asked the participants whether they used feminine care products (including tampons, menstrual pads, panty liners, vaginal douches, sprays, powders, wipes, moisturizes/lubricants, vaginal tablets, shaving cream, and others), in the past 3 months. The use of FHPs other than menstrual products (i.e., tampons, pads, or panty liners) was classified based on the number of types used, including no, one type only, or more than two types.
A self-administered menstrual diary was employed to keep records of menstruation length, duration of heavy bleeding, and the number of tampons or pads used during day and night. Participants were instructed to fill out the form and return it to the clinic at the last visit. Tampons and/or pads were classified based on their absorbency. Tampons included light tampons, regular tampons, super tampons, and super plus tampons. Pads included mini pads, regular pads, maxi pads, super pads, and panty liners.
Participants were also asked to complete questionnaires about VOC exposure-related activities right after urine sample collection. The questions included paint or gasoline use (e.g., storage of items in home, recent fill-ups); source of drinking water (from a private well); use of deodorizers at home (e.g., mothballs, moth crystals, toilet deodorizers); use of natural gas for cooking or baking; spending time at a swimming pool, in a hot tub or a steam room; use dry cleaning solvents or visit a dry cleaning shop or wear clothes that had been dry cleaned; use of fingernail polish or visit a nail salon. Background VOC exposure was defined as whether or not involved in one of the VOC exposure-related activities.
Statistical analyses
Detection rates, medians, and interquartile ranges (IQRs) for both volume-based (ng/mL) and specific gravity-adjusted (ng/g) concentrations of 98 VOCs found in 99 urine samples were calculated overall and by study visit. We examined temporal changes in VOC concentrations during one menstrual cycle and the associations between the use of menstrual products and urinary VOC concentrations. We compared urinary VOC concentrations for tampon users and pad users. For VOCs with detection rates >50% across all study visits, we utilized repeated measure analysis of variance to assess variations in VOC concentrations before and after the menstrual period. To examine the associations between the use of menstrual products and urinary VOC concentrations during each woman's period, we used generalized estimating equations (GEEs) with adjustments for race, study visits, time-varying background VOC exposure, and use of other FHPs at baseline. VOCs with detection rates >50% across all study visits were included in the analysis. Age, education, employment status, and smoking variables were not considered because there is little variation in our study population.
In secondary analyses, VOC concentrations measured in menstrual products were linked to urinary VOC concentrations in women. VOC levels in tampons and pads were calculated as the product of the mean VOC concentration per unit of each menstrual product and the number of menstrual products used during the period. These VOC concentrations were measured in our previous assessments of products in the U.S. market.9 The number of products was computed using weights to reflect the product mass,9 specifically, 0.8, 1.0, 1.5, and 2.0 for light, regular, super and superplus tampons, and 0.5, 0.8, 1.0, and 1.5 for panty liners, mini, regular, and maxi pads. GEEs were implemented to examine the relationships between VOCs in products and urinary VOC concentrations in women, after controlling for race, study visits, time-varying background VOC exposure, and use of other FHPs at baseline. Analyses were conducted using SAS version 9.4 (SAS Institute, Inc.).
Results
Basic characteristics for the study population
Table 1 shows characteristics for participants at baseline and during repeated samplings. The median age was 23 (IQR: 22, 25) years. Most women had bachelor's degrees or higher, registered as students at the University of Michigan, never smoked, and did not live with someone who currently smoked cigarettes. All women used tampons or pads during menstruation, and most reported using other FHPs regularly at enrollment. During their menstrual period, 59.1% of the women used pads or panty liners only, 22.7% used tampons only, and 18.2% used both types. More than half of the participants reported VOC exposure-related activities, e.g., using dry cleaning solvents, or breathing fumes from paints or fingernail polish (Table 2).
Table 1.
