Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Vitamin B6 for nausea and vomiting in pregnancy
- Ellen C G Grant (8 February 2004)
-
Treating Nausea & Vomiting in Pregnancy
- Michael D Croft, Shenley, Herts WD7 9LP (8 February 2004)
-
Depth of Learning
- Ed Peile (10 February 2004)
-
Nof 1 Trials in general practice
- Roger Gadsby (11 February 2004)
-
n=1 trial
- Allan J Savory (12 February 2004)
-
Length of Treatment Periods
- Martin Phillips (13 February 2004)
-
Treating Nausea and Vomiting During Pregnancy: n of 1 trial
- Anthony M Barnie-Adshead (19 February 2004)
-
MYMOP: correct references
- Charlotte Paterson (26 February 2004)
|
|
|||
|
Ellen C G Grant, Physician and medical gynaecologist 20 Coombe Ridings, Kingston-upon-Thames, Surrey, KT2 7JU, UK
Send response to journal:
|
The authors have chosen to use an n of 1 trial of vitamin B6 treatment for pregnancy sickness because of an interest in complementary or herbal medicine. Unlike much of alternative medicine, which often lacks a detailed scientific background, nutritional medicine is based on fundamental evidence based medical research. In the 1960s it was discovered that pregnancy and oral contraceptives lowered zinc, raised copper levels and caused vitamin B6 deficiencies.1,2 Many women are currently deficient in these and other essential nutrients like magnesium.3 Pregnancy, cortisol, oestrogens and oral contraceptives can induce vitamin B6 deficiency by increasing the activity of the enzyme tryptophan oxygenase, which requires vitamin B6 as a co-factor. Functional deficiency of vitamin B6 can impair the decarboxylation of dopa to dopamine. Vitamin B6 (pyridoxine) is converted into the active form, pyridoxal 5 phosphate, by riboflavin (vitamin B2) and magnesium. Vitamin B6 is required for normal essential fatty acid (EFA) desaturation and elongation. Zinc and magnesium are also required for EFA desaturation and such co-factor deficiencies block omega-6 and omega-3 EFA pathways causing deficiencies even when intake of first stage linoleic acid and alpha linolenic acid would otherwise be adequate. Repletion of cofactors is important because attempting to use only omega-6 EFAs or only omega-3 EFAs as a pharmaceutical4 can cause deficiencies in the other pathway. Zinc is necessary for the transport of vitamin B6 across cell membranes into the cell. Deficiency impedes the absorption of zinc and may impair cellular immunity and diminish antibody responses. A woman with vitamin B6 deficiency will consequently have disturbed protein, fat and carbohydrate metabolism and immunity.4-6 Biolab functional blood tests find B vitamins are most the most commonly deficient vitamins. A majority of patients are vitamin B6 deficient (an individual test costs £15) but they are usually given supplements with a group B vitamin complex to prevent supplementation of one B vitamin causing a deficiency of another B vitamin. For example, repletion of vitamin B 6 deficiency will increase the demand for vitamin B2. Why therefore would an unscreened pregnant woman be expected to become asymptomatic when supplemented with only one essential nutrient for two days out of five? Primary dysmenorrhoea, pregnancy sickness, premenstrual syndrome, and menopausal flushing, are usually warning signs of multiple biochemical upsets and an inability to cope with changes in hormone levels. Common nutritional deficiencies increase adverse reactions to foods and chemicals.3 Therefore giving sporadic or continuous single nutrient supplementation, or even pharmaceutical drugs, is unlikely to prevent the recurrence of daily symptoms. Pregnancy sickness should be taken seriously as a signal of nutrient deficiencies. It should not be regarded as an opportunity for randomised trials of pharmaceutical drugs or unmonitored single nutritional supplements. Pregnancy is about feeding a growing foetus with optimal quantities of nutrients. Preconceptional and maternal nutritional deficiencies are extremely common.6 Without evidence based advice on essential nutrient supplementation and the implementation of low allergy dieting, a pregnant woman suffering from nausea and vomiting is unlikely to be treated efficiently and she will not be able to achieve adequate blood levels of nutrients for her foetus. The effects may last for generations as has been indicated by animal studies.7 1. How does the evidence obtained from the n of 1 trial differ from the evidence of randomised controlled trials? The n of 1 trial is an inappropriate way to treat symptoms of pregnancy sickness with a single essential nutrient supplement. If such symptoms were caused by a single nutrient deficiency, or if use of a single nutrient like vitamin B6 given in a relatively low dose, had pharmaceutical actions, continuous treatment would be more likely to show an effect in a standard randomised controlled trial. 