Maternal Smoking and Infant Birth Weight

1
Maternal Smoking and Infant Birth Weight

• Lucinda England, MD, MSPH
• National Institute of Child Health and Human
  Development (NICHD)

2

• Lucinda England, MD, MSPH
• Post-doctoral Fellow in the Division of
  Epidemiology, Statistics, and Prevention
  Research, NICHD
• Interests smoking and fetal growth, smoking and
  preeclampsia
• This research was conducted while the author was
  an Epidemic Intelligence Officer at the Centers
  for Disease Control and Prevention

3
Learning Objectives

• Understand the relationship between tobacco
  exposure during pregnancy and fetal growth
• Understand the limitations of methods currently
  available for measuring tobacco exposure

4
Learning Objectives (Continued)

• Appreciate the importance of understanding the
  nature of the relationship between exposure and
  outcome

5
Smoking during pregnancy increases likelihood of

• Low birth weight
• Intrauterine growth restriction
• Preterm delivery
• SIDS
• Maternal complications

6
Quitting during pregnancy improves birth weight,
but...

• Only one-in-four pregnant smokers quits
  successfully
• One-in-three pregnant smokers cuts back
• It is unknown if cutting back is beneficial

7
How do we measure tobacco exposure?

• 1. Self report Inexpensive, but
• has several limitations a) Some may
  not report use honestly
• b) Recall bias
• c) People tend to round to the
  nearest pack or half-pack

8
How do we measure tobacco exposure?

• 2. Biomarkers Objective way to
• validate and quantify exposure
• a) Thiocyanate
• b) Carbon monoxide
• c) Cotinine

9
Biomarkers cotinine

• Primary metabolite of nicotine
• Popular biomarker because it
• -Has long half life (20 hours)
• -Can be detected in serum, urine, or saliva
• -Is stable in stored specimens

10
Unresolved research issues

• 1. Can cotinine concentration be used to
  predict how much a woman smokes?
• 2. Does a pregnant woman who cuts back improve
  her infants birth weight?

11
To address these issues, we need to know the
following

• 1. What is the nature of the relationship
  between cotinine and cigarettes per day?
• 2. What is the nature of the relationship
  between tobacco exposure and birth weight?

12
Methods
13

• Data source A collaborative project between the
  CDC and three state health departments
• Study population Self-reported smokers attending
  public prenatal clinics from 1987-1990 and giving
  birth to live, term, singleton infants

14

• Outcome birth weight
• Exposure tobacco
• Measured in two ways
• Self-reported cigarettes smoked per day
• Urine cotinine concentration
• Obtained at two points in time
• At entry into prenatal care
• In the 3rd trimester

15
Analysis I Relationship between cigarette
smoking and urine cotinine

• General linear models were used to quantify the
  amount of cotinine variability explained by self
  reported cigarettes smoked per day.

16
Analysis II Tobacco exposure and birth weight

• General linear models were used to generate mean
  adjusted birth weight.
• Birth weight was plotted against cigarettes per
  day and against urine cotinine concentration.

17
Results
18
Study population

• 5,572 women screened
• 4,005 women eligible
• 3,395 women had at least one set of corresponding
  measures of exposure
• 2,481 women still smoking in 3rd trimester

19
Cigarettes per day and urine cotinine (See Figure
1)

• Median urine cotinine concentration increased as
  cigarettes per day increased.
• Urine cotinine concentration reached a plateau at
  gt 15 cigarettes per day.

20
Cigarettes per day and urine cotinine (See Figure
1)

• Urine cotinine concentration varied considerably
  at all levels of cigarette smoking.
• The amount of variation in urine cotinine
  concentration explained by cigarettes per day
  (r2) was only 14 percent.

21
Figure 1. Box and whiskers plot of urine cotinine
by cigarettes per day
Click for larger picture
22
Tobacco exposure and birth weight

• Mean adjusted birth weight decreased as
  cigarettes per day increased (See Figure 2).
• The relationship was not linear the sharpest
  decline in birth weight seen at low levels of
  smoking.
• A similar pattern seen with cotinine
  concentration (see Figure 3).

23
Figure 2. Mean adjusted infant birth weight by
3rd trimester smoking
Click for larger picture
24
Figure 3. Mean adjusted infant birth weight by
3rd trimester cotinine
Click for larger picture
25
Discussion
26
Answers to research questions

• 1. What is the relationship between urine
  cotinine and cigarettes smoked per day?
• Answer Urine cotinine concentration does not
  correlate closely with cigarettes per day.

27
Answers to research questions

• 2. What is the relationship between tobacco
  exposure and birth weight?
• Answer Birth weight declines sharply at low
  levels of exposure, then levels off.

28
Study Limitations

• We dont know time elapsed between last cigarette
  smoked and urine collection.
• We dont have information on types of cigarette
  smoked or how cigarettes were smoked.

29
Conclusions

• Our ability to quantify tobacco exposure is
  limited both self-reported cigarette use and
  urine cotinine have serious limitations.
• More research is needed to uncover the mechanisms
  through which tobacco causes fetal growth
  restriction.

30
Conclusions

• Studies of the effects of smoking reduction on
  health outcomes are needed before recommending
  reduction as an alternative to quitting.

31
Suggested readings

• England et al. Measures of maternal tobacco
  exposure and
• infant birth weight at term. Am J Epidemiol,
  2001153954-60.
• England et al. Effects of smoking reduction
  during pregnancy
• on the birth weight of term infants. Am J
  Epidemiol, (in press).
• Ellard GA et al. Smoking during pregnancy the
  dose
• dependence of birthweight deficits. Br J Obstet
  Gynaecol,
• 1996 103806-13.
• Klebanoff MA et al. Serum cotinine concentration
  and self-
• reported smoking during pregnancy. Am J
  Epidemiol, 1998
• 148259-62.
• Haddow JE et al. Cigarette consumption and serum
  cotinine
• in relation to birth weight. Br J Obstet
  Gynaecol, 198794678-81.