Participant Characteristics of 25 Naturally Cycling Women at the University of Michigan School of Public Health at Visit 1
| Characteristic | Visit 1 (n = 25) |
|---|---|
| Median (IQR) or n (%) | |
| Age, years | 23 (22–25) |
| Race | |
| White | 11 (44.0) |
| Black | 6 (24.0) |
| Asian | 8 (32.0) |
| Highest grade of education | |
| Some college or associate degree | 2 (8.3) |
| Bachelor's degree | 16 (66.7) |
| Master's degree or higher | 6 (25.0) |
| Missing | 1 |
| Employment | |
| Student | 20 (87.5) |
| Staff | 3 (12.5) |
| Missing | 2 |
| Smoking status | |
| Never smoke | 23 (95.8) |
| Former smoke | 1 (4.2) |
| Current smoke | 0 (0) |
| Missing | 1 |
| Use of tampons, pads, or panty liners | |
| No | 0 (0) |
| Yes | 24 (100) |
| Missing | 1 |
| Use of products other than tampons or pads | |
| No | 9 (37.5) |
| One type only | 10 (41.7) |
| More than two types | 5 (20.8) |
| Missing | 1 |
| Use of tampons or pads during period | |
| Pad or liner only | 13 (59.1) |
| Tampon only | 5 (22.7) |
| Both tampon and pad | 4 (18.2) |
| Missing | 2 |
| Duration of menstrual period | 5 (5–7) |
| Duration of heavy bleeding | 2 (2–3) |
Open in a new tab
First-morning-void urine samples were collected four times across two menstrual cycles, including 7 days before the start of menstruation (visit 1), 3 days after the onset of period, or the end of heavy bleeding (visit 2); 7 days after the onset of period or the end of menstruation (visit 3), and 7 days after the end of the period (visit 4).
IQR, interquartile range.
Table 2.
Specific Gravity and Background Volatile Organic Compound Exposure in 25 Naturally Cycling Women at the University of Michigan School of Public Health During the Follow-Up Visits
| Characteristic | Visit 1 (n = 24) | Visit 2 (n = 25) | Visit 3 (n = 25) | Visit 4 (n = 25) |
|---|---|---|---|---|
| Median (IQR) or n (%) | Median (IQR) or n (%) | Median (IQR) or n (%) | Median (IQR) or n (%) | |
| Specific gravity | 1.013 (1.007–1.016) | 1.012 (1.008–1.017) | 1.010 (1.006–1.015) | 1.013 (1.007–1.017) |
| Background VOC exposure | ||||
| No | 7 (33.3) | 11 (47.8) | 9 (39.1) | 10 (45.5) |
| Yes | 14 (66.7) | 12 (52.2) | 14 (60.9) | 12 (54.5) |
Open in a new tab
First-morning-void urine samples were collected four times across two menstrual cycles, including 7 days before the start of menstruation (visit 1), 3 days after the onset of period or the end of heavy bleeding (visit 2), 7 days after the onset of period or the end of menstruation (visit 3), and 7 days after the end of the period (visit 4).
VOC, volatile organic compound.
VOCs in urine samples
Of 98 target VOCs, we detected 36 compounds in urine samples from our study participants, including 9 alkanes, 5 aldehydes, 12 aromatics, 2 esters, 2 halohydrocarbons, 2 ketones, 1 terpene, and 3 others (Supplementary Table S1). Among them, hexane, n-nonane, hexanal, nonanal, benzene, toluene, p-isopropyltoluene, 2-butanone, and methyl isobutyl ketone had detection rates >50%. Median (IQR) and the range of specific gravity-adjusted concentrations are displayed in Supplementary Table S2. Hexane and 2-butanone had the highest concentrations, with medians and maximums of 1.3 (IQR: 0.5, 4.3) ng/mL and 491.9 ng/mL, as well as 2.0 (IQR: 0.9, 4.2) ng/mL and 37.2 ng/mL, respectively. Other compounds showed much lower urinary levels. VOC concentrations by study visits are summarized in Table 3. Changes in detection rates and urinary concentrations at menstrual cycle did not reach statistical significance.
Table 3.
