Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Cut caffeine in pregnancy?
- Johanna M Geleijnse, P.O. Box 8129, 6700 EV Wageningen, The Netherlands (4 November 2008)
-
Therapeutic use
- Neal Devitt MD (4 November 2008)
-
Fast caffeine clearance may be a surrogate variable for smoking.
- Martijn B. Katan (6 November 2008)
-
Half-life a proxy for clearance?
- Joseph F Standing (7 November 2008)
-
Let´s focus on tobacco not caffeine
- Hannu Vierola (7 November 2008)
-
Caffeine for premature infants: time for a revisit
- Roy K Philip, Adnan Mirza and Margo Dunworth (12 November 2008)
-
FSA advice on caffeine
- Andrew Wadge (13 November 2008)
-
Need for a good night!
- Isabelle Marc, Frederic Series (12 December 2008)
-
Caffeine - not just coffee
- Dr Euan H M Paul (15 December 2008)
-
Re: Therapeutic use
- Dr Viera Scheibner PhD (8 January 2009)
-
Maternal Caffeine intake during pregnancy and risk of fetal growth restriction: a large prospective observational study
- Justin C Konje, Neelam Potdar, Justin Konje and Marcus Cooke (Leicester), Darren Greenwood, Alastair Hay and Janet Cade (Leeds) on behalf of CARE Study Group (22 January 2009)
|
|
|||
|
Johanna M Geleijnse, nutritional epidemiologist Division of Human Nutrition, Wageningen University, P.O. Box 8129, 6700 EV Wageningen, The Netherlands
Send response to journal:
|
From their prospective study of maternal caffeine intake during pregnancy and fetal growth restriction, the CARE Study Group concluded that caffeine intake before conception and during pregnancy should be reduced (1). The UK Food Standards Agency (FSA) responded immediately, advising pregnant women to drink no more than two cups of coffee or four cups of tea a day. The impact of such an advice is huge given the fact that coffee and tea are widely consumed beverages, and women are generally concerned about their pregancy outcome. This calls for careful consideration of the abovementioned study from which the risk estimations were derived. Observational epidemiological studies are prone to bias and confounding. When studying nutrient intakes in relation to pregnancy outcomes, there is a large number of factors that should be dealt with in the design and/or statistical analysis of the study. Factors to be considered include maternal health consciousness, changes in body weight and smoking during pregnancy, and pregnancy-related health complaints such as hypertension, diabetes, and nausea. These factors, in my opinion, have not adequately been addressed in the paper by the CARE Study Group. To illustrate, it is unclear how nausea during pregnancy was measured and how adjustments were made in the multivariable model. The commonly occuring phenomenon of nausea and vomiting could play a central role in the complex biological pathways leading to pregnancy outcome. It has been hypothesized that nausea stimulates early placental growth, and morning sickness has been associated with positive pregnancy outcome and a decreased risk of low birth weight (2). Stein and Susser postulated that nausea in pregnancy may create an erroneous association between caffeine consumption and miscarriage (3), which may also hold for the present study on fetal growth restriction. Nausea is associated with increasing hormone levels during a normal pregnancy and is less common in pregnancies with a poor outcome (3). It should be noted that nausea is not merely a confounder that can be put in the model as a dummy variable, but a major determinant (antecedent) of nutrient intakes during pregnancy, including caffeine. In the present study it would have been more appropriate to perform a stratified analysis and examine the relation of caffeine with fetal growth restriction separately for women