Indirect Assessment of Oxygen Requirements in

Transcription

1 Arh. Dis. Childh., 1966, 41, 25. ndiret Assessment of Oxygen Requirements in Newborn Babies by Monitoring Deep Body Temperature J. W. SCOPS and QBAL AHMD From the Nuffield Neonatal Researh Unit, nstitute of Child Health, Hammersmith Hospital, London The general availability of pure 2 in ylinders gave mediine a powerful therapeuti tool. Until 1952 oxygen was used freely in high onentrations in many premature baby units, in the knowledge that many babies were hypoxi and on the assumption that it would in any ase do no harm. The demonstration that high onentrations of inspired 2 ould lead to retrolental fibroplasia in premature babies (Patz, Hoek, and De La Cruz, 1952) led to more aution, and it beame a generally aepted pratie to restrit added 2 so that inspired onentrations never exeeded 4% (Kinsey, 1956). This seemed partiularly reasonable in view of the demonstration (Campbell, 196) that 2 onentrations of 4% would be able to overome any diffusion defet whih at that time was thought to be the main ause of hypoxia in hyaline membrane disease (Craig, Fenton, and Gitlin, 1958). However, it has subsequently been shown that adherene to this limit may lead to an inreased mortality in premature babies with respiratory distress (Avery and Oppenheimer, 196) and to a higher inidene of neurologial abnormality of the type assoiated with hypoxia (MDonald, 1963). Warley and Gairdner (1962) and Gupta (personal ommuniation) have shown that individual babies with the respiratory distress syndrome may be grossly hypoxi in onentrations of inspired 2 as high as 4%, and that inreasing the 2 onentration to levels previously onsidered dangerous may be neessary to ahieve anything like a satisfatory P2 in the baby's arterial blood. There is now evidene from kittens that it is high 2 tensions in the blood, rather than high onentrations of ambient 2, that lead to retrolental fibroplasia (Ashton, 1964). The liniian is therefore in a dilemma: individual babies are liable to retrolental fibroplasia if exposed to high 2 onentrations, while other babies will Reeived April 26, suffer death or neurologial damage if high 2 onentrations are withheld. What is more, both situations may our at different times in the same baby. Assessing a baby's 2 requirements is made very diffiult by lak of data on whih to base a judgement. The most reliable way of determining 2 requirements is by diret measurement of Po2 in arterial blood. This arries with it the disadvantage of having to obtain arterial blood from a sik infant and of needing speial failities and equipment. The liniian has to depend on signs suh as the presene or absene of yanosis, respiratory irregularity, restlessness (rarely present in hypoxi premature babies), and frequeny of apnoei spells. Any or all of these signs may be useful but none an be said to be entirely reliable. For instane, babies who are grossly hypoxi are not always obviously yanosed. Additional information would undoubtedly be valuable, and we believe that in ertain irumstanes this is provided by monitoring the retal temperature, whih an then be used as a guide to 2 requirements (Davis and Tizard, 1961). Theoretial Bakground t has been shown in babies (Oliver and Karlberg, 1963) as in many newborn mammals, e.g. at (Hill, 1959), rabbit (Adamsons, 1959), monkey (Dawes, Jaobson, Mott, and Shelley, 196), guinea-pig (Dawes and Mestyan, 1963), that moderate hypoxia impairs the ability to inrease heat prodution in response to a ool environment, resulting in a fall in deep body temperature. Fig. 1 shows a model of the situation in homeothermi newborn animals. n the temperature range known as the neutral range, 2 onsumption and, therefore, heat prodution is minimal. The environmental temperature below whih the oxygen onsumption starts to 25 Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

