In: Magnes Res 1991 Sep-Dec;4(3-4):137-52
Summary:The sudden infant death syndrome (SIDS) remains a leading cause of death during the first year. The common epidemiological and pathological data which characterize SIDS include the curve for age at death (with 3 months as modal age), the stigmata of early maternal intrauterine injury, the seasonal predominance in winter, and the absence of an adequate cause of death at autopsy. Some data characterize risk factor subgroups: for example low socioeconomic level, environmental pollution, stress, and mistakes in baby care. Symptoms before death may be lacking, they may be common and non-specific, or rarely they may be acute, corresponding to apparent life-threatening events (ALTE). SIDS may be a magnesium-dependent disease of the transition from chemical to physical thermoregulation. This theory originates from a synthesis of our present knowledge of SIDS, maternal magnesium status, and thermoregulation in the baby. It is consistent with all the epidemiological and pathological prerequisites characterizing SIDS. It eliminates the hiatus between relatively minor thermal stress and induced lethal thermal stroke. Logical scepticism about the role of an implausible lethal superacute magnesium deficiency is no longer justified with regard to well established chronic marginal magnesium deficiency. Further experimental and clinical research will be interesting, i.e. ex vivo studies on brown adipose tissue (BAT) and magnesium deficiency under various conditions of thermal exposure. But even now the theory leads to three therapeutic consequences: (1) the need to define the importance of magnesium deficiency in diagnosis and treatment of ALTE; (2) an assessment of the use of new techniques of rewarming (i.e. extracorporeal circulation) in hypothermia cases to distinguish cot death from "apparent death"; (3) investigation of the prevention of SIDS with magnesium through a blinded and randomized multicentre prospective cooperative study of magnesium supplementation in pregnant and lactating women, followed not only in the mother, fetus, and neonate at birth, but also through the first year of life.
Key words: Apparent life-threatening events, brown adipose tissue, cold stress, cot death, heat stroke, magnesium, near-miss sudden infant death syndrome, sudden infant death syndrome, thermal regulation, uncoupling protein.
The sudden infant death syndrome (SIDS) has been defined since 1970 as the "the sudden death of any infant which is unexpected by history and in which a thorough postmortem examination fails to demonstrate an adequate cause of death"4. SIDS or cot death (crib death) remains a leading cause of infant death (rate of occurrence between 1.5 and 3 per 1000 live births) and has generated hundreds of hypothesis. But any viable hypothesis for the cause of SIDS must take into account its unique epidemiological features and autopsy findings2-5.
The purpose of this study was:
(1) To delineate the epidemiological and pathological data, distinguishing between common characteristics concerning the whole SIDS population and particular data concerning subgroups of SIDS.
(2) To stress the points which indicate that hypothermia, hyperthermia, and magnesium deficit may be causes of SIDS.
(3) To suggest, after a study on the relationship between magnesium and thermoregulation, a new hypothesis on SIDS in agreement with all these prerequisites. SIDS could be a magnesium-dependent disease of the transition from chemical to physical thermoregulation.
(4) To work out further experimental and clinical research including: a SIDS prevention cooperative study on maternal physiological magnesium supplementation, a very cautious approach to the difficult diagnosis of death in hypothermic forms of SIDS, and lastly the evaluation and control of the magnesium disturbances in "near-miss SIDS".
In all cases of SIDS there is concern with certain epidemiological data on the characteristics of the death, on the possibility of intrauterine injury, and on seasonal factors2-8.
Characteristics of the death
The infants die quickly and silently during their sleep. Death is also unexpected: at the usual feeding time, the baby is discovered dead in his cot. But if death occurs during sleep, there is no significant difference in the time of its occurrence between 20.00 and 08.00 or 08.00 and 20.00 hours2-10.
The curve for age at death among SIDS victims is unique. In comparison with the curves for 21 other known causes of infant mortality, the curve for cot death babies appears significantly different. This curve is independent from other risk factors including race, gender, birth weight, seasonality, maternal age, and month of gestation. Although the aetiology of cot death may be multifactorial, there is probably an underlying mechanism related to growth and development in the majority of cases which makes these particular infants highly susceptible between the ages of 2 and 4 months, with 3 months as modal age2-11. The incidence decreases until the age of 6 months and the disease becomes exceptional during the second semester of life2-11. On the other hand, some SIDS deaths occur among newborns and comprise as much as 10% of all neonatal mortality. Neonates present a probably small but significant segment of the total SIDS population. By the design of studies, these early neonatal SIDS deaths have been ignored until now7,12.
Higher incidence rates for SIDS have been observed in infants of diminished birth weight. Several recent analytical studies, for example the NICHD cooperative study13, controlling a variety of potential confounders of birth size, have estimated the likelihood of SIDS deaths in relation to the four variables of birth size: birth weight, birth length, head circumference, and gestational age. SIDS may be closely associated with symmetrical intrauterine growth retardation because of reduction in both birth weight and length even after controlling gestational age and sex of the infant. This suggests that mechanisms responsible for growth retardation in this situation begin early in pregnancy. Post-term infants were at lower risk13-15. The notion of early intrauterine injury as playing some part in the genesis of SIDS relies also on subtle morphological differences between SIDS necropsies and controls. Minor adverse influences during intrauterine development produce some dysplastic and dysmorphic lesions. The difference between the numbers of flaws or minor anomalies in SIDS and controls is impressive. This is another piece of evidence for the emerging concept ascribing a role in SIDS genesis to difficulties arising before birth7,16,17. The nature of this intrauterine risk is speculative. Although the risk of SIDS recurrence in families is greater than for all children, i.e. 1 in 100 instead of 1 in 500, these infants still have a 99% chance of not having SIDS. Heritability is not a major factor. The role of constitutional factors appears of very little importance3-5,7,8,18. Among the environmental factors, various pollutants such as maternal smoking in particular and alcoholism--both factors leading to Pb and Cd exposure--or high altitude exposure are not commonly found in all SIDS cases3-5,8,9,19,25. An early and permanent nutritional factor appears as the most probable hypothesis.
All the epidemiological studies on SIDS show that it occurs predominantly in the winter months. This seasonal factor is observed in all cases of infantile mortality but not during the neonatal period. Common respiratory infections are only represented in about 50% of cases. The seasonal variation of SIDS is thought to be linked to low temperature, either indirectly through an increased frequency of infections or directly through cold stress9-10,26-28.
The single most important finding in the extensive pathological studies of SIDS remains the absence of an adequate cause of death at autopsy6,7.
Besides the common characteristics concerning the whole SIDS population, some features are peculiar to certain subgroups of SIDS. These are environmental risk factors, symptoms before death, and post-mortem features. Environmental risk factors are mainly socioeconomic factors, pollutants, and climatic and local factors.
SIDS is more common in the offspring of families of low socioeconomic status. So called racial susceptibility to SIDS, as well as low maternal age, could be imputed to these socioeconomic disadvantages as well to related cultural factors2-5,7-9.
From epidemiological data, both environmental and social risk factors are known, but the combination of risk indicators suggest that exposure to pollution may be important9,21,29. Among the environmental factors that may participate in intrauterine growth retardation, we have pointed out the deleterious influence of maternal smoking which induces passive cigarette smoking for the baby postnatally3,9,19,22,25,30. Smoking causes carbon monoxide pollution31,32 and Cd exposure20. Alcohol addiction induces its own toxicity9,24 and is a possible factor in lead exposure9,20,21,23,33. Maternal substance abuse--and opioid addiction particularly--increases the risk of SIDS. Moreover endogenous opiates have been involved in SIDS, but there are also links between narcotic abuse and socioeconomic features9,24,32,34-42. Several iatrogenic factors may be considered, phenothiazines in particular5,9,42,43.
Climatic and local data
Apart from seasonal influences, some clinical features may intervene in the genesis of SIDS. Increased altitude predisposes an infant to SIDS. In the same populations post-neonatal deaths from all other causes combined were found to decrease with increasing altitude. Perhaps this is due to in utero hypoxia, which is capable of inducing an early intrauterine insult related to SIDS but which does not result in increased mortality in general44. Ambient temperature may be a factor either of cold or heat stress. But this climatic factor seems hierarchically less important than infant care practices in relation to thermoregulation in babies. An example of this is low incidence of SIDS in Hong Kong45 which has a hot humid climate but where most infants sleep supine. This has tended to confirm the suggestion that the prone position may be an important risk factor46-48, infants sleeping in procubitus being more likely to become hyperthermic than those sleeping in the supine position, other thermoregulation factors being equal. Overheating and the prone position are independently associated with an increased risk of SIDS. Heat loss can occur by convection, radiation, evaporation and conduction27,83-87. Particular cultural combinations of infant care practices (sleeping position, clothing, bedding and room heating) may facilitate a thermal stress in the aetiology of SIDS. Procubitus, overwrapping, excessive clothing, and sometimes an unusually warm environment, possibly combined with mild infection or overlying, could produce heat stroke and hyperthermic SIDS49-57. Conversely underwrapping, insufficient clothing, cold environment , or chilly air coming through an open window could induce cold stroke and hypothermic SIDS26-28,58-61.
