Ann Thorac Surg 2004;77:66-71
© 2004 The Society of Thoracic Surgeons
Original article: cardiovascular
Early postoperative body temperature and developmental outcome after open heart surgery in infants
Serena M. Cottrell, MRCPCHa,
Kevin P. Morris, MDa*,
Paul Davies, PhDb,
David C. Bellinger, PhDc,f,
Richard A. Jonas, MDd,g,
Jane W. Newburger, MDe,h
a Department of Paediatric Intensive Care, Birmingham Children's Hospital, Steelehouse Lane, Birmingham, B4 6NH, United Kingdom
b Department of Statistical Advisory Service, Birmingham Children's Hospital, Birmingham, United Kingdom
c Department of Neurology, Boston Children's Hospital, Boston, Massachusetts, USA
d Department of Cardiovascular Surgery, Boston Children's Hospital, Boston, Massachusetts, USA
e Department of Cardiology, Boston Children's Hospital, Boston, Massachusetts, , USA
f Department of Neurology, Harvard Medical School, Boston, Massachusetts, USA
g Department of Surgery, Harvard Medical School, Boston, Massachusetts, USA
h Department of Pediatrics, Harvard Medical School, Boston, Massachusetts, USA
Accepted for publication July 18, 2003.
* Address reprint requests to Dr Morris, Pediatric Intensive Care Unit, Birmingham Children's Hospital, Steelehouse Lane, Birmingham, West Midlands B4 6NH, UK
e-mail: kevin.morris{at}bch.nhs.uk
 |
Abstract
|
|---|
BACKGROUND: Experimental data have suggested that early postoperative temperature management after cerebral ischemia may alter neurologic outcome. We explored whether minor deviations in early postoperative body temperature after infant heart surgery affects developmental outcome.
METHODS: In a study of infants undergoing repair of congenital heart disease, 95% of whom had a period of deep hypothermic circulatory arrest, postoperative temperature data were collected following cardiac surgery. Subjects were infants who had been enrolled in one of two prospective randomized single-center trials. Development was tested at age one year (the Bayley Scales of Infant Development) and at four years (Wechsler Preschool and Primary Scale of Intelligence, including Full Scale IQ, a Verbal IQ, and a Performance IQ).
RESULTS: Perioperative temperature data were reviewed in 329 patients, of whom 244 (74%) were evaluated at age one year and 156 (48%) were evaluated at four years. The temperature profile was recorded during the rewarming phase and for 36 hours postoperatively on the Intensive Care Unit. There were no significant associations between postoperative temperature and any of the neurodevelopmental tests at age one or four years. A further analysis assessing the percentage of time over specific temperature cutoff points of 37.5°C, 38°C, 38.5°C, and 39°C, revealed no significant effect.
CONCLUSIONS: Neurodevelopmental outcome at one and four years after repair of complex congenital heart disease was not significantly affected by the early postoperative body temperature profile of the infant when a management strategy aiming for normothermia is employed.
|
Introduction
|
|---|
Induced hypothermia has been used for many years to protect the brain from ischemic insult that may occur during cardiac surgery. Deep hypothermia, using temperatures as low as 12°C, is used to protect the brain from ischemia reperfusion injury during circulatory arrest or low-flow cardiopulmonary bypass surgery. Little attention, however, has been focused on the influence of postoperative temperature management on the development of delayed or secondary brain injury and subsequent neurologic outcome. In the present study, we explored the influence of body temperature in the first 36 hours after reparative infant heart surgery on neurodevelopmental outcomes at ages one and four years. Using prospectively collected data from two randomized trials [1–3], we retrospectively reviewed temperatures recorded in the medical record to explore the effects of temperature on later outcomes.
|
Patients and methods
|
|---|
Subjects
We reviewed the medical records of patients enrolled in prospective trials examining the effect of total circulatory arrest [1, 4–6] and pH strategy [2, 3, 7] on neurodevelopmental outcomes. From infants enrolled in these trials, we selected for this review patients in whom [1] detailed postoperative temperature data were available, and [2] developmental evaluation at ages one year and(or) four years were performed. Each patient was given a code number to protect patient confidentiality. Parents of patients enrolled in each trial had given informed consent for participation in accordance with guidelines of the Institutional Review Board of Boston Children's Hospital.