Detection Rates, Minimum, Median, 25th and 75th Percentiles, and Maximum Specific Gravity-Adjusted Urinary Concentrations (ng/g) of Volatile Organic Compounds Measured in 99 Urine Samples
| VOC | Visit 1 | Visit 2 | Visit 3 | Visit 4 | p | ||||
|---|---|---|---|---|---|---|---|---|---|
| % > LOD | Median (IQR) | % > LOD | Median (IQR) | % > LOD | Median (IQR) | % > LOD | Median (IQR) | ||
| Alkanes | |||||||||
| Hexane | 91.7% | 1.6 (0.7–13.3) | 84.0% | 1.0 (0.4–2.9) | 84.0% | 0.9 (0.4–3.0) | 92.0% | 1.5 (0.8–4.4) | 0.11 |
| n-Nonane | 66.7% | 0.02 (<LOD–0.08) | 64.0% | 0.01 (<LOD–0.04) | 60.0% | 0.01 (<LOD–0.07) | 64.0% | 0.02 (<LOD–0.05) | 0.62 |
| n-Tetradecane | 20.8% | <LOD | 36.0% | <LOD (<LOD–0.02) | 28.0% | <LOD (<LOD–0.02) | 24.0% | <LOD | NA |
| n-Undecane | 8.3% | <LOD | 32.0% | <LOD (<LOD–0.02) | 24.0% | <LOD | 12.0% | <LOD | NA |
| n-Tridecane | 4.2% | <LOD | 4.0% | <LOD | 12.0% | <LOD | 0% | <LOD | NA |
| n-Pentadecane | 8.3% | <LOD | 4.0% | <LOD | 4.0% | <LOD | 4.0% | <LOD | NA |
| n-Octane | 0% | <LOD | 4.0% | <LOD | 0% | <LOD | 8.0% | <LOD | NA |
| n-Hexadecane | 0% | <LOD | 4.0% | <LOD | 0% | <LOD | 4.0% | <LOD | NA |
| n-Dodecane | 4.2% | <LOD | 0% | <LOD | 0% | <LOD | 0% | <LOD | NA |
| Aldehydes | |||||||||
| Hexanal | 83.3% | 0.2 (0.1–0.3) | 92.0% | 0.2 (0.07–0.3) | 92.0% | 0.2 (0.1–0.3) | 96.0% | 0.2 (0.1–0.3) | 0.98 |
| Nonanal | 70.8% | 0.1 (<LOD–0.3) | 84.0% | 0.1 (0.04–0.2) | 80.0% | 0.1 (0.04–0.2) | 88.0% | 0.1 (0.07–0.2) | 0.97 |
| Heptanal | 29.2% | <LOD (<LOD–0.08) | 48.0% | <LOD (<LOD–0.09) | 28.0% | <LOD (<LOD–0.09) | 36.0% | <LOD (<LOD–0.09) | NA |
| Butanal | 29.2% | <LOD (<LOD–0.20) | 24.0% | <LOD | 24.0% | <LOD | 44.0% | <LOD (<LOD–0.20) | NA |
| Octanal | 0% | <LOD | 8% | <LOD | 0% | <LOD | 4% | <LOD | NA |
| Aromatics | |||||||||
| Benzene | 50.0% | <LOD (<LOD–0.02) | 76.0% | 0.02 (0.01–0.05) | 68.0% | 0.02 (<LOD–0.04) | 80.0% | 0.03 (0.008–0.06) | 0.34 |
| Toluene | 100% | 0.1 (0.05–0.3) | 84.0% | 0.06 (0.02–0.1) | 92.0% | 0.07 (0.04–0.1) | 84.0% | 0.05 (0.03–0.2) | 0.05 |
| Styrene | 41.7% | <LOD (<LOD–0.05) | 64.0% | 0.009 (<LOD–0.04) | 60.0% | 0.009 (<LOD–0.04) | 56.0% | 0.002 (<LOD–0.04) | NA |
| p-Isopropyltoluene | 66.7% | 0.02 (<LOD–0.05) | 76.0% | 0.02 (0.01–0.05) | 84.0% | 0.03 (0.02–0.08) | 68.0% | 0.02 (<LOD–0.06) | 0.10 |
| o-Xylene | 25.0% | <LOD (<LOD–0.008) | 36.0% | <LOD (<LOD–0.008) | 20.0% | <LOD | 28.0% | <LOD (<LOD–0.008) | NA |
| Ethylbenzene | 25.0% | <LOD | 40.0% | <LOD (<LOD–0.02) | 32.0% | <LOD (<LOD–0.01) | 32.0% | <LOD (<LOD–0.02) | NA |
