Magnesium Research (1995) 8, 3, 191-206
'In order that a book exudes truth,
we must stuff ourselves with its
subject right over the ears'.
'Pour qu'un livre sue la vérité,
il faut être bourré de son sujet
jusque par-dessus les oreilles'.
(August 6 1857).
Summary: The quality criteria of biomedical journals--and of 'Magnesium Research' as such--are being given a new insight. This has practical implications for the contributors and the editors.
General considerations on the patterns of evaluation of scientific papers highlight five types of errors, that may be incurred in submitted manuscripts.
1. The information conveyed may remain ambiguous: for example, lack of discrimination between acute and chronic patterns, or confusion between the clinical and toxicological consequences of pharmacological and physiological studies: for example it is a real scientific fraud to identify the absent toxic effects of physiological magnesium supplementation with those of high pharmacological magnesium doses...which in fact may induce toxicity. There should be no confusion between in vitro and in vivo data, with a good understanding of the systemic neuroendocrine metabolic and renal regulations and of the multiple local targets concerned. It is always important to discriminate between the two types of deficit: deficiency due to insufficient intake which merely requires oral physiological supplements and depletion related to a dysregulation which requires more or less specific correction of its causal dysregulation.
2. Insufficient information retrieval, frequently with consultation of one database only. Because of indexing omissions and word usage idiosyncrasies, no literature search can retrieve all papers. Monographs and books of proceedings are rarely mentioned in databases and therefore escape consultation. It is obvious, besides, that some of the quoted references have sometimes not been read, but only the title (and in the best cases also the abstract), occasionally with the remaining misprints. A good specific and general knowledge of the background of the study is necessary.
3. Basic methodological errors. Coexistence does not mean causality. Analogous patterns do not demonstrate an identical aetiopathogenesis. The complexity of biology must not be disregarded just because the present trend focuses on one aspect of knowledge at the expense of many others.
4. Thought processes must be unbiased. Citation of supportive papers to the prejudice of unsupportive constitutes a real ethical fraud. It seems also very important to submit for publication papers with negative results as well as ones with positive results. In studies on pharmacological indications for magnesium, choice of the magnesium salt used ought to be justified and the efficiency must be evaluated vs reference treatment.
5. Observance of the formal regulations is frequently neglected, in the presentation of manuscripts particularly.
Some guidelines should therefore be added to the directions to contributors. Before establishing valuable protocols and in order to write up well structured introductions, discussions and conclusions, the authors should have a comprehensive view of their subjects, that is to say an overall knowledge of the previous general and specific publications related to the topic which must be read. Title, conclusions and abstract ought to be informative and unequivocal. Both supportive and unsupportive data should be taken into account in the discussion. References should be duly consulted or mentioned as 'cited in'. There should be strict observance of the presentation of the manuscripts according to the directions to contributors. Studies resulting in negative results should not be disregarded. Reciprocally, the editorial policy requires that editors and referees ought to be strict as regards the quality criteria. Although editors and peer reviewers are in no position to detect basic fraud they can, however, highlight errors, whether due to simple oversight or to more subtle ethical or scientific pseudo-frauds. Both conclusive and inconcern papers may deserve publication: positive and negative results equally concern public health.
To conclude, the editorial board must be ready to reject dubious manuscripts but must at the same time keep their minds open to consider in a positive light innovative papers. But all opinions may be discussed and the letter to the editor allows permanent debating.
Key words: Biomedical journals, editorial policy, evaluation, magnesium, peer-review, quality criteria
The primary purpose of biomedical literature is the dissemination of knowledge from basic scientists to physicians and reciprocally by, insuring efficient communication from scientists to scientists, from scientists to medical practitioners, among practitioners, and from practitioners to scientists. This is why every biomedical journal should clearly define its publication policy and state in its instructions for authors the criteria for scientific rigour and clinical pertinence that will be applied by the Editorial Board in selecting articles for publication1.
Magnesium Research, the official organ of the international Society for the Development of Research on Magnesium, welcomes reports pertaining to all disciplines and concerns both scientists and physicians. After several years of publication, its editorial policy now demands that its quality criteria should be given a new insight. This has practical implications for the contributors and for the Editorial Board.
The aims of this article are as follows. (i) To enhance the quality criteria: use of unequivocal and heuristic concepts, good documentation, strict methodology, complete discussion, and compliance with formal regulations, by quoting a certain number of errors which should be avoided: ambiguous or insufficient information, methodological mistakes, thought process biases and overlooking of formal rules (ii) To define practical consequences for both contributors and Editorial Board by establishing complementary guidelines for the use of the authors and by issuing editorial policy criteria that aim to ensure scientific rigour.
It is necessary to successively highlight the importance of five types of quality criteria: clear concepts, good documentation, strict methodology, comprehensive discussion, and correct presentation, to avoid ambiguity, insufficient knowledge, methodological errors, sins of omission and formal negligence.
Five important notions need to be emphasized: distinction between acute and chronic patterns, discrimination between physiological and pharmacological data, differentiation between in vitro and in vivo studies, specification of two types of deficit: deficiency and depletion, and discrepancy between an epidemiological interpretation and a physiological or pathophysiological finding.
It is, for example, necessary to distinguish between physiological data observed in overt acute magnesium deficiency and the physiological consequences of chronic magnesium deficiency.
The physiological properties of magnesium are now well known. They cover development, neuromuscular, cardiovascular, renal, bone and endocrine apparatus, reproduction, coagulation, humoral balance, and immune and other defence processes. They exhibit a variety of antistress functions: thermoregulation and anti-allergic, anti-toxic, anti-hypoxic, anti-inflammatory systems, through anti-oxidant properties particularly. These physiological properties are relevant to the multiple effects observed in vivo during acute magnesium deficiency and to their specific reversibility through oral magnesium supplementation with physiological doses2.
The experimental and clinical forms of chronic marginal magnesium deficiency are becoming better identified, but they differ from overt signs of acute deficiency3. For example, acute experimental magnesium deficiency in the rat may induce decreases in skeletal muscle and myocardial magnesium and in serum alkaline phosphatase. This causes non-significant modifications of total cholesterol, increase of insulin secretion, arterial hypotension with hyperreceptivity of adrenergic agonists and infertility.
Conversely, this is not observed in chronic marginal experimental magnesium deficiency: magnesium concentration is not significantly modified in skeletal muscle and myocardium4 nor is serum alkaline phosphatase5. But total cholesterol increases significantly6 and insulin secretion decreases7. As a rule blood pressure is not modified8,9 unless late nervous, renal or arterial secondary alterations, uncontrolled by magnesium supplementation, cause an increase10. Adrenergic receptivity is decreased, as in ageing11,12. Pregnancy is possible, but the fertility is reduced and there is a higher mortality rate in the offspring3,13,14.
It should not be permitted today to extrapolate from physiological data observed in overt acute magnesium deficiency to physiological consequences of chronic magnesium deficiency, which constitute the main indication for epidemiological trials 15. For example, this background invalidates the use of skeletal magnesium concentration as a relevant marker of chronic marginal deficiency4. It explains why measurement of total cholesterol should be included in the follow-up of the effects of magnesium supplementation in dyslipidaemias6,16 and allows us to consider the part played by chronic magnesium deficiency in the physiopathology of ageing12,17.