with and without nausea. In women who experience nausea, a low intake (in fact: lowered intake) of caffeine is much more frequent and likely to be related to the severity of nausea in a dose- reponse manner. Finding associations for caffeine in women without nausea would have strengthened the conclusions. With regard to other confounders, it appears for smoking that only adjustment for baseline status was made and not for changes in status and amount of cigarettes smoked during pregnancy. This could have influenced the findings, since smoking is associated with the intake of caffeinated beverages and also with offspring birth weight. Furthermore, the role of health consciousness of the mother was not addressed. The investigators report that they conducted sensitivity analyses to test the robustness of the results, but the paper provides no insight into the impact of confounders or effect modifiers on the data. For these reasons I conclude that the present study does not provide a strong scientific basis for cutting down coffee and tea consumption during pregnancy. Johanna M. Geleijnse, PhD Competing interests: None References 1. CARE Study Group. Maternal caffeine intake during pregnancy and risk of fetal growth restriction: a large prospective observational study. BMJ 2008;337;a2332. 2. Huxley RR. Nausea and vomiting in early pregnancy: its role in placental development. Obstet Gynecol. 2000;95:779-782. 3. Stein Z, Susser M. Miscarriage, caffeine, and the epiphenomena of pregnancy: the causal model. Epidemiology 1991;2:163-167. Competing interests: None declared |
|||
|
|
|||
|
Neal Devitt MD, physician La Familia Medical Center 87505
Send response to journal:
|
My wife regularly consumed more than 3 cups of coffee daily during her pregnancies. Our babies were 9 and 10 pounds. Perhaps for us caffeine had a positive effect in reducing fetal weight and the risk of birth complications. Competing interests: None declared |
|||
|
|
|||
|
Martijn B. Katan, Professor of Nutrition VU University, Institute of Health Sciences, De Boelelaan 1085, Amsterdam, 1081 HV The Netherlands
Send response to journal:
|
The CARE study group reported that caffeine intake in pregnant women was associated with lower birth weight. The effect was especially pronounced in women with rapid caffeine clearance (Table 5). The authors speculate that this points to a role for caffeine metabolites.
There is, however, a simpler explanation. Caffeine clearance is sped up, and caffeine half life is shortened, by cigarette smoking [1,2]. The apparent effect of caffeine clearance on birth weight could therefore be due to confounding by cigarette smoking. The authors may argue that adjustment for smoking reduced but did not eliminate the effect of caffeine (table 3). However, exposure of the foetus to cigarette smoke cannot be measured with 100% precision, and therefore residual confounding by smoking may still be present even after statistical adjustment. It would be helpful if the authors would provide a table of caffeine intake versus smoking, so that the potential for confounding can be evaluated. References 1. Murphy TL, McIvor C, Yap A, Cooksley WG, Halliday JW, Powell LW. The effect of smoking on caffeine elimination: implications for its use as a semiquantitative test of liver function. Clin Exp Pharmacol Physiol. 1988 Jan;15(1):9-13 2. Joeres R, Klinker H, Heusler H, Epping J, Zilly W, Richter E. Influence of smoking on caffeine elimination in healthy volunteers and in patients with alcoholic liver cirrhosis. Hepatology. 1988 May-Jun;8(3):575-9. Competing interests: None declared |
|||
|
|
|||
|
Joseph F Standing, Postdoctoral Fellow Department of Pharmaceutical Biosciences, Uppsala Universitet, SE 751 24, Sweden.