Symptoms before death
From 25 to 50% of SIDS cases are unexpected: no premonitory symptoms are observed before sudden death. But in 50 to 75% of cases symptomatology may be present during the days or hours which precede death. Common symptoms include poor feeding, irritability, drowsiness, sudden onset of rapid and noisy breathing, hypothermia or hyperthermia, signs of minimal respiratory or digestive infections, periodic episodes of pallor or cyanosis, and excessive sweating which may appear to be the result of dysautonomic symptoms due to either sympathic or parasympathic hyperactivity. It seems impossible to consider the onset of these non-specific and common symptoms as "sinister symptoms" announcing near sudden death. They may be observed in a minority of cases (around 30%) in conditions of stress9,25,62-65. In a very small number of cases, the symptoms observed before death include acute symptoms which were so severe that the concept of "near-miss SIDS" was originated to describe those infants who had been resuscitated. The reason for including the near-miss cases in the SIDS group are insufficient. Though the main epidemiological data are common to both groups, and though subsequent occurrence of sudden death occurs in some of these infants, the trend is comparatively weak (0.6-4% of cases); that is to say the risk for SIDS is approximately seven times higher than for other infants. Near-miss SIDS infants constitute a mixed population with apparently life-threatening events (ALTE), among which only a few have a truly unexplained episode of near death. The label "near-miss" is most of the time the consequence of a judgment made by anxious parents. Since cardiac death through ventricular fibrillation is practically instantaneous, it is far more likely that a mother will observe her child dying of a respiratory than of a cardiac death. This will allow her to intervene and to save the infant, creating a new near-miss. ALTE represent a subgroup at a somewhat higher risk for SIDS which may provide highly relevant information as to the role of respiratory abnormalities in the genesis of SIDS (apnoea theory), but this group may also constitute a source of quite misleading information if data are uncritically extrapolated to the entire SIDS problem3-6,9,25,65,66.
If the SIDS victim is observed in the room where the death occurred, it is important, before the body is examined, to note environmental factors and the types of infant care practices: ambient temperature, bedding, clothing and sleeping position.
In all cases, before necropsy, inspection shows pallor--sometimes with an erythematous face--or cyanosis. Weight, length and head circumference may show growth retardation. It is of great importance to assess the cooling of the body by taking deep rectal temperature, taking account of the postmortem delay. This evaluation is complicated by the fact that body weight, clothing and ambient temperature must be taken into account, but it allows discrimination to be made between hypothermia or hyperthermia at the moment of the death. If delay after death is less than a day it is possible to take samples for microscopical, bacteriological or virological and biochemical studies2,6,7,68-71. There are classically three main morphological features of the SIDS necropsy: pulmonary congestion, pulmonary oedema, and thoracic petechiae which resemble those found in experimental anaphylactic shock. In a complicated double-blind procedure, a group of three pathologists established that the two pulmonary characteristics were not helpful discriminators between SIDS cases and "dead controls". But thymic petechiae were noted in 52% of cases of cot death. The long-held notion that thoracic petechiae are distinctive, unique, and a singular feature of SIDS is only true for about half of cot death victims, but the other half does not have thymic petechiae7. To support the "apnoea hypothesis", seven so called markers of chronic hypoxia have been described2,68,69 but some of them have not been confirmed and the others, i.e. brown fat alterations or brain stem gliosis, may be present for reasons other than hypoxaemia7. Each SIDS theory emphasizes particular pathological features which in practice are observed in subgroups of SIDS. The main observations involve the nervous, respiratory, cardiac and endocrine systems. Special attention to the brain stem, hypothalamus, and central and peripheral autonomic nervous apparatus shows subtle maturational abnormalities. In the respiratory tract, where minor infection is common, surfactant may [be] altered. In the myocardium the presence or absence of abnormalities in the cardiac conduction system have been reported, as well as in the vascular apparatus (either hypoplasia or hyperplasia of the glomic tissue in the carotid body) In the endocrine system increased chromaffin tissue in the adrenal medulla or Langerhans islet cell hyperplasia of the pancreas may be observed2,6,7,32,68-73. All these data, complemented by cautious extrapolation from in vivo observation of polygraphic and cardiopneumographic patterns in ALTE, point to two mechanisms of death: respiratory and cardiac. Through immaturity of the brain stem and possibly an associated obstructive factor, sudden prolonged spontaneous apnoea induces hypoxia. SIDS infants then succumb to a more prolonged irreversible apnoea. Heart rate abnormalities and QT prolongation measured with Holter recordings may precede instantaneous and silent death by ventricular fibrillation. Hypotheses of developmental or dysautonomic abnormalities (either sympathic or vagal hyperactivity) have been presented, but strangely enough nothing has been found concerning various electrolytic imbalances which, however, are well-established causes for these arrhythmias2,6,7,32,65,68-79.
Among the hypotheses of causes of SIDS it seems important to group together hypothermic and hyperthermic forms as "dysthermic", and to reassess what evidence may support the neglected theory of magnesium deficiency.
The epidemiological and clinical datas which stress the importance of dysthermic forms of SIDS have been highlighted above. It seems that postmortem hyperthermias are more frequent than hypothermic forms80. The curve for age at death among SIDS cases might be correlated with the time of conversion from chemical to physical thermogenesis81,82. Paradoxically, early studies on brown fat in SIDS dealt with their relationship with hypoxia81,83 or with dysautonomia84. It seems more logical to consider their key role in thermal dysregulation. Brown adipose tissue (BAT) is highly specialized for heat production, its mitochondria showing a variable degree of uncoupled oxidative phosphorylation dependent on a 32 Kd mitochondrial protein (UCP). Human UCP gene is now isolated and is assigned to chromosome 4 in q 31. The gradual depletion of BAT starts just after the inevitable exposure to cold at birth and continues during the first year. Thus its disappearance correlated well with the time of conversion from chemical to physical thermogenesis. Intrauterine growth retardation in a premature infant decreases BAT, UCP and the ability to regulate body temperature when exposed to cold. BAT heat production is sympathetically regulated by a complex neurohormonal control system, involving hypothalamus, thyroid, and adrenal glands. SIDS dysthermias may be the result of an impaired control of BAT thermoregulatory mechanisms leading to a modified temperature set-point85.
A few interesting studies on the morphological and chemical alterations of BAT in SIDS allow us to propose that hypothermic forms may be induced by functional failure of BAT and hyperthermic forms by inappropriate functional excess of chemical thermogenesis. Although immaturity91 or fat retention2,7,83,92 may appear as stigmata of hypofunction, fat depletion may correspond either to an insufficient supply inducing hypothermia or to inappropriate lipolysis inducing hyperthermia28,93-95. BAT may be particularly vulnerable to infarction on account of its high metabolic activity and rich capillary plexus96. Brown fat also appears to be a favoured substrate for various virus infections97,99 and a target for maternal alcoholism98 and maternal magnesium deficiency99. Thus the hiatus between the factors of chilling or warming and lethal outcome from cold or heat stroke could be due to a previous disorder of thermal regulation, and primarily to a functional alteration of BAT.
Animal experimentation had shown that heat stroke of the same degree induces sudden death among kittens aged four weeks or less and only hyperthermia among older kittens100. It is suggested that hyperthermic SIDS and febrile convulsions may be equivalent responses to a particular level of heat stress according to age. Most cot deaths occur before 6 months age and febrile convulsions after that age101. The dysthermic theory of SIDS agrees with the conflicting data on the beneficial102 or deleterious74 influence of active sleep (or REM sleep) on the aetiopathogenesis of SIDS. During active sleep, there is an increase in factors which oppose the development of hyperpyrexia. Thus REM sleep can protect against the hyperpyretic form of SIDS and be noxious in the case of the hypothermic form103.
For about 20 years Caddell has been accumulating evidence on the aetiopathogenic role of acute magnesium deficiency in SIDS104-110. Acute experimental and clinical magnesium deficiency may induce a pattern of pseudo-anaphylactic shock. The clinical and pathological anatomical characteristics are those of vagal shock and seem analogous to lesions observed in SIDS, especially those relating to the respiratory and cardiac apparatus. It is interesting to note that the usual absence of any allergen, which is the traditional argument against an immunologic theory, loses its value when it is no longer a question of a true anaphylactic shock but rather of a pseudoallergic anaphylactoid shock due to magnesium deficit, such as can be observed in anaesthesiology or in certain cases of sudden death in adults111. The high metabolism of early baby growth can increase magnesium needs. The lower the weight and age of the animal at the onset of magnesium deficiency the shorter the life and the more abrupt the death105. Clinically neonatal and postneonatal hypomagnesaemia is frequent112 and the parenteral magnesium retention test shows that more than 50% of infants and postpartum women at risk have increased retention. A retrospective study of magnesium therapy in 200 premature neonates with apnoea neonatorum seems to indicate that magnesium supplementation is associated with a reduction in respiratory problems, i.e. apnoea and respiratory distress syndrome, in this population. But this study has been criticized. It was retrospective, not blinded, and not randomized. Magnesium therapy was either intramuscular or oral. The criteria by which private physicians decided on the dose of magnesium and the other confounding variables for apnoea and prematurity were not known. Lastly the urine response to a parenteral magnesium load is discussed as though it were an accurate index of magnesium status in children113,114.