2 26 l >.. n V u C% o lx -1 AR " 15 /o 15/O2 \ Neutral Temp. Sopes and Ahmed Range FG. 1.-A diagram of the relation between metaboli rate, retal temperature, and environmental temperature in the newborn mammal. rise is known as the ritial temperature. Below this environmental temperature, 2 onsumption must be inreased if the deep body temperature is to be maintained. t follows that if the ability to inrease 2 onsumption is impaired by hypoxia the deep body temperature will fall. f the hypoxia is orreted, the 2 onsumption, and later the deep body temperature, will rise. Conversely, the ability to maintain the deep body temperature at environmental temperatures below the neutral range implies that the baby is not seriously hypoxi. This argument is known to apply to a normal newborn baby breathing air and low oxygen mixtures, respetively (Oliver and Karlberg, 1963). t is reasonable to suppose that in babies with inadequate respiratory exhange (respiratory distress, pneumonia, et.) the argument will also apply as between oxygen-enrihed mixtures and air, and that information so obtained might be of use linially. Given a onstant thermal environment the neessary information an be obtained by measuring the deep body temperature or, more ertainly, by measuring the 2 onsumption diretly (see Tizard, 1964). Methods The observations reported here were made on babies in the Neonatal Ward of the Hammersmith Hospital. This ward admits newborn babies needing speial are, the majority of whom are premature, but a substantial minority are term babies with a wide variety of linial onditions. Over the past 18 months, more than 3 oxygen onsumption studies have been made on over 1 of these babies in an investigation into temperature ontrol in newborn babies, under various onditions of sikness and health. The babies reported in this paper were studied in the ourse of this investigation, and all showed linial or laboratory evidene of hypoxia. n many of them, hypoxaemia was onfirmed by measurement of arterial Po2. The ommonest linial ondition was that of 'hyaline membrane disease' (the respiratory distress syndrome of the newborn). This group of hypoxi babies inludes those with birth weights varying between 78 and 3, g. and gestations between 28 weeks and 41 weeks. n a minority of studies a standard ward inubator (Oxygenaire) was used. To prevent undue flutuations in the environmental temperature, the heat ontrol was heked over the period of observation and adjusted manually where neessary. Retal temperature was measured by an eletrial thermometer (thermistor or thermoouple) inserted at least 1 m. into the hild's olon (Karlberg, 1949) and strapped in plae with adhesive tape. 2 onentrations in the inubator were frequently heked, using a paramagneti 2 analyser (Bekman). Throughout the period of observation, nursing proedures were redued to a minimum to prevent flutuations in environmental temperature and 2 onentration. n the majority of the studies, the baby was plaed in a speially onstruted inubator in whih minute by minute 2 onsumption is measured and reorded. With this inubator, one an selet any environmental temperature and any 2 onentration whih then stay onstant to within -1 C. or -1% 2, over the period of observation. Heat exhange by radiation is ontrolled by opaque walls at the temperature of the ambient air. Retal temperature was monitored as before and the hild's linial ondition (olour, respiration, ativity) was onstantly wathed. Where arterial Po2 was measured, blood was taken from an indwelling umbilial artery atheter into a heparinized syringe, the Po2 being measured immediately on a Clark's eletrode (Bekman maro eletrode, sample size 2 ml. blood). Results On numerous oasions in the ourse of the wider investigation, reduing the environmental temperature to below the ritial temperature produed a negligible or only a small inrease in 2 onsumption, with the result that retal temperature began to fall. On at least 21 oasions, inreasing the onentration of 2 in the atmosphere by 2% or more, without hanging other environmental Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

3 ndiret Assessment of Oxygen Requirements in Newborn Babies onditions, aused a substantial inrease in the rate of oxygen onsumption. The normal baby's metaboli response in these irumstanes is not affeted by inreasing the atmospheri 2 onentration (Oliver and Karlberg, 1963; Sopes, unpublished data), so that this substantial rise suggested that the babies onerned were hypoxi before the added 2 was given. n 11 of the babies onerned, the fall in body temperature was prevented or the rate of fall was redued during the period onerned, usually 1 to 2 minutes. n the other 1 no hange in the rate of fall ould be seen in this short period, despite the inrease in oxygen onsumption. The lower environmental temperature hosen was usually between 26 and 29 C., and the inrease in oxygen onsumption effeted by adding 2 was from 1 to 5%. n babies in whom there was undoubted severe anoxia (see individual ases), the inrease was between 36 and 5%. These observations are seen in better perspetive when individual ases are desribed in detail. n some of these ases oxygen onsumption together with retal