| p-Xylene m-Xylene | 33.3% | <LOD (<LOD–0.009) | 36.0% | <LOD (<LOD–0.01) | 32.0% | <LOD (<LOD–0.008) | 24.0% | <LOD | NA |
| Naphthalene | 16.7% | <LOD | 24.0% | <LOD | 44.0% | <LOD (<LOD–0.04) | 24.0% | <LOD | NA |
| m-Cresol | 8.3% | <LOD | 16.0% | <LOD | 8.0% | <LOD | 8.0% | <LOD | NA |
| n-Propylbenzene | 0% | <LOD | 0% | <LOD | 4% | <LOD | 4% | <LOD | NA |
| Tert-butylbenzene | 4.2% | <LOD | 0% | <LOD | 0% | <LOD | 0% | <LOD | NA |
| 1,2,4-Trimethylbenzene | 8% | <LOD | 0% | <LOD | 4% | <LOD | 0% | <LOD | NA |
| Esters | |||||||||
| n-Butyl acetate | 12.5% | <LOD | 12.0% | <LOD | 24.0% | <LOD | 20.0% | <LOD | NA |
| Ethyl acetate | 8.3% | <LOD | 8.0% | <LOD | 12.0% | <LOD | 24.0% | <LOD | NA |
| Halohydrocarbons | |||||||||
| Methylene chloride | 58.3% | 0.13 (<LOD–0.65) | 48.0% | <LOD (<LOD–0.4) | 44.0% | <LOD (<LOD–0.2) | 32.0% | <LOD (<LOD–0.09) | NA |
| Tetrachloroethene | 0% | <LOD | 4% | <LOD | 0% | <LOD | 0% | <LOD | NA |
| Ketones | |||||||||
| 2-Butanone | 83.3% | 1.7 (1.1–3.1) | 100% | 1.4 (0.9–2.5) | 100% | 2.7 (1.5–3.9) | 96.0% | 1.2 (1.7–3.3) | 0.37 |
| Methyl isobutyl ketone | 87.5% | 0.1 (0.08–0.2) | 92.0% | 0.2 (0.09–0.2) | 88.0% | 0.2 (0.1–0.3) | 92.0% | 0.1 (0.09–0.2) | 0.44 |
| Terpenes | |||||||||
| Limonene (R)-(+) | 29.2% | <LOD (<LOD–0.04) | 16.0% | <LOD | 24.0% | <LOD | 28.0% | <LOD (<LOD–0.03) | NA |
| Others | |||||||||
| Methyl t-butyl ether | 25.0% | <LOD | 8.0% | <LOD | 8.0% | <LOD | 12.0% | <LOD | NA |
| Tetrahydrofuran | 4.2% | <LOD | 0% | <LOD | 4% | <LOD | 0% | <LOD | NA |
| Methacrylonitrile | 4.2% | <LOD | 0% | <LOD | 0% | <LOD | 0% | <LOD | NA |
Open in a new tab
For VOCs with detection rates >50% across all study visits, we utilized repeated measure ANOVA to assess variations in VOC concentrations before and after menstrual period. For VOCs with lower detections, it is not feasible to conduct statistical testing.
ANOVA, analysis of variance; LOD, limit of detection; NA, not applicable.
Associations between use of tampons or pads and urinary VOC concentrations
Effect estimates of associations between the use of menstrual products and specific gravity-adjusted VOC concentrations are presented in Table 4. After controlling for race, study visits, background VOC exposure, and use of other FHPs, women who used tampons only during the period had significantly higher urinary concentrations of 2-butanone (β = 1.58 log ng/g, 95% confidence interval [CI]: 0.16–3.00, p = 0.03) and methyl isobutyl ketone (β = 0.63 log ng/g, 95% CI: 0.03–1.22, p = 0.04), compared with those who used pads/liners only.
Table 4.