Although acute and chronic magnesium deficiencies are specifically reversible through oral magnesium supplementation with physiological doses, the experimental and clinical symptoms may differ. The typical pattern of chronic magnesium deficiency is latent, whereas overt signs are observed in acute magnesium deficiency.
Pharmacological effects of magnesium are observed irrespective of the magnesium status. They are established either in vitro, in situ, or in vivo when a parenteral or an oral intake is high enough to exceed any homeostatic capacity which may prevent magnesium overload18. These basic differences between pharmacological and physiological magnesium actions lead to discriminate between the two types of magnesium load tests. Clinical efficiency of parenteral magnesium administration should not be used as a diagnostic tool attesting to magnesium deficiency. Conversely physiological oral doses of magnesium are totally devoid of the pharmacodynamic effects of parenteral magnesium and without clinical effects when magnesium status is normal. Correction of symptoms by this oral magnesium load constitutes the best proof that they were due to magnesium deficiency15, 18-21.
But the main consequence of the differentiation between the pharmacological properties of magnesium is to allow us to distinguish between two different types of therapy with magnesium: atoxic physiological oral therapy and pharmacological magnesium therapy which may induce toxicity18,19,21. Today the main form of magnesium therapy is oral magnesium physiological supplementation of magnesium deficiency. The palliative oral doses meant to balance magnesium deficiency are obviously devoid of any toxicity since their purpose is to restore to normal the insufficiency of magnesium intake 18,19,21. In order to use the pharmacological properties of magnesium, no matter what the magnesium status is, it is necessary to go beyond the mechanisms of magnesium homeostasis to induce a therapeutic magnesium overload. Large doses of magnesium given orally are advisable where there are chronic indications and the parenteral route is suitable for acute applications. Both types of pharmacological magnesium treatment are capable of inducing toxicity18,19,21. It is a real scientific fraud and an ethical misconduct to fail to differentiate between the absent toxic consequences of a physiological magnesium supplementation and the effects with those of high pharmacological magnesium doses which are potentially dangerous.
It is always dangerous to extrapolate from pharmacological to physiological data. For example, a pharmacological magnesium load may induce death through peripheral and not central mechanisms18,19. But sudden death in magnesium-deficient rats results from brain dysfunction and not from simple apnoea due to tonic contraction of the respiratory muscles22. This physiological study presents data establishing that death due to magnesium deficiency is caused by central and not peripheral mechanisms that is the contrary of the pharmacological mechanism. This shows that the main nervous target for magnesium is not identical in the fields of pharmacology and physiology23.
In vitro and in vivo studies
It is erroneous to extrapolate from in vitro data or from in situ or other pharmacological manipulations to in vivo physiological significance.
For example although vascular tone in vitro, is increased by lowering magnesium, in vivo, acute magnesium deficiency may induce vasodilatation and hypotension 10, 12. A quantitative coronary angiographic study fails to provide any correlation between coronary vascular tone and serum magnesium concentrations in patients with ischaemic heart disease24.
It is very important to be aware of the existence in vivo of multiple systematic (neuroendocrine/metabolic) and local (direct or mediated, cellular and subcellular) regulations of magnesium status21,25-28. Thus in the course of magnesium deficiency due to an insufficient magnesium intake, some reactions do not mirror responses observed with magnesium excess. For example while calcitonin and nitric oxide are released in the case of magnesium excess18, 25,28-30, no decrease is observed in magnesium deficiency but exactly the opposite28: increased release of calcitonin25 and calcitonin gene related peptide31 and of nitric oxide32.
It is very important to stress the basic differences between the two types of magnesium deficit: magnesium deficiency and magnesium depletion18,19,21. In the case of magnesium deficiency, the disorder corresponds to an insufficient magnesium intake: it merely requires oral physiological magnesium supplementation. In the case of magnesium depletion the disorder which induces magnesium deficit is related to a dysregulation of the control mechanisms of magnesium metabolism: either failure of the mechanisms which ensure magnesium homeostasis or intervention of endogenous or iatrogenic perturbing factors of magnesium status. Magnesium depletion requires more or less specific correction of its causal dysregulation.
This differential diagnosis between these two types of magnesium deficit is of major importance in case of primary magnesium deficit. It is also of interest in the case of secondary magnesium deficit. Although they both initially always require their own specific aetiological treatment, this major therapeutic measure cannot always be performed. It may not be possible to get an alcoholic to quit drinking or to manage diabetes mellitus perfectly. In such cases, a careful analysis of the mechanisms inducing the magnesium deficit can lead to effective therapy. In chronic alcoholism, magnesium deficiency secondary to an insufficient magnesium intake is frequent 19,33, whilst in diabetes mellitus the part of the deficiency appears usually less important than the dysregulations inducing magnesium depletion19,34. In chronic alcoholism oral magnesium supplementation frequently constitutes an interesting adjuvant treatment particularly for the prevention of alcoholic encephalopathy18,19,26,28,33,35,36, while it is much more rarely useful in diabetes mellitus18,37,38. However, in magnesium depletion it is always wise to maintain a sufficient magnesium intake to provide substrate for correcting the dysregulation substrate. It may thus constitute a necessary adjunct to the treatment18,19. As a rule, in front of each type of magnesium deficit, whether primary or secondary, it is imperative to determine the respective roles of magnesium deficiency and of magnesium depletion. For example, primary deficit in sports medicine seems to be more often related to depletion than to deficiency39 while in ageing the two mechanisms are usually associated12.
Each of these two types of deficit corresponds to a very different experimental model. The usual experimental method where magnesium deficit is induced by a low level of magnesium in the diet typically constitutes a model of primary magnesium deficiency, with specific reversibility of the symptoms by simple physiological oral magnesium supplementation18, 19, 21.
Conversely experimental models of magnesium depletion are not controlled by physiological oral magnesium intake18, 21, 28. Various types of more or less severe magnesium depletion can be used: genetic models (in rats and mice), and acquired models: either secondary to an irreversible (or partially reversible) cause (such as traumatic brain injury) or reversible. In the latter case, the models associate a low intake with diverse types of stress17, 18, 21, 28, 40, 41.
Physiological oral magnesium supplementation of maternal magnesium deficiency appears to be of major importance not only for the comfort of the mother and the development of the baby at birth but to prevent disturbances of transition from chemical to physical thermoregulation14, 42 and, later neurological cardiovascular and metabolic troubles19, 26, 28, 43. Undetected and untreated marginal maternal deficiency may cause problems during early development. Such deficiency could presage such serious later problems as sudden infant death syndrome14,42, neurolability, some forms of infantile convulsions or psychiatric disturbances19,26,28, and even adult cardiovascular disease, syndrome X and non-insulin-dependent diabetes mellitus28, 43. The protocol of a multi-centre study of maternal magnesium supplementation should be followed not only on the mother, the fetus and the neonate at birth, but also on the child throughout life from infancy to older age14, 28, 43.
As with the two types of deficit, deficiency and depletion, a similar discrimination should be applied to magnesium overload. It is important to distinguish between magnesium overload due to an excessive supply--usually iatrogenic--which may be simply controlled by reducing the magnesium supply and magnesium overload due to dysregulation of magnesium status, requiring more or less specific correction of its causal regulation: for example haemodialysis for palliation of renal failure.