There is only slight epidemiological support for the above thesis. Thus a direct correlation between soft water and SIDS and an inverse correlation between the magnesium concentration in drinking water and SIDS have been shown in one study115. There are some data concerning sudden death in adults, but two major factors may intervene in the relationship between water hardness and sudden cardiac death: magnesium concentration and corrosivity. It is necessary to appreciate the importance of this latter116, which may be in opposition to the findings of Caddell, who extrapolated analogically--as was also done later117--from experiments carried out in weanling rather than newborn rats with superacute (magnesium intake: 3 ppm) or acute (magnesium intake: 50 ppm) magnesium deficiency, to neonates and infants who are much younger and who are never subject to such drastic reductions in magnesium intake. An animal model of cot death should be related to the early mortality rate of the offspring, which in rats is the first week mortality. Experimental studies of magnesium deficiency during pregnancy have mainly studied its teratogenic effects together with problems of parturition and delayed postpartum uterine involution118. In contrast to the usual priorities for other essential nutrients such as potassium suggested by the term "fetal parasitism", the fetus is more severely affected than the mother during maternal magnesium deficit119,120. Very few animal studies have concerned the effect of maternal magnesium intake on the viability of the offspring. If some of the live offspring are eaten by the mother the remainder of the litter die spontaneously. Maternal magnesium deficiency has a marked effect on viability. Maternal magnesium deficit affects the condition of the offspring more severely than that of the mother. The reverse is observed during lactation99,120. Further research is needed on the effect of chronic suboptimal maternal magnesium intake121 on the mortality rate and the quality of the offspring over the duration of pregnancy and lactation.
Postmortem diagnosis of alterations in magnesium status is difficult105,111,116K. Plasma magnesium may sharply increase after stress, for example in ALTE122, and after death, even with magnesium deficiency. Magnesium deficiency has no significant effect on liver magnesium, particularly in the fetus and neonate111,123. Sudden death of adults is accompanied by a significant fall in myocardial magnesium levels. The myocardium seems more vulnerable than skeletal muscle to magnesium loss111,116,124,125. As in adults and because of the heterogeneity of the myocardium116 it seems advisable to section the apex of left ventricle when investigating myocardial magnesium content. Bone is the best tissue for determining whether or not a young mammal has adequate stores. But the irregular distribution of magnesium within a single bone warrants the study of one entire bone, such as the sternum105. Magnesium levels in the vitreous humor are rarely depressed, so their measurement does not provide valid data on which to base the evaluation of magnesium status. Normal vitreous magnesium is not incompatible with an aetiopathogenic role of magnesium in SIDS105,111. Thus measurements of myocardial magnesium and bone magnesium should be routine in SIDS. But these only constitute two items in the overall magnesium status and are not specific investigations of the whole magnesium pool.
The best tool to diagnose experimental or clinical magnesium deficiency is an oral magnesium loading test. At physiological dose level, oral magnesium supplementation is totally devoid of the pharmacodynamic effect of parenteral magnesium. Correction of symptoms constitutes the best proof that they were due to magnesium deficiency111,126. If maternal magnesium deficiency is the cause of newborn and infant magnesium deficiency, maternal magnesium supplementation should be the key to establishing the magnesium theory. Indirect support for these data is given in a large scale prospective follow-up study of 7870 pregnancies over the period from 1964-1970 in which 20 German centres participated. Among 500 variables evaluated it turned out that women with stomach complaints taking magnesium-containing antacids had a surprisingly low rate of early and late abortions and their babies showed low perinatal mortality127. Several prospective double blind studies in Germany and in Hungary on the effect of magnesium supplementation during pregnancy were focused on the consequences of maternal magnesium deficiency on the mother and on the baby at birth. Among the positive effects of maternal magnesium supplementation, six significant results concerned risk factors for SIDS: a longer gestational age, greater birth weight, crown-heel length, and head circumference, APGAR scores less often below 8, and a reduction in the incidence of intrauterine growth retardation. It now appears worthwhile to study the effects of maternal magnesium supplementation during the first year of development of the infant128-131. Cooperative, prospective, blinded, randomized, preventive studies of maternal supplementation should also include a follow up of neonates and infants. Other studies should concern ALTE, all the more so since polygraph studies allow observation of correlations between magnesium and episodes of sleep apnoea132.
In place of an inexplicable superacute severe magnesium deficiency as a possible, but unlikely, cause of SIDS, these new data allow the substitution of the well-established condition of chronic maternal magnesium deficiency, which may be palliated by simple oral physiological magnesium supplementation.
Artificial hyperthermia in man induces negative magnesium balance133 and conversely hypermagnesaemia brings about hypothermia111,134. Body temperature may be regulated through adaptative variations in magnesaemia134. Magnesium activates heat defences. Antihyperthermic magnesium effects might be due to a central sedative action of magnesium on the hypothalamus134 or to a peripheral mechanism,29 reduction of muscular excitability--a factor of physical thermoregulation111,135,136--and of chemical catecholaminergic stimulation of chemical thermoregulation. Acute magnesium deficiency should induce an increase in energy expenditure which may partly depend on central and peripheral neuromuscular hyperexcitability111,135,136. Since the increase in basal metabolism does not seem to be due to an increase in thyroid function137, it could result from a catecholamine activation of BAT111,138. Failure of these central and peripheral antipyretic magnesium effects may favour hyperthermia. Magnesium deficit may well constitute a major element in the aetiopathogenesis of various types of malignant hyperthermia (malignant neuroleptic syndrome, malignant anaesthetic hyperthermia, and heat stroke). It might also play a role in cases of hyperthermic SIDS111.
The magnesium requirement in a cold environment is markedly increased139,140. Cold exposure induces an adaptive rise in serum magnesium. The increase follows a linear relationship to fall in rectal temperature. When the regulation of temperature is poorly developed and particularly at a very young age i.e. puppies vs dogs), the rise in serum magnesium is greatest141. This increase in serum magnesium during efficient adaptation to cold may be due to mobilization of labile magnesium from its stores, mainly bone140. Hypothermic peripheral tissues such as skin could be a source of a transient rise in serum magnesium, but this has only been shown in the turtle, a poikilothermic animal. Dietary intake increases by as much as 60% during cold weather and there may be hypersecretion of adrenaline during cold exposure, but these mechanisms do not occur during hibernation111,141,144.
Various points of similarity between the hypothermic form of SIDS and hibernation have been emphasized61, but hibernation is life-protecting while SIDS is a lethal event. This discrepancy could be explained by a difference in magnesium status: in hibernation, adaptative hypermagnesaemia is a prominent feature, but this is impossible in the face of the chronic magnesium deficiency which may lead to SIDS.
Failure of the adaptational mobilization of magnesium may be
observed during cold stress. For example, in fasted sheep severe
cold exposure induces hypo- rather than
hypermagnesaemia145,146. The incidence of acute
clinical hypomagnesaemia in cattle is sometimes associated with
the transference of the animal onto early spring and late autumn
pastures, when the cattle are subject to sudden extremes in
temperature. Several mechanisms intervene:
(1) Low magnesium intake, since in fed sheep a cold stress capable of causing hypomagnesaemia in fasted sheep induces no variations in magnesaemia146;
(2) Intense hyperadrenalinaemia which induces hypomagnesaemia--and not hypermagnesaemia usually caused by moderate adrenaline hypersecretion--through stimulation of lipolysis;
(3) magnesium chelation by non-esterified free fatty acids;
(4) uptake in adipocytes111,145,146; and
(5) a reduction in the ability of the liver to catabolize citrates during cold stroke may induce magnesium deficit due to an excess of endogenous citrates111,147.
The mechanisms of the beneficial effects of hypermagnesaemia in cold adaptation may come from cytoprotective properties and activation of thermogenesis, mainly chemical, in babies. The well known cytoprotective properties of magnesium are linked to its stabilizing effects on cellular and subcellular membranes, along with its metabolic inhibiting, antihypoxic, and cryoprotective functions111. That is why it is used as an element in solutions need[ed] for perfusing organs destined for transplant surgery and sometimes in open-heart surgery111.
Too few studies have investigated the role of magnesium in BAT physiology. More than 15 years have elapsed since studies were carried out by two groups on some consequences of acute in vivo magnesium deficiency on BAT in adult rats. Mitochondrial swelling and increased adenosine 3',5' monophosphate (cAMP) concentration were observed and appeared to be due to hypersecretion of noradrenaline and adrenaline148,149. The first studies on the relationship between chronic magnesium deficiency and cold stress were carried out in Canada. In adult rats with chronic magnesium deficiency cold stress enhances the development of focal myocardial necrosis, while magnesium supplements are protective150,152.