Coefficients and 95% Confidence Intervals of Associations Between Use of Menstrual Products and Specific Gravity-Adjusted Urinary Volatile Organic Compound Concentrations During Menstrual Cycles from General Estimating Equation Models
| VOCs | Tampons only | Pads or liners only |
|---|---|---|
| β (95% CI) p value | β (95% CI) p value | |
| Hexane | 0.56 (−0.84 to 2.00) | Ref |
| p = 0.43 | ||
| n-Nonane | −1.52 (−7.03 to 3.98) | Ref |
| p = 0.59 | ||
| Hexanal | 1.28 (−0.71 to 3.27) | Ref |
| p = 0.21 | ||
| Nonanal | −0.16 (−0.78 to 0.46) | Ref |
| p = 0.61 | ||
| Benzene | −0.23 (−1.58 to 1.12) | Ref |
| p = 0.74 | ||
| Toluene | −0.52 (−2.18 to 1.14) | Ref |
| p = 0.54 | ||
| p-Isopropyltoluene | −0.06 (−1.32 to 1.23) | Ref |
| p = 0.93 | ||
| 2-Butanone | 1.58 (0.16 to 3.00) | Ref |
| p = 0.03 | ||
| Methyl isobutyl ketone | 0.63 (0.03 to 1.22) | Ref |
| p = 0.04 |
Open in a new tab
Women who used both tampons and pads/liners were not included in the comparison.
CI, confidence interval.
Associations between VOCs in products and urinary VOC concentrations in women
Effect estimates of urinary VOC concentrations associated with changes in VOC levels in menstrual products are displayed in Table 5. After multivariable adjustment, a 1 ng increase of n-nonane in products was associated with higher urinary concentrations (β = 6.60 log ng/g, 95% CI: 1.63–11.57, p = 0.009). Similar results were observed for benzene (β = 0.85 log ng/g, 95% CI: 0.13–1.58, p = 0.02) and toluene (β = 0.13 log ng/g, 95% CI: 0.03–0.22, p = 0.009).
Table 5.
Coefficients and 95% Confidence Intervals of Associations Between Chemicals Detected in Products and Specific Gravity-Adjusted Urinary Volatile Organic Compound Concentrations During Menstrual Cycles from General Estimating Equation Models
| Urinary VOCs | 1 ng increase in chemicals in the products |
|---|---|
| β (95% CI) p value | |
| Hexane | 0.31 (−1.62 to 2.24) |
| p = 0.75 | |
| n-Nonane | 6.60 (1.63 to 11.57) |
| p = 0.009 | |
| Hexanal | −0.0001 (−0.001 to 0.001) |
| p = 0.89 | |
| Nonanal | 0.0001 (−0.003 to 0.004) |
| p = 0.94 | |
| Benzene | 0.85 (0.13 to 1.58) |
| p = 0.02 | |
| Toluene | 0.13 (0.03 to 0.22) |
| p = 0.009 | |
| p-Isopropyltoluene | 0.03 (−0.005 to 0.06) |
| p = 0.09 | |
| 2-Butanone | −0.007 (−0.02 to 0.01) |
| p = 0.47 | |
| Methyl isobutyl ketone | −0.01 (−0.03 to 0.004) |
| p = 0.13 |
Open in a new tab
Bold text indicates a statistically significant difference with a p-value less than 0.05.
Discussion
In this pilot study, 36 VOCs were detected in urine samples from a small cohort of reproductive-aged women during longitudinal follow-up during one menstrual cycle. Tampon users had higher concentrations of 2-butanone and methyl isobutyl ketone than those who used pads/liners during their period, and estimated levels of n-nonane, benzene, and toluene in the menstrual products were associated with urinary levels of these VOCs. To our knowledge, this is the first study to examine longitudinal changes of urinary VOC concentrations during the menstrual cycles, and the first to investigate associations between VOC levels in menstrual products and urinary VOC concentrations during the menstrual period. The clinical implications of exposure to VOCs through FHPs are unknown but may suggest menstrual products as a potential source of toxic chemicals.
