bmj.com Rapid Responses for Wisborg et al., 326 (7386) 420

Rapid Responses to:

PAPERS:
Kirsten Wisborg, Ulrik Kesmodel, Bodil Hammer Bech, Morten Hedegaard, and Tine Brink Henriksen
Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study
BMJ 2003; 326: 420 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Moderation rather than abstinence in pregnancy: Arpan Dutta (22 February 2003)

What about Tea Intake and Oral Tobacco Chewing: Monika Malhotra, Jai B. Sharma (23 February 2003)

What about caffeine-free coffee?: Ludwig TSOI (24 February 2003)

Coffee consumption and perinatal outcome: The authors should adjust for history of drug abuse.: Michael Sindos, Narendra Pisal and Stavroula Michala (24 February 2003)

Statistical concerns: Adam Jacobs (25 February 2003)

Coffee and stillbirths, via the effect of coffee on total homocysteine?: Dag S. Thelle, Benedicte Paus (26 February 2003)

Re: coffee, homocysteine and stillbirths: Richard G Fiddian-Green (28 February 2003)

Caffeine�s role in pregnancy outcome � a complex picture?: Marcus S. Cooke, Sara Kirk, Janet Cade, Mark D. Evans, Darren Greenwood, Justin Konje, Joseph Lunec, Nigel Simpson, James Walker and Christopher P. Wild (21 March 2003)

Bitter experience: Sean Tierney (13 June 2003)

Moderation rather than abstinence in pregnancy 22 February 2003

Arpan Dutta,
4th Year Medical Student
Warrington Hospital, North Cheshire Hospitals NHS Trust, Lovely Lane, Warrington, Cheshire, WA5 1QG
Send response to journal:
Re: Moderation rather than abstinence in pregnancy

Editor- Wisborg et al�s study on maternal caffeine intake and stillbirth and infant mortality (1) has created interest from the British press appearing this evening on ITN News.(2)
However, this study though prospective does not follow caffeine consumption through to delivery or stillbirth, instead it was only looked at 16 weeks gestation when the booking visit would occur. This does not prove exposure to caffeine was throughout pregnancy. The results show that the 95% confidence interval in some cases incorporated 1.0 or a figure less than this and there is the added factor respondent recall bias that the authors note would affect data from a questionnaire study like this one. Some results from this study would suggest that a moderate intake of caffeine reduced stillbirth and infant death. So is there a clinically significant difference?
Previous studies have found association with lowered birth weight and smaller head circumference but no association with miscarriages or premature births.(3) Perhaps the advice of moderation in pregnancy rather than abstinence is a more palatable strategy for pregnant women from the advice of this study.
Arpan Dutta, 4th Year Medical Student, Warrington Hospital, North Cheshire Hospitals NHS Trust, Lovely Lane, Warrington, Cheshire, WA5 1QG
md0u9146@liv.ac.uk
1. Wisborg K, Kesmodel U, Bech BH, Hedegaard M, Henriksen TB. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study BMJ 2003;326:420
2. Coffee warning to mums-to-be. ITN (ITV) News, 21 Feb 2003
3. Watkinson B, Fried PA. Maternal caffeine use before, during and after pregnancy and effects upon offspring. Neurobehav Toxicol Teratol 1985; 7: 9-17
Competing interests: � None declared

What about Tea Intake and Oral Tobacco Chewing 23 February 2003

Monika Malhotra,
Senior Research Associate
Maulana Azad Medical College, New Delhi 110002,India,
Jai B. Sharma
Send response to journal:
Re: What about Tea Intake and Oral Tobacco Chewing