Heroux et al. published a seminal paper on the long term effect of suboptimal dietary magnesium on Mg and Ca contents of organs, on cold tolerance, and on life span in rats153. Within the first 69 days typical signs of magnesium deficiency were evident in rats acclimatized to 28° C and 6° C. Severe reduction in magnesium content was observed except in brown fat, heart, thymus, and duodenum. It is very interesting that in magnesium deficiency there exist complex homeostatic mechanisms to maintain normal--and even increased--concentrations of magnesium in the tissues during magnesium deficit, which are of vital importance111,154. Between 69 and 517 days the rats had apparently become adapted to the low magnesium diet, with the same relative stability of magnesium content in BAT and an increase in its CA concentration. But the tolerance to severe cold stress at -20° gradually decreased with the duration of magnesium deficiency. Three days after the onset of the stress, all rats acclimatized to 28° showed an increase in the magnesium content of brown fat. In contrast all groups acclimatized to 6° showed a significant decrease in the magnesium content of brown fat after the severe stress. All magnesium-deficient groups acclimatized to 28° showed a degree of hypothermia which gradually increased with the duration of magnesium deficiency. There was a definitely decreasing trend in cold resistance. After cold acclimatization at 6°, the resistance to cold stress was better. Longevity was reduced by chronic marginal magnesium deficiency. At 69 days, there was a higher incidence of lesions and more severe lesions in the heart of magnesium-deficient rat after cold stress. But after adaptation to the magnesium-deficient diet (at 240 and 517 days) the cold stress increased the degree of hypothermia in older rats, but not the incidence of the heart lesions. Reduction in longevity might first depend on deleterious effects on the heart, and then on inadequate thermal regulation153.
Only one study99 concerns BAT in the offspring of magnesium-deficient dams. Newborn rats from dams which had received a magnesium-deficient diet from days 5-12 of pregnancy showed a high death rate. Histological examination showed swelling of the mitochondria and a loss of lipid inclusions in BAT. A link is proposed between the high death rate and the alterations in BAT, since this tissue plays an important role in temperature regulation, especially after birth99. Sadly the numerous subsequent investigations on the consequences of maternal magnesium deficiency and on the relationship between magnesium and lipid metabolism carried out in that laboratory have no longer concerned BAT123,138,155,157.
Various in vitro biochemical data may provide new insights on the link between magnesium and BAT functions. Mg2+ may intervene at each step of chemical thermogenesis: membranous lipoprotein lipase, which induces the capture by adipocytes of circulating triglycerides; microsomal and mitochondrial forms of several enzymes of glycerolipid synthesis; and enzymes which play a role in the lipolytic events following the binding of noradrenaline to the cell membrane receptor (adenylate cyclase, protein kinase, triglyceride lipase, and phosphodiesterase in membranes, cytosol, microsomes, and mitochondria particularly)86,158-161. Uncoupling protein (UCP), which occurs exclusively in BAT mitochondria, conducts protons and dissipates energy, thereby producing heat. The H+ uniport through UCP is inhibited by purine nucleotides and activated by fatty acids. The link between magnesium and nucleotide binding is controversial. It has been claimed to be absent162 or present. Mg2+ appears to be a negative modulator of purine nucleotide inhibition of H+ (and Cl-) transport through UCP163. Mg2+ in BAT of cold-acclimatized hamsters decreases the affinity and the maximum number of available adenosine binding sites164, while in BAT of cold-stressed hamsters it has no effect on the affinity but increases the number of GDP sites165. The binding of GDP to untreated mitochondria may represent thermogenic activity at the time of death whereas after treatment with Mg2+ it may more closely reflect total thermogenic capacity of the mitochondrion166. UCP belongs to the family of mitochondrial ion carriers and the effect on the transport of ions other than H+ should receive consideration. For example according to the Mg2+ carrier-brake hypothesis the Mg-regulated K+-H+ antiporter is the homeostatic mechanism which prevents mitochondrial swelling and lysis. A physiological consequence of respiration-linked proton ejection is the establishment of a large electrochemical cation uptake by mitochondria and subsequently the uptake of salts and water. Matrix Mg2+ determines the amount of matrix K+ at which the K+ cycle is in a steady state and thus determines matrix volume. BAT mitochondrial volume homeostatis seems regulated by matrix Mg2+.167-169 Lastly the synthesis of UCP may be affected by magnesium deficiency, which plays an important role in protein synthesis111.
To sum up, magnesium status intervenes through different mechanisms in the regulation of both warm and cold exposure. Conversely chronic magnesium deficiency may favour both cold and heat stroke.
This new theory considers SIDS as a magnesium-dependent disease of the transition from chemical to physical thermoregulation. It originates from a synthesis of our present knowledge on SIDS, maternal magnesium status, and the relationship between magnesium and thermoregulation in the baby.
First of all this hypothesis on SIDS agrees perfectly with all the necessary epidemiological and pathological prerequisites which characterize SIDS.
The curve for age at death corresponds to the transition between chemical and physical thermoregulation81,82. Chronic marginal maternal magnesium intake, present in all social classes but more frequent in lower socioeconomic groups, constitutes a well known experimental and clinical factor in reduction of gestational age, symmetrical growth retardation with decreased birth weight, birth length, and head circumference, and a significant increase in Apgar score128-131. All accessory factors such as stress, alcohol addiction, CD and lead exposures, may induce secondary magnesium deficit111. Symptoms before death may be exactly those of marginal magnesium deficiency, either totally latent, or symptomatic with signs analogous to "sinister symptoms" and also manifested in ALTE(111). Dysautonomic symptoms are most often amphotonic with either vagal or sympathetic predominance111,154,170. Sleep disorganization is very well known during primary chronic magnesium deficit in the adult111,154,171,172 and is also observed in the infant132. Lengthening of the corrected QT interval may constitute a classical stigma of chronic magnesium deficit111,173-176.
Hypothermia and hyperthermia theories on SIDS could separately explain the hypothermic and hyperthermic clinical forms of SIDS, but the considerable hiatus between such commonplace elements as inappropriate baby caring or climatic factors conducive of dysthermia (even when combined) and the lethal event make it difficult to retain these hypotheses. The large gap between the magnesium intake of the babies and the occurrence of a superacute lethal form of magnesium deficiency makes one question the mechanism of Caddell's theory on SIDS. But all these discrepancies are resolved by the magnesium-dependent dysthermoregulation theory. Chronic maternal magnesium deficiency as early as the beginning of the pregnancy and sometimes during lactation constitutes a too often neglected nutritional factor in early intrauterine injury and in symmetrical growth retardation particularly99,118-121,123,126-131,155-157. Among its possible noxious consequences it may induce alterations in thermoregulation, and particularly on BAT in the newborn and infant99,158-169. This theory rules out the hiatus between the banality of thermal stress and induced lethal thermal stroke. In winter, chilly air from an open window, underwrapping and under-clothing26-28,58-61 may induce hypothermic SIDS because of a magnesium-dependent failure of cold regulation. Conversely, and still in winter, a "common cold" overwrapping, over-clothing, and a prone sleeping position49-57 may induce hyperthermic SIDS because of a magnesium-dependent failure in heat regulation. Scepticism about the role of an implausible lethal superacute magnesium deficiency is logical. But it is out of place with regard to the well established condition of chronic marginal magnesium deficiency in humans, and particularly in pregnant women. In developed countries, recommended daily dietary amounts of magnesium have been set at 6 mg/kg. d. This intake prevents negative magnesium balance177,178 and chronic magnesium deficiency, though with a usual intake around 4 mg/kg. d121,126 chronic deficiency is relatively common. In general, some differences exist between the various nutritionists regarding the special requirements of women during pregnancy: either 1 x RDA (minimum requirement, with an "adaptive" mechanism) or optimal requirement (for example 1.5 or more x RDA). Present knowledge on the experimental and clinical consequences of marginal magnesium deficiency on the mother and fetus99,118-121,123,126-131,155-157 together with the absence of toxicity of physiological oral magnesium intake ( less 2 RDA)111,126,177,178, leads to the conclusion that a daily supplement of 300 mg in pregnant women is highly recommendable.
This new theory on SIDS should generate further experimental and clinical research. A large gap exists between the classical but scanty in vivo data on the importance of magnesium in thermoregulation and the highly elaborate in vitro biochemical and molecular studies, particularly on BAT function. Ex vivo data on BAT from magnesium-deficient animals might bring further information on the relationship between magnesium and thermal regulation. Magnesium-deficient animals should be studied at thermoneutrality, in cold and warm environments, under cold and heat stress, and with or without cold and warm acclimatization. These animal models might indicate the most important targets of magnesium-dependence, allowing us to distinguish labile from stable alterations according to postmortem delay, particularly on BAT. After these preliminary studies, the effect of marginal maternal magnesium deficiency in rats appears to be the cornerstone of the hypothesis. It seems possible to obtain, by inducing marginal magnesium deficiency of dams throughout pregnancy and lactation, newborn rats without malformations but with a high rate of mortality. BAT from these newborns will be studied morphologically--by light and electron microscopy complemented by tissue cartography from a nuclear microprobe--and biochemically. Sympathetic stimulation of BAT may also be studied.
