Objective. The objective of this study was to compare radiant warmer and incubator care for preterm infants from birth with respect to temperature control and weight gain.
Methods. Sixty preterm infants <33 weeks' gestation were randomized at birth to radiant warmer or incubator care. The initial goal was to maintain abdominal temperature at 36.8°C in both groups and axillary temperature at 36.8 to 37.3°C; air servocontrol was used for incubator infants. Infants in both groups received added humidity for 5 days if their weight was <1000 g and for 3 days if they weighed between 1000 and 1249 g. During a 3-hour period on days 1 to 7, recordings of abdominal, forehead, and foot temperatures were obtained. The percentage of the recording time during which the abdominal temperature was in the target range of between 36°C and 37.5°C was determined as an indicator of temperature control. Weight gain from birth to 1800 g was compared. Secondary outcomes included fluid balance and clinical events.
Results. There were 30 infants in each group; 48 were <1500 g (of whom 17 were <1000 g). There were no significant differences in birth weight, gestation, gender, or illness severity scores in the 2 groups. Significant differences in temperature control were noted on day 1. Although admission temperatures were similar, lower abdominal temperatures were noted in the first 2 hours of life in the incubator group (medians were 36.6°C and 35.9°C in the radiant warmer and incubator groups, respectively). Similarly, mean abdominal temperatures during the 3-hour recording on day 1 were lower in the incubator group, and infants in this group spent a significantly greater percentage of the recording time with temperatures outside the target range (17.3% compared with 0.88%). Other temperature recordings from the forehead and foot were not significantly different in the groups. Fluid intakes were higher for infants under radiant warmer on days 2, 3, and 4, and the difference amounted to a mean of 12.8 mL/kg/d. Maximum sodium levels in the first week were similar in the 2 groups. Mean weight gain was 17.4 g/kg/d for the radiant warmer group and 17.1 g/kg/d for the incubator group; days to regain birth weight and length of hospital stay were not significantly different. Greater numbers of infants in the radiant warmer group required phototherapy, and adverse events (which included death, necrotizing enterocolitis, chronic lung disease, grade 3 or 4 intraventricular hemorrhage, periventricular leukomalacia, or retinopathy requiring laser treatment) were less frequent in the radiant warmer group (1 infant compared with 8 in the incubator group; relative risk 0.1; 95% confidence intervals: 0.01–0.82).
Conclusions. This study has shown differences in abdominal temperatures on day 1 and outcome, although the latter finding should be viewed with caution because of the sample size. The results indicate benefits for the initial use of the radiant warmer after birth. Although fluid requirements were higher in the radiant warmer group for days 2 through 4, the increased fluid volumes were given without apparent adverse effect.
Provision of an optimal thermal environment for the preterm infant has long been regarded as a priority in neonatology. Mortality rates were shown by Silverman et al1 to be affected by environmental temperatures. More recently, body temperature has been used for prediction of mortality in the Score of Neonatal Acute Physiology.2 In the long term, weight gain has been shown to be more rapid when the preterm infant is cared for in the thermoneutral zone,3 presumably because more metabolic capacity is available for growth, rather than being expended on temperature control.
Clinically, both overhead radiant warmers and closed incubators can be used to maintain body temperature and limit the metabolic rate in preterm infants.4 Randomized studies comparing the devices thus far have involved stable, preterm infants observed for short periods of time (1 hour-3days).5 The main difference to emerge was increased insensible water losses in radiant warmer-nursed infants. Clinical outcomes relating to infants nursed on one device or the other from birth are not available.
The aim of this study was to compare the 2 warming methods in preterm infants from birth. Primary outcomes were temperature control and weight gain.
Sixty preterm infants <33 weeks' gestation <1750 g were randomized to an incubator (Air Shields double-walled [C550 or Isolette, Vickers Medical, Hatboro, PA] or a radiant warmer [Fisher & Paykel IW930 “Cosy Cot,” Fisher & Paykel, Auckland, New Zealand]). Infants were weighed and stratified randomization was conducted on arrival to the nursery. Randomization block size was 4; a computer program was used for this purpose. Each infant was given a numbered card that had the group assignment concealed beneath an adhesive label. Cards were kept in a locked box, which was opened for randomization. Consent for the study was obtained after randomization, and the design was approved by the Ethics Committees of the Health Authority and Middlemore Hospital.