The article by Wisborg et al (1) highlights the fact that pregnant ladies need to be more careful and responsible about their dietary intake,smoking habits,alcohol intake and coffee intake as all these may be associated with poorer perinatal outcome. The authors found almost doubling in the rate of stillbirths in the women taking more than 8 cups of coffee.
Although the authors considered smoking and alcoho; habits into consideration, but they did not consider dietary habits and profession into consideration which could have affected the stillbirth rate in the women as dietary habits are known to affect the nutritional status of women indirectly affecting the stillbirth rate (2).Indian women almost exclusively drink tea with less of tea leaves and more of milk and sugar with much less caffeine intake and it is unlikely that it will affect the stillbirth rate. Also some Indian women have habit of chewing tobacco and others have the habit of eating soil ( pica)which may also have an effect on the perinatal outcome. However as suggested by studies before coffee intake causes higher spontaneous abortion rate. We agree with the authors that coffee and tea intake should be cut down to the minimum for better perinatal outcome and the women should be advised to take adequate diet in iron , calories and proteins to avoid anaemia and malnutrition during pregnancy as anaemia is associated with poor perinatal outcome (4).
References:
1.Wisborg K, Kesmodel U, Bech BH,Hedegaard M,Henriksen TB. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study. BMJ 2003; 326 : 420.
2.Sharma JB,Soni D,Murthy NS,Malhotra M. Effect of dietary habits on prevalence of anemia in pregnant women of Delhi. J Obstet Gynaecol Res 2003; 29 (2); 73-78.
3.Cnattingus S,Sigorello LB,Anneren G,Clausson B,Ekbom A,Ljunger E. et al. Caffeine intake and the risk of first trimester spontaneous abortions.N Engl J Med 2000; 393: 1819-25.
4. Malhotra M,Sharma JB,Batra S,Sharma S,Murthy NS,Arora R. Maternal and perinatal outcome in varying degrees of anemia. Int J Gynecol Obstet 2002;79: 93-100.
Competing interests: � None declared

What about caffeine-free coffee? 24 February 2003

Ludwig TSOI,
Senior Medical Officer
A&E Department, North District Hospital, HONG KONG
Send response to journal:
Re: What about caffeine-free coffee?

In response to Wisborg's article (1) of drinking coffee during pregancy related to stillbirth, it raises more issues than what it settles.
Firstly, the caffeine issue is not settled. As we all know coffee contains not only caffeine, but a mixture of different ingredients (milk, sugar, stabilizers, flavor, and other alkaloids from coffee beans). Attributing the result of drinking coffee to just one constituent seems illogical. If the authors want to prove what they contend, they should contrast the group taking 8 cups of ordinary coffee to those taking 8 cups of decaffeinated coffee.
Secondly, the issue of tea, chocolate and cola has been played down by the authors. In many parts of the world, these beverages are consumed in a large scale and high dose, like the traditional Chinese tea in Hong Kong. Yet, Hong Kong has one of the lowest perinatal mortality rate in the world. In Japan and China, tea is regarded as part of the healthy diet to be consumed regularly. The same issue also applies to cola in America and many parts of the world. So what will be the combined effect of coffee, tea, chocolate and cola should the woman be taking marginal levels of coffee (4-7 cups).
One of the gold standards in establishing causal relationship between two factors is to show the dose-dependent relationship. Your study can only showed that, statistically speaking, there is a one-off increase in stillbirth after 8 or more cups of coffee. To better illustrate the dose- dependent relationship, it would be desirable to subdivide the last group into those taking 8-11 cups, and those taking 12 cups or more.
Caffeine is also used widely to treat migraine headache. Should caffeine lead to stillbirth in the form of coffee, it should carry the same risk for women taking it for their migraine. Should we ban women from taking this seemingly harmless drug lest caffeine leads to stillbirth. Very often women in their early pregnancy are unaware of it, morning sickness complicating migraine would make them take their caffeine- containing tablets as before. Should we add a label to these prescriptions?
Lastly, I doubt very much about the estimation of the number of cups of coffee per day (and hence the dose of caffeine). The size of the cup, and the extent to which a cup is filled, all affect the actual amount of coffee taken. A large cup filled to the top and a small cup filled half way up can make a difference more than 100%. Putting 2 heaps of coffee and 3 heaps of coffee can make a difference of 50%. Drinking one brand of ordinary coffee versus another brand of black coffee can change the actual amount of caffeine dramatically. So how accurate can your estimation be?
Reference: (1) Wisborg K at el. BMJ 2003;326:420. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study
Competing interests: Coffee, tea, chocolate and cola lover.