It seems necessary to underscore the discrepancy between a hypothetical epidemiological interpretation and a proven physiological or pathophysiological proven fact9, 44. For example studies which credit the myth that magnesium is a major antihypertensive nutrient45-50 have confused epidemiological and pathophysiological data9,44. Epidemiological studies are necessarily interpreted on the grounds of well established pathophysiological bases. Therefore, as magnesium is erroneously considered to be a major antihypertensive nutrient, it is presented in the Honolulu heart study51 as ranking first among other multiple dietary factors showing inverse associations with blood pressure. The same epidemiological data could be differently interpreted in the light of our background knowledge of the relationship between magnesium and blood pressure9,10,16,44,52-64 and by using the interesting concept of multicolinearity associated with the data concerning the relation between blood pressure and Ca, P, K and particularly Na sensitivity51,65. Though magnesium does not appear to be a major antihypertensive nutrient, magnesium deficit, through its numerous harmful actions, both directly on the nephrocardiovascular apparatus and indirectly through the neuroendocrine apparatus, may sometimes behave as a cofactor of a hypertensive agent in particular cases9,44.
Water hardness and the magnesium content of water have frequently been confused as factors of hypertension46-49. Although a positive correlation between the hardness of water and cardiovascular mortality may exist9,46-49,66,67 or may not9,66,67, it may be due to various mechanisms. If magnesium appears as the prominent protective factor in drinking water, other protective factors may also be involved9. The calcium level in drinking water may be important, in advance of the consumer particularly, since it acts as a crucial anticorrosive substrate of the protective coating deposited on the pipes by well balanced water9. But in the water story noxious factors may also intervene. Corrosivity, well defined in the Ryznar index, seems the prominent harmful factor in drinking water9. Epidemiological data confusing hard water and its magnesium content constitute a scientific mistake.
One might think it is redundant to insist on such elementary notions as the difference between acute and chronic patterns, between pharmacology and physiology, in vitro and in vivo studies, deficiency and depletion, epidemiology and physiopathology, but it is surprising to see that such basic concepts are so frequently overlooked, not only in submitted manuscripts but even in articles published in prestigious peer-reviewed journals.
When writing his historical novel Salammbo, Gustave Flaubert emphasized the importance of ample documentation: "In order that a book exudes truth, we must stuff ourselves with its subject right over the ears (Pour qu'un livre sue la vérité, il faut être bourré de son sujet jusque par-dessus les oreilles)68. Advisable with a novel, this requirement is even more imperative with scientific papers.
When dealing with a given subject too many authors fail to acquire an extensive knowledge of the published data in the field, both specific and general. Such failure of cognition appears to be this inevitable consequence of the immense size and rapid growth of scientific literature.
The first cause may be insufficient information retrieval, frequently due to consultation of one database only. Obviously, no computerized search can retrieve all papers, because of indexing omissions and word usage idiosyncrasies69. Monographs and books of proceedings are rarely mentioned in databases and therefore escape consultation. For example, a recent study on the links between fatigue and magnesium status was empirically started from the good results obtained by some general practitioners70. If the authors had conducted a careful literature search and taken into account well established data which they overlooked they would have developed better methodology. Simple consultation of the subject index (asthenia, fatigue) of one monograph19 and one book of proceedings71 on magnesium would have shown the place of this symptom in the established body of clinical, paraclinical, pathophysiological and therapeutic knowledge on chronic primary magnesium deficiency72. A thorough study of the symptomatology of magnesium deficiency would have shown that although the psychiatric forms of magnesium deficiency may fit into a neurotic pattern, they never result in dementia19,28. But the documentation being faulty, a study on magnesium deficiency in chronic schizophrenia, relying on an obsolete paper73 showing that psychotic behaviour had been reported in 50 per cent of subjects with hypomagnesaemia73, erroneously concluded that 'attempts should be made to correct the underlying magnesium deficiency which may potentially (?) attenuate psychotic symptomatology'74!
Insufficient consultation: It is frequently obvious that some of the quoted references have not been read through but only the title, and perhaps the abstract75.
It is well known that online access to databases for titles only is cheaper than for titles and abstracts! It is therefore imperative that both titles and abstracts should be informative and unequivocal.
Structured abstracts may substantially increase the amount and quality of information immediately available from simple consultation1,76. Although interesting, these structured abstracts should not be too much encouraged in so far as they might discourage the reader from perusing the whole articles!
Insufficient appraisal: The authors must appraise each reference critically and efficiently. The fact that a previous paper has been published in a peer-reviewed journal is not sufficient to ascertain its quality: the standards for selection may have been casual. Lax standards place readers at the mercy of the biases and speculations of the authors of each reference, who may undercut the available evidence1. The citation of each reference involves the responsibility of the person who quotes. Before being placed in a bibliography a paper must be appraised efficiently and critically, which requires real expertise from the people who are going to use the citation in a manuscript.
To sum up, the quality of the documentation ought to rely on good information retrieval and evaluation.
Basic methodological errors are frequently observed.
Among the main ones we find:
The methodological mistake of using the coexistence of two phenomena as a demonstration for a link between them is too frequently observed. This type of error can be illustrated by the following examples. Since in a number of diseases such as alcoholism and diabetes mellitus magnesium deficit and hypertension may coexist 'the hypomagnesaemic state is presented as probably more than coincidental in the aetiology of hypertensive state'46-50!; or again, as there is usually an inverse correlation between the level or arterial blood pressure and the hardness of drinking water, it is suggested that an increased intake of magnesium may have a beneficial effect on blood pressure46-50!. A remarkable epidemiological study in 2633 adults showed that a high dietary magnesium intake is associated with better lung function and a reduced risk of airway hyperreactivity and wheezing. It might therefore be inferred that a low magnesium intake, i.e. magnesium deficiency, could be involved in the aetiology of asthma and chronic obstructive disease77. But this remains to be proven. Coexistence does not mean causality. Only the specific reversibility of the symptoms through the control of the magnesium deficit will really demonstrate the existence of an aetiopathogenic link. A study showing that plasma, erythrocyte and mononuclear leucocyte concentrations of magnesium do not significantly differ from those of healthy controls disagrees with the hypothesis of the existence of magnesium deficiency in asthmatic patients78.
The clinical and pathological anatomical characteristics of superacute magnesium deficiency shock and lesions observed in the sudden infant death syndrome seem analogous. For over 20 years Caddell has attempted to accumulate more evidence to establish the validity of her hypothesis79,80 but still finds it difficult because the data do not constitute a pathophysiological demonstration14.
Content analysis may be an interesting strategy used in information retrieval technique to identify important but neglected problems. Such strategies do not rely on having prior expert knowledge of a given field, but instead analyse several analogies: for example, co-occurrence of words in titles, and co-citation of patterns of papers within that field. Of particular interest are 'complementary-but-disjoint' pairs of research topics that show certain related patterns of word usage, yet lack any cross-citation and have not been related directly to each other by scientists69.
This technique has provided guidance in studies speculating on links between magnesium and migraine81 and between magnesium and neurologic disease69.
The complexity of biology must not be disregarded just because the present trend focuses on one aspect of knowledge at the expense of many others
For example many recent studies have emphasized the role of NMDA receptor-mediated excitotoxicity in the pathogenesis of a variety of acute and chronic neurological diseases and upon the role of endogenous magnesium ions in regulating this process: about 900 records in Medline deal with magnesium and NMDA receptors, usually in vitro28,69.