We hypothesized that VOC levels would increase when women used a large number of FHPs during the first 3 days of menstruation (heavy bleeding period, visit 2) and then decrease at the end of the period (visit 3) and go back to the background level (visits 1 and 4). On the contrary, we did not observe such temporal patterns in urinary VOC measurements. We do not have enough evidence to suggest an association between use of menstrual products and VOCs. Several reasons may explain the observed null findings. First, urinary concentrations capture a wide range of VOC sources that include not only FHPs, but other personal care products as well as environmental sources such as indoor and ambient air pollutants; exposures from these other sources would occur regardless of the use of FHPs. Second, we focused on parent compounds in urine in this pilot study. Most VOCs are quickly transformed into more water-soluble metabolites and excreted in urine.12 Therefore, we may not detect parent VOCs due to the use of FHPs if the dominant fraction of VOCs were metabolized. Furthermore, VOCs can be detected in breath, blood, and urine samples. Although urinary VOCs can provide a relatively convenient and good indication of very recent exposure, these different biomarkers might yield different results. Finally, we cannot rule out the possibility of loss of target VOCs via volatilization or possible contamination with environmental VOCs during the collection, storage, transport, and analysis of the urine samples. While we believe that the first reason, namely, the contribution of non-FHP-related VOC sources, is the likely reason for the null findings, future studies that consider biotransformation of VOCs and other urine or blood biomarkers are warranted.
Our prior research quantified VOCs in menstrual pads and tampons.9 Mean levels across different brands of tampons in the U.S. market were 1.87 ng/g for 2-butanone, 1.45 ng/g for methyl isobutyl ketone, 0.40 ng/g for n-nonane, 27.17 ng/g for hexanal, and 1.34 ng/g for toluene; while for sanitary pads, concentrations averaged 0.57 ng/g for 2-butanone, 0.17 ng/g for methyl isobutyl ketone, 0.17 ng/g for n-nonane, 2.98 ng/g for hexanal, and 1.52 ng/g for toluene. It is consistent with other studies conducted for VOC measurements in sanitary pads.7,8 However, in our previous analysis of NHANE data, we failed to observe a relationship between VOCs and the use of tampons or pads.10 The null association between tampon/pad use and VOCs might result in the single biological sample not reflecting the episodic use of FHPs during the menstrual cycle, that is, missing the time window of exposure. In the present study, we provided menstrual products to study participants and used menstrual diaries to record tampon/pad use. The longitudinal study design also enabled us to explore the potential contributions of menstrual products to VOC exposure in women.
VOCs can exert toxic effects. Several of the detected VOCs, including hexane, benzene, toluene, and methyl isobutyl ketone, are considered carcinogenic and reproductive toxicants.13 Adverse reproductive outcomes associated with toluene exposure, for example, include reduced fertility, spontaneous abortion, and congenital malformation.14 Although no reproductive toxicity was observed for n-butanone exposure, human and laboratory animal studies suggest potential associations between 2-butanone and neurotoxicity.15 Animal studies suggest that methyl isobutyl ketone may damage the liver and kidneys.16,17 Methyl isobutyl ketone has also been found as chemicals known to cause reproductive toxicity.5
Vaginal and vulvar tissues are expected to have higher uptake and may be more vulnerable than other tissues to the effects of toxic chemicals. Vaginal and vulvar tissues are more permeable than exposed skin due to differences in tissue structure, occlusion, hydration, and susceptibility to friction.18,19 The mucosa of vaginal and vulvar epithelia have high permeability to contaminants given the absence of a keratinized stratum corneum and loosely packed skin layers.20–22 Arteries, blood vessels, and lymphatic vessels are abundant in the walls of the vagina, which allows for direct uptake of chemicals through peripheral circulation.23
Given the toxicity of certain VOCs and the unique characteristics of the vaginal and vulvar tissues, potential VOC exposure through this exposure pathway should be minimized. We did not collect information on health outcomes and did not evaluate potential links between urinary VOC concentrations and adverse health effects in this pilot study. Using reasonable exposure scenarios, however, our previous research did not support any meaningful noncancer or cancer health risks related to VOC exposure from lifetime use of most menstrual pads and tampons.9 Nonetheless, more than 60% of our study participants used multiple products regularly, for example, pads, tampons, washes, and wipes. It is also estimated that women have an average of 456 menstrual periods in their life, which translate to more than 9000 tampons used in a lifetime. Therefore, it is still unclear whether the use of FHPs may increase health risks.
Strengths of our study include the analysis of 98 target VOCs in repeatedly collected urine samples across the menstrual cycles. In addition, we also collected longitudinal data on VOC exposure-related activities, which may confound the associations between the use of FHPs and VOCs. Finally, we closely monitored the use of tampons or pads during the period by directly providing the menstrual products and implementing a menstrual diary.