Coffee consumption and perinatal outcome: The authors should adjust for history of drug abuse. 24 February 2003

Michael Sindos,
Research Fellow
The Whittington Hospital, London, N19 5NF, UK,
Narendra Pisal and Stavroula Michala
Send response to journal:
Re: Coffee consumption and perinatal outcome: The authors should adjust for history of drug abuse.

Editor- We read with interest the article of Wisborg et al (1). The authors support that high maternal coffee consumption during pregnancy is associated with an increased risk of stillbirth but not with infant death. They mention that women with a high intake of coffee are more likely to be smokers and to have a high intake of alcohol. The authors correctly adjusted the results for smoking and drinking habits. However, they do not provide any information about drug abuse amongst these women. It can be hypothesised that consumption of eight or more cups of coffee is suggestive of addictive behaviour. The use of illegal drugs such as cocaine is associated with an increased incidence of parallel cigarette and alcohol use (2,3). The adverse effect of maternal use of heroin, cocaine, crack cocaine and benzodiazepines in pregnancy has been adequately documented (3-5). The use of cannabis may not be a major prognostic factor regarding the outcome of pregnancy, but is an indicator of low socio-economic status and use of other harmful drugs (2).
The authors draw the profile of pregnant women that consume high amounts of coffee. They tend to be older, more often multiparous, more likely to be single, less likely to be students and they had fewer years of education. Interestingly this description matches with the profile of drug users (3,4). For the above reasons we believe that the exclusion of the information of drug abuse from the study is a methodological error.
References:
1. Wisborg K, Kesmodel U, Bech BH, Hedegaard M, Henriksen TB. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study.BMJ 2003;326:420
2. Miller JM Jr, Boudreaux MC.A study of antenatal cocaine use-chaos in action.Am J Obstet Gynecol 1999;180:1427-31
3. Singer L, Arendt R, Song LY, Warshawsky E, Kliegman R. Direct and indirect interactions of cocaine with childbirth outcomes. Arch Pediatr Adolesc Med 1994;148:959-64
4. Martinez A, Larrabee K, Monga M.Cocaine is associated with intrauterine fetal death in women with suspected preterm labor. Am J Perinatol 1996;13:163-6
5. McCarthy JE, Siney C, Shaw NJ, Ruben SM. Outcome predictors in pregnant opiate and polydrug users. Eur J Pediatr 1999 ;158:748-9
Competing interests: None declared

Statistical concerns 25 February 2003

Adam Jacobs,
Director
Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
Re: Statistical concerns

Wisborg et al found an association between maternal coffee consumption and risk of stillbirth, and claim that 'after adjustment for potential confounding factors the association remained significant.' However, I am not convinced that the data they present supports that claim.
The claim of a significant association appears to be based on the statistical significance for the odds ratio that compares the highest consumption group with the zero consumption group. This odds ratio is in any case only marginally significant at the 5% level, as the 95% confidence interval extends as far as 1. More importantly, the authors do not present the results of an overall test of differing risk of stillbirth among all the coffee consumption groups.
From a statistical point of view, it is not good practice to rely on pairwise comparisons between specific groups if the overall group effect is not significant. We are not told whether it is significant or not, but given that neither the low nor medium consumption group differs from the zero consumption group, and the high consumption group differs only marginally from the zero consumption group, my guess is that it isn't.
Competing interests: None declared

Coffee and stillbirths, via the effect of coffee on total homocysteine? 26 February 2003

Dag S. Thelle,
Professor of CVD epidemiology and prevention
Institute of Cardiovascular Diseases, Sahlgrenska University Hospital. S-416 85 G�teborg,
Benedicte Paus
Send response to journal:
Re: Coffee and stillbirths, via the effect of coffee on total homocysteine?