Infants with major congenital abnormalities, congenital infections, or those who were not born in the hospital were excluded from the study.
The study had 80% power to detect a 5% difference in the percentage time that the abdominal skin temperature was <36°C on any day in the first week of life, and an 80% power to detect an 8% difference in weight gain (50 g over a 1-month period).
The initial goal was to maintain abdominal skin temperature at 36.8°C and to keep the axillary temperature in the range of 36.8°C to 37.3°C. Readings from abdominal thermistors (midline, midway between umbilicus and xiphisternum when supine; paravertebral, lower thoracic region when prone) were displayed continuously for infants in both the incubator and radiant warmer groups from admission. In addition, admission temperature and hourly skin readings were recorded on nursing observation charts. Axillary temperature was monitored at least every 4 hours. Changes to set temperatures were made using an algorithm. For infants nursed under the radiant warmer, if the axillary recording was >37.3°C, the abdominal set temperature was decreased by 0.1°C and the axillary reading was rechecked within 30 minutes. For axillary temperatures below 36.8°C, the set temperature was increased by 0.1°C. Infants in incubator care were preferentially nursed using air servocontrol because this provides a more stable thermal environment.4 Initial air temperature settings ranged from 34°C for those >1500 g to 37°C (with added humidity) for the smallest infants. Handling of infants was achieved through the portholes with cuffs applied. Target abdominal skin temperature was 36.8°C for the incubator group; axillary temperatures were monitored at least every 4 hours, and more frequently if the abdominal readings were below the target level. The air temperature was increased or decreased by 0.1°C if the axillary reading was outside the 36.8°C to 37.3°C range. When axillary temperatures were consistently maintained in the range described, the set point was reduced with increasing postnatal age and maturity.
Apart from the temperature control procedures described above, readings were obtained from thermistors applied to the abdomen (or lower back when prone), forehead, and sole of the foot. The thermistors were covered with reflective foil discs and readings were obtained at 5-second intervals for a limited period of 3 hours. This period was chosen because of the need to ensure continued skin contact for the thermistors and to provide sufficient monitoring equipment for the infants studied. Measurements were obtained in the morning (0900–1200) for the first 7 days of life (day 1 readings were obtained as soon after birth as possible) and recorded on computer. The air temperature and humidity inside the incubator were recorded, as was the ambient air next to the infant on the radiant warmer. For this purpose, an absorption-based polymer thermistor/humidity sensor (Hy-Cal HIH, Honeywell Inc, Morristown, NJ) was suspended in the middle of the incubator 10 cm above the infant. The temperature of the inside of the incubator door was recorded to indicate periods of door opening. Handling of infants under radiant warmers was more difficult to quantify but was measured for infants receiving added humidity as described below. Nursing and medical procedures were conducted as required, regardless of whether the temperatures were being recorded.
The percentage of the temperature recordings that were below 35.5°C, between 35.5 and 35.99°C, and over 37.5°C was determined for each patient on each day. The mean skin temperature was calculated according to a weighted formula, which makes use of the abdominal, foot, and head temperatures.6
Humidification (relative humidity: 70%–80%) was provided for infants with birth weight <1250 g. Infants <1000 g received 5 days humidification; those 1000–1249 g, received 3 days of added humidity. For infants assigned the radiant warmer, a removable tent-like structure made of transparent polyethylene 35-μm thick with flaps at either end was placed over the infant. Humidified air was provided using a humidifier (Fisher & Paykel HC 200 or MR730, Fisher & Paykel, Auckland, New Zealand). A probe measuring relative humidity was placed under the tent. The tent was either completely removed for nursing procedures (eg, airway suctioning) or the end flaps opened for more minor events. In an attempt to quantify handling of infants under radiant warmers, a drop in relative humidity of at least 20% back to the ambient room levels was regarded as indicating a period of handling.