Operative methods
After induction of anesthesia, the infants were surface cooled with ice packs to the head, a cooling mattress, and low ambient room temperature. Cardiopulmonary bypass and core cooling to a rectal temperature of 18°C or lower were initiated as soon as the bypass cannulas were in place. Any infant with a lowest rectal temperature more than 18°C was excluded from the trials. After surgery, the infants were rewarmed to a rectal temperature of at least 35°C before the cessation of bypass. Other details of operative and anesthetic protocols are provided elsewhere [1–3, 6, 8, 9].
Temperature data
The temperature profile was analyzed using two separate phases: (1) the rewarming phase while in the operating room and (2) the Intensive Care Unit (ICU) phase. Temperatures (rectal and tympanic) were measured every 15 to 30 minutes in the operating room phase and speed of rewarming was calculated in degrees Celsius per minute. Tympanic temperature values were used in the analysis of the rewarming phase because we felt tympanic temperature would most closely approximate to brain temperature and, hence, was most relevant for later neurodevelopmental outcome. In the ICU, temperatures were recorded to the nearest 30 minutes and were mainly rectal measurements. A combination of a three point moving average and linear interpolation was used to calculate a temperature profile with values at 1.5-hour intervals for each case to minimize the bias resulting from any missing values (calculated temperature profile).
Developmental testing
Infants in both studies had been prospectively tested at approximately age one year with the Bayley Scales of Infant Development. This test results in two scores, the Mental Developmental Index (MDI), which examines precognitive function, and the Psychomotor Developmental Index (PDI), which assesses gross and fine motor function [10, 11]. Children enrolled in the Boston Circulatory Arrest Study, all of who had D-transposition of the great arteries (D-TGA), were tested at age four years with the Wechsler Preschool and Primary Scale of Intelligence Revised [12]. This includes a Full-Scale Intelligence Quotient (FSIQ), Verbal IQ (VIQ), and Performance IQ (PIQ). The results of neurodevelopmental testing in these two study cohorts as they pertain to the earlier studies have been reported elsewhere [1–8].
Statistical analysis
General linear model analysis of variance was used to study the relationships between the neurodevelomental indices (MDI, PDI, FSIQ, VIQ, and PIQ) and (1) the speed of rewarming for the operative phase, and (2) the temperature profiles at 90-minute intervals in the ICU phase. Adjustments were made for the effects of different treatment groups (total circulatory arrest vs low flow bypass, 
-stat vs pH-stat), diagnosis, and circulatory arrest times.
Within a subgroup of patients at greater risk of neurologic injury, ie, those with 40 minutes or more of deep hypothermic circulatory arrest (DHCA), we used a Pearson's correlation coefficient to analyze the influence of postoperative temperature on the MDI and PDI scores. Pearson's correlation coefficients were also used to analyze a relationship between the neurodevelopmental indices and the percentage of time above the following specific temperature points: 37.5°C, 38°C, 38.5°C, and 39°C.
|
Results
|
|---|
We performed a chart review on 352 patients enrolled in the two prospective randomized trials [1–3]. Three hundred and thirty five infants (95%) had a period of DHCA during surgery. Perioperative temperature data were available in 329 patients (93%), of whom 244 (74%) underwent neurodevelopmental assessment at one year of age and 156 (47%) were evaluated at four years of age. In the remaining patients, either temperature data or neurodevelopmental follow-up data were unavailable. The initial rewarming period in the operating room was very rapid, followed by a slow stabilization to normothermia over the following 2 to 4 hours (Figs 1 and 2).
The mean speed of rewarming was 0.6 degrees per minute with an interquartile range of 0.53 to 0.61 degrees Celsius per minute. Only 3.3% of patients had a temperature of greater than 37.5°C in the final stages of rewarming. The maximum temperature reached during rewarming was 38.2°C (rectal temperature).
View larger version (18K):
[in this window]
[in a new window]
|
Fig 1. Temperature versus time, showing both the rewarming phase in the operating room and the ICU phase for 36 hours following open heart surgery. This figure demonstrates the minimum value (Min), 25th, 50th (median), 75th percentiles, and the maximum value (Max). Total cases 328. (ICU = intensive care unit.)
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig 2. Temperature versus time showing the ICU phase only. This figure demonstrates the 10th, 25th, 50th (median), 75th, and 90th percentiles. Total cases 328. (ICU = intensive care unit.)
|
|
Management of temperature deviations was not standardized by protocol during either study. However, according to usual clinical practice, large deviations prompted use of topical cooling in the case of pyrexia and the use of a warming device in the presence of a low body temperature. This resulted in a relatively narrow range of temperature while the infants were on the ICU. Only 0.4% of all the recorded temperatures were 39°C or above, 1.6% were 38.5°C and above, and only 6% were 38°C or above. The maximum deviation occurred at the 12 hour time point with 15% of the temperature values reaching 38°C or above. Mild hypothermia was common in the first 3 hours in the ICU with 42% of readings at or below 36°C. However, overall fewer than 3% of temperature recordings were below 35.5°C.