Coffee and reproductive health
The paper by Wisborg et al should be assessed in light of so far twenty published observational studies assessing the association between caffeine and/or caffeine containing beverages and hazards of reproductive health(1- 21). Thirteen of these studies showed a positive relationship between the intake of caffeine, caffeine metabolites or caffeine beverages and reproductive hazards such as spontaneous abortions, low birth weight or growth retardation (5-9, 12-14,16, 18-21) whereas seven did not show any significant associations (2-4,10,11,15,17). The seven inconclusive studies were all retrospective with regard to exposure ascertainment, whereas the nine prospective studies all demonstrated an association between coffee/caffeine or caffeine metabolite exposure and spontaneous abortion or retarded foetal growth. We are not aware of any formal meta-analysis assessing these studies, but the validity of prospective studies is likely to be higher than restrospective studies. It is notoriously difficult to link nutritional habits to health outcome at an individual level, but Wisborg et als paper supports the hypothesis that coffee/caffeine exposure is associated with increased risk of reproductive hazards.
This association might be related to the effect of coffee on total homocysteine (tHcy) levels. Four randomised intervention trials have shown that intake of both filtered and unfiltered coffee at an ordinary consumption level affects the tHcy levels (22-25). The more coffee, the higher the tHcy levels. tHcy levels have long been associated with a increased risk of reproductive hazards such as spontaneous abortions, intrauterine fetal death, or fetal growth restriction (26-31). Based upon the above, we suggest that the observation of increased risk of reproductive hazards associated with the intake of coffee is mediated via the association between coffee consumption and tHcy levels.
References
1. Wisborg K, Kesmodel U, Hammer Bech B, Hedegaard M , Brink Henriksen T, Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study BMJ 2003;326:420
2. Vollset SE, Refsum H, Irgens LM, Emblem BM, Tverdal A, Gjessing HK, Monsen AL, Ueland PM. Plasma total homocysteine, pregnancy complications, and adverse pregnancy outcomes: The Hordaland Homocysteine Study. Am J Clin Nutr 2000; 71: 962-68.
3. Linn S, Schoenbaum SC, Monson RR, Rosner B, Stubblefield PG, Ryan KJ. No association between coffee consumption and adverse outcomes of pregnancy. N Engl J Med 1982 ; 306:141-5
4. Kurppa K, Holmberg PC, Kuosma E, Sax�n L. Coffee consumption during pregnancy and selected congenital malformations: a nationwide case- control study. Am J Public Health 1983 ; 73:1397-9
5. Furuhashi N, Sato S, Suzuki M, Hiruta M, Tanaka M, Takahashi T. Effects of caffeine ingestion during pregnancy. Gynecol Obstet Invest 1985; 19:187-91
6. Srisuphan W, Bracken MB. Caffeine consumption during pregnancy and association with late spontaneous abortion.Am J Obstet Gynecol 1986 ; 154:1 14-20
7. Martin TR, Bracken MB.The association between low birth weight and caffeine consumption during pregnancy. Am J Epidemiol 1987 ; 126:813-21
8. Caan BJ, Goldhaber MK. Caffeinated beverages and low birthweight: a case-control study Am J Public Health 1989 ; 79:1299-300
9. Armstrong BG, McDonald AD, Sloan M. Cigarette, alcohol, and coffee consumption and spontaneous abortion. Am J Public Health 1992 ; 82:85-7
10. Mills JL, Holmes LB, Aarons JH, Simpson JL, Brown ZA, Jovanovic- Peterson LG, Conley MR, Graubard BI, Knopp RH, Metzger BE. Moderate caffeine use and the risk of spontaneous abortion and intrauterine growth retardation. JAMA 1993 ; 269:593-7
11. Fortier I, Marcoux S, Beaulac-Baillargeon L. Relation of caffeine intake during pregnancy to intrauterine growth retardation and preterm birth. Am J Epidemiol 1993 ; 137:931-40
12. Stanton CK, Gray RH. Effects of caffeine consumption on delayed conception. Am J Epidemiol 1995; 142:1322-9