For incubator infants, the incubator humidification mechanism was used. Relative humidity was measured with the same probe as for the radiant warmer.
Fluids, Electrolyte Balance, and Feeds
Initial intake was 10% dextrose water for the first 48 hours. Parenteral nutrition commenced on day 3 with a standard formula containing 10% dextrose, 22.4 g/L of amino acids and electrolytes, including 30 mmol/L of sodium salt (Baxter Healthcare, Auckland, New Zealand). Twenty percent intralipid at 1 g/kg/d was begun on day 3 and increased by 1 g/kg/d to a maximum of 3 g/kg/d. Infants >30 weeks' gestation were commenced on 60 mL/kg/d, and advanced by 30 mL/kg/d, until they reached 160 mL/kg/d. Those 30 weeks' gestation or less commenced at 90 mL/kg/d, and increased 30 mL/kg/d, until an intake of 160 mL/kg/d was attained. This regime was followed for infants in the radiant warmer and incubator groups unless changes were required. Changes were based on results of electrolytes (target sodium values were 135–145 mmol/L), urea and creatinine, urine output (target 2–3 mL/kg/hour), and blood pressure.
Oral feeds were commenced as soon as possible after birth and increased by 30 mL/kg/d as tolerated to a target of 160 mL/kg/d. Intravenous intake was reduced accordingly. Where possible, expressed breast milk was used and fortifier (Cow and Gate Nutricia, Zoetermeer, Holland) was added once full feeds were achieved. Nutriprem (Nutricia) was the infant formula used when insufficient breast milk was available. Feeds were increased to 180 mL/kg/d if weight gain was below 15 g/kg/d averaged over 6 days. Other nutritional supplements included a multivitamin preparation (Vitadol C, Karicare, Douglas Pharmaceuticals, Auckland, New Zealand) and oral iron (3 mg/kg/d).
In addition to weight on admission, alternate day weights were obtained. The weight gain was calculated from the nadir to 1800 g. Respiratory care was predominantly nasal prong continuous positive airway pressure-based, following that of the New York Presbyterian Hospital.
Student's t test was used for normally distributed data. Nonparametric data were tested by the Mann Whitney Utest and Spearman's correlation coefficient; for proportions the χ2 test or Fisher's exact test were used. Binary logistic regression and discriminant analysis were used to determine factors associated with outcome. Analysis of variance for repeated measures was used to compare serial data.
Of the 60 cases, there were 17 infants <1000 g (9 warmer, 8 incubator), 31 weighing between 1000 and 1499 g (15 warmer, 16 incubator) and 12 between 1500 and 1749 g (6 warmer, 6 incubator). The 2 groups are compared in Table 1.
Of 71 eligible infants, 61 were randomized and 60 were studied (1 infant was randomized but consent for the study refused); in 5, randomization was not conducted and 5 were not randomized because of unavailability of monitoring equipment.
The mean values for abdominal skin temperature were similar on admission in the 2 groups (P = .89; Table 2). The mean value of the lowest temperature recorded between admission and 2 hours of age was significantly lower in the incubator group (P = .007). Likewise, a greater proportion in the incubator group had a recording below 36°C between 0 and 2 hours (data was not available from 1 patient in each group). Mean values for the lowest temperature after age 2 hours and up to age 4 hours were also lower in the incubator group (P = .01).
A significant correlation was noted between birth weight and the lowest temperature recorded between 0 and 2 hours (P = .003). Infants were divided into 3 groups depending on birth weight (<1000 g, 1000–1499 g, and 1500–1749 g) and the lowest temperatures recorded between 0 and 2 hours and 2.01 and 4 hours were compared using analysis of variance for repeated measures. This showed significantly lower temperatures in the incubator group (P = .049) and that lower temperatures were recorded in those with lower birth weights (P = .01).