temperature was being measured, in others the retal temperature alone was reorded while the baby was in the nursing inubator. Case 1 (Fig. 2). This oloured baby was born at 41 weeks' gestation weighing 2,98 g. The mother was given pethidine, 1 mg. intramusularly, one hour before delivery. During labour the foetal heart rate fell to 1 and delivery was assisted by use of the vauum extrator. At birth the hild breathed at one. At 1 minute the hild was in good ondition with a heart rate of 148 per minute, rying, pink, and with good limb tone. Having been born at 11 p.m. she spent the night in the nursery and appeared quite well, but at 8 a.m. next morning she was found to have a retal temperature of 31 1 C. There was no distress or yanosis and apart from her low temperature no abnormality was found on linial examination. When the baby was plaed in the oxygen onsumption inubator she had already been warmed to 33 C. At an environmental temperature of 33 C., breathing air, her retal temperature was rising steadily (Fig. 2). When the environmental temperature was lowered to 28 C., the expeted inrease in 2 onsumption did not our, and the rate of rise of retal temperature fell. The environmental temperature was then raised first to 33 C. and then to 35 C., and the atmosphere was enrihed to 5% oxygen, without ausing any rise in oxygen onsumption. When the environmental temperature was again redued to 28 C., there was a substantial inrease in 2 onsumption. Replaing the atmosphere with air at 28 C. resulted in a fall in 2 onsumption to near basal level, and a fall in retal temperature. nrihing the inspired air one more to 5% 2 again enabled the hild to inrease her 2 onsumption and to maintain the body temperature. t -w a &- 35. U 34 W e o ot v u 36,-34. x wi *,.. FG. 2.--(Case 1 in text.) Retal temperature, ambient oxygen onentration, rate of oxygen onsumption, and environmental temperature, over a 2a-hour period, in a 2,98 g. baby. was onluded that the baby, who showed no yanosis or respiratory diffiulty, was hypoxi in air but not in 5% 2. Her subsequent progress was uneventful and at 1 day of age she was able to maintain her retal temperature spontaneously in air at 28 C. t is noteworthy that all these hanges ould be inferred by observation of the retal temperature, but more rapid and definite information was obtained by measuring oxygen onsumption. Case 2 (Fig. 3). This baby was born prematurely (34 weeks' gestation) and weighed 2,7 g. At birth she breathed at one and at 1 minute had a heart rate of 12, good limb tone, and was pink in olour. At 2 minutes, however, there was a generalized toni onvulsion followed by apnoea, neessitating resusitation. mmediately after spontaneous respiration had restarted at 5 minutes there was mild sternal reession. At 4 hours the respiratory rate had risen to 64, and there was mild reession and grunting. No ause was found for the onvulsion-the CSF was not blood stained. At 22 hours, when the baby was plaed in the 2 onsumption inubator, she still had mild respiratory distress. At an environmental temperature of 34-5' C., 27 Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

4 28 a% i -._ in C Retal Temp. 3C. L 36C. Sopes and Ahmed.' Y 8- C oo nv. Temp. FG. 3.-(Case 2 in text.) The effet on metaboli rate and retal temperature of variation in inubator temperature and ambient oxygen onentration. ah measurement was made over a 1-minute period. Po2 measured while the hild breathed air was 47 mm. Hg. breathing 41/ 2, the retal temperature was rising slowly. On hanging the environmental temperature to 29 C., while the hild was breathing 41 2, there was an inrease in 2 onsumption (Fig. 3), but at this relatively ool temperature the retal temperature fell slightly. While breathing air at the same ool temperature, the 2 onsumption rate fell to near basal levels, and the rate of fall of retal temperature was aelerated. t was onluded that this hild was hypoxi in air and in fat the arterial Po2 estimated at the end of the observation, when the hild was breathing air, was 47 mm. Hg. She was kept in 4% oxygen for the next 24 hours and subsequently had an uneventful reovery. 