13. Dlugosz L, Belanger K, Hellenbrand K, Holford TR, Leaderer B, Bracken MB. Maternal caffeine consumption and spontaneous abortion: a prospective cohort study. Epidemiology 1996 ; 7:250-5
14. Fenster L, Hubbard AE, Swan SH, Windham GC, Waller K, Hiatt RA, Benowitz N. Caffeinated beverages, decaffeinated coffee, and spontaneous abortion Epidemiology 1997 ; 8:515-23
15. Santos IS, Victora CG, Huttly S, Carvalhal JB. Caffeine intake and low birth weight: a population-based case-control study.Am J Epidemiol 1998 ; 147:620-7
16. Parazzini F, Chatenoud L, Di Cintio E, Mezzopane R, Surace M, Zanconato G, Fedele L, Benzi G. Coffee consumption and risk of hospitalized miscarriage before 12 weeks of gestation.Hum Reprod 1998 ; 13:2286-91
17. Fenster L, Quale C, Hiatt RA, Wilson M, Windham GC, Benowitz NL. Rate of caffeine metabolism and risk of spontaneous abortion.Am J Epidemiol 1998; 147:5 503-10
18. Klebanoff MA, Levine RJ, DerSimonian R, Clemens JD, Wilkins DG. Maternal serum paraxanthine, a caffeine metabolite, and the risk of spontaneous abortion. N Engl J Med 1999 ; 341:1639-44
19. Eskenazi B, Stapleton AL, Kharrazi M, Chee WY. Associations between maternal decaffeinated and caffeinated coffee consumption and fetal growth and gestational duration.Epidemiology 1999 ; 10:242-9
20. Cnattingius S, Signorello LB, Anner�n G, Clausson B, Ekbom A, Ljunger E, Blot WJ, McLaughlin JK, Petersson G, Rane A, Granath F. Caffeine intake and the risk of first-trimester spontaneous abortion.N Engl J Med 2000 ; 343:1839-45
21. Wen W, Shu XO, Jacobs DR, Brown JE. The associations of maternal caffeine consumption and nausea with spontaneous abortion.Epidemiology 2001 ; 12:38-42
22. Grubben MJ, Boers GH, Blom HJ, Broekhuizen R, de Jong R, van Rijt L, de Ruijter E, Swinkels DW, Nagengast FM, Katan MB. Unfiltered coffee increases plasma homocysteine concentraitions in healthy volunteers: a randomized trial. Am J Clin Nutr 2000; 71: 480-86
23. Urgert R, van Vliet T, Zock PL, Katan MB. Heavy coffee consumption and plasma homocysteine: a randomized controlled trial in healthy volunteers. Am J Clin Nutr 2000; 725: 1107-10.
24. Christensen B, Mosdol A, Retterstol L, Landaas S, Thelle DS.Abstention from filtered coffee reduces the levels of homocysteine and cholesterol - a randomized, controlled trial. Am J Clin Nutr 2001 ; 74;3:302-7.
25. Strandhagen E, Landaas S, Thelle DS. Folate supplement reduces the homocysteinraising effect of filtered coffee. A randomised placebo controlled trial. Eur J Clin Nutr (in press)
26. Nelen WL, Blom HJ, Steegers EA, den Heijer M, Thomas CM, Eskes TK. Homocysteine and folate levels as risk factors for recurrent early pregnancy loss. Obstet Gynecol 2000; 95: 519-24.
27. Wouters MG, Boers GH, Blom HJ, Trijbels FJ, Thomas CM, Borm GF, Steegers-Theunissen RP, Eskes TK. Hyperhomocysteinemia: a risk factor in women with unexplained recurrent early pregnancy loss. Fertil Steril 1993; 60: 820-25.
28. Goddijn-Wessel TA, Wouters MG, van de Molen EF, Spuijbroek MD, Steegers-Theunissen RP, Blom HJ, Boers GH, Eskes TK. Hyperhomocysteinemia: a risk factor for placental abruption or infarction. Eur J Obstet Gynecol Reprod Biol 1996; 66: 23-29.
29. Sorensen TK, Malinow MR, Williams MA, King IB, Luthy DA. Elevated second-trimester serum homocyst(e)ine levels and subsequent risk of preeclampsia Gynecol Obstet Invest 1999; 48: 98-103.
30. Leeda M, Riyazi N, de Vries JI, Jakobs C, van Geijn HP, Dekker GA. Effects of folic acid and vitamin B6 supplementation on women with hyperhomocysteinemia and a history of preeclampsia or fetal growth restriction. Am J Obstet Gynecol 1998; 179: 135-139.
31. Vollset SE, Refsum H, Irgens LM, Emblem BM, Tverdal A, Gjessing HK, Monsen AL, Ueland PM. Plasma total homocysteine, pregnancy complications, and adverse pregnancy outcomes: The Hordaland Homocysteine Study. Am J Clin Nutr 2000; 71: 962-68.
Competing interests: � None declared