Nervous hyperexcitability due to magnesium deficiency depends on modification of the turnover of the monoamines serotonin, acetyl choline, catecholamines (dopamine and noradrenaline mainly) and excitatory amino acids (aspartic and glutamic acids mainly), with a decreased turnover of inhibitory amino acids (γ-aminobutyric acid and taurine mainly)26,28. As a great number of in vitro studies on magnesium and NMDA receptors have suggested that the latter had predominant and almost exclusive importance, their in vivo role has often been overestimated82. Although hyper NMDA receptivity enters into the mechanisms of nervous hyperexcitability due to magnesium deficiency as confirmed in vivo83, hyperreceptivity of other non-NMDA receptors of excitatory amino acids, particularly may also intervene84. It is not because the current trend is towards NMDA receptors that the interest of many other mechanisms should be underestimated26,28.
Magnesium is usually granted an important role in cell membrane functions. But as nowadays a great many papers on the Na/K-ATPase pump are being published the tendency is to give this priority in explaining the consequences of magnesium deficiency on K and Na metabolism and to overlook the other mechanisms. Here again we have an example of over-simplification. On the local membranous target, magnesium may act on sodium and potassium through various mechanisms: paracellular and cellular. In the cell several magnesium-dependent mechanisms may intervene: not only the Na-K+-ATPase pump, but also the K and Na channels, Na-K-2Cl co-transport, Na-H antiport, H-K pump, Na-Ca antiport and Na-Mg exchange85. These complex local pathways do not rule out the important role of systemic neuro-endocrine-metabolic regulation of the ionic effects of magnesium disturbances7,19,21.
To sum up, methodological rigour is necessary and demands that some elementary rules would be respected.
Thought processes should be unbiased. Sins of omission are too often observed: preferential citations and truncated discussions, unexplained choice of the drug used and inadequate protocols.
Citation of supportive papers while unsupportive papers are forgotten constitutes a real ethical fraud. Whereas there is a considerably higher frequency of citation of supportive papers, unsupportive papers are partially and even totally ignored or excluded. In the latter case it is a real conspiracy of silence, a genuine scientific 'omerta'86. This though process bias is highly detrimental for public health: therapeutic decision, meta analyses and systematic reviews become misleading. For example in order to determine whether lowering cholesterol values significantly prevented coronary heart disease and death, a comparison was made of frequency of citation with outcome of 22 controlled cholesterol-lowering trials using coronary heart disease or death, or both, as end points. No evidence was found that lowering serum cholesterol concentrations reduced mortality or prevented coronary disease. Claims of the opposite seem to be based on the preferential citation of supportive trials87. Some trials advocating a beneficial effect of parenteral magnesium in acute myocardial infection could be due to selective identification of positive studies88, but early use of intravenous magnesium for this indication is still controversial18,88-91.
Studies on the pharmacological indications of magnesium preparations have shown that the choice of one among the various sorts available has definite implications. It still happens too often that there are no pharmacological or clinical data to back up the choice. Among numerous magnesium salts, curiously the most widely used preparations are those containing magnesium sulphate. However among the soluble salts MgSO4 has the least advantageous pharmacological properties. Its absorption, cellular penetration, membrane effects and antihypoxic properties are poor. Comparative studies between SO4Mg and CL2 have indicated that better absorption and retention are observed with CL2 than with SO4Mg. Moreover sulphate infusion increases the urinary excretion of magnesium, calcium and potassium19,92. Study of the effects of CL2 and of SO4Mg on the ionic transfer components through the isolated human amniotic membrane, has shown different actions. MgCl2 interacts with all the exchangers while the effects of MgSO4 are limited to paracellular components without interaction with cellular components (excepted the antiport Na/H). Besides, MgCl2 increases the ionic flux ratio of this asymmetric membrane while MgSO4 decreases it, until it reaches a value close to 192.
The sulphate anion has unfavourable effects in comparison with other ions used in magnesium therapy93,94.
When developing an experimental or clinical protocol it is very important to compare the results not only vs placebo, but also vs the reference drug if one is used. For example, phenytoin induces quicker cervical dilatation, reduced fall in haematocrit after delivery, lower incidence of hot flushes, dyspnoea and weakness in the mother than magnesium infusions, while less side effects on fetal heart tracing are observed 18.
To sum up it is necessary to avoid sins of omission such as preferential citations, unexplained choice of the drug used and failure to mention of the beneficial effects of reference drugs.
Observance of the formal regulations is frequently neglected, in the presentation of the manuscripts particularly. Simple reading of the directions to contributors which define the philosophy of a journal would suffice to facilitate the work of the Editorial Board.
Emphasis laid on these different types of quality criteria for biomedical manuscripts may appear elementary and even redundant. But in fact it is necessary as such prerequisites sometimes seem to be forgotten even in well known peer-reviewed journals.
Awareness of the quality criteria for manuscripts to be published in biomedical journals leads to two types of conclusion: (i) When such shortcomings as the aforementioned are observed, complementary guidelines would be issued and given to the contributors; (ii) Reciprocally the Editorial Board should see that there is no laxity in adherence with the revised bulk of requirements, should be strict on possible scientific misconducts and should be ready to consider any well conducted study, even if inconclusive.
These will relate to general and specific considerations.
Their aim is to ensure that contributors submit manuscripts in strict accordance with the basic ethical and scientific principles and particularly with the five notions previously attached to quality criteria but too often neglected: unequivocal and heuristic concepts, extensive knowledge of the subject, strict methodology, unbiased thought processes, and observance of the formal regulations.
Unequivocal and heuristic concepts: This requires, for instance, explicit discrimination between acute and chronic patterns, pharmacological and physiological properties, in vitro and in vivo data, deficiency and depletion, epidemiology and physiopathology.
Although apparently obvious, such distinctions are not always found in all articles or even reviews49,50,94, no matter how prestigious the journal.
Extensive knowledge of the subject: This rests first on an overall search for information. A literature search from only one database is most often insufficient69. Online access to several databases is often necessary. But much interesting information can only be retrieved in monographs and books of proceedings19, 71, 95-99. Next to a literature search comes a critical appraisal which is obviously impossible if only the title (and possibly the abstract) have been consulted! Assuming that a reference is not available, it should be mentioned in the bibliography 'cited in' (see ref. 68). Each citation of an article involves the responsibility of the author who refers to it. Insufficient evaluation is a factor for perpetuating errors. For example although plasma magnesium is usually normal in hypertension9,54 many papers (e.g. 49) in support of the notion that patients with hypertension show hypomagnesaemia continue to cite an obsolete paper100 with too small a number of cases, where the determination of magnesium was carried out with an old colorimetric method and where 'the hypertensive subjects were drawn from a low economic bracket compared to controls', a socioeconomic reason which sufficed to create magnesium deficiency...irrespective of the blood pressure disturbances! This constitutes a case where critical appraisal is clearly lacking.
Strict methodology: Among the main methodological errors we have highlighted: (i) lack of discrimination between co-existence and causality, (ii) misleading causal interpretations of analogous patterns, (iii) and oversimplification of biology.
The importance of the oral physiological magnesium load test may illustrate the significance of these first two principles. The only proof of a causal link between an insufficient intake of magnesium and an experimental or clinical pattern analogous with that of magnesium deficiency is its specific reversibility by simple physiological oral magnesium supplementation19-21.
To exemplify the third principle it may be added that failure to observe from strict methodology may sometimes be due to the trite necessity of getting indispensable grants and therefore fitting in expected patterns...a financial 'avatar' which may open a path for oversimplification.