However, our findings have a number of limitations. The pilot study recruited a small sample cohort of women from the University of Michigan SPH, and thus results may not be generalizable to the general population. Future longitudinal studies with a larger sample size are warranted to confirm our findings. Second, we excluded women with irregular menstrual cycles. Since VOCs may have a negative impact on menstrual cycle characteristics,24 we may underestimate the true contributions of menstrual products to VOC concentrations detected in women. Furthermore, we did not observe changes in VOC concentrations across the study visits. It is possible that fluctuations in sex hormones may have an impact on VOC concentrations. We did not recruit or engage women who did not use tampons or pads as our reference group. Nonetheless, women served as their self-controls as we collected urine samples from women with two samples during period, one sample during luteal phase, and one sample during follicular phase. Additionally, it is not easy to recruit nonusers or women who used menstrual products other than tampons or pads (e.g., menstrual cups). Additionally, we failed to collect information on many other covariates, such as body weight in this pilot study, and thus our findings need to be interpreted with caution. Finally, urinary concentrations may not be the optimal biomarker, however, blood sampling is invasive and infeasible for this type of study.
Conclusions
In the present study, we detected 36 VOCs in a small cohort of women, including hexane, n-nonane, hexanal, nonanal, benzene, toluene, p-isopropyltoluene, 2-butanone, and methyl isobutyl ketone. We did not observe significant longitudinal changes in urinary VOC concentrations across the menstrual cycles. Compared with women who used pads or liners during the period, tampon users had significantly higher 2-butanone and methyl isobutyl ketone concentrations. Higher n-nonane, benzene, and toluene estimated from menstrual products were associated with higher urinary concentrations in women. While our findings from this pilot study do not support the hypothesis that the use of menstrual products increases urinary VOC concentrations during the period, they do suggest that tampons may contribute to higher exposure to 2-butanone and methyl isobutyl ketone than sanitary pads. Future studies with repeated measurements and a larger sample size are warranted to confirm our results.
Supplementary Material
Supplemental data
Suppl_TableS1.docx (16KB, docx)
Supplemental data
Suppl_TableS2.docx (14.3KB, docx)
Authors' Contributions
All authors conceptualized and designed the study. N.D. and S.K.P. formulated the methods for statistical analyses. N.D. conducted the statistical analyses and drafted the initial article. All authors reviewed the article, had significant input into the editing and interpretation of data, and read and approved the final article.
Disclaimer
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
Support for this research was provided by grant P30ES017885 from the National Institute of Environmental Health Sciences (NIEHS), National Institutes of Health. This study was also supported by grants from the NIEHS R01-ES026964, and by the Center for Disease Control and Prevention (CDC)/National Institute for Occupational Safety and Health (NIOSH) grant T42-OH008455.
Supplementary Material
Supplementary Table S1
Supplementary Table S2
References
- 1.Wang S, Ang HM, Tade MO. Volatile organic compounds in indoor environment and photocatalytic oxidation: State of the art. Environ Int2007;33:694–705. [DOI] [PubMed] [Google Scholar]
- 2.Rowe BL, Toccalino PL, Moran MJ, Zogorski JS, Price CV. Occurrence and potential human-health relevance of volatile organic compounds in drinking water from domestic wells in the United States. Environ Health Perspect2007;115:1539–1546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Konkle SL, Zierold KM, Taylor KC, Riggs DW, Bhatnagar A. National secular trends in ambient air volatile organic compound levels and biomarkers of exposure in the United States. Environ Res2020;182:108991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.National Research Council (US) Committee on Contaminated Drinking Water at Camp Lejeune. Contaminated Water Supplies at Camp Lejeune: Assessing Potential Health Effects. Washington (DC): National Academies Press (US); 2009. 3, Systemic Exposures to Volatile Organic Compounds and Factors Influencing Susceptibility to Their Effects. Available from: https://www.ncbi.nlm.nih.gov/books/NBK215288 Accessed January17, 2020. [PubMed]
- 5.Agency for Toxic Substances and Disease Registry (ATSDR). Substance Priority List. https://www.atsdr.cdc.gov/spl/index.html#2019spl Accessed January17, 2020.