Send response to journal:
|
The findings relating to caffeine half-life in this paper maybe misleading as the authors state “half-life is a proxy for clearance” [1]: ke=Cl/Vd t1/2 = ln2/ke AUC=Dose/Cl ke = elimination rate constant, Cl = clearance, Vd = volume of distribution, t1/2 = elimination half-life, AUC = area under the concentration-time curve. Exposure to caffeine is measured by AUC. Imagine two mothers (one big, one small) with the same Cl. The small one will have a lower Vd and therefore higher ke, her t1/2 will be smaller, but their baby’s exposure (AUC) to caffeine will be the same. Another interpretation of the finding “short caffeine half-life is linked to growth restriction”[1] could be "smaller women have smaller babies". The authors also state they were unable to measure the rate of formation or elimination of caffeine metabolites [1]. If metabolites were measured in saliva, it is possible to estimate their formation and elimination rates using population pharmacokinetic modelling (even with only two samples per subject). To attempt to link caffeine Cl and birth weight, the pharmacokinetic data collected ought to have been analysed using pharmacokinetic modelling. It would seem that the ABPI’s concern over a lack of pharmacokinetic literacy in the UK [2] is justified. References 1. CARE Study Group Maternal caffeine intake during pregnancy and risk of fetal growth restriction: a large prospective observational study. BMJ 2008; 337: a2332. 2. Association of the British Pharmaceutical Industry, Gaps In Skills Supply Places UK Biomedical Competitiveness At A Tipping Point, 2008. Available at: http://www.abpi.org.uk/press/press_releases_08/131008.asp Competing interests: I receive postdoctoral fellowship funding from Pfizer to do pharmacokinetic research. |
|||
|
|
|||
|
Hannu Vierola, Senior Obstetrician Iimed Ltd
Send response to journal:
|
Nonsense for obstetricians! Truly we should focus our studies on cigarettes, alcohol and drugs in pregnancy. Pregnant women who use tobacco or drugs or both ususally drink more coffee-maybe here we can see any clinical effect of this study. Warnings about usual 2-3 cups/day coffee drinking during pregnancy make young pregnant women think that "coffee is dangerous cigarettes not"-it is easier to stop drinking coffee than quit smoking! The press like to publish this kind of studies with very little practical meaning. In Finland 5-20 % of pregnant women have pathological glucose tolerance--and we know that caffeine prevents insulin tolerance and is prevention for diabetes, parkinsonims etc. Why speak of 20-30 g lower weight with caffeine use while we see tobacco babies 200-500 g smaller.. Competing interests: None declared |
|||
|
|
|||
|
Roy K Philip, Consultant Paediatrician Regional Maternity Hospital, Limerick, Ireland, Adnan Mirza and Margo Dunworth
Send response to journal:
|
The CARE Study Group article1 reiterating the association of fetal growth restriction to caffeine consumption during pregnancy could stimulate a search for alternative options to treat apnoea of prematurity (AOP) in neonatal units. Methylxanthines are universally used in the management of AOP for more than 2 decades, caffeine citrate being the most commonly used agent. No other stage in life is caffeine therapeutically used on a per kilogram basis more than prescribed to the premature infants during their most vulnerable period of protoplasmic growth. Current dosage regimens for AOP treatment expose a 25 to 32 week gestation premature infant to a much higher mg/kg/day of xenobiotic than even the high level of exposure from dietary caffeine sources during pregnancy as quoted in the CARE study. Usual therapeutic range reached among premature infants is 18-23 mg/L. As maintenance dose of 5-10 mg/kg/day of caffeine citrate (2.5 – 5 mg base) is often prescribed to 500 g to 1500 g neonates (equivalent to a 150 - 300 mg/day derived from tea or coffee for a 60 kg woman during 2nd trimester of pregnancy, even assuming comparable pharmacokinetics), CARE group observation that the fetal growth restriction was consistent across all trimesters1 is truly worrying. Similar observation reported in the past2 also suggests a need to reflect upon the prolonged neonatal caffeine prescription to extremely low birth weight (ELBW) and very low birth weight (VLBW) infants. Absence of cytochrome P450 1 A2 in the fetus, awareness that caffeine is primarily metabolized in human liver1, diuretic, hyperglycemic and catabolic effect of caffeine3, all could suggest the