Unbiased thought processes: Three main sins of omission should be banned:1 deliberate scientific 'omerta' expressed in the form of preferential citations and discussions 86-88, (ii) unexplained choice of the drug used92-94, and (iii) development of inappropriate protocols.
Compliance with the formal regulations: Courtesy and efficiency should be sufficient to ensure strict observance of the formal regulations of each journal. Each contributor should be faithful to the philosophy of the journal and abide by specific rules of material presentation of the manuscripts. This elementary adherence to the directions to contributors is however most frequently neglected!
Three types of specific considerations ought to be added. They concern:
1. Title, conclusion and abstract.
2. Introduction, methods, discussion and bibliography.
3. Submission of interesting well-conducted studies irrespective of their results.
Title, conclusion and abstract: These three basic elements of an article should not be ambiguous. For example, a double blind, placebo, cross-over study entitled 'Efficacy of potassium and magnesium in essential hypertension' showed that potassium (60 mM/day) lowered arterial blood pressure in patients with mild hypertension. Giving magnesium (20 mM/day) either alone or in combination with potassium did not have any effect63. The title was equivocal and insufficiently informative. 'Efficacy of a high dose of potassium and inefficacy of a physiological magnesium supplementation in essential hypertension' would have been more appropriate.
Another example of a very interesting article whose title is ambiguous is: 'Effects of magnesium sulphate and nitric oxide in pulmonary hypertension induced by hypoxia'101. This title would have been clearer if two words had been added: 'Effects of magnesium sulphate infusion and nitric oxide inhalation in...'.
An identical comment may concern the title of the important study 'Can magnesium sulphate reduce the risk of cerebral palsy in very low birthweight infants'102. The addition of two words would have brought more clarity: 'Can maternal magnesium sulphate infusion reduce...'. If articles from symposia sponsored by drug companies are not peer-reviewed, the number of misleading titles is significantly increased. Journal editors should maintain editorial control over contributions from these symposia103.
Structured abstracts may increase the information immediately available1,76 but the shorter ones including objective, design (without detailed methods and results, or references) and conclusions, appear sufficiently informative and yet are not a disincentive to perusing the articles.
Methodology, introduction, discussion and bibliography: Before establishing valuable protocols and in order to write up well structured introductions, discussions and conclusions, the authors should have a comprehensive view of their subjects19, 69, 71, 95-100, followed by careful critical appraisal18,86-91. But it is not always the case. A randomised double blind placebo-controlled trial has shown that replacing common salt by low sodium, high potassium, high magnesium natural mineral salt could offer a valuable non-pharmacological approach to lowering blood pressure in older people with mild to moderate hypertension104. These findings are very interesting. But when the discussion concerns a hypothetical contribution of the small increase of the magnesium intake to the reduction in blood pressure, only papers supporting a so-called role of magnesium in the control of hypertension are cited, irrespective of the unsupportive papers 8-10, 16, 44, 52-64.
A similar criticism concerns a prospective study assessing the frequency of low serum or erythrocyte magnesium in a large group of patients with idiopathic mitral valve prolapse and the effects of oral physiological magnesium supplementation105. Methodology and results of this research would have been better if established after covering the bulk of the documentation. But evidently several quoted references were not duly consulted which induced lacunae and errors. Overlooking part of the literature previously published brings to disregard major items--for example, as here, clinical and electromyographic criteria of tetany--which would have given to the study a much wider scope of interest through a better utilisation of the observed data integrated into an appropriate protocol106.
Submission of well conducted studies, irrespective of the result: Investigators should be encouraged to recognize the seriousness of publication bias whereby research with statistically significant results has priority, and to submit well conducted studies directed at important questions no matter what the outcome107. Investigators are more responsible for the bias than editors. The presence of a null result is the most frequent reason for failing to write up the study107. There is an association between results reported to be significant and publication of the study originating primarily with the authors108.
Similar general and specific rules for the editorial board must reciprocally correspond to these new general and specific considerations of the instructions.
The editorial policy requires that editors and referees ought to be strict as regards compliance of the contributors with all the quality criteria.
Two specific aspects of the rigour of editorial board should be highlighted:
(i). Detection of scientific misconduct: Editors and peer reviewers are in no position to detect intentional fraud, that is to say research in which deliberately falsified or unsubstantiated data were used. It has been shown that fraudulous research, even after retraction, goes on disseminating errors109-112. But editors and peer reviewers are able to detect several types of ethical and scientific misconduct from authorship practices113 to subtle pseudo-frauds which have already been commented upon. This constitutes one of the main responsibilities of the editorial board.
(ii). Publication of interesting well conducted studies irrespective of the results: Journal editors must convey to the authors that their acceptation of a good paper is not linked with the nature of the results: well conducted studies should be published no matter what the outcome: positive, null or negative, with or without statistically significant results 101, 107, 108, 114. Magnesium Research sees that this editorial policy is followed23,24,64,115.
The editorial board cannot accept manuscripts which do not adhere to the policy, philosophy and material instructions well defined in directions to contributors and their complementary guidelines.
Editorial decision must be objective and dispassionate without conflicting factors such as personal relationships, academic competition or financial considerations116.
However, all opinions may be discussed, and the Letters to the Editor and the Commentaries on recent advances allow permanent debating.
1 . Haynes, R.B. (1990): Loose connections between peer-reviewed clinical journals and clinical practice. Ann. Intern. Med. 113, 724-728.
2. Durlach, J. (1988): The properties of magnesium: biochemical functions, cellular functions and physiologic properties. In: Magnesium in clinical practice, ed. J. Durlach, pp. 7-15. London: John Libbey.
3. Kubena, K.S. & Durlach, J. (1990): Historical review of marginal intake of magnesium in chronic experimental magnesium deficiency. Magnes. Res. 3, 219-226.
4. Nowitzki-Grimm, S., Grimm, P., Thoni, H. & Classen, H.G. (1991): Diagnosis of magnesium status. Magnes.-Bull. 13, 107-115.
5. Cerklewski, F.L., Wan, D.T.Y. & Leklem, J.E. (1992): Effect of moderate zinc and magnesium deficiency in the rat (abstr. 2531). FASEB J. A 1373.
6. Rayssiguier, Y., Gueux, E. Durlach, V., Durlach, J., Nassir, F. & Mazur, A. (1992): Magnesium and the cardiovascular system: I. New experimental data on magnesium and lipoproteins. In: Molecular biology in atherosclerosis, ed. M.J. Halpern, pp. 507-512. London: John Libbey.
7. Durlach, J. (1988): Regulation of the calcium and potassium problems caused by magnesium deficit. In: Magnesium in clinical practice, pp. 24-28. London: John Libbey.
8. Nishiyama, S., Saito, N. & Konishi, Y. (1989): Effect of severe and moderate magnesium deficiency on blood pressure, cardiac function and regional blood flow in male rats. In: Magnesium in health and disease, eds. Y. Itokawa & J. Durlach, pp. 253-260. London: John Libbey.
9. Durlach, J., Bara, M. & Guiet-Bara, A. (1989): Magnesium in drinking water: Its importance in cardiovascular risk. In: Magnesium in health and disease, eds. Y. Itokawa & J. Durlach, London: John Libbey. 173-182.