6- t -S2-2 Case 3 (Fig. 4). This baby was born after 35 weeks' gestation weighing 1,96 g. There had been a small ante-partum haemorrhage at 27 weeks. At 35 weeks the mother was examined under anaesthesia and the membranes were ruptured. The ord prolapsed and the baby was, therefore, delivered at one by lower segment aesarean setion. At 1 minute, the baby was apnoei, limp, and blue, with a heart rate of 66. His airways were aspirated and his trahea intubated, oxygen being given by intermittent positive pressure. By 1 hour of age there was obvious subostal reession on inspiration and he had developed severe respiratory distress, whih lasted 5 days. Over this period his Po2 was onsistently low (see Fig. 4) unless very high onentrations of oxygen were given (7-8%). At 4 days of age he was tested in the oxygen onsumption inubator. At an environmental temperature of 34 9 C., breathing 77% 2, his basal metaboli rate was measured; at this time the retal temperature was rising slowly. On reduing the environmental temperature to 28-2 C., while he was breathing 77% 2, there was a rise in 2 onsumption, though the retal temperature fell slightly. When tested in 54% 2 and 42% 2 for C- 37 C. 4) a W L 36 'C. - N- - on U, m nv. Temp. FG. 4.-(Case 3 in text.) The effet on metaboli rate and retal temperature of variation in inubator temperature and ambient oxygen onentration. ah measurement was made over a 1-minute period. Po2 while breathing 7% 2 was 9 mm. Hg; Po2 while breathing 54 / 2 was 54 mm. Hg; and Po2 while breathing was 22 mm. Hg. 1-minute periods (Fig. 4), the 2 onsumption rate was redued in steps to near basal levels, and the rate of fall of retal temperature was inreased. Subsequently, giving him 85% 2, still at an environmental temperature of 28 2 C., there was a substantial rise in 2 onsumption and the retal temperature remained stable. Arterial Po2 estimations just before and just after the testing period were 22 mm. Hg in 34%O2 54 mm. Hg in 54% 2, and 9 mm. Hg in 7%2- On the sixth day of life he had reovered from his respiratory distress and made good progress. Case 4 (Fig. 5). This baby was born by breeh delivery at 29 weeks' gestation weighing 1,18 g. At 1 minute she was apnoei, limp, and blue, with a heart rate of 8. She was, therefore, intubated and artifiially ventilated by intermittent positive pressure. She developed severe respiratory distress, ompliated by repeated apnoei spells, and needed very high onentrations of environmental oxygen for the first few days of life. At 48 hours an exhange transfusion was performed for jaundie of prematurity. Unlike most babies with respiratory distress, her respiratory diffiulties persisted for weeks. At 48 hours, at the height of her illness, her Po2 in air was only 19 mm. Hg and in 86% 2 only 54 mm. Hg. She was kept in over 6%2 until tested on the eleventh day in the oxygen onsumption inubator. She had a satisfatory metaboli response to ooling to 29 7 C. (Fig. 5b) while breathing 62% 2 and 45% 2, but in 35% 2 there was no response. She was maintained at 55% 2 On the seventeenth day she had a satisfatory metaboli response to ooling to 28-8 C. (Fig. 5) while breathing 47% 2 and 37% 2, but not in 23% 2 She was now kept in 4% O.. On the 23rd day (Fig. 5d) vj ' a) Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

5 P 2 5i4mm.Hg * in 36/o2 d 8 in nv. Tzmp. ndiret Assessment of Oxygen Requirements in Newborn Babies (a) (b) Aged 48 Hr Aged Days mm.hg ir mcy (d) u~ 4 o m so C. (e) () Aged 17 Days 1- N Aged 23 Days Aged 26 Days Aged 31 Days --,1a o4 N 2- O MM CO O ) _V _LN _u m mlt C C C. nv. Temp. FG. 5.-(Case 4 in text.) Changes in oxygen requirements over the period of 1 month in a baby with severe and persistent respiratory distress. (a)-(f) stages as stated in the figure. she had a satisfatory response in 3% 2 but was yanosed in air and so was maintained at 3-35% oxygen. At 26 days (Fig. 5e) there seemed to be some response at 23% 2 but at 31 days (Fig. 5f) she still needed 26% 2. Her subsequent reovery was gradual. Her eyes have been examined regularly by an ophthalmologist, and at 41 months of age no evidene of retrolental fibroplasia has been seen. Case 5 (Fig. 6). This baby was born at home to an unmarried mother who had onealed her pregnany. Gestation was unertain but birth weight was 1,64 g. The baby was thought to be dead at birth and was left lying wet and unovered in a old room. When he was noted to be gasping he was transferred to the Neonatal Ward, where he was found to have a retal temperature of 29 5 C. He was a small oedematous immature baby, breathing spontaneously but irregularly, and without any obvious reession or grunting. n the oxygen onsumption inubator at an environmental temperature of 31 * 8 C. (retal temperature had by now risen to 31 C.) and breathing 33% 2, the retal temperature was rising steadily, but the 2 onsumption rate was very low (3 ml./kg.min.). On replaing with air at the same temperature, the hild beame slightly 'dusky', the 2 onsumption rate fell, and the retal r- r~-rt BC. (f) < 318 ' Cyanoti Attak...o*s*.. **. 