2. How would you apply this evidence to other patients presenting with nausea and vomiting in pregnancy? I would not apply this evidence to other patients as I do not expect this trial to give a meaningful result. Even if sporadic use of vitamin B6 helped this patient I would not assume all patients were vitamin B6 deficient. 3. Can you think of other clinical dilemmas that would be helped by the use of n of 1 trials? N of 1 trials could be of use in individuals who were not pregnant and who were willing to test short acting pharmaceutical drugs. Examples would be testing the effects of antihistamines, anti-emetics or analgesics for hay fever, sea sickness or otherwise untreatable chronic pain, respectively. Parenteral nutrients could be beneficial in the short term for some patients, for example magnesium for eclampsia but, as nutritional deficiencies can be diagnosed accurately, failing to investigate and replete seems unethical. Diagnosable nutrients deficiencies are not suitable for either type of trial. There is a need for nutritional analyses to become more widely available. 1 Halsted HJ, Hackly BM, Smith JC. Plasma zinc and copper in pregnancy and after oral contraception. Lancet 1968:2:278-83. 2 Rose DP. The effects of gonadol hormones and oral contraceptives on tryptophan metabolism. In: Eds Salhanick HA, Kipnis DM, Vande Weile RL. Metabolic effects of gonadal hormones and contraceptive steroids. London-- New York: Plenum Press, 1969 pp352-65. 3 Grant ECG. The pill, hormone replacement therapy, vascular and mood over-reactivity, and mineral imbalance. J Nutr Environ Med 1998;8:105-116. 4 Anthony H, Birtwhistle S, Eaton K, Maberly J. Environmental Medicine in Clinical Practice. BSAENM Publications 1997, pp173-4. 5 Grant ECG. The influence of hormones on headache and mood in women. Hemicrania 1975;6:2-10. 6 Horrobin DF. Gamma linolenic acid: an intermediate in essential fatty acid metabolism with potential as an ethical pharmaceutical and as a food. Rev Contemp Pharmacother 1990:1:1-45. 6 Barnes B, Grant ECG et al. Nutrition and preconception care. Lancet 1985;2:1297. 7 Passwater RA, Cranton EM. In: Trace elements, hair analysis and nutrition. Keats, New Canaan, Connecticut: 1983:291-303. Competing interests: None declared |
|||
|
|
|||
|
Michael D Croft, salaried GP Gateways Surgery, Shenley, Herts WD7 9LP
Send response to journal:
|
1] This is an ambiguous question. If you mean "how does the evidence obtained from [this] n of 1 trial differ from the evidence of [other]RCTs [trying to answer the same questions on efficacy of vit B6 on N&V in early pregnancy]", the answer is we dont know, because you have not given us the results of the trial. If you mean "how does the evidence obtained from the n of 1 trial [in general] differ from the evidence of RCTs [in general]" then one has to say it has hugely restricted external validity. Internal validity of this one is pretty good, according to the design information we have been given (though a blinded pharmacist would have helped), so once the code has been broken, one would have an evidence base to answer the question, "Did B6 therapy help Mrs Reynolds with her symptoms in the particular weeks under study in her early first pregnancy?" Subsequently one could use this evidence to inform a decision to continue or withhold B6 in this particular patient, provided one felt one could extrapolate its application to later/different weeks in the same pregnancy,or the same weeks in a later pregnancy established with the same partner. The evidence from RCTs are off the shelf and may fit Mrs Reynolds, but this trial is made-to-measure and therefore suits better, particularly as she can choose her own outcome measures. The outcomes of the quoted RCTs are not huge, but may indicate a potential delivery of benefit to patients, though we are not told of the characteristics of the populations studied. 