Abdominal Temperature—Percentage Within Target Range
Day 1 Percentage Data for Abdominal Temperature
The median time after birth of commencing the 3 hour recordings was similar in the 2 groups (median in the warmer group was 16 hours, whereas that in the incubator group was 19 hours; P = .3).
The percentage of readings during the 3 hour monitoring period that were below 35.5°C on day 1 was higher in the incubator group (P = .016; medians were 3.9% and 0.8%). Likewise the percentage between 35.5°C and 35.99°C on day 1 was higher in the incubator group (median values were 17.3% in the incubator group and 0.88% in the radiant warmer group; P = .04). There was no significant difference in the percentage of readings above 37.5°C (medians were 0% in both groups; P = .19).
Week 1 Percentage Data for Abdominal Temperature
Using analysis of variance for repeated measures there were no significant differences between days 2 and 3, 3 and 4, 5 and 6; consequently results are shown for days 1, 3, 5, and 7 (Fig 1).
There were no statistically significant differences within subjects (P > .05). Comparing the incubator and radiant warmer groups between subject differences in the percentage of readings, below 35.5°C did not reach statistical significance (P = .06), although a greater percentage of readings between 35.5 and 35.99°C was obtained in the incubator group (P = .008). The percentage of readings above 37.5°C was similar in the 2 groups (P = .77).
Recordings of Abdominal, Forehead, and Foot Temperatures During Week 1
The mean, median, and 95th percentiles for abdominal temperature were significantly higher (P < .05) in the radiant warmer group on day 1 (P = .039 for medians; Fig 2). Comparing median day 1 abdominal temperatures for infants <1250 g (this group received care in a humidified environment), there was a trend toward lower temperatures in the incubator group (P = .058). For infants 1250 g and above (nonhumidified), there was no significant difference in day 1 abdominal temperatures between the incubator and warmer groups (P = .75).
The recordings from the forehead and foot were not significantly different in the 2 groups. Mean skin temperatures were not significantly different between the 2 groups on any of the days.
Apart from day 1 abdominal temperatures, there were no differences between the 2 groups on days 3, 5, and 7 when compared using 2-way analysis of variance. The coefficients of variation for abdominal temperatures were similar in the 2 groups; on day 1 it was 0.65 in the radiant warmer group and 0.53 in the incubator group (P= .38). On day 3, the figures were 0.61 and 0.60 (P = .59), respectively, and 0.48 and 0.46 on day 5 (P = .95). Likewise, day 7 figures were not significantly different (0.66 for the radiant warmer group and 0.52 incubator group;P = .25).
Abdomen-Foot Temperature Gradient in Week 1
The differences between the abdominal and foot temperatures were determined. Overall, there was no significant difference between the incubator and radiant warmer groups in the first week (P = .16; analysis of variance for repeated measures).
When the incubator infants were analyzed separately, there was a trend toward an increased trunk-foot gradient with age in the first week (P = .08 using analysis of variance for repeated measures). There was, however, a difference in the effect of age in the 2 groups. The radiant warmer group showed less gradient from trunk to foot through the first week than the incubator group (P= .05; analysis of variance for repeated measures).
Incubator Door Opening and Handling of Infants Under Radiant Warmers
On average, the incubator doors were opened 1.17 times per hour during normal clinical cares (range: 0–7) in the first week. The period of opening varied considerably in duration and on effect on the infants' temperature (data not shown).
For infants receiving humidification, the median number of handling episodes (as assessed by a 20% drop in relative humidity) was 0.73 occasions per hour for the radiant warmer group compared with 0.79 for the incubator group (P = .2).
Twenty-nine infants received added humidification. Relative humidities for days 1 to 5 were not significantly different in the 2 groups (between subject variation P = .83) and the median values in both groups were between 60 and 70%. In the nonhumidified group, there was no significant variation between subjects (P = .35).