potential for short term growth concern due to the current practice of postnatal caffeine for AOP. Even though 18-21 month follow up of caffeine exposed infants for the treatment of AOP did not show reduction in weight or head circumference4, somatic effect of dose and duration specific exposure is not ascertained. Our observation suggesting a clinically and statistically significant reduction in the weight gain of a four year (January 2003 to December 2006) prospective cohort of ELBW and VLBW infants exposed to a higher duration and dose of caffeine for the treatment of AOP is summarized in table 1. Our regional maternity unit with an in-birth rate of 5,400 per year from a fairly fixed Caucasian population (ethnic minorities accounting for <5% of antenatal booking) has an antenatal steroid uptake of 94%. Excluded from the analysis were those with major congenital anomalies, exposure to diuretics or postnatal steroids and culture proved sepsis. Decision to use the 5 mg/kg/day or 10 mg/kg/day of maintenance dose was at the discretion of the treating clinical team. Maternal exposure to caffeine / tobacco / steroids or neonatal morbidity variables was not controlled. Our results, acknowledging the absence of a placebo group and lack of correction for post-natal variables that perhaps prompted the clinical team to use a higher dose for some infants still merits the suggestion of a postnatal growth restriction secondary to caffeine exposure. Relative safety of caffeine compared to other drugs used in the treatment of AOP perhaps makes clinicians less vigilant to limit the duration of treatment thus leading to probable unnecessary exposure with questionable benefit. Recent advances in respiratory support such as nasal continuous positive pressure (NCPAP) with facilities for back-up breaths / synchronization and nasal intermittent positive pressure ventilation (NIPPV) with additional synchronization need to be explored further along with postural adjustments / changes to infant care procedures, to avoid the undue reliance on neonatal caffeine use. Our observation is not powered to establish causation however; the tabulated association along with the antenatal reports1, 2 warrants case control studies to address this under-perceived element of suboptimal postnatal weight gain among premature infants. Reference: 1. CARE Study Group. Maternal caffeine intake during pregnancy and risk of fetal growth restriction: a prospective observational study. BMJ 2008;337:2332-43 2. 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 3. Aranda JV, Turmen T, Sasyniuk BI. Pharmacokinetics of diuretics and Methylxanthines in the neonate. Eur J Clin Pharmacol. 1980;18:55-63. 4. Schmidt B, Roberts RS, David P, et al. Long-term effects of caffeine therapy for apnoea of prematurity. N Eng J Med. 2007; 357:1893-902. Table 1. Mean daily weight gain of ELBW and VLBW infants treated with two different dosage schedules of caffeine citrate for the treatment of apnoea of prematurity (AOP). _________________________________________________________________________________ Dose 5 mg/kg/day 10 mg/kg/day p 5mg/kg/day 10mg/kg/day p (2.5 mg/kg base) (5 mg/kg base) (2.5 mg/kg base) (5 mg/kg base) Number (n = 230) (n = 122) (n = 202) (n = 150) __________________________________________________________________________________ Days of 14 – 28 14 - 28 29 – 56 29 – 56 treatment Gestation 28 ± 3.4* 27.6 ± 3.2 NS 27.5 ± 3.2 28.1 ± 3.5 NS at birth Birth weight 1024 ± 422 968 ± 414 NS 955 ± 403 970 ± 393 NS (Grams) IMV / SIMV 16 ± 10 14 ± 11 NS 21 ± 16 19 ± 13 NS Days NCPAP 22 ± 16 25 ± 12 NS 33 ± 19 39 ± 14 NS days kcal / kg** 122 ± 14 119 ± 16 NS 123 ± 14 121 ± 16 NS Weight gain 16.9 ± 13 14.04 ± 12.2 0.04 14.7 ± 10.3 10.5 ± 9.1 0.01 Per day (grams) * Mean ± SD, NS – not significant, IMV – intermittent mandatory ventilation, SIMV – synchronized intermittent mandatory ventilation, NCPAP – nasal continuous positive airway pressure, kcal - kilocalories, ** enteral or parenteral route. Competing interests: None declared |
|||
|
|
|||
|
Andrew Wadge, Chief Scientist Food Standards Agency, WC2B 6NH
Send response to journal:
|
The Food Standards Agency’s new advice to women on the amount of caffeine they should consume during pregnancy was not based on the results of the FSA-funded research alone. The independent experts on the Committee on Toxicity, after considering the results of this new study as well as all of the other studies that have been published in peer-reviewed literature on this subject, concluded that caffeine intake during pregnancy is associated with an increased risk of fetal growth restriction. On the basis of this risk assessment considering all of the available information, the Food Standards Agency has changed its advice. Competing interests: None declared |