10. Rayssiguier, Y., Mbega, J.D., Durlach, V., Gueux, E., Durlach, J., Giry, J., Dalle, M., Mazur, A., Lautant, P. & Berthelot, A. (1992): Magnesium and blood pressure. I. Animal studies. Magnes. Res. 5, 139-146.
11. Rayssiguier, Y., Durlach, J., Guiet-Bara, A. & Bara, M. (1990): Aging in magnesium status. In: Metal ions in biology and medicine, eds. P. Collery, L.A. Poirier, M. Manfait & J. C. Etienne, pp. 62-66. London-Paris: John Libbey Eurotext.
12. Durlach, J., Durlach, V., Bac, P., Rayssiguier, Y., Bara, M. & Guiet-Bara, A. (1993): Magnesium and aging. II. Clinical data: etiological mechanisms and physiopathological consequences of the magnesium deficit in the elderly. Magnes. Res. 6, 379-394.
13. Fehlinger, R. (1990): Therapy with magnesium salts in neurological diseases. A critical appraisal. Magnes.-Bull. 12, 35-42.
14. Durlach, J., Durlach, V., Rayssiguier, Y., Ricquier, D., Goubern, M., Bertin, R., Bara, M., Guiet-Bara, A., Olive, G. & Mettey, R. (1991): Magnesium and thermoregulation. I. Newborn and infant. Is sudden infant death syndrome a Mg-dependent disease of the transition from chemical to physical thermoregulation? Magnes. Res. 4, 137-152.
15. Durlach, J., Durlach, V., Rayssiguier, Y., Bara, M. & Guiet-Bara, A. (1992): Methodological prerequisites and systemical approach to the study of magnesium supplementation in marginal magnesium deficiency. In: Metal ions in biology and medicine, Vol. 2, eds. J. Anastassopoulou, P. Collery, J.C. Etienne & Theophanides, pp. 383-388. London-Paris: John Libbey Eurotext.
16. Durlach, J., Durlach, V., Rayssiguier, Y., Bara, M. & Guiet-Bara, A. (1992): Magnesium and the cardiovascular system. II. Clinical data.. A critical review. In: Molecular biology of atherosclerosis, ed, M.J. Halpern, pp. 513-521. London: John Libbey.
17. Rayssiguier, Y., Durlach, J., Gueux, E., Rock, E. & Mazur, A. (1993): Magnesium and aging. I. Experimental data: importance of oxidative damage. Magnes. Res. 6, 369-378.
18. Durlach, J., Durlach. V., Bac, P., Bara. M. & Guiet-Bara, A, (1994): Magnesium and therapeutics. Magnes. Res. 7, 313-328.
19. Durlach, J. (1988): Magnesium in clinical practice. London, Paris: John Libbey, 360p.
20. Durlach, J. (1991): Magnesium: clinical forms of primary magnesium deficiency. In: Modern lifestyles, lower energy intake and micro-nutrient status, ed. K. Pietrzik, pp. 156-167. London, Berlin, New York: Springer.
21. Durlach, J. (1993): Present and future of magnesium research. J. Jpn. Soc. Magnes. Res. 12, 113-135.
22. Nakamura, M., Abe, S., Goto, Y. & Chismaki, A.(1995): Sudden sound-induced death in magnesium deficient rats after repetitive episodes of seizures resulting from brain dysfunction. Magnes. Res. 8, 47-54.
23. Durlach, J. (1995): Editorial. Magnes. Res. 1-4.
24. Herzog, W.R., Materese, F.J. & Atar, D. (1995): Reactivity of coronary arteries in relation to serum Mg levels in patients with ischemic heart disease: a quantitative angiographic study. Magnes. Res. 8, 57-64.
25. Durlach, J. (1988): Regulation of magnesium metabolism. In: Magnesium in clinical practice, ed. J. Durlach pp. 20-39. London: John Libbey.
26. Durlach, J., Poenaru, S., Rouhani, S., Bara, M. & Guiet-Bara, A. (1987): The control of central neural hyperexcitability in magnesium deficiency. In: Nutrients and brainfunction, ed. W.B. Essman, pp. 48-71. Karger: Basel.
27. Durlach, J., Bara, M. & Guiet-Bara, A. (1990): Magnesium and its relationship to oncology. In: Metal ions in biological systems, Vol. 26, eds. H. Sigel & A. Sigel, pp. 549-578. New York, Basel: Marcel Dekker.
28. Durlach, J. & Bac, P. (1996): Mechanisms of action on nervous system in magnesium deficiency and dementia. In: Mineral and metal neurotoxicology, eds. M. Yasui, K. Ota, M.J. Strong & A. Verity. Boca Raton, FLA (in press): CRC Press.
29. Kemp, P.A., Gardiner, S.M., March. J.F. & Rubin, P. (1994): Effects of N6-nitro-L arginine methyl ester on regional haemodynamic responses to MgSO4 in conscious rats. J. Pharmacol. 111, 325-331.
30. Izzo, A.A., Gaginella, T.S., Mascolo, N. & Capasso, F. (1994): Nitric oxide as a mediator of the laxative action of magnesium sulfate. Br. J. Pharmacol. 113, 228-232.
31. Weglicki, W.B., Tong Mak, I. & Phillips, T.M. (1994): Blockade of cardiac inflammation in Mg2+, deficiency by Substance P receptors inhibition. Circ. Res. 74, 1009-1013.
32. Rayssiguier, Y, Mazur, A., Gueux, E. & Rock, E. (1994): Mg deficiency affects lipid metabolism and atherosclerosis processes by a mechanism involving inflammation and oxidative stress. (abstr.). Magnes. Res. 7(Supp. 1), 46-47.
33. Durlach, J. (1988): Chronic alcoholism. In: Magnesium in clinical practice, ed. J. Durlach, pp. 120-133. London: John Libbey.
34. Durlach, J. (1988): Magnesium deficit in diabetes mellitus. In: Magnesium in clinical practice, ed. J. Durlach, pp. 156-169. London: John Libbey.
35. Gullestad, J., Dolva, L.O, Soyland, E., Manger, A.T., Falch, D., Veirod, M.B. & Kjekshus, J. (1991): Effects of oral magnesium treatment in chronic alcoholics. In: Magnesium a relevant ion, eds. B. Lasserre & J. Durlach, pp. 405-409. London: John Libbey.
36. Gullestad. L., Olva, L.Ø., Manger, A.T.. Falch, D. & Kjekshus, J. (1992): Oral magnesium supplementation improves metabolic variables and muscle strength in alcoholics. Alcoholism: Clin. Exp. Res. 16, 986-990.
37. Joslyn, S., Lynch, C., Wallace, R., Olson, D. & Van Hoesen, C. (1990): Relation between diabetes mellitus mortality rates and drinking water magnesium levels in Iowa. Magnes. Trace Elem. 9, 94-100.
38. Gullestad, L., Jacobsen, T. & Dolva, L.O. (1994): Effects of Mg treatment on glycemic control and metabolic parameters in NIDDM patients. Diabetes Care 17, 460-461.
39. Rayssiguier, Y., Guezennec, C.Y. & Durlach, J. (1990): New experimental and clinical data on the relationship between magnesium and sport. Magnes. Res. 3, 93-102.
40. Bac, P., Herrenknecht, C., Binet, P. & Durlach, J. (1993): Audiogenic seizures magnesium-deficient mice: effects of magnesium pyrrolidone-2-carboxylate, magnesium acetyltaurinate, magnesium chloride and vitamine B-6. Magnes. Res. 6, 11-19.