Removal of the humidity tent resulted in a mean abdominal temperature fall of 0.48°C (standard deviation: 0.22) and replacing the tent was associated with a mean abdominal temperature rise of 0.57°C (standard deviation: 0.26). Eight and a half percent of occasions on which the tent was removed resulted in the abdominal temperature falling below 36°C, and in 4.8% of cases replacement of the tent was associated with a rise above 37.5°C.
Fluids and Electrolytes
Daily fluid intakes (mL/kg) were compared. There was no significant difference on day 1 (P = .18), although infants under radiant warmer care required greater fluid volumes on days 2, 3, 4 (analysis of variance for repeated measuresP = .014). The mean daily difference in fluid intake was 12.8 mL/kg/d. There was no significant difference in fluid intake between days 5 and 7 (P = .82). Considering the period after the first week until the end of the study, median fluid intakes were 159 mL/kg/d (interquartile range: 152–170) in the warmer group and 155 mL/kg/d (145–167) in the incubator group (P = .15).
There were no significant differences in the maximum or minimum serum sodium in the first week of life (P = .77). There was a nonsignificant trend toward greater numbers of infants with serum sodium above 150 mmol/L in the warmer group (Table 4). Subgroup analysis did not reveal significant differences in maximum serum sodium; median values in the <1000 g infants were 144 mmol/L and 146 mmol/L in the radiant warmer and incubator groups, respectively (P = .93). The maximum weight loss was similar in both groups (Table 3).
Days to regain birth weight, average length of stay, and weight gain were statistically similar in the 2 groups (Table 3). Median caloric intakes in the warmer and incubator groups for the duration of the study were 102 and 100 kcal/kg/d, respectively (P = .41).
When infants were grouped according to birth weight (<1000 g, 1000–1499 g, 1500–1749 g), there was a significant association between birth weight group and fluid intake on days 1 to 7 (P < .001), with infants <1000 g receiving more fluid than those >1000 g (an average of 22 mL/kg/d more in the warmer group and 14 mL/kg/d more in the incubator group). There were also significant associations between birth weight group and maximum weight loss (P = .036), and birth weight group and weight gain per kg (P < .001). There was a trend toward higher serum sodium in the first week of life and birth weight group (P = .089).
These are summarized in Table 4. Based on bilirubin treatment guidelines,7 larger numbers of the radiant warmer group required phototherapy (P = .01). This was not associated with fluid intake, weight loss, or serum sodium levels. Overall adverse outcome (defined as death, intraventricular hemorrhage grade 3 or 4, periventricular leukomalacia, necrotizing enterocolitis, chronic lung disease, or retinopathy of prematurity requiring treatment), occurred in 1 of the warmer group infants and 8 of the incubator group (P = .026; relative risk 0.1, 95% confidence interval: 0.01–0.82). Discriminant analysis identified birth weight, method of warming, and sex as variables predictive of adverse outcome.
Details of the cases with adverse outcome are as follows. The 1 infant in the warmer group weighed 740 g at birth and had retinopathy grade 3. The 8 in the incubator group included 3 with necrotizing enterocolitis, one of whom died (birth weights 845 g, 690 g, and 1295 g), 2 with chronic lung disease (birth weights 730 g and 1075 g), 1 with periventricular leukomalacia (birth weight 1120 g) and 2 others who died (1 with grade 4 intraventricular hemorrhage and 1 with volvulus not associated with malrotation).
Previous studies have compared radiant warmers and incubators for maintenance of temperatures in the preterm infant.8–13Most were of crossover design, enrolled older stable infants, and were for short periods of time. Furthermore, several of the studies are >20 years old—electronic design has changed significantly since then. Historically, the main difference to emerge was increased insensible water loss under the radiant warmer (weighted mean difference 0.94 g/kg/d).5 Some workers have noted greater oxygen consumption using radiant warmers, although a meta-analysis showed no significant difference.5
The present study was conducted from birth on a range of preterm infants in an intensive care setting and primary outcomes were temperature control and weight gain. Secondary outcomes included important neonatal morbidities and mortality.