33/ AR 5/o 33/o 5/o 4 4 9p.m. 2 TM FG. 6.-(Case 5 in text.) nvironmental temperature, retal temperature, ambient oxygen onentration, and rate of 2 onsumption in the ourse of rewarming a hypothermi premature baby. temperature no longer rose. Giving 5o 2 was assoiated with a rise in retal temperature and in rate of 2 onsumption. As the body temperature rose the environmental temperature was inreased to C. There was at this stage no differene in rate of rise of retal temperature or of rate of 2 onsumption whether the hild was given 5% 2 or 33% 2. At 1.15 p.m. the hild had a yanoti episode assoiated with a fall in 2 onsumption rate and in the rate of rise of retal temperature. He was subsequently given 5% 2. t was not onsidered justifiable to test this baby's 2 onsumption at a lower environmental temperature. He died 4 hours later after an episode of apnoea and was found at neropsy to have a pulmonary haemorrhage-a ommon finding in old babies. Again an aurate reord of the retal temperature gave the same information as measurement of 2 onsumption. Case 6 (Fig. 7). This baby was born prematurely (probably 28 weeks' gestation) weighing 1,2 g. At birth, respiration started spontaneously, but within minutes there was rapid respiration with grunting and reession. He went on to develop severe linial respiratory distress together with episodes of apnoea. At 24 hours in the oxygen onsumption inubator at an environmental temperature of 32 C., breathing 4% 2, the hild's retal temperature was gradually falling, 29 Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

6 3 *6 35. uw ix 34. o o5 C 2 -- o _ nvironmental Temp..o-32OC-32--C C- 4"/o2 7%2 Po2 86mm.Hg Pa2 3mm. AR Pa /. 2 *.. %%%la ** * * OsSO. 3pm p.m p.m. 2 FG. 7.-(Case 6 in text.) nvironmental temperature, ambient oxygen onentration, arterial Po,, retal temperature, and rate of 2 onsumption in a 1,2 g. premature baby who subsequently died with an intraventriular haemorrhage. indiating that this temperature was below the neutral range for this hild. When breathing air there was a small fall in oxygen onsumption rate, and when breathing 7% 2 the rate returned very nearly to that obtaining in 4. However, neither of these hanges was refleted in the reord of the retal temperature whih ontinued to fall slowly. stimations of arterial Po2 while the hild breathed the respetive 2 onentrations were arried out during or immediately after the test period and are entered in Fig. 7. Although hypoxia was abolished by giving 7%2 (in fat a dangerously high P2 was ahieved), this was not refleted in the rate of oxygen onsumption. This hild's failure to ahieve a metaboli response was learly not due only to hypoxia. He subsequently died, and at neropsy he was found to have pulmonary ateletasis with hyaline membrane and an intraventriular erebral haemorrhage. Case 7 (Fig. 8). This baby was born after 33 weeks' gestation weighing 2,1 g. She developed respiratory distress and was treated with added oxygen. At 24 hours she still had rapid respirations and sternal reession. During a test period, the inubator temperature was maintained at 31.7 C. While she was breathing air her retal temperature was falling. When the 2 onentration of the atmosphere was inreased to 4-5%, the retal temperature stabilized. Reduing the 2 onentration one more to 21% was followed by a fall in retal temperature. nvironmental 2 onentrations of 5% and 32% were assoiated with a stable retal temperature. t was onluded that this hild needed at least 32 ' 2 in her atmosphere. She subsequently made a good reovery. Sopes and Ahmed. U - 4a : o u u ~ TM (hours) FG. 8.-(Case 7 in text.) Variation of retal temperature with 2 onentration of inspired air in a premature baby with respiratory distress. nubator temperature was maintained at 31 7 C. Case 8 (Fig. 9). This baby, the smaller of twins, was born at 33 weeks' gestation weighing 92 g. She developed respiratory distress within 2 hours of birth. At 24 hours, while breathing 3' 2 and with the inubator temperature maintained at 33 C. (as it was throughout the period of observation), her retal m 3 6-1! * o 34 : 6-4- x AG (hours) 1 * FG. 9.-(Case 8 in text.) Variation of retal temperature with 2 onentration of inspired air in a premature baby with respiratory distress. nubator temperature was maintained at 33 C. Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