It will be of great interest to use modern biochemical and molecular techniques to investigate current indices of chemical thermogenesis in BAT sections from newborn rats born after chronic magnesium deficiency. Validation of these data should be made in relation to any postmortem delay87-96,179.
Since the human amnion constitutes a valuable and sophisticated model for studying the electrophysiological and ultrastructural membranous effects of magnesium and of several pollutants180-184, it seems of interest to use this technique to study ex vivo the effect of marginal magnesium deficiency during pregnancy on rat amnion. It would also be of interest to determine some markers of marginal maternal magnesium deficiency on this amniotic membrane. The same processes could also be applied to maternal exposure to pollutants.
Clinical research on SIDS
The magnesium-dependent dysthermoregulation theory requires extreme caution when assessing a dead baby's core temperature. It seems important in this respect not only to measure deep rectal temperature using a low reading thermometer, but also to evaluate oesophageal temperature, which may be considered as the reference body core temperature. Peritympanic temperature seems very close to oesophageal temperature and is much easier to measure185. The importance of this type of evaluation is first of all to discriminate between hypothermic and hyperthermic forms71,80, and then to make a rapid diagnosis of death in cases of hypothermia. Studies on severe accidental hypothermia have shown that the external appearance of these patients may be identical to that of cadavers: cold and pallid skin, cyanotic lips and extremities, mydriasis with loss of corneal and pupillary reflex, respiratory and cardiac arrest. Even with profound hypothermia, some of these patients have been rescued by modern treatments. The only definite criterion for diagnosis of death in these cases of severe hypothermia is failure to respond to resuscitation and rewarming185-189.
Clinical research in ALTE
The magnesium-dependent dysthermoregulation theory demands that careful investigation into the most accurate criteria for thermal regulation and magnesium status should be carried out.
Investigations on thermal regulation might include firstly extremely precise evaluation of body core temperature: measurement of deep rectal temperature with low reading thermometers and of peritympanic (or for lack of equipment sometimes oesophageal) temperature185-189; and secondly, attempts to investigate basal BAT functions through non-invasive procedures using thermography, thermometry, computerized tomography, and cautious administration of noradrenaline or modifications of ambient temperature89,190-197. Morphological and biochemical studies on BAT biopsies could be envisaged.
The investigations on magnesium status will be analogous to those described in adults and children, with adjustment for age. Magnesium-dependent symptoms may involve many functions, but especially the nervous and cardiovascular apparatuses. These symptoms are without specificity. Specific magnesium investigations mainly involve free or total magnesium in plasma, erythrocytes, lymphocytes, and urine, but none of these magnesium measurements constitute an exploration of the magnesium pool111,126.
In the final analysis, correction of the clinical abnormalities with an oral physiological magnesium load is the best demonstration of the role of magnesium deficiency in the pathogenesis of a particular clinical disorder111,126. In case of ALTE, dysthermic, cardiovascular and nervous symptoms (and particularly lengthening of QTc, sleep disorganization and apnoeas) should be carefully monitored. The magnesium load should also initiate continuous physiological magnesium supplements which might constitute a specific treatment of ALTE due to chronic magnesium deficiency.
The magnesium-dependent thermal dysregulation theory on SIDS leads to three therapeutic consequences.
(1) Magnesium therapy may constitute a simple treatment of ALTE due to chronic marginal magnesium deficiency. Only systematic research and specific treatment of magnesium deficiency will reveal the importance of magnesium deficiency in the pathogenesis of what may perhaps be a heterogeneous disease.
(2) If some hypothermic cases of cot death are only "apparent death" it becomes very important to consider the diagnosis of death with the same caution as exists in the case of known accidental hypothermia. It seems wise to transfer the baby rapidly to a well equipped emergency unit where rewarming extracorporeal circulation using femoro-femoral bypass may be started. It might be advisable to cautiously add low doses of magnesium to other helpful treatments such as glucose, naloxone and thiamine. Heroic measures, including open thoracotomy and irrigation with warm saline, might be discussed, for example when there is failure to warm the core temperature at a rate of 1° C per hour185-189.
(3) The main consequence of the magnesium-dependent thermal dysregulation theory of SIDS is to develop a simple preventive treatment for this disease. A multicentre cooperative prospective study, blinded and randomized, of magnesium supplementation in pregnant and lactating women should be started. This simple and cheap supplementation with oral physiological doses of magnesium (300 mg/day) is ethically justifiable. In most obstetric departments magnesium supplementation is not systematic. Therefore randomization would not raise any ethical problem with regard to the control group. With the supplemented group, the suggested physiological load (300 mg/day) is without toxicity111,177,178. Furthermore the beneficial effects of magnesium supplementation are well established for the mother, for fetal development, and for the baby at birth127,131. But follow up of the baby over the first year must be made in order to discover the results on apnoea neonatorum, respiratory distress syndrome and SIDS. It will thus be possible to check the validity of the hypothesis and its importance. Magnesium-dependent hypothermic and hyperthermic forms of SIDS may be two subgroups of SIDS or concern the whole population. Supplementation trials are necessary to answer these major questions.
1. Bergmann, A.B., Beckwith, J.B. & Ray, G.G. (1970):Sudden infant death syndrome, pp. 17-18. Washington DC: University of Washington Press.
2. Naeye, R. L. (1977): The sudden infant death syndrome. A review of recent advances. Arch. Pathol. Lab. Med. 101, 165-167.
3. Wagner, M., Samson-Dollfus, D. & Mallet, E. (1984): Facteurs de risque de la mort subite inexpliquée du nourrisson. Gazette Méd. 91, 59-62.
4. Russel-Jones, D.L. (1985): Sudden infant death in history and literature. Arch. Dis. Child. 60, 278-281.
5. Dehan, M. (1986): La mort subite du nourrisson. Concours Méd. 108, 1603-1610.
6. Schwartz, P.J. (1987): The quest for the mechanisms of the sudden infant death syndrome: doubts and progress. Circulation 75, 677-683.
7. Valdes-Dapena, M. (1988): Sudden infant death syndrome. Overview of recent research developments from a pediatric pathologist's perspective. Pediatrician 15, 220-230.
8. Dehan, M. & Gilly, M., eds. (1989): Mort dubite du nourrisson. Progrès en pédiatrie, vol. 6. Paris: Doin Editeurs.
9. Kahn, A., Rebuffat, E., Blum, D., Sottiaux, M. & Muller, M.F. (1989): Aspects épidémiologiques de la mort subite du nourrisson. Ibid, pp. 13-31.
10. Bouvier-Colle, M.H., Hatton, F & Inzan, J. (1989): Situation épidémiologique de la mort subite du nourrisson en France. Ibid, pp. 34-38.
11. Goldberg, J., Hornung, R., Yashimata, T. & Wehrmacher, W. (1986): Age at death and risk factors in sudden infant death syndrome. Aust. Paediatr. J. 22 (suppl. 1), 21-28.
12. Polberger, S. & Svenningsen, N.W. (1985): Early neonatal sudden infant death and near death of fullterm infants in maternity wards. Acta Paediatr. Scand. 74 , 861-868.
13. Van Belle, G., Hoffman, H. & Peterson, D. (1988): Intrauterine growth retardation and the sudden infant death syndrome. In: Sudden infant death syndrome: risk factors and basic mechanisms, eds. Harper R.M. & Hoffman H.J. pp 203-219. New York: PMA Publishing Corporation.
14. Novenius, S.G. (1987): Sudden infant death syndrome in Sweden in 1973-1977 and 1979. Acta Paediatr. Scand. (suppl.) 333, 1-138.
15. Buck, G.M., Cookfair, D.L., Michalek, A.M., Nasca, P.C., Standfast, S.J., Sever, L.E. & Kramer, A.A. (1989): Intrauterine growth retardation and risk of sudden infant death syndrome (SIDS). Am. J. Epidemiol. 129, 874-884.
16. Vawter, F.F. & Kozakewich, H.P.W. (1983): Aspects of morphologic variation amongst SIDS victims. In Sudden infant death syndrome, eds J.T. Tildon., L.M. Roeder & A. Steinschneider, pp. 133-134. New York: Academic Press.
17. Moltz, G. & Hartman. H. (1984): Dysmorphism, dysplasia and anomaly in sudden infant death. N. Engl. J. Med. 26, 259.
18. Dehan, M. & Gilly, R. (1989): Mort subite du nourrisson: prise en charge. Progrès en pédiatre, vol. 6. pp. 317-324. Paris: Doin Editeurs.
19. Peterson, D.R. (1983): Epidemiology of the sudden infant death syndrome: progress, prospects. In Sudden infant death syndrome eds. J.T. Tildon, L.M. Roeder & A. Steinschneider, pp. 89-97. New York: Academic Press.