Temperature control was assessed in terms of maintenance of target temperatures and variability. The incubator group took longer to warm up after admission on day 1; putting lower birth weight infants at greater risk. This may be significant as low temperatures on admission to the nursery have been associated with a higher mortality in preterm infants14 and presumably the duration of cold stress is important. There were also significant differences in abdominal temperature on day 1; the greater percentage of recording with abdominal temperatures below 36°C on day 1 and in the first week in the incubator group were most noticeable. Is an abdominal temperature below 36°C important physiologically? A recent meta-analysis showed a substantial and significant reduction in mortality when preterm infants in incubators had abdominal temperature servocontrolled to 36°C rather than being nursed at lower temperatures.15 The relative risk was 0.72 (95% confidence intervals: 0.54–0.97) with a risk difference of –12.7%. These findings were more pronounced in very low birth weight infants. Work by Silverman et al16and others6 indicates that an even higher abdominal temperature (36°C–37°C) is closer to the thermoneutral zone. It is likely that the incubator-nursed infants in the present study spent more time outside the thermoneutral zone than the radiant warmer group on day 1. The poorer outcomes in the incubator group may be related to this fact (especially as illness severity scores were similar in both groups on admission). Discriminant analysis indicated that adverse outcome was related to birth weight, type of warming device, and gender. However, a larger study with more power would be needed to examine mortality and specific morbidities.
The upper limit of the desired target range for abdominal temperature was chosen as 37.5°C because of the finding of raised core temperatures above this level.6 The present study indicated that both groups spent very small percentages of the recording times with temperatures above this value.
Earlier work showed that mortality was reduced with improved servocontrol.17 However, it is difficult to relate these earlier findings to current thermoregulation, and variability of temperature recordings (both mean skin and abdominal) in the present study was not significantly different in the 2 groups. The coefficient of variation for both groups during routine clinical care was somewhat higher than that reported by Chessex et al,18 but the latter report concerns stable preterm infants, and incubator opening was controlled to a single period of 2 minutes during 4 hours recording. Most of the temperature variability probably occurs with perturbation of the physical environment, eg, door opening or changes in humidity.4,,19
The number of occasions of incubator door opening in the present study (mean of 1.17 per hour) was in keeping with the 30 per 24 hours reported by other workers4 and fewer than the 2 per hour noted by Chessex et al18 for standard clinical care. In the present study, it was not possible to quantify the number of handling episodes experienced by infants under radiant warmers. The number of occasions the tent was removed for humidified infants was used as a proxy for this and although it was found to be statistically similar to the number of handling events experienced by the humidified incubator group, it is probably an underestimate as some handling occurred through the end flaps of the tent.
Variability of skin temperature in the radiant warmer group was not significantly altered by humidification, although abdominal skin temperature changed by approximately 0.5°C each time the tent was removed or replaced. Heater output was observed to increase markedly when the humidity tent was removed, but the rapid response time of the warmer may have limited the temperature swings.
The calculated mean skin temperatures were not different on day 1, although abdominal temperatures differed in the 2 groups. This may relate to a different temperature distribution and there was a nonsignificant trend toward higher foot temperatures in the incubator infants on day 1. Other workers have noted an increasing gradient between trunk and foot temperatures in incubator infants after birth.19 Although we found such a trend in the incubator group, it was not apparent in the radiant warmer group. Differences in heat distribution between incubators and radiant warmers have been noted previously, although their significance is uncertain.4 These differences in trunk-foot gradient with time and heating device make interpretation in relation to thermal stress difficult.