2] You could offer other patients the same n of 1 trial, though if you are going to do this you would be better off doing a properly funded prospective RC crossover trial with other clinicians and getting evidence which is more likely to be generalisable. Which this evidence is not. It is, in effect, anecdotal. 3] Suitable conditions for n of 1 trials would include those where there are chronic symptoms which need to be suppressed, occurring over a period of days, weeks or months, using drugs with suitable lengths of action and with minimal side-effect profiles. These are particularly applicable to alternative medical remedies, and would inform on the use of these in such conditions (in individual patients) as pre-menstrual syndrome and chronic daily headache. Competing interests: None declared |
|||
|
|
|||
|
Ed Peile, Professor of Medical Education Warwick Medical School, University of Warwick, CV4 7AL
Send response to journal:
|
I commented last week on the breadth of learning from cases. I am struck this week by the depth aspect. Only two responses so far to the issues raised about N of 1 trials, but both very thought-provoking. The N of 1 trial is a good example of a phrase that may trip lightly off the tongue as we rationalise our decision to treat our patient in a rather idiosyncratic way. The authors of the case report were clearly working to completely different standards in trying to obtain objective evidence of benefit from their experimental treatment of the patient. The responses to their article help us all to think more deeply about the circumstances in which we can evaluate a single patient response to treatment. This little corner of practice-based evidence is relevant to all who treat patients, and it is worth the effort to think about it. Competing interests: I act as an editorial adviser to BMJ and I have received honaria for work on interactive case learning |
|||
|
|
|||
|
Roger Gadsby, GP Nuneaton, Warwickshire
Send response to journal:
|
GP Commentators Response I think the relative lack of response to this second part of the case history reflects the fact that most GP's have little experience of n of 1 trials in their clinical practice. Also I think many GP's in the UK feel frightened of suggesting any sort of therapy, be it tablets, nutritional therapy or vitamins, to a woman in early pregnancy, even in the form of a n of 1 trial, for fear that such a therapy might cause a foetal abnormality, and they might be blamed. The difficult problem is to know if there are effective therapies to treat pregnancy sickness, and if these effective therapies have also been shown not to increase the risk of foetal abnormalities. Competing interests: I am trustee of the charity pregnancysicknesssupport which has received an educational grant from Duchesnay Inc |
|||
|
|
|||
|
Allan J Savory, Community Pharmacist John Savory Pharmacy
Send response to journal:
|
In this article you ask what other use n=1 trials might be. Here is a practical example. In New Zealand the government drug buying agency reimburses Pharmacy for the lowest priced brand of a medicine available (ususlly a generic copy). If a patient prefers a dearer brand they are required to pay the difference. A few years ago one of our patients expressed reservations about the effecacy of the fully subsidised generic tablet of Isosorbide mononitrate (ISMN) he was taking when compared with the original brand and wondered if he should pay the premium. In order to test his contention we established a short n=1 trial supervised by his wife and incorporated the three brands of ISMN then available. All brands were similar in apperance and he, but not his wife, was blinded to the brand taken. He then, like Mrs Reynolds, kept a diary of his physical condition. Although as Michael Croft suggests the external validity of the trial was not very robust it was not intended to be, however the internal validity proved to be adequate. After two months when we reviewed and compared the calendar of drug administration and the calender of physical condition we were able to establish in the patients own mind "that there was no bloody difference" He continued to take the fully subsidised generic brand. I believe that this shows that n=1 trials can be used to establish which alternative treatments best suit a particular patient. Editorial note
Competing interests: None declared |
|||
|
|
|||
|
Martin Phillips, SpR Occupational Medicine Business Healthcare, Mansfield Woodhouse, Notts, NG19 9AQ