20. Erickson, M.M., Poklis, A., Ganther, G.E., Dickinson, A.W. & Hillman, L.S. (1983): Tissue minerals levels in victims of sudden infant death syndrome. I Toxic metals. Lead and cadmium. Pediatr. Res. 17, 779-783.
21. Lyngbye, T., Hansen, O.N., Vangbert, L. & Grandjean, P. (1985): Lead as a cause of SIDS. N. Engl. J. Med. 313, 954-955.
22. Toubas, P.L., Duke, J.C., Caffree, M.A., Mattice, C.D., Bendell, D & Orr, W.C. (1986): Effect of maternal smoking and caffeine habits on infantile apnea: retrospective study. Pediatrics 78, 159-163.
23. Drasch, G.A., Ktetschmer, E. & Lochner, C. (1988): Lead and sudden infant death. Investigations on blood samples of SID babies. Eur. J. Pediatr, 147, 79-84.
24. Scher, M.S.S., Richardson, G.A., Coble, P.A., Day, N.L. & Stoffer, D.S.S. (1988): The effects of prenatal alcohol and marijuana exposure: disturbances in neonatal sleep cycling and arousal. Pediatr. Res. 24, 101-105.
25. Demarquez, J.-L., Carre, M. & Paty, J. (1989): Un concept noveau: les malaises graves du nourrisson. Progrès en Pédiatrie, vol. 6, pp. 189-205. Paris: Doin Editeurs.
26. Bonser, R. & West, R. (1979): Sudden infant death and low temperature. Lancet ii, 1379.
27. Cornwell, A.C. (1979): Sudden infant death syndrome. A testable hypothesis and mechanism. Int. J. Neurosci. 10, 31-44.
28. Dunne, K.P. & Matthews, T.G. (1988): Hypothermia and sudden infant death syndrome. Arch. Dis. Child. 63, 438-440.
29. Hoppenbrouwers, T., Calub, M., Arakawa, K. & Hodgman, J.E. (1981): Seasonal relationship of sudden infant death syndrome and environmental pollutants. Am. J. Epidemiol. 113, 623-635.
30. Bergman, A.B. & Wiesner, L.A. (1976): Relation of passive cigarette smoking to sudden infant death syndrome. Pediatrics 58, 665-668.
31. Variend, S. & Forrest, A.R.W. (1987): Carbon monoxide concentrations in infant death. Arch. Dis. Child. 62 417-418.
32. Voyer, M. & Valleur-Masson, D. (1989): Etiologie des morts subites du nourrisson. Progrès en pédiatrie, vol. 6. pp. 135-160, Paris: Doin Editeurs.
33. Vives, J.F., Bellet, H., Lapinski, H., Mirouze, D., Richard, J.L., Hirsch, J.L., Soulayrac, M., Mathieu-Daude, P., Vallat, G, & Michel, H. (1980): Alcoolisme chronique et intoxication saturnine. Gastroenterol. Clin. Biol. 4, 119-122.
34. Rajegowda, H.K., Kandall., S.R. & Falliglia, H. (1978): Sudden unexpected death in infants of narcotic-dependent mothers. Early Human Dev. 2, 219-222.
35. Kuich, T.E. & Zimmerman, D. (1981: Could endorphin be implicated in SIDS? N. Engl. J. Med. 304, 973.
36. Bergsrom, L., Lagercrantz, H. & Terenium, L. (1984): Post-mortem analysis of neuropeptides in brains from sudden infant death victims. Brain Res. 323, 665-668.
37. Hindmarsh, K.W./ & Sankaran, K. (1985): Endorphins and the neonate. Can. Med. Assoc. J. 132, 331-334.
38. Sankaran, K., Hindmarsh, K.W., Wallace, S.M., Mckay, R.J. & O'Donnell, M. (1986): Cerebrospinal fluid and plasma beta-endorphin concentrations in prolonged infant apnea (Near-Miss Sudden Infant Death Syndrome). Dev. Pharmacol. Ther. 9, 224-230.
39. Myer, E.C., Morris, D.L., Adams, M.L., Brase, D.A. & Dewey, W.L. (1987): Increased cerebrospinal fluid beta-endorphin immunoreactivity in infants with apnea and in siblings of sudden infant death syndrome. J. Pediatr. 11, 660-666.
40. Pinto, F., Torrioli, M.G., Casella, G., Tempesta, E. & Fundaro, C. (1988): Sleep in babies born to chronically heroin addicted mothers. A follow up study. Drug Alcohol Depend. 21, 43-47.
41. Ramadabran, K. & Moore, B.E. (1988): Sudden infant death syndrome and opioid peptides from milk. Am. J. Dis. Child, 142, 12-13.
42. Monin, P. (1989): Medicaments et malaises. Progrès en Pédiatrie, vol. 6, pp. 345-352. Paris: Doin Editeurs.
43. Kahn, A. & Blum, D. (1983): Phenothiazines and sudden infant death syndrome. Pediatrics 70, 75-78.
44. Gett, A.G. & Hill, H.F. (1982): Sudden infant death syndrome: incidence at various altitudes. Devel. Med. Child. Neurol, 24, 61-68.
45. Lee, N.N.Y., Chan, Y.T., Davies, D.P., Lau, E. & Yip, D.C.P. (1989): Sudden infant death syndrome in Hong-Kong: confirmation of low incidence. B.M.J. 298, 721.
46. Senecal, J., Roussey, M., Defawe, G., Delahaye, M. & Piquemal, B. (1987): Procubitus et mort subite du nourrisson, Arch. Fr. Pediatr. 44, 131-136.
47. Beal, S. (1988): Sleeping position and SIDS. Lancet ii, 512
48. Engelberts, A.C. & de Jonge, G.A. (1990): Choice of sleeping position for infants: possible association with cot death. Arch. Dis. Child. 65, 462-467.
49. Laplane, R. (1945): Coup de chaleur hivernal. Nourrisson 33, 49-53.
50. Bacon, C., Scott., D. & Jones, P. (1979): Heatstroke in well-wrapped infants. Lancet i, 422-425.
51. Stanton, A.N., Scott, D.J. & Downhame, M.A.P.S. (1980): is overheating a factor in some unexpected infant deaths? Lancet i, 1054-1057.
52. Stanton, A.N. (1984): Overheating and cot death. Lancet ii, 1199-1201.
53. Thach, B.T. (1986): Sudden infant death syndrome. Old causes rediscovered? N. Engl. J. Med. 315, 126-128.
54. Bass, M., Kravath, R.E. & Glass, L. (1986): Death-scene investigation in sudden infant death. N. Engl. J. Med. 315, 100-105.
55. Nelson, E.A.S., Taylor, B.J. & Weatherall, I.L. (1989): Sleeping position and infant bedding may predispose to hyperthermia and the sudden infant death syndrome. Lancet i, 199-201.
56. Fleming, P.J., Gilbert, R., Azay, Y., Berry, P.J., Rudd, P.T. & Stewart, A. (1990): Interaction between bedding and sleeping position in the sudden infant death syndrome: a population based case-control study. B.M.J. 301, 85-89.
57. Lequien, P. (1991): Position de sommeil et risque de mort subite chez le nourrisson. Le Concours Med. 113, 1811-1813.
58. Mann, T.P. (1955): Hypothermia in the newborn: a new syndrome. Lancet i, 613-614.
59. Mann, T.P. & Elliott, R.I.K. (1957): Neonatal cold injury due to accidental exposure to cold. Lancet i, 229-234.
60. Collier, H.O.J. (1971): Sudden death in infancy, Lancet i, 343.
61. Pearson, R.D. & Greenaway, A.C. (1990): Sudden infant death syndrome and hibernation: is there a link? Med. Hypotheses 31, 131-134.
62. Stanton, A.N., Downham, M.A.P.S., Oakley, J.R., Emery, J.L. & Knowelden, J. (1978): Terminal symptoms in children dying suddenly and unexpectedly at home: Preliminary report of the DHSS multicentre study of postneonatal mortality. B.M.J. 11, 1249-1251.
63. Valman, B. (1985): Preventing infant deaths. B.M.J. 290, 339-340.
64. Kahn, A., Van de Merckt, C., Dramaix, M., Magrez, P., Blum, D., Rebuffat, E. & Montauk, L. (1987): Transepidermal water loss during sleep in infants at risk for sudden death. Pediatrics 80, 245-250.
65. Gilly, R. (1987): Le syndrome de la mort subite manquée du nourrisson. Presse Med. 16, 419-420.
66. Dehan, M. & Gilly, R. (1989): La mort subite du nourrisson: un drame familial et un dêfi collectif. Progrès en Pédiatrie, vol. 6, pp. 1-8. Paris: Doin Editeurs.
67. Lequien, P., Carpentier, C., Dumoulin, M. & Pierrat, V. (1989): Mort subite du nourrisson et infection. Progrès en Pédiatrie, vol. 6. pp. 161-170. Paris: Doin Editeurs.