41. Bac, P., Herrenknecht, C., Binet, P. & Durlach, J. (1993): Effects of various magnesium salts on the action of systemic kainic acid in magnesium deficient rats: a new model of accelerated hippocampal aging-like injury? (abstr.). Magnes. Res. 6, 300-301.
42. Goubern, M., Rayssiguier, Y., Miroux, B., Chapey, M.F., Ricqier, D. & Durlach, J. (1993): Effect of acute magnesium on the masking and unmasking of the proton channel of the uncoupling protein in rat brown fat. Magnes. Res. 6, 135-143.
43. Durlach, J. (1993): Death from infancy to older age and maternal magnesium deficiency. How long should be the follow up of the consequences of undernutrition in pregnancy be continued? Magnes. Res. 6, 297-298.
44. Durlach, J., Durlach, V., Rayssiguier, Y., Bara, M. & Guiet-Bara, A. (1992): Magnesium and blood pressure. II. Clinical studies. Magnes. Res. 5, 147-153.
45. Charbon, G.A. (1986): Effects of magnesium on central hemodynamics and central circulation. Magnes. Bull. 8, 230-236,
46. Karppanen, H. (1986): Epidemiological aspects of magnesium deficiency in cardiovascular diseases. Magnes. Bull. 199-203.
47. Altura, B.M. & Altura, B.T.(1990): Role of magnesium in pathogenesis of hypertension. relationship to its actions on cardiac and vascular smooth muscle. In: Hypertension: pathophysiology, diagnosis and management, eds. J.H. Laragh & B.M. Brenner. Vol. 1, pp. 1003-1025. New York: Raven Press.
48. Altura, B.M. & Altura, B.T. (1991-1992): Cardiovascular risk factors and magnesium: relationships to atherosclerosis, ischemic heart disease and hypertension. Magnes. Trace Elem. 10, 182-192.
49. Altura, B.M., Altura, B.T., Gebrewold, A., Ising, H. & Gunther, T. (1992): Noise-induced hypertension and magnesium in rats. J. Appl. Physiol. 72, 194-202.
50. Rude, R.K. (1992): Magnesium deficiency and diabetes mellitus. Causes and effects. Postgraduate Med. 92, 217-224.
51. Joffres, M.R., Reed, D.M. & Yano, K. (1987): Relationship of magnesium intake and other dietary factors to blood pressure: the Honolulu heart study. Am. J. Clin. Nutr. 45, 469-475.
52. Lau, K. & Oasa, C.(1984): Interactions between magnesium and blood pressure. Adv. Exp. Med. Biol. 178, 275-290.
53. Gunther, T., Ising, H., Babisch, W. & Vormann, J. (1984): Magnesium intake and blood pressure of spontaneously hypertensive rats. Magnes. Bull. 6, 120-123.
54. Harlan, W. R., Hull, A. L. & Scmouder, R. L. (1984): Blood pressure and nutrition in adults: the National Health and Nutrition Examination Survey. Am. J. Epidemiol. 120, 17-28.
55. Boston, J.H., Beauchene, R.E. & Cruikshank, D.P. (1987): Prospective study of erythrocyte magnesium in teenage pregnancy: relationship to blood pressure and pregnancy induced hypertension (abstr.). J. Am. Coll. Nutr. 6, 41.
56. Ku, D.D. & Hyung, S.A. (1987): Magnesium deficiency produces endothelium-dependent vaso-relaxation in canine coronary arteries. Pharmacol. Exp. Ther. 241, 961-966.
57. Olhabhery, J., Reyes, A.J., Acosta-Barrius, T.N., Leary, W.P. & Queiruga, C. (1987): Pilot evaluation of the putative antihypertensive effect of magnesium. Magnes. Bull. 9, 181-184.
58. Nishiyama, S., Saito, N. & Konishi, Y. (1987): The mechanism of lowered blood pressure in magnesium deficient rats and magnesium deficient rats fed cadmium. J. Jpn. Soc. Mg Res. 6, 93-94.
59. Cohen, L. (1988): Magnesium and hypertension. Magnes. Bull. 8, 93-94.
60. Tillman, D.M. & Semple, P.F. (1988): Calcium and magnesium in essential hypertension. Clin. Sci. 75, 395-402.
61. Whelton, P.K. & Klag, M.J. (1989): Magnesium and blood pressure: review of the epidemiologic and clinical trial experience. Am. J. Cardiol. 63, 26G-30G.
62. Rudnicki, M., Junge, J., Frohlich, A., Ornwold, K. & Fischer-Rasmussen, W. (1990): Magnesium supplement in pregnancy-induced hypertension. A clinico-pathological study. A.P.M.I.S. 98, 1123, 1127.
63. Patki, S., Singh, J., Gokhale, S.V., Bulak, P.M. & Shroti, D.S. (1990): Efficacy of K and Mg in essential hypertension: a double-blind controlled crossover study. BMJ. 301, 521-523.
64. Plum-Wirell, M., Stegmayr, B.G. & Wester, P.O. (1994): Nutritional magnesium supplementation does not change blood pressure nor serum or muscle K and Mg in untreated hypertension. A double-blind cross-over study. Magnes. Res. 7, 277-283.
65. Reed, D., McGee, D., Yano, K. & Hankin, J. (1985): Diet, blood pressure and multicollinearity. Hypertension 7, 405-410.
66. Cardoso, S.M. (1994): Magnesium in tap water (abstr.) Magnes. Res. 7(Supp. 1), 27-28.
67. Sarros, G., Nastos, A., Nastou, H., Sarrou, V., Fotopoulos, N., Anastassopoulous, J. & Kriticos, A. (1994): Magnesium in drinking water and mortality in Greek cities (abstr.). Magnes. Res. 7(Supp. 1), 28-29.
68. Flaubert, G. Letter to Ernest Feydeau (dated August 6 1857). Cited in: Troyat, H. (1988): Flaubert Flammarion, Paris p.183.
69. Smalheiser, N.R. & Swanson, D.R. (1994): Assessing a gap in the biomedical literature: Magnesium deficiency and neurologic disease. Neurosc. Res. Commun. 15, 1-9.
70. Cox, L.M., Campbell, M.J. & Dowson, D. (1991): Red blood cell magnesium and chronic fatigue syndrome. Lancet 337, 757-760.
71. Lasserre, B. & Durlach, J. (1991): Magnesium: a relevant ion. London: John Libbey, 432 p.
72. Durlach, J. (I 992): Chronic fatigue syndrome and chronic primary magnesium deficiency. Magnes. Res. 5, 68.
73. Hall, R. C. W. & Joffe, J. R. (1973): Hypomagnesemia. Physical and psychiatric symptoms. JAMA 53, 159-164.
74. Kanofsky, J.D. & Sandyk, R. (1991): Magnesium deficiency in chronic schizophrenia. Intern. J. Neuroscience 61, 87-90.
75. Haynes, R.B., McKibbon, K.A., Walker, C.J., Ryan, C., Fitzgerald, D. & Ramsdem, M. (1990): Online access to MEDLINE in clinical settings. A study of the use and usefulness. Ann. Intern. Med. 112, 78-84.
76. Haynes, R.B., Mulrow, C.D., Huth, E.J., Altman, D.J. & Gardner, M.J. (1990): More informative abstracts revisited. Ann. Intern. Med. 113, 69-76.
