Shorter Menstrual Cycles Associated with Chlorination By-Products in Drinking Water Gayle Windham, K Waller, M Anderson, L Fenster, P Mendola, S Swan California.

Shorter Menstrual Cycles Associated with Chlorination

By-Products in Drinking Water

Shorter Menstrual Cycles Associated with Chlorination

By-Products in Drinking Water

Gayle Windham, K Waller, M Anderson,

L Fenster, P Mendola, S SwanCalifornia Department of Health Services

Environmental Health Investigations Branch

(Work supported in large part by USEPA)

TTHM Levels and Menstrual CyclingTTHM Levels and Menstrual CyclingBackground• Found association of SAB and high consumption of water

with high TTHM and BDCM levels.• Other studies confirm increased risk of adverse pregnancy

outcomes with chlorination by-products.

Purpose• THMs associated with other reproductive endpoints? • Identify possible mechanisms for the SAB effects. • Examine THM exposure and menstrual cycle function in a

study of 400 premenopausal women.

Women’s Reproductive Health Study Methods

Women’s Reproductive Health Study Methods

• Prospective Study of Kaiser members in Santa Clara County

• Women “at risk” of pregnancy, married, ages 18-39

• Urine collected daily (about 6 months) to measure:– Steroid metabolites (E1C, PdG) — menstrual

cycle function– hCG — early pregnancy loss

• Data collected 1990-91 (follow-up 2 years later)

Data Collection InstrumentsData Collection Instruments

• Screening Interview — phone– eligibility criteria

• Baseline Questionnaire — phone– # glasses unheated tap water at home– # glasses hot tap at home– # glasses bottled water/day (and brand)– # showers at home (and length)/week– Demographics and covariates

• Daily Diary — self-administered

Individual Hormone ResultsIndividual Hormone Results

DLT DLT DLT DLT DLT

DLT DLT DLT DLT DLT

DLT DLT DLT DLT DLT

4

8

60

120

Progesterone

Estrogen

anovulatory

0

0

50

150

0

E1C to PdG Ratio

Assignment of THM LevelAssignment of THM Level

• Geocode residence address to identify water utility company

• Obtain utility THM monitoring data (total and individual)

• Cycle-specific THM levels estimated:– 90-day window, starting 60 days before cycle start– average all utility monitoring points– weighted if moved– average two closest to window, if none during

• Woman-level TTHM from average of cycle-specific levels

Results — Subject Characteristics (N=401)Results — Subject Characteristics (N=401)

% TTHMCharacteristic N % >80 µg/LAge 21-29 years 137 34.2 2

30-34 years 159 39.6 335-39 years 105 26.2 4

Race White 283 70.6 3Hispanic 52 13.0 4Other 66 16.4 3

Education No College 89 22.2 3Some College 151 37.7 2College Graduate 161 40.2 4

Pregnancies 0 48 12.0 4≥1, no losses 256 63.8 2≥1, ≥1 loss 97 24.2 4

Menstrual Cycle Length Parameters byTotal Trihalomethane Level

Menstrual Cycle Length Parameters byTotal Trihalomethane Level

TTHM Level (µg/L)0-40 >40-60 >60

Cycle LengthMean (s.e.) 29.7 (0.26) 29.3 (0.28) 28.7 (0.28)Adj. Difference (CI) Ref -0.50 (-1.1, 0.11) -1.1 (-1.8, -0.40)

Follicular PhaseMean (s.e.) 16.9 (0.27) 16.5 (0.29) 16.0 (0.30)Adj. Difference (CI) Ref -0.39 (-0.98, 0.20) -0.94 (-1.6, -0.24)

Luteal PhaseMean (s.e.) 12.9 (0.09) 12.8 (0.11) 13.0 (0.10)Adj. Difference (CI) Ref -0.08 (-0.33, 0.18) 0.07 (-0.20, 0.35)

Short cycle(<24 days)

AOR=1.5

Long cycle(>36 days)AOR=0.60

Long FP(>24 days)AOR=0.37

Short FP(<12 days)AOR=1.2

Short Luteal(<11 days)AOR=0.56

0-40 µg/L41-61 µg/L>60 µg/L

14

12

10

8

6

4

2

0

Menstrual Cycle Outcomes byTTHM Level

Menstrual Cycle Outcomes byTTHM Level

% A

ffec

ted

Length Parameters by Total Trihalomethane Daily Consumption Level

Length Parameters by Total Trihalomethane Daily Consumption Level

TTHM Consumption Level (µg/day)0 >0-40 >40

Cycle LengthN 449 717 458

Mean (s.e.) 29.8 (0.39) 29.43 (0.28) 28.5 (0.33)Adj. Difference (CI) Ref -0.23 (-1.2, 0.77) -1.1 (-2.2, -0.06)

Follicular PhaseN 402 676 436

Mean (s.e.) 17.1 (0.43) 16.6 (0.30) 15.8 (0.34)Adj. Difference (CI) Ref -0.32 (-1.4, 0.77) -1.1 (-2.2, 0.03)

Length Parameters by Quartiles of Chloroform and Brominated Trihalomethane Levels

Length Parameters by Quartiles of Chloroform and Brominated Trihalomethane Levels

Quartile of Exposure

1 2-3 4Mean in Days Difference Difference

(s.e.) (95% CI) (95% CI)

Cycle LengthChloroform 29.6 (0.30) -0.43 (-0.99, 0.13) -0.30 (-1.0, 0.40)Sum of Brominated 30.0 (0.34) -0.72 (-1.4, -0.04) -1.2 (-2.0, -0.40)

Follicular PhaseChloroform 16.8 (0.31) -0.42 (-0.96, 0.12) -0.13 (-0.82, 0.56)

Sum of Brominated 17.2 (0.35) -0.66 (-1.3, 0.02) -1.1 (-1.9, -0.29)

TTHM Levels and Menstrual CyclingTTHM Levels and Menstrual Cycling

Strengths• Prospective study• TTHM exposure from existing records in narrow time frame• Information on some water use patterns• Endpoint determined from biologic measures (vs. self-report)• Adjusted for several potential confounders• First study of reproductive function in non-pregnant women

Limitations• Potential exposure misclassification (utility average)• No information on exposure outside the home• Study sample may not represent full range of cycle variability

Conclusions and SummaryConclusions and Summary

• Increasing TTHM level is associated with decreasing cycle length, primarily in the follicular phase

• This decrease is associated with the brominated THMs, primarily chlorodibromomethane

• Combining consumption and TTHM level reveals similar, but not greater, associations

• Accounting for showering also does not reveal stronger associations

Confirm effects on ovarian function in other studies