68. Naye, R. (1980): La mort subite du nourrison. Pour la Science 32, 26-33.
69. Valdes-Dapena, M. (1982): The pathologist and the sudden infant death syndrome. Am. J. Pathol. 106, 118-131.
70. Imbert, M.-C. & Gilly, J. (1989): Mort subite du nourrisson: apport de l'anatomie pathologique. Progrès en Pédiatrie, vol. 6. pp. 97-108. Paris: Doin Editeurs.
71. Durigon, M. (1989): Mort subite du nourrisson: aspects medico-légaux. Progrès en Pédiatrie, vol 6, pp. 109-119. Paris: Doin Editeurs.
72. Boulley, A.M., Dehan, M. & Dommergues, J.P. (1980): Mort subite du nourrisson et nésidioblatose. Arch. Fr. Pediatr. 37, 553-554.
73. Kopp, N., Chigr, F., Najimi, M., Denoroy, L., Jordan, D., Awano, K., Coquerel, A., Chayvialle, J.A. & Gilly, R. (1989): Le système nerveux central dans la mort subite inexpliquée du nourrisson: données anatomocliniques. Progrès en Pédiatrie, vol. 6, pp. 131-132. Paris: Doin Editeurs.
74. Samson-Dollfus, D. (1989): Polygraphie et risque de mort subite du nourrisson. Ibid, pp. 208-223.
75. Messer, J. & Haddad, J. (1989): Apports de la cardiorespirographie dans les malaises graves du nourrisson. Ibid, pp. 225-233.
76. Lucet, V. & Cheron, G. (1989): Exploration des anomalies du rythme cardiaque. Ibid, pp. 235-252.
77. Cauchemez, B., Peirano, P. & Samson-Dollfus, D. (1989): Le système nerveux autonome dans le syndrome de la mort subite du nourrisson: analyse du rythme cardiaque et de la variabilité sinusale sur les enregistrements Holter d'enfants décédés du syndrome. Arch. Mal. Coeur Vaiss. 82, 745-752.
78. Stramba-Badiale, M. (1990: Risques cardio-vasculaires dans les morts subites du nourrisson. Réunion du Groupe d'Etude de Langue Francaise sur la Mort Subite du Nourrisson (Nice: 19-20 Oct.) Livre des Rapports, 19-20.
79. Lombroso, C.T. & Lerman P. (1976): Breath-holding spells (cyanotic and pallid infantile syncope). Pediatrics 39, 563-581.
80. Pfeifer, K. (1980): Bedentung der rektal Temperaturmessung und der Umgebmigs-Untersuchungen beim plotzlichen Kindertot. Dtsch. Med. Wochenschr. 105, 1065.
81. Heim, T. (1971): Thermogenesis in the newborn infant. Clin. Obset. Gynecol, 14, 790-820.
82. Stern, L. (1977): Thermoregulation in the newborn infant: physiologic and clinical considerations. Acta. Paediatr. Belg. 30, 3-14.
83. Naye, R.M. (1974): Hypoxemia and the sudden infant death syndrome. Science 186, 837-838.
84. Mettey, R. (1987): Mort subite inexpliquée du nourrisson, tissu adipeux brun, hypertonie sympathique. Arch. Fr. Pediatr. 44, 398.
85. Girardier, L. & Stock, M.J. eds. (1983): Mammalian thermogenesis. London: Chapman & Hall.
86. Portet, R. (1983): Le tissu adipeux brun. Pour la Science 66, 64-71.
87. Ricquier, D., Barlet, J.-P., Garel. J.-M., Combes-George, M. & Dubois, M.P. (1983): An immunological study of the uncoupling protein in brown adipose tissue mitochondrial. Biochem. J. 210, 859-866.
88. Lean, M.E.J., & James, W.P.T. (1983): Uncoupling protein in human brown adipose tissue mitochondria. Isolation and detection by specific antiserum. Febs Lett. 163, 235-240.
89. Lean, M.E.J., Trayhurn, P., Murgatroyd, P.R. & Dixon, A.K. (1987): The case for brown adipose tissue function in humans: biochemistry, physiology and computed tomography. In Recent advances in obesity research V, eds, E.M. Berry, S.H. Blonheim, H.E. Eliahou, & E. Shafrir, pp. 109-116, London: John Libbey.
90. Cassard, A.M., Bouillaud, F., Mattei, M.-G., Hentz, E., Raimbault, S., Thomas, M. & Ricquier, D. (1990): Human uncoupling protein gene: structure, comparison with rat gene and assignment to the long arm of chromosome 4. J. Cell. Biochem. 43, 255-264.
91. Menget, A., Coulon, S., Coulon, G., Oppermann, A. & Raffi, A. (1988): L'autopsie dans la mort subite et inexpliquée du nourrisson. Particularité du poumon et de la graisse brune périsurrénalienne. Progrès en Néonatalogie vol. 2, pp. 154-162. Basle: Karger.
92. Valdes-Dapena, M., Gillane. M.M. & Catherman, R. (1976): Brown fat retention in sudden infant death syndrome. Arch. Pathol. Lab. Med. 100, 547-549.
93. Aherne, W. & Hull, D. (1966): Brown adipose tissue and heat production in the newborn infant. J. Pathol. Bacteriol. 91, 223-234.
94. Emery, J.L. & Dinsdale, F. (1978): Structure of periadrenal fat in childhood in both expected and cot deaths. Arch. Dis. Child. 53, 154-158.
95. Lean, M.E. J. & Jennings, G. (1989): Brown adipose tissue activity in pyrexial cases of cot death. J. Clin. Pathol, 42, 1153-1156.
96. Stephenson, T.J. & Variend, S. (1987): Visceral brown fat necrosis in postperinatal mortality. J. Clin. Pathol. 41, 597-598.
97. Grist, N.R. & Urquart, G.E.D. (1987): Brown fat and sudden death. J. Clin. Pathol. 40, 896-900.
98. Huttunen, P., Kortelainen, M.-L. & Hirvonen, J. (1989): Foetal and lactational exposure to alcohol increases oxidative capacity of brown adipose tissue in the rat. A possible relation to cot death. Br. J. Exp. Pathol. 70, 691-695.
99. Gunther, T., Dorn, F. & Merker, H.J. (1973): Embryotoxic effects produced by magnesium deficiency in rats. Z. Klin. Chem. Klin. Biochem, 11, 87-92.
100. Sunderland, R. & Emery, J.L. (1981): Febrile convulsions and cot death. Lancet ii, 176-178.
101. Lennox, M.A. Sibley, W.A. & Zimmerman, J.M. (1954): Fever and febrile convulsions in kittens: a clinical electroencephalographic and histopathological study. J. Pediatr. 45, 179-190.
102. Watanabe, K., Inokuma, K. & Negoro, T. (1983): REM sleep prevents sudden infant death syndrome. Eur. J. Pediatr. 140, 289-292.
103. Stothers, J.K. & Warmer, R.M. (1984): Thermal balance and sleep state in the newborn. Early Human Dev. 9, 313-322.
104. Caddell, J.L. (1972): Magnesium deprivation in sudden unexpected infant death. Lancet i, 258-262.
105. Caddell, J.L. (1978): Exploring the magnesium-deficient weanling rat as an animal model for the sudden infant death syndrome: physical, biochemical electrocardiographic and gross pathologic changes.
********REFERENCES 106-147 MISSING ********
148. Gunther, T., Schmalbeck, J. & Merker, H.J. (1973): Gehalt des braunen Fettgervebes der Ratte an Noradrenalin und zyklischem AMP im Magnesiummangel. Z. Klin. Chem. Biochem. 11. 233-236.
149. Rayssiguier, Y., Larvor, P. & Durlach, J. (1974): Carence en magnesium chez le rat et variations de certains parametères en rapport avec le métabolisme lipidique. C. R. Soc. Biol. 168, 736-738.
150. Mishra, R.K. (1960): Studies on experimental magnesium deficiency in the albino rat. Rev. Can. Biol. 19, 127-180.
151. Heggtveit, H.A., Herman, L. & Mishra, R.K. (1964): Cardiac necrosis and calcification in experimental magnesium deficiency, Am. J. Pathol. 45, 757-782.
152. Heggtveit, H.A. (1965): The cardiomyopathy of magnesium deficiency. In Electrolyte and cardiovascular diseases, vol. 1, ed. F. Bajusz, pp. 204-220. Basel: Karger.
153. Heroux, O., P.D. & Heggtveit, A. (1977): Longterm effect of suboptimal dietary magnesium on magnesium and calcium content of organs, on cold tolerance and on lifespan and its pathological consequences in rats. J. Nutr. 107, 1640-1652.
154. Durlach, J., Poenaru, S., Rouhani, S., Bara, M. & Guiet-Bara, A. (1987): The control of central neural hyperexcitability in magnesium deficiency. In Nutrients and brain function, ed. W.B. Essman, pp. 48-71, Basel: Karger.