Re: coffee, homocysteine and stillbirths 28 February 2003

Richard G Fiddian-Green,
None
None
Send response to journal:
Re: Re: coffee, homocysteine and stillbirths

The associations between coffee consumption and reproductive health (1,2)add weight to the proposal that phosphodiesterase inhibitors and other means of enhancing the metabolic effects of cAMP may cause an impairment of mitochondrial oxidative phosphorylation (3-7). The putative effect is likely to be dose-related and most likely to be evident when ATP resynthesis by oxidative phosphorylation is impaired.
The rise in homocysteine might be a reflection of a compensatory mechanism designed to preserve the adenine nucleotide pool during anaerobiosis. Loss of adenine nucleotide has serious consequences for even in optimal circumstances it is said to take 300 days to replenish the pools by de novo synthesis. If some degree of impairment perists it can be expected to take much longer and in some circumstaances to be unattainable. The risk of developing organ dysfunctions and failures in patients whose mitochondrial oxidative phosphorylation is impaired and especially in those whose adenine nucleotide pool is depleted can be expected to be high. Stillbirth may be one of the consequences.
1. Coffee and stillbirths, via the effect of coffee on total homocysteine? Dag S. Thelle, Benedicte Paus (26 February 2003)
2. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study Kirsten Wisborg, Ulrik Kesmodel, Bodil Hammer Bech, Morten Hedegaard, and Tine Brink Henriksen BMJ 2003; 326: 420
3.Consequences of vascular or mitochondrial disorders? Richard G Fiddian-Green (27 February 2003)
4. Spontaneous loss of early pregnancy and risk of ischaemic heart disease in later life: retrospective cohort study Gordon C S Smith, Jill P Pell, and David Walsh BMJ 2003; 326: 423-424
5. Madness, hyperhomocysteinemia, metabolic rate and body temperature Richard G Fiddian-Green bmj.com/cgi/eletters/325/7378/1433#28469, 6 Jan 2003
6.pH/calcium regulation of homocysteine metabolism Richard G Fiddian-Green bmj.com/cgi/eletters/325/7374/1202#28569, 9 Jan 2003
7. Homocysteine: an adenine nucleotide trap? Richard G Fiddian-Green bmj.com/cgi/eletters/325/7374/1202#28517, 7 Jan 2003
Competing interests: � None declared

Caffeine�s role in pregnancy outcome � a complex picture? 21 March 2003

Marcus S. Cooke,
Lecturer
University of Leicester, Leicester Royal Infirmary, Clinical Sciences Building, Leicester, LE2 7LX,
Sara Kirk, Janet Cade, Mark D. Evans, Darren Greenwood, Justin Konje, Joseph Lunec, Nigel Simpson, James Walker and Christopher P. Wild
Send response to journal:
Re: Caffeine�s role in pregnancy outcome � a complex picture?