|||
|
|
|||
|
Isabelle Marc, Assistant Pr, Dpt Pediatrics, ULaval 2705 boulevard Laurier, Québec, Québec G1V 4G2, Frederic Series
Send response to journal:
|
Physical and hormonal changes that occur during pregnancy can affect sleep duration and architecture and stability. Changes in sleep stages occur as early as 11-12 weeks of gestational age. Pregnant women spend less time in deep sleep with more light sleep with an increase in time spent awake during the night (decrease in sleep efficiency) compared to pre-pregnancy sleep. In some studies, complications as pre-eclampsia, diabetes and fetal growth restriction have been associated with sleep disturbances and snoring. As caffeine consumption can be a behavioral consequence in response to disrupted sleep and sleepiness in pregnant woman, can the authors assess or discuss this issue in their study? Isabelle Marc, MD, PhD
References 1- The Influence of reproductive status and age on women's sleep. Lee KA, Baker FC, Newton KM, Ancoli-Israel S. J Womens Health (Larchmt). 2008 Sep;17(7):1209-14 2- Sleep disturbances increase interleukin-6 production during pregnancy: implications for pregnancy complications. Okun ML, Hall M, Coussons-Read ME. Reprod Sci. 2007 Sep;14(6):560-7. Competing interests: None declared |
|||
|
|
|||
|
Dr Euan H M Paul, Scientific Advisor CoSIC Chipping Norton OX7 5NQ
Send response to journal:
|
We bring to your attention the news item on your home page reporting on the study by Konje et al on CAFFEINE intake during pregnancy. It seems somewhat unfair that the picture next to this item is a cup of black coffee, when the focus of the study was caffeine from all sources. Further, and of particular relevance, 62% of dietary caffeine intake in this cohort came from tea! Perhaps you may consider adding a cup of tea and a can of cola to make your visual representation more accurate. Competing interests: None declared |
|||
|
|
|||
|
Dr Viera Scheibner PhD, Scientist/Author Retired Blackheath NSW Australia
Send response to journal:
|
Dear Editor, What a perfect textbook example of the paradoxical effect of the dosage! Three cups of coffee obviously had the opposite effect compared with just one cup of coffee, hence the lady's babies were heavier when she consumed three cups of coffee a day than if she had only consumed one cup of coffee a day, during her pregancies. It seems to me that she had empirically hit on the right dosage. This natural phenomenon could be therapeuticaly useful and should be further researched. After all, the paradoxical effect of stimulants, such as amphetamins, is already used in the alleviation of symptoms in some cases of ADHD. Caffeine side effects appear less harmful. Competing interests: None declared |
|||
|
|
|||
|
Justin C Konje, Professor of Obstetrics and Gynaecology University of Leiccester, LE2 7LX, Neelam Potdar, Justin Konje and Marcus Cooke (Leicester), Darren Greenwood, Alastair Hay and Janet Cade (Leeds) on behalf of CARE Study Group
Send response to journal:
|
The Editor BMJ We would like to thank Geleijnse, Katan and Standing for their comments on our paper on the association between caffeine and fetal growth restriction [1] We agree with Geleijnse that observational epidemiological studies are prone to bias and confounding, but our study used detailed, objective and validated measures of exposure [2]. We prospectively quantified caffeine intake and included aspects of metabolism and smoking in our model. We agree that there are a large number of parameters that could potentially affect the outcome in studies such as ours, and should therefore be factored into study design and analysis of results. Indeed we included several such confounders in our analyses. For example, maternal weight, smoking, nausea and pregnancy-related complications of diabetes and hypertension were included. Sensitivity analyses were also undertaken excluding women who experienced pregnancy-related hypertension and diabetes and the results were, again, unchanged. There is no published evidence that nausea stimulates placental growth, but rather that the hormones which are associated with nausea and vomiting in early pregnancy are associated with placental growth. As Geleijnse rightly points out, Stein and Susser [3] only