Send response to journal:
|
Should each treatment period be six days long, with a second washout day at the end of the period? Without this final washout day the benefits (if any) of treatment on days 4 & 5 could be detected in the first two days of the next treatment period, which randomly could be placebo days. Days 1 & 2 - randomly treatment or placebo
Competing interests: None declared |
|||
|
|
|||
|
Anthony M Barnie-Adshead, Retired GP Retired GP
Send response to journal:
|
Vitamin B6 is a significant subject for a trial of treatment for nausea and vomiting of pregnancy (NVP). The dose of vitamin B6 10mg three times daily is appropriate. This dosage emphasises a warning of using medicines bought over the counter. I noticed two years ago that I could buy in one shop, 300mg tablets of vitamin B6, and one year ago I bought in the same shop 90, 100mg vitamin B6 tablets with the direction for use "as a food supplement for adults take one tablet daily with a meal, as directed by a health care professional". It would apparently be easy to take a higher dose of vitamin B6 than is recommended in this trial. The dosage for medication obtained over the counter should be more specific for use as safe effective treatment for NVP. Although I have no experience or knowledge of an n of 1 trial, it would appear that to have a reliable result from the symptoms used in the validated patient generated outcome measure - measure yourself medical outcome profile (MYMOP)each symptom should be clearly defined. Mrs Reynolds has chosen dizziness as a symptom, which is not necessarily related to NVP and is therefore not relevant. It is surprising that she does not include nausea which many women describe as the more distressing and certainly the more persistent symptom than vomiting (1). Vomiting can be easily quantified into the number of vomits per day, nausea can also be quantified if Mrs Reynolds keeps a structured daily diary of her symptoms, the hours of nausea per day can be calculated. These two symptoms are then appropriate for her MYMOP. Daily activity and general wellbeing are very relevant measures but will be too vague to give clear cut information. The number of activities Mrs Reynolds failed to acomplish each day, eg shopping, cooking, housework, childcare, time lost from outside employment, should be mentioned in daily activities. Specific items to define general wellbeing could include hours spent resting during the day, number of meals eaten, quantity of fluids consumed and mood, particularly feelings of depression due to the NVP. The ordinal scale of 0-6 from each of the 4 measures, nausea, vomiting, daily activities and wellbeing, as shown clearly by the figure in your article with relevant well defined criteria, can be used for a more accurate calculation of the total daily NVP in the n of 1 trial. (1) Smith C, Crowther C, Beilby J, Dandeaux J 'Impact of Nausea and Vomiting of Women: a burdon of early pregnancy: Aust NZ J Obstet Gynaecol 2000; 40; 4; 397-401. Competing interests: I am a trustee of the pregnancy sickness support charity, which has received an unrestricted education grant from Duchesnay inc, Canada. |
|||
|
|
|||
|
Charlotte Paterson, Research Fellow, MRC Health Services Research Collaboration Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Rd, Bristol BS8 2PR
Send response to journal:
|
Measure Yourself Outcome Profile, MYMOP, is a patient generated, or individualised, problem specific outcome measure. Readers who wish to find out more about it cannot do so from the incorrect reference given on 7th February, but can do so from the references below (1,2), and from the website http://www.hsrc.ac.uk/mymop. MYMOP is suitable for adaption to a diary format, and the case of Ms Reynolds is a good illustration of this application. Our own experience at the MRC Health Services Research Collaboration is that MYMOP, adapted to a diary format, was feasible, acceptable and useful in an n-of-1 pilot study of different types of knee supports for osteoarthritis of the knee. Yours sincerely
1. Paterson C. Measuring outcome in primary care: a patient-generated measure, MYMOP, compared to the SF-36 health survey. British Medical Journal 1996;312:1016-20. 2. Paterson C,.Britten N. In pursuit of patient-centred outcomes: a qualitative evaluation of MYMOP, Measure Yourself Medical Outcome Profile. J Health Serv Res Policy 2000;5:27-36. Competing interests: None declared |
|||