7 ndiret Assessment of Oxygen Requirements in Newborn Babies 31 Raising the onentration of Disussion Our data onfirm the thesis that hypoxia in a temperature was falling. oxygen in the inspired air to 4% and then 6% was assoiated with a stabilization and then a rise in retal temperature: reduing the 2 onentration to between 35 and 4/ was aompanied by a fall in retal temperature, whih was reversed by one more raising the 2 onentration to 6%. At 39 hours of age when the hild was linially somewhat better, reduing the 2 onentration to 3/ was no longer aompanied by a fall in retal temperature whih in fat ontinued to rise. Case 9. This hild was born after 32 weeks' gestation and weighed 1,86 g. at birth. At 1 minute respiration was established and the hild was in good ondition. At 2 hours of age she had several episodes of apnoea, and then developed inreasing dyspnoea. At 4 hours of age her respiratory rate was 8, and there was grunting, subostal reession, and frequent apnoei episodes. At this time arterial Po,, while the hild breathed 1o oxygen, was 42 mm. Hg, and Po2 was 35 mm. Hg. When aged 28 hours she was tested in the 2 onsumption inubator. Breathing 91% oxygen at an environmental temperature of 34 6 C. her 2 onsumption rate was 5-75 ml./kg./min. Reduing the ambient temperature to 27 * 9 C., while she breathed 91% 2 was assoiated with an inrease in 2 onsumption rate to 7-3 ml./kg./min., and an arterial Po2 measured at this time was 44 mm. Hg. At the same ool temperature, the ambient oxygen onentration was redued to 62%. This was assoiated with a fall in 2 onsumption rate to 5 9 ml./kg./min. and a fall in retal temperature. Her arterial P2 measured simultaneously was 32 mm. Hg. She was nursed in 95% oxygen and an environmental temperature of 34 C. but died at 48 hours after an apnoei episode. Neropsy showed pulmonary ateletasis and hyaline membrane. Case 1. This hild was born by breeh delivery after 33 weeks' gestation weighing 1,86 g. At 1 minute he was yanosed, apnoei, limp, and unresponsive, with a heart rate of 6. He was intubated and given artifiial respiration, and spontaneous respiration started by 4 minutes, but at 1 minutes there was already subostal reession, and he went on to develop obvious grunting and dyspnoea. At 13 hours he was tested in the 2 onsumption inubator. At an environmental temperature of 35 * 4 C. his 2 onsumption rate was 4-7 ml./kg./min. At an environmental temperature of 26 6 C., breathing 35% 2, his metaboli rate rose to 6-9 ml./kg./min. and a simultaneous arterial Po2 was 64 mm. Hg. At the same ool temperature, breathing air, the metaboli response was redued to a rate of 6 2 ml./kg./min., at a time when arterial Po2 was 46 mm. Hg. n this hild the metaboli response to old was impaired but not abolished at a Po2 of 46 mm. Hg. He subsequently reovered from his respiratory distress. newborn baby impairs or abolishes his ability to ahieve a metaboli response to ooling. n some of the babies (e.g. Cases 3, 4, 9, and 1) hypoxaemia was proved by measurement of arterial Po2; in all the others there was good irumstantial and linial evidene of hypoxia. When a baby has a poor or absent metaboli response to ooling, and when the response is restored by enrihing the atmosphere with oxygen (all other environmental fators remaining onstant), it suggests (1) that he was hypoxi before the added oxygen, and (2) that adding oxygen has orreted the hypoxia. A failure of metaboli response, whih is not orreted by enrihing the oxygen in his atmosphere, may be explained in one of two ways: first, the enrihment ahieved may have been insuffiient, or the failure may have been due to other fators. Among the other fators that we have found to inhibit a metaboli response to ooling are symptomati hypoglyaemia, various drugs, erebral damage, and severe hypothermia. t follows that if an attempt to restore the metaboli response by putting the baby in a high onentration of oxygen fails, it may be dangerous (in terms of Po2 levels whih, when higher than normal, an ause retrolental fibroplasia) to inrease the 2 onentration still further. f, however, adding 2 restores the metaboli response, and the ambient O, onentration is then kept at not more than 2% above the level at whih the response is impaired, probably there is little danger of ausing retrolental fibroplasia, beause even in babies with normal lungs, the inspired 2 onentration at whih there is danger of ausing retrolental fibroplasia (4%o) is at least 2% above that at whih the metaboli response is impaired (<15% 2). n newborn small animals the metaboli response is impaired at inspired 2 onentrations below 15% and