postulated that nausea may create an erroneous association between caffeine and miscarriage. However, this was not supported by any data, and this association was mainly for miscarriage – an outcome measure which was not investigated in our study. We did, however assess self-reporting (number of days) nausea, in all three trimesters of pregnancy and used the data in the multivariable model. As an effect modifier, we modelled this using an interaction term (which is better than simply stratifying, as suggested). There was a stronger association for women suffering nausea than for those who did not suffer nausea. As 80% of our women suffered some nausea in the first trimester, these results should be relevant to a majority of women. Since we examined the association between caffeine intake, throughout pregnancy, with fetal growth, and nausea is much less of a problem after the first trimester, the persistence of the association would strengthen our conclusions. As for adjustment for smoking, Geleijnse is correct that this is very important, which is why we used an objective biomarker of smoking (salivary cotinine). We did repeat this measure in a large sub-sample and when we took this into account in a further analyses (results not published), the conclusions were unchanged. Qualitative assessment to study the role of health consciousness of the mother was beyond the scope of this study and perhaps future studies might be able to investigate this. With respect to the letter from Katan, it has been reported that after maternal smoking, high concentrations of nicotine and cotinine have been detected in the fetal circulation [4] possibly because of increased placental permeability to these compounds [5,6]. Indeed, the effects of maternal smoking have been demonstrated in the placenta and fetus at anatomic, enzymatic and cellular level. Any residual confounding effect of smoking on the fetus after considering maternal smoking as a covariate in the statistical model is therefore highly unlikely. In this study, adjustment for maternal caffeine reduced but did not eliminate the effect of caffeine on fetal growth. Joseph Standing makes a reference to the caffeine half-life used in our study and suggests that this could potentially be misleading. We are aware of the relationship between half-life and clearance. As both the fetus and the placenta lack the principal metabolising enzyme for caffeine, fetal exposure will not be constant, but dependent upon maternal caffeine clearance. We agree that “smaller women are likely to have smaller babies”, but this factor had been included in the identification of fetal growth restriction using the customized birth weight centile calculator. We measured various metabolites of caffeine which are currently being analyzed taking into account the mother's weight at booking for the assessment of caffeine clearance. This will subsequently be included in pharmacokinetic modelling to assess the individual contribution of caffeine and its metabolites to our findings We welcome these comments, but do not feel that the conclusions from our work, which was commissioned by the FSA, are weakened in any way by the comments made from the correspondences. Nor do we feel that these comments would impact on the FSA's recommendation to restrict maternal caffeine intake, as this was not only based on our findings relating to birth weight, but on an exhaustive meta-analysis of all published studies of both birth weight and miscarriage. References 1. CARE Study Group Maternal Caffeine intake during pregnancy and risk of fetal growth restriction: a large prospective observational study. BMJ 2008;337:a2332 2. Boylan SM, Cade JE, Kirk SFL et al. Assessing caffeine exposure in pregnant women. British Journal of Nutrition 2008; 100:875-882 3. Stein Z, Susser M. Miscarriage, caffeine and the epiphenomena of pregnancy. Epidemiology 1991; 2:163-167 4. Luck, W., Nau H., et al. Extent of nicotine and cotinine transfer to the human fetus, placenta and amniotic fluid of smoking mothers. Dev Pharmacol Ther, 1985; 8: 384-95. 5. Lambers, D. S. and Clark K. E. The maternal and fetal physiologic effects of nicotine. Semin Perinatol, 1996; 20: 115-26. 6. Jauniaux, E., Gulbis B., et al. Maternal tobacco exposure and cotinine levels in fetal fluids in the first half of pregnancy. Obstet Gynecol, 1999; 93: 25-9. Neelam Potdar, Justin C Konje, Marcus S Cooke (Leicester), Darren Greenwood, Alastair W Hay, Janet Cade (Leeds) For CARE Study Group Competing interests: None declared |
|||