Mental and psychomotor developmental indexes at one-year follow-up
Of the three hundred and twenty nine children in whom perioperative temperature data were available, 244 (74%) underwent developmental evaluation at age one year. Neither speed of rewarming on bypass nor temperature at any time in the postoperative period was significantly associated with the MDI or PDI at age one year. Figures 3 and 4
demonstrate the scatter plot for the MDI and the PDI scores, respectively, at one of these time points (12 hours following admission to ICU). The other time points show similar distributions. Additionally, analyses within a subgroup of patients most at risk of brain injury, having received 40 minutes or more of DHCA, showed no correlation between postoperative temperature and the MDI or PDI scores following surgery. Similarly, the percentage of time during which body temperature exceeded cutoff points (37.5°C, 38°C, 38.5°C, and 39°C) was not associated with any of the neurodevelopmental outcome scores.
View larger version (24K):
[in this window]
[in a new window]
|
Fig 3. MDI at age 1 year versus temperature at 12 hours post surgery. Total cases 244; group mean (SD) 105.1 (15.0). (MDI = Mental Developmental Index; SD = standard deviation.)
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig 4. PDI at age 1 year versus temperature at 12 hours post surgery. Total cases 244; group mean (SD) 95.3 (16.9).(PDI= Psychomotor Developmental Index; SD = standard deviation.)
|
|
Full scale IQ tests at four years of age
IQ scores at four years of age were only available from patients enrolled in the circulatory arrest trial [1, 5], all of whom had D-TGA. Of one hundred and fifty eight children in whom perioperative temperature data were available, 156 (99%) were evaluated at age four years. We explored the relationship to four-year FSIQ, VIQ, and PIQ in this population. There was no evidence of any relationship or correlation between the speed of rewarming in the operating room or the temperature profile while on the ICU and FSIQ, VIQ, or P IQ (Figs 5–7)).
Additionally, there was no association between any of these indices and percentages of time above specific temperature cutoff points (37.5°C, 38°C, 38.5°C, and 39°C).
View larger version (18K):
[in this window]
[in a new window]
|
Fig 5. FSIQ at age 4 years versus temperature at 12 hours post surgery. Total cases 156; group mean (SD) 92.0 (16.0) (FSIQ = Full-Scale Intelligence Quotient; SD = standard deviation.)
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig 6. VIQ at age 4 years versus temperature at 12 hours post surgery. Total cases 156; group mean (SD) 95.1 (15.0). (SD = standard deviation; VIQ = Verbal Intelligence Quotient.)
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig 7. PIQ at age 4 years versus temperature at 12 hours post surgery. Total cases 156; group mean (SD) 91.6 (14.5). (PIQ = Performance Intelligence Quotient; SD = standard deviation.)
|
|
|
Comment
|
|---|
As the mortality for cardiac surgery in infants has declined, interest in neurologic morbidity following surgeryhas heightened. Most studies have explored the influence of intraoperative management strategies on neurologic outcome, but few have assessed the influence of events occurring in the postoperative period. Among factors in the postoperative period that could influence subsequent neurodevelopmental outcome, hypothermia has been hypothesized to be neuroprotective, and hyperthermia to be detrimental. Our study examines the influence of postoperative temperature management on severity of brain injury and subsequent neurologic outcome following infant cardiac surgery. We found that, in infants undergoing cardiac repair using hypothermic cardiopulmonary bypass techniques, the postoperative temperature variation observed in our study did not have an important effect on neurodevelopmental outcomes at ages one or four years. However, the study is limited by the unit policy of intervening to avoid wide variations from normothermia.
A body of evidence suggests that a period of hyperthermia may be detrimental following brain injury, whether this is hypoxic, ischemic, or traumatic in origin. Furthermore, mild hyperthermia is common following cardiopulmonary bypass. Evidence from animal models of brain injury also indicates that postinjury hyperthermia is detrimental to neurodevelopmental outcome [13, 14] even when the onset of the hyperthermia is delayed by 24 hours [14]. Conversely, a number of animal studies suggest that induction of moderate hypothermia following injury may be neuroprotective [15–18]. The degree of neuroprotection from hypothermia in animal models appears to be related to the degree of hypothermia, the duration of hypothermia, and the time of onset following the primary insult to the brain [15–18].