is abolished when the inspired 2 onentration falls below 8% (Adamsons, 1959; Hill, 1959). These perentages would orrespond very roughly with arterial Po2 of 75 and 3 mm. Hg, respetively. Our data, admittedly inadequate (Cases 3, 4, 9, and 1), suggest that in the human baby the metaboli response is seriously impaired at an arterial P2 of 45 to 55 mm. Hg and abolished at a Po2 of about 3 mm. Hg. One would expet yanosis at this P2, but Gupta (personal ommuniation) has repeatedly observed dark blue arterial blood with P2 levels of less than 3 mm. Hg in babies who are linially pink. n fat, dead babies often remain pink if kept in oxygen (Tizard, 1964). More Po2 data of this sort are being sought. t is obviously Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

8 32 Sopes and Ahmed unethial to test the situation artifiially in normal babies and one must wait for ases in whih disease has produed the test situation. ven in the absene of a metaboli hamber, useful information may be gained by monitoring the retal temperature, provided the inubator is equipped to maintain a very stable environment, though some aution and understanding are neessary for the interpretation of the results. Most ommerial inubators fail to provide the exating onditions needed for this test. n partiular, babies lose heat by radiation in single-walled inubators, so that an air temperature within the so-alled neutral zone does not guarantee a neutral thermal environment. Over the test period the baby must be kept at a onstant environmental temperature just below the neutral range whih varies in individual babies. n most small premature babies C. is below the neutral range and in most full-term babies 29-3 C. is below the neutral range (Adams, Fujiwara, Spears, and Hodgman, 1964a, b; Bruk, 1961, and personal observations). These environmental temperatures are unaeptably low for nursing some sik babies, so the test period may have to be limited. Monitoring the retal temperature by repeated insertion of a linial thermometer is not very satisfatory. The hanges sought are small, and fators suh as inadequate time for equilibration, and irregular depth of insertion (Karlberg, 1949), may lead to inauraies. We prefer a thermistor or thermoouple inserted to about 1 m. and strapped in plae. A devie for automati reording is obviously an advantage. ven with good failities and equipment one must be areful that large hanges in heat lost or gained by evaporation or radiation (Agate and Silverman, 1963) are not affeting temperature hange. f, in an environmental temperature below the neutral range, a hild's retal temperature is stable or rising, it suggests he is not seriously anoxi (unless the anoxia is hroni (Bruk, Adams, and Bruk, 1962).) f a falling retal temperature is orreted by inreasing the ambient 2 onentration, without other environmental hange, it suggests he was previously anoxi and that this has been orreted. f the retal temperature is falling and that fall is not orreted by inreasing ambient oxygen, it may mean that still higher onentrations are needed, or that some fator other than hypoxia is preventing a metaboli response (Case 6, Fig. 7). Caution and thought are needed in deiding the next step in suh a ase. We have found it safe and onvenient to make hanges in 2 onentration in steps of 1-2%. n many newborn animals heat is produed by non-shivering thermogenesis (Sopes and Tizard, 1963). There is good evidene that in the rabbit a substantial amount of this heat is produed in brown fat (Dawkins and Hull, 1964) and in the human baby there is irumstantial evidene that the same thermogenesis may apply (Dawkins and Sopes, 1965). A 'small for dates' baby with hypothermia, admitted to this hospital reently, had no metaboli response to old at an age and in a state when he would be expeted to have no brown fat (Aherne and Hull, 1964). There was no metaboli response to ooling and no measurable glyerol in his blood. Two weeks later, when he was well overed with fat, he showed a normal metaboli response to ooling, and the ooling was assoiated on this oasion with a rise in plasma glyerol. Most newborn babies have brown fat in the axilla (Aherne and Hull, 1964), and it is possible that monitoring axillary temperature (near the site of heat prodution in the brown fat) may be a more sensitive index of metaboli response than monitoring retal temperature. Observations on these lines are now in progress. t is lear that monitoring retal temperature an give valuable information for assessing oxygen requirements in babies. t is espeially