Marcus S. Cooke, Sara Kirk*, Janet Cade*, Mark D. Evans, Darren Greenwood*, Justin Konje, Joseph Lunec, Nigel Simpson*, James Walker* and Christopher P. Wild*.
University of Leicester, and *University of Leeds.
We have read with interest the paper by Wisborg et al., (1) and the rapid responses received to date. These clearly illustrate the problems in exploring any potential link between caffeine and pregnancy outcome, such as coffee consumption as a proxy marker, and single timepoint assessments of intake.
For decades, there has been concern as to whether pregnant women should avoid caffeine-containing beverages during pregnancy, although this was largely based on animal studies and circumstantial evidence, rather than proof derived from human studies (2). Reports of Caffeine intake during pregnancy has been associated with congenital malformation, low birth weight or spontaneous abortion (SAB), but with varying and often contradictory conclusions (3). Nevertheless, consumption of caffeine is common during pregnancy (4), and is not exclusively from tea and coffee; chocolate/cocoa, cola and energy drinks, and an appreciable number of non- prescription drugs may also contribute significantly (5).
Whilst it appears that moderate caffeine consumption (150mg/day) has no effect on SAB (6), doses in excess of 300mg/day may be associated with an increased risk of SAB and low birth weight, a conclusion supported by the findings of the Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (7). The disparate conclusions from the human epidemiological studies may be explained by a number of methodological limitations, as discussed below.
Reliance upon simply recording the number of tea or coffee servings consumed per day is the largest single source of error in epidemiological studies, particularly as there are numerous sources of caffeine.
Furthermore, there is a significant mis-classification of exposure when coffee consumption alone is used as a surrogate measure of caffeine intake (8), and this has been a major impediment to the assignation of a threshold level of caffeine intake, levels above which may represent a significant risk during pregnancy.
Another major limitation of previous epidemiological studies is the reliance on birth weight as the end-point for assessing fetal growth. It is well recognised that the diagnosis of restricted fetal growth cannot simply be assigned by birthweight alone, as variation in birthweight naturally occurs due to fetal factors such as sex, and maternal factors such as height, weight, parity, and ethnicity (9). Studies relying on a birthweight cut-off, even when limiting this to those babies born at term, will inevitably include a large number of apparently appropriate-for- gestational-age babies that in reality have not achieved their growth potential, and indeed will exclude growth-restricted babies that apparently reach a �normal� weight at birth (10).
Failure to account for inter-individual variation in the metabolism of caffeine is another potential limitation. Of the four primary routes of caffeine metabolism in humans, 3�-demethylation is quantitatively the most important reaction, with CYP1A2 being identified as the enzyme responsible. Studies have shown there to be a wide inter-individual variation in CYP1A2, although the genetic basis for this is not understood (11).
Smoking is an important confounding factor as, on average, smokers consume more caffeine than non-smokers and, smoking increases the rate at which caffeine is metabolised (12). Several studies have shown a positive association between smoking, SAB and low birth weight (reviewed in 13,14).
Associated with smoking, alcohol has well established teratogenic properties and may, in itself, be associated with low birth weight (14). Alone, or in combination, clearly these factors may complicate meaningful data interpretation.