In a study by Shum-Tim and colleagues [13] examining the importance of postoperative body temperature, three groups of piglets underwent 100 minutes of deep hypothermic circulatory arrest. Postoperatively, they were assigned to different brain temperature targets for a 24-hour period (40°C, 37°C, and 34°C). Neurologic status was assessed by means of a neurologic deficit score, an overall performance rating, and histologic damage. Neurologic injury was significantly increased in a group assigned to a postoperative brain temperature of 40°C, compared to a group maintained at a brain temperature of 34°C.
Human studies have also supported an effect of temperature on risk of brain injury. Preliminary evidence suggests that induced hypothermia may result in improved neurologic outcome in adults following out-of-hospital cardiac arrest [19, 20]. Epidemiological data have shown that a maternal fever in excess of 39°C during labor is associated with an increased incidence of spastic cerebral palsy [21]. Hyperthermia is an independent risk factor for poor outcome following traumatic brain injury in children [22], and in adults after cardiac arrest and stroke [23, 24]. Additionally, the incidence of pyrexia following brain injury in adults is common and frequently untreated [25].
The current study should be interpreted in light of certain limitations. Because temperature recordings were obtained from the hospital records retrospectively, the timing of temperature recording and the method of measurement of temperature could not be standardized. Animal data suggest brain temperature can be up to 2°C above that of the rectal temperature and 1°C above esophageal temperature in piglets [13]. The brain temperature of infants can only be estimated and factors such as uneven cooling of the brain during surgery may be important. Failure to detect an effect of postoperative temperature on later neurodevelopmental outcome may be explained, in part, by the small variation in temperature in infants cared for in the intensive care unit and the avoidance of prolonged periods of DHCA, at the study institution. Although some variations in temperature were recorded, heating and cooling devices were used whenever excessively high or low values were observed. This resulted in episodes of hypothermia or hyperthermia being of short duration. Furthermore, we might have expected to see the greatest effects of perioperative temperature regulation on later developmental outcome in patients with more than 40 minutes of total circulatory arrest. Because of the small numbers of patients in our study, with the combination of long circulatory arrest times and high or low temperatures, we had limited statistical power when analyzing temperature effects in this subgroup. We were unable to adjust for all factors that might have affected neurodevelopmental outcome, such as an interaction between low cardiac output and temperature regulation, hematocrit, and preoperative neurologic status, due to the retrospective nature of the study. There are also limitations in the measurement of neurocognitive functioning in infants and children compared with an adult population. The Bayley developmental assessments show good concurrent validity but poor predictive validity. Four-year developmental outcomes were available only from children with D-TGA enrolled in the circulatory arrest trial.
Identification of risk factors for brain injury in infants undergoing open heart surgery is fraught with difficulty as genetic, antenatal, preoperative, perioperative, and postoperative factors may all contribute to neurodevelopmental outcome. We designed this study to minimize variability as much as possible. As all the data were collected from a single center, the same management protocols with the exception of the bypass strategy were applied in each case. In addition, as all patients had been enrolled in one of two randomized clinical trials, prospectively collected data on a broad range of variables were readily available. However, the infants with four-year follow-up data were predominantly white children with a diagnosis of D-TGA which may limit generalizing the results to a population with a wider racial mix and with mixed diagnoses. Patients with D-TGA are particularly optimal for study of neurodevelopmental outcomes because of their low incidence of coexisting anomalies and infrequent hemodynamic instability following surgery [4]. Despite studying a relatively homogeneous group we found large variation in developmental scores, with values ranging from 50 to 140. Approximately 20% of the infants tested at one year had IQ scores below 80, representing two standard deviations below the mean of a normal population. Thus, there is a need to identify methods to improve neurodevelopmental outcome in these infants.
In summary, our data do not provide evidence to support the hypothesis that minor or brief fluctuations around normothermia in the postoperative period are a factor in determining secondary brain injury in infants undergoing complex cardiac surgery. We continue to exercise careful control of temperature postoperatively, especially in the management of certain postoperative arrhythmias (junctional ectopic tachycardia), and in patients known to have suffered a gross cerebral ischemic insult, such as a postoperative cardiac arrest. We suggest that a randomized-controlled trial incorporating more diverse temperature targets than the current study, as well as a population at greater risk, would be needed to investigate a possible neuroprotective effect of mild or moderate hypothermia.
|
Acknowledgments
|
|---|
This research was supported by the U.S. National Institutes of Health (Grants RO1 HL41786