satisfatory sine it depends on the physiologial response of the baby rather than on arbitrarily determined 'levels'. The information, if it is to be used linially, must obviously be assessed in relation to the total linial situation of the baby inluding its disease, olour, respiration, and general linial state. Conlusion A baby's need for oxygen enrihment of the inspired air an be assessed by measuring his ability to produe heat. ndiret evidene of this ability an be gained by monitoring the retal temperature in ertain thermal onditions, and this information an be valuable linially. For a disussion of the limitations of this method and the preautions neessary the reader is referred to the text. We are grateful to the trustees of the Sir William Coxen trust fund for equipment and laboratory failities, and to the Nuffield Foundation and to H. J. Heinz and Co. Ltd. for grants whih made this work possible. We also wish to thank Dr. J. M. Gupta who performed the Po2 estimations and to Professor J. P. M. Tizard and Dr. J. A. Davis for advie and help in the preparation of this paper. RFRNS Adams, F. H., Fujiwara, T., Spears, R., and Hodgman, J. (1964a). Gaseous metabolism in premature infants at C. ambient temperature. Pediatris, 33, 75. -,,- and - (1964b). Temperature regulation in premature infants. ibid., 33, 487. Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by

9 ndiret Assessment of Oxygen Requirements in Newborn Babies 33 Adamsons, K., Jr. (1959). Breathing and the thermal environment in young rabbits. J. Physiol. (Lond.), 149, 144. Agate, F. J., Jr., and Silverman, W. A. (1963). The ontrol of body temperature in the small newborn infant by low-energy infrared radiation. Pediatris, 31, 725. Aherne, W., and Hull, D. (1964). The site of heat prodution in the newborn infant. Pro. roy. So. Med., 57, Ashton, N. (1964). Retrolental fibroplasia in kittens. Personal ommuniation to Tizard (1964). Avery, M.., and Oppenheimer,. H. (196). Reent inrease in mortality from hyaline membrane disease. J. Pediat., 57,553. Bruk, K. (1961). Temperature regulation in the newborn infant. Biol. Neonat. (Basel), 3, 65. -, Adams, F. H., and Bruk, M. (1962). Temperature regulation in infants with hroni hypoxemia. Pediatris, 3, 35. Campbell,. J. M. (196). Respiratory failure; the relation between oxygen onentrations of inspired air and arterial blood. Lanet, 2, 1. Craig, J. M., Fenton, K., and Gitlin, D. (1958). Obstrutive fators in the pulmonary hyaline membrane syndrome in asphyxia of the newborn. Pediatris, 22, 847. Davis, J. A., and Tizard, J. P. M. (1961). Pratial problems of neonatal paediatris onsidered in relation to animal physiology. Brit. med. Bull., 17, 171. Dawes, G. S., Jaobson, H. N., Mott, J. C., and Shelley, H. J. (196). Some observations on foetal and new-born rhesus monkeys. J. Physiol. (Lond.), 152, , and Mestyan, J. (1963). Changes in the oxygen onsumption of new-born guinea-pigs and rabbits on exposure to old. ibid., 168, 22. Dawkins, M. J. R., and Hull, D. (1964). Brown adipose tissue and the response of new-born rabbits to old. ibid., 172, , and Sopes, J. W. (1965). Non-shivering thermogenesis and brown adipose tissue in the human new-born infant. Nature (Lond.), 26, 21. Hill, J. R. (1959). The oxygen onsumption of new-born and adult mammals. J. Physiol. (Lond.), 149, 346. Karlberg, P. (1949). The signifiane of depth of insertion of the thermometer for reording retal temperature. Ata paediat. (Uppsala), 38, 359. Kinsey, V.. (1956). Retrolental fibroplasia: ooperative study of retrolental fibroplasia and the use of oxygen. Arh. Ophthal., 56, 481. MDonald, A. D. (1963). Cerebral palsy in hildren of very low birth weight. Arh. Dis. Childh., 38, 579. Oliver, T. K., Jr., and Karlberg, P. (1963). The effet of environmental temperature and 15% oxygen on the gaseous metabolism of newborn infants. Ata paediat. (Uppsala), Suppl. 14, p. 51. Patz, A., Hoek, L.., and De La Cruz,. (1952). Studies on the effet of high oxygen administration in retrolental fibroplasia.. Nursery observations. Amer. J. Ophthal., 35, Sopes, J. W., and Tizard, J. P. M. (1963). The effet of intravenous noradrenaline on the oxygen onsumption of new-born mammals. J. Physiol. (Lond.), 165, 35. Tizard, J. P. M. (1964). ndiations for oxygen therapy in the newborn. Pediatris, 34, 771. Warley, M. A., and Gairdner, D. (1962). Respiratory distress syndrome of the newborn-priniples in treatment. Arh. Dis. Childh., 37, 455. Arh Dis Child: first published as /ad on 1 February Downloaded from on 21 August 218 by guest. Proteted by