77. Britton, J. (1994): Dietary magnesium, lung function, wheezing and airway hyperreactivity in a random population sample. Lancet 344, 357-362.
78. De Walk, H.W., Kok, P.T.M., Struyvenberg, A., Van Rijn, H.J.M., Halboom, J.R.E., Kreukniet, J. & Lammers, J.W.J. (1993): Extracellular and intracellular magnesium concentrations in asthmatic patients. Eur. Respir. J. 6, 1122-1125.
79. Caddell, J.L. (1972): Magnesium deprivation in sudden infant death. Lancet i, 258-262.
80. Caddell, J.L, (1993): Hypothesis: possible links between the respiratory distress syndrome of the premature neonate. the sudden infant death syndrome and magnesium deficiency shock. Magnes. Res. 6, 25-32.
81. Swanson, D.R. (1988): Migraine and magnesium: eleven neglected connections. Perspect. Biol. Med. 31, 526,-557.
82. Depoortere, H., Francon, D. & Llopis, J. (1993): Effects of a magnesium deficient diet on sleep in rats. Neuropsychobiology 27, 237-245.
83. Nakamura, M., Abe, S., Goto, Y., Chishakis, S., Akazawa, K. & Kato, M. (1994): In vivo assessment of prevention of white-noise-induced seizure in Mg-deficient rats by NMDA receptor blockers. Epilepsy Res.17, 249-256.
84. Nakamura, M., Abe, S. & Akazawa, K. (1995): AMPA - but not NMDA receptor blocker NBQX prevents seizure induction in Mg-deficient rats. Magnes. Res. 8, 55-56.
85. Bara, M., Guiet-Bara. A. & Durlach, J. (1993): Regulation of Na and K pathways by magnesium in cell membranes. Magnes. Res. 167-177.
86. Apfelbaum, M. (1992): Omerta. Chole-Doc. 12, 1-2.
87. Ravnskov, U. (1992): Cholesterol lowering trials in coronary heart disease: frequency of citation and outcome. BMJ 305, 15-19.
88. Egger, M. & Smith, G.D. (1995): Misleading meta-analysis. Lessons from 'an effective, safe, simple intervention that wasn't'. BMJ 310, 752-754.
89. Yusuf, S. & Flather, M. (1995): Magnesium in acute myocardial infarction Br. Med. J. 310, 751-752.
90. Antman, E.M., Lau, J., Berkey, C., McIntosh, M., Chalmers, T.C. & Mosteller, F. (1994): Large vs small trials of acute myocardial infarction: big numbers do not tell the whole story. Circulation 90, abstr. 1743.
91. Fernandes, S.J. (1994): Early intravenous magnesium administration in acute myocardial infarction. Magnes. Res. 7, 341-343.
92. Bara, M., Guiet-Bara, A. & Durlach, J. (1994): Comparative effects of MgCl2 and MgSO4 on the ionic transfer components through the isolated human amniotic membrane. Magnes. Res. 7, 11-16.
93. Bara, M., Guiet-Bara, A. & Durlach, J. (1992): A new method of in vitro prescreening evaluation of several magnesium salts. Methods Find. Exp. Clin. Pharmacol. 14, 305-310.
94. McLean, R.M. (1994): Magnesium and its therapeutic uses: a review. Am. J. Med. 96, 63-76.
95. Itokawa, Y. & Durlach, J. (1989): Magnesium in health and disease. p. 432. London: John Libbey.
96. Sigel, H. & Sigel, A. (1990): Metal ions in biological systems Vol. 26 Compendium on Mg. New York: Marcel Dekker. 744 pp.
97. Huguet, C. & Coppenet, M. (1992): Le magnésium en agriculture. Versailles: INRA. 275 pp.
98. Birch, N.J. (1993): Magnesium and the cell. London: Academic Press 289 pp.
99. Fazekas, T., Selmeczi, B. & Stefanovits, P. (1994): Magnesium in biological systems. Environmental and biomedical aspects. Budapest: Akademiai kiado 335 pp.
100. Albert, D.G., Morita, Y. & Iseri, L.T. (1958): Serum magnesium and plasma sodium in essential vascular hypertension. Circulation 17, 761-764.
101. Ryan, C.A., Finer, N.N. & Barrington, K.J. (1994): Effects of MgSO4 and NO in pulmonary hypertension induced by hypoxia in newborn piglets. Arch. Dis. Child 71, F 151-F 155.
102. Nelson, K.B. (1995): Can MgSO4 reduce the risk of cerebral palsy in very low birthweight Infants? Pediatrics 95, 263-269.
103. Bero, L.A., Galbraith, A. & Rennie, D. (1992): The publication of sponsored symposia in medical journals. N. Engl. J. Med. 237, 1135-1140.
104. Geleijnse, J.M., Witteman, J.C.M., Bak, A.A.A. Den Breijen, J.H. & Grobbee, D.E. (1994): high potassium, high magnesium salt in older subjects with mild to moderate hypertension. BMJ 309, 436-440.
105. Coghlan, H.C. & Natello, G. (1991-1992): Erythrocyte Mg in symptomatic patients with primary valve prolapse: relationship to symptoms, mitral leaflet thickness. joint hypermotility and autonomic regulation. Magnes. Trace Elem. 10, 205-214.
106. Durlach, J. (1994): Primary mitral valve prolapse: a clinical form of primary magnesium deficit. Magnes. Res. 7, 339-340.
107. Easterbrook, P.J., Berlin. J.A., Goplan, R. & Matthews, D.R. (1991): Publication bias in clinical research. Lancet 337, 867-872.
108. Dickersin, K., Min, Y.I. & Meinert, C.L. (1992): Factors influencing publication of research results. Follow-up of applications submitted to two institutional review boards. JAMA 267, 374-378.
109. Culliton, B.J. (1983): Coping with fraud: the Darsee case. Science 220, 31-35.
110. Maisonneuve, H. (1985): Faut-il arbitrer les publications? L'arbitrage est-il scientifique? Presse Med. 14, 873-874.
111. Maisonneuve, H. (1990): L'expertise avant publication. Presse Med. 19, 1254-1256.
112. Snodgrass, G.L. & Pfeifer, M.P. (1992): The characteristics of medical retraction notices. Bull. Med. Libr. Assoc. 80, 328-334.
113. Yankhauer, A. (1990): Editor's report: Scientific misconduct and the responsibility of journal editors. Am. J. Public Hlth. 80, 339-400.
114. Schulman, K., Sulmasy, D.P. & Roney, D. (1994): Ethics, economics and the publication policies of major medical journals. JAMA 272, 154-156.
115. Fruhwald, F.M., Dusleag, J., Fruhwald, S.M., Grisold, M., Gasser, R. & Klein, W. (1993): A physiological oral magnesium supplement does not influence total serum magnesium, left ventricular ejection fraction and prognosis in patients with dilated cardiomyopathy. Magnes. Res. 6, 251-255.
116. International Committee of Medical Journal Editors (1993): Conflict of interest. Ann. Intern. Med. 118, 646-647.
All articles by Dr. Durlach are copyrighted, and permission is granted to Web users only to make single hard copies for personal use. Additional reprints should be obtained from the originating journals. Excerpts may be used by the media with attribution to Dr. Durlach.
This page was first uploaded to The Magnesium Web Site on April 15, 1996