We represent a consortium of scientists and clinicians, funded by the UK Food Standards Agency, to study the effects of caffeine upon pregnancy. Unlike many previous investigations of caffeine and reproductive effects, this is a prospective study that includes biomarkers of caffeine intake and metabolism. The study will consider all sources of caffeine intake, not just tea and coffee, along with relevant confounding factors, similar to a recent report by Bracken et al. (15). Whilst this recent publication, and our investigation, represent significant improvements on previous studies, by addressing major confounding factors, and refining endpoint measurements, we will extend this research by examining a potential influence of inter-individual variability in response to caffeine consumption (CYP1A2 phenotyping). The considerable size of this study (3000 pregnant women) is achieved through recruitment of volunteers from three of the largest maternity units within the UK. It is hoped that the research will help to reduce the uncertainties in the risk assessment and provide a robust basis for advice to pregnant women on the issue of caffeine consumption.
References
1. Wisborg K, Kesmodel U, Bech BH, Hedegaard M, Henriksen TB. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study. BMJ 2003; 326 : 420
2. Golding J. Reproduction and caffeine consumption � a literature review. Early Hum. Development 1995; 43 : 1-14.
3. Christian MS, and Brent RL. Teratogen update: evaluation of the reproductive and developmental risks of caffeine. Teratology 2001; 64 : 51 -78.
4. Hill RM, Craig JP, Chaney MD. et al. Utilisation of over-the-counter drugs during pregnancy. Clin. Obstet. Gynecol., 1977; 20: 381-94.
5. Morris, MB. and Weinstein, L. Caffeine and the fetus: is trouble brewing? Am. J. Obstet. Gynecol. 1981; 140: 607-610.
6. Klebanoff, MA., Levine, RJ., DerSimonian, R. et al. Maternal serum paraxanthine, a caffeine metabolite, and the risk of spontaneous abortion. N. Engl. J. Med. 1999 341, 1639-1644.
7. Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment. Statement on the reproductive effects of caffeine. 2001; October, http://www.food.gov.uk/science/ouradvisors/toxicity/statements/cotstatements2001/caffeine
8. Brown J, Kreiger N, Darlington GA. et al Misclassification of exposure: coffee as a surrogate for caffeine intake. Am. J. Epidemiol. 2001; 153: 815-20.
9. Gardosi J, Chang A, Kalyan B, Sahota D, Symonds EM. Customised antenatal growth charts. Lancet 1992, Feb 1; 339 (8788): 283-7
10. Mongelli M, Gardosi J. Reduction of false-positive diagnosis of fetal growth restriction by application of customized fetal growth standards. Obstet. Gynecol. 1996; 88: 844-8.
11. Oscarson M, Ingelman-Sundberg M, Daly AK et al (2000) Human cytochrome P450 (CYP) alleles [web site]. http://www.imm.ki.se/CYPalleles/
12. Benowitz NL, Hall SM and Modin G. (1989) Br. J. Med., 298, 1075-1076.
13. Craemer DW. and Wise LA. The epidemiology of recurrent pregnancy loss. Seminars in Reprod. Med., 2000; 18 : 331-9.
14. Gardella JR. and Hill JA. Environmental toxins associated with recurrent pregnancy loss. Seminars in Reprod. Med., 2000; 18: 407-24.
15. Bracken, MB, Triche, EW, Belanger, K, Leaderer BP. Association of maternal caffeine consumption with decrements in fetal growth. Am. J. Epidemiol., 2003; 157: 456-66.
Competing interests: � None declared

Bitter experience 13 June 2003

Sean Tierney,
Consultant Surgeon
Adelaide & Meath Hospital Dublin
Send response to journal:
Re: Bitter experience

I am not an expert but my reading in this area does suggest that women may find that their taste for coffee/tea or other drinks may be reduced during pregnancy (particularly early pregnancy) because women become much more aware of (and averse towards) bitter flavours such as the alkaloids in both tea and coffee.
It has also been suggested that a sensitivity to alkaloids or bitter substances may also be an evolutionary advantage associated with morning sickness during pregnancy.
Have the authors considered the possibility that those who shun coffee during pregnancy might also avoid other bitter substances which might explain the observed effect?
Competing interests: None declared