Target levels of humidity were achieved with both warming devices. Insensible water losses were presumed to be higher in the radiant warmer group because of the higher fluid requirement. Higher insensible losses have been noted previously with radiant warmers and are more pronounced the lower the birth weight.9,,21,22 Despite this, there was no difference in weight loss, days to regain birth weight, serum electrolytes, or the number of infants requiring treatment for patent ductus arteriosus in the present study. This suggests increased fluids can be given to offset these losses in the first week without significant clinical effect. Maximum serum sodium in the first week was similar in the groups, although there was a trend to greater numbers of infants having sodium above 150 mmol/l in the radiant warmer group. Although maintaining fluid balance in preterm infants can be challenging, the percentage of infants <1500 g with sodium levels >150 mmol/l observed in this report (16%) is similar to that noted elsewhere.23 The significance of mild hypernatremia is unclear, and it is not associated with intraventricular hemorrhage.24,,25
Weight gain has been noted to be increased when infants are nursed in the thermoneutral range.3 The similar weight gain, length of stay, and caloric intake observed in the 2 groups in the current work suggests that the overall thermal environment taken over the whole period of the study was equivalent in the groups (although it appeared to differ on day 1).
There have been no previous reports of increased use of phototherapy in relation to radiant warmers. Dehydration and weight loss have been associated with use of phototherapy,26,,27 but logistic regression of data in the present study showed no statistical link with fluid intake, weight loss, or serum electrolytes (results not shown). The finding is currently unexplained.
Infection is also an important clinical outcome. Previous reports have noted increased handling of infants under radiant warmers but no changes in rates of skin colonization.28–30 The present study showed no statistical increase in blood culture positive infections, in the use of antibiotics, or in other clinical features of infection, eg, skin sepsis, respiratory infection, or diarrhea.
This study showed early benefits in the use of radiant warmers—these included more rapid warming and improved maintenance of abdominal temperature on day 1 extending to the end of the first week. A disadvantage was the higher fluid requirement, although this did not seem to affect clinical outcomes. Longer term outcomes, such as weight gain and length of stay, seemed to be similar. The significance of the finding of increased major morbidities with the use of incubators needs additional work and would best be answered by a larger multicenter study. A number of other issues, including access to the infant by staff and parents, exposure to light and noise, and cost have not been addressed in this study.
Funding for the study was obtained from Fisher & Paykel, manufacturers of the radiant warmer.
- Received August 30, 2000.
- Accepted February 5, 2001.
- Address correspondence to Michael P. Meyer, MD, MBChB, Neonatal Unit, Middlemore Hospital, Hospital Rd, Otahuhu, Auckland, New Zealand. E-mail:
Dr de Klerk is currently with the Department of Pediatrics (Neonatology), Maimonides Medical Center, Brooklyn, New York.
- Silverman WA,
- Fertig JW,
- Berger AP
- Richardson DK,
- Gray JE,
- McCormick MC,
- Workman K,
- Goldmann DA
- Glass L,
- Silverman WA,
- Sinclair JC
- ↵Bell EF, Glatzl-Hawlik MA. Environmental temperature control. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology. 2nd ed. Philadelphia, PA: WB Saunders; 1998:716–727
- ↵Flenady VJ, Woodgate PG. Radiant warmers vs incubators in newborn infants. In: Cochrane Library. Issue 2, Oxford, England: Update Software; 2000
- Malin SW,
- Baumgart S
- ↵Perlman M, Kiripalani H, Schmidt B, James A. Guidelines for Neonatal Treatment. Toronto, Canada: The Hospital for Sick Children; 1985
- LeBlanc M
- Marks KH,
- Gunther RC,
- Rossi JA,
- Maisels MJ
- Marks KH,
- Nardis EE,
- Momin MN
- ↵Sinclair JC. Servo-control of abdominal skin temperature at 36°C in low birth weight infants. In: Cochrane Library. Issue 2. Oxford, England: Update Software; 2000
- Wu PKY,
- Hodgeman J
- Omar SA,
- DeCristofaro JD,
- Agarwal BI,
- La Gamma EF
- Dykes FD,
- Lazzara A,
- Ahmann P,
- Blumenstein B,
- Schwartz J,
- Brann AW
- Wu PYK,
- Hodgman JE,
- Kirkpatrick BV,
- White NB Jr.,
- Bryla DA
- Wood B,
- Culley P,
- Roginski C,
- Powell J,
- Waterhouse J
- Davenport SE
- Chang CT,
- Glass L,
- Evans HE,
- Pierog SH
- Copyright © 2001 American Academy of Pediatrics