155. Gunther, T., Ising, H., Mohr-Nawroth, F., Chahoud, I. & Merker, H.J. (1981): Embryotoxic effects of magnesium deficiency and stress on rats and mice. Teratology, 225-237.
156. Vormann, J., Forster, R. & Gunther, T. (1983): Foetal and maternal magnesium metabolism: effect of magnesium deficiency and isoproterenol. J. Clin. Chem. Clin. Biochem, 21, 765-773.
157. Gunther, T., Vormann, J., Hollriegl, V. & Gossrau, R. (1991): Effect of magnesium deficiency and salicylate on lipid peroxidation in vivo. Magnesium Bull. 13, 26-29.
158. Darnley, A.C., Carpenter, C.A. & Saggerson, E.D. (1988): Changes in activities of some enzymes of glycerolipid synthesis in brown adipose tissue of cold-acclimated rats. Biochem. J. 15, 351-355.
159. Beleznai, Z., Szalay, L. & Janczik, V. (1988): Ca2+ and Mg2+ as modulators of mitochondrial L-glycerol-phosphate dehydrogenase. Eur. J. Biochem. 170, 631-636.
160. Swick, A.G. & Swick, R.W. (1988): Changes in GDP binding to brown adipose tissue mitochondria and the uncoupling protein. Am. J. Physiol. 255, E865-E870.
161. Rayssiguier, Y., Noe, L., Etienne, J., Gueux, E., Cardot, P. & Mazur, A. (1991): Effects of magnesium deficiency on post-heparin lipase activity and tissue lipoprotein lipase in the rat. Lipids (in press).
162. Klingenberg, M. (1988): Nucleotide binding to uncoupling protein. Mechanism of control by protonation. Biochemistry 27, 781-791.
163. Jezek, P., Houstek, J. & Drahota, Z. (1988): Alkaline pH, membrane potential. Magnesium cations are negative modulators of purine nucleotide inhibition of H+ and Cl- transport through the uncoupling protein of brown adipose tissue. J. Bioenerg. Biomembr. 20, 603-622.
164. Unelius, L., Mohell, N. & Nedergaard, J. (1990): Cold acclimation induces desensitization to adenosine in brown fat cells without changing receptor binding. Am. J. Physiol. 258, C818-C826.
165. Henningfield, M.F. & Swick, R.W. (1989): Magnesium alters number of brown adipose tissue (BAT) mitochondrial GDP binding sites (abstract). FASEB J. 3, A346.
166. Henningfield, M.F. & Swick, R.W. (1989): The concentration of uncoupling protein correlates with Mg2+ activated binding of GDP. Biochem. Cell. Biol. 67, 108-112.
167. Di Resta, D.J., Kutschke, K.P., Hottois, M.D. & Garlid, K.D. (1986): K+-H+ exchange and volume homeostasis in brown adipose tissue mitochondria. Am. J. Physiol. 251, R787-R793.
168. Jezek, P., Beavis, A.D., Di Resta, D.J., Cousino, R.N. & Garlid, K.D. (1989): Evidence for two distinct chloride uniport pathways in brown adipose tissue mitochondria. Am. J. Physiol. 257, C-1142-C1148.
169. Garlid, K.D. (1990): New insights into mechanisms of anion uniport through the uncoupling protein of brown adipose tissue mitochondria. Biochem. Biophys. Acta 1018, 151-154.
170. Durlach, J. (1981): Déficit magnésique, tétanie et dystonie neurovégétative Magnesium Bull. 3, 121-136.
171. Poenaru, S., Durlach, J., Rouhani, S., Reba, A., Auphelle, D. & Iovino, M. (1983): Analyse électroclinique de 100 cas de tétanie par déficit magnésique. Magnesium Bull. 5, 19-23.
172. Poenaru, S., Rouhani, S., Durlach, J., Lemos, M., Magalhaes, J., Durlach V. & Gueux, E. (1989): Magnesium and monoaminergic neurotransmitters: elements of human and experimental pathophysiology. In Magnesium in health and disease, eds. Y. Itokawa & J. Durlach, pp. 291-297. London: John Libbey.
173. Davis, W.H. & Ziady, F. (1978): The effect of oral magnesium chloride therapy on the QTc and QUc intervals of the electrocardiogram. S. Afr. Med. J. 53, 591-593.
174. Aikawa, J.K. (1981): Magnesium: its biologic significance. Boca Raton: CRC Press.
175. Krasner, B.S., Girdwood, R. & Smith, H. (1981): The effect of slow releasing oral magnesium chloride on the QTc interval of the ECG during open heart surgery. Can. Anaesth. Soc. J. 4, 329-333.
176. Reddy, C.V.R., Kyok, J.P., Khan, R.G. & El-Sherif, N. (1984): Repolarization alternans associated with alcoholism and hypomagnesemia. Ann. J. Cardiol. 53, 390-391.
177. Durlach, J. (1989): Recommended dietary amounts of magnesium: Mg RDA. Magnesium Res. 2, 195-203.
178. Durlach, J. Mareschi, J.P. (1991): Recommended dietary amounts for magnesium II. Updated European consensus and future prospects. In Magnesium: a relevant ion, eds. B. Lasserre & J. Durlach, pp. 39-49. London: John Libbey.
179. Lean, M.E.J., James, W.P.T., Jennings, G. & Trayhurn, P. (1986): Brown adipose tissue uncoupling protein content in human infants, children and adults. Clin. Sci. 71, 291-297.
180. Bara, M., Guiet-Bara, A. & Durlach, J. (1985): Monovalent transfer through isolated human amnion: a new pharmacological model. Methods Find Exp. Clin. Pharmacol. 7. 209-216.
181. Bara, M., Guiet-Bara, A. & Durlach, J. (1989): A qualitative theory of the screening-binding effects of magnesium salts on epithelial cell membranes: a new hypothesis. Magnesium Res. 2, 243-248.
182. Guiet-Bara, A., Bara, M. & Durlach, J. (1990): Ultrastructural effects of magnesium on human amniotic epithelial cells. Magnesium Res. 3, 23-29.
183. Guiet-Bara, A., Bara, M., Durlach, J. (1990): Magnesium: A competitive inhibitor of lead and cadmium. Ultrastructural studies on the human amniotic epithelial cell. Magnesium Res. 3, 31-36.
184. Durlach, J. (1990): Données nouvelles sur les mécanismes membranaires d'antagonisme entre le magnésium et l'alcool: implications physiologiques et therapeutiques. Bull. Acad. Natl. Med. 174, 897-908.
185. Binet, J.P. (1991): A propos des hypothermies accidentelles. Rev. Praticien 5, 489-496.
186. Nesemann, M.E., Busch, H.M. & Gunersen, A.L. (1983): Asystolic cardiac arrest in hypothermia. Wis. Med. J. 82, 19-20.
187. Zell, S.C. & Kurtz, K.J. (1985): Severe exposure hypothermia: a resuscitation protocol. Ann. Emergency Med. 14, 339-345.
188. Althaus, U., Aeberhard, P., Schupbach, P., Nachbur, B.H. & Muhlemann, W. (1985): Management of profound accidental hypothermia with cardiorespiratory arrest. Ann. Surg. 195, 492-495.
189. Walpoth, B.H., Locher, T., Leupi, F., Schpupbach, P., Muhlemann, W. & Althaus, U. (1990): Accidental deep hypothermia with cardiopulmonary arrest: extra corporeal blood rewarming in 11 patients. Cardiothorac. Surg. 4, 390-393.
190. Kalberg, P., Moore, R.E. & Oliver Jr, T.K. (1962): The thermogenic response of the newborn infant to noradrenaline. Acta Paediatr. Scand, 51, 284-292.
191. Silverman, W.A., Zamelis, A., Sinclair, J.C. & Agate, F.J. (1964): Warm nape of the newborn. Pediatrics 33, 984-987.
192. Glass, L., Silverman, W.A. & Sinclair, J.C. (1968): Effect of the thermal environment on cold resistance and growth of small infants after the first week of life. Pediatrics 41, 1033-1046.
193. Bolton, D.P.G., Fox, A.M. & Kennaird, D.L. (1970): Preliminary observations on the application of thermography to the study of brown adipose tissue in the human new-born (abstract). J. Physiol.(Lond) 208, 23P-24P.
194. Grausz, J.P. (1970): Interscapular skin temperatures in the newborn infant. J. Pediatr. 76, 752-756.
195. Rylander, E., Pribylova, H. & Lind, J. (1972): A thermographic study of infants exposed to cold. Acta Paediatr. Scand. 61, 42-48.
196. Leibel, R.L., Berry, E.M. & Hirsch, J. (1987): In vivo evidence for catechol-responsive brown adipose tissue in obese patients. In Recent advances in obesity research V, eds. E.M. Berry, S.H. Blonheim, E.H. Eliahou, & E. Shafrir, pp. 117-123, London: John Libbey.
197. Lean, M.E.J. (1989): Brown adipose tissue in humans. Proc. Nutr. Soc. 48, 243-256.
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 January 2, 1996