- 6.Scranton A. Potential health effects of toxic chemicals in feminine care products. Missoula, MT: Women's Voices for the Earth, 2013. [Google Scholar]
- 7.Park CJ, Barakat R, Ulanov A, et al. . Sanitary pads and diapers contain higher phthalate contents than those in common commercial plastic products. Reprod Toxicol2019;84:114–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kim HY, Lee JD, Kim J-Y, et al. . Risk assessment of volatile organic compounds (VOCs) detected in sanitary pads. J Toxicol Environ Heal Part A2019;82:678–695. [DOI] [PubMed] [Google Scholar]
- 9.Lin N, Ding N, Meza-Wilson E, et al. . Volatile organic compounds in feminine hygiene products sold in the US market: A survey of products and health risks. Environ Int2020;144:105740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ding N, Batterman S, Park SK. Exposure to volatile organic compounds and use of feminine hygiene products among reproductive-aged women in the United States. J Women's Heal2020;29:65–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Feminine Hygiene—United States | Statista Market Forecast. Available at: https://www.statista.com/outlook/cmo/tissue-hygiene-paper/feminine-hygiene/worldwide Accessed January11, 2021.
- 12.Lin CC, Huang CN, Wang JD, et al. . Exposure to multiple low-level chemicals in relation to reproductive hormones in premenopausal women involved in liquid crystal display manufacture. Int J Environ Res Public Health2013;10:1406–1417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Silva CL, Passos M, Câmara JS. Solid phase microextraction, mass spectrometry and metabolomic approaches for detection of potential urinary cancer biomarkers—A powerful strategy for breast cancer diagnosis. Talanta2012;89:360–368. [DOI] [PubMed] [Google Scholar]
- 14.The Proposition 65 List | OEHHA. Available at: https://oehha.ca.gov/proposition-65/proposition-65-list Accessed January11, 2021.
- 15.Bukowski JA. Review of the epidemiological evidence relating toluene to reproductive outcomes. Regul Toxicol Pharmacol2001;33:147–156. [DOI] [PubMed] [Google Scholar]
- 16.Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological Profile for 2-Butanone, 2020. https://www.atsdr.cdc.gov/ToxProfiles/tp29.pdf Accessed November1, 2020.
- 17.Stout MD, Herbert RA, Kissling GE, et al. . Toxicity and carcinogenicity of methyl isobutyl ketone in F344N rats and B6C3F1 mice following 2-year inhalation exposure. Toxicology2008;244:209–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Nemec MD, Pitt JA, Topping DC, et al. . Inhalation two-generation reproductive toxicity study of methyl isobutyl ketone in rats. Int J Toxicol2004;23:127–143. [DOI] [PubMed] [Google Scholar]
- 19.Farage M, Maibach HI. The vulvar epithelium differs from the skin: Implications for cutaneous testing to address topical vulvar exposures. Contact Dermatitis2004;51:201–209. [DOI] [PubMed] [Google Scholar]
- 20.Hussain A, Ahsan F. The vagina as a route for systemic drug delivery. J Control Release2005;103:301–313. [DOI] [PubMed] [Google Scholar]
- 21.Lesch CA, Squier CA, Cruchley A, Williams DM, Speight P. The permeability of human oral mucosa and skin to water. J Dent Res1989;68:1345–1349. [DOI] [PubMed] [Google Scholar]
- 22.Squier CA. Penetration of nicotine and nitrosonornicotine across porcine oral mucosa. J Appl Toxicol1986;6:123–128. [DOI] [PubMed] [Google Scholar]
- 23.Squier CA, Hall BK. The permeability of skin and oral mucosa to water and horseradish peroxidase as related to the thickness of the permeability barrier. J Invest Dermatol1985;84:176–179. [DOI] [PubMed] [Google Scholar]
- 24.Richardson JL, Illum L. (D) Routes of delivery: Case studies. (8) The vaginal route of peptide and protein drug delivery. Adv Drug Deliv Rev1992;8:341–366. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental data
Suppl_TableS1.docx (16KB, docx)
Supplemental data
Suppl_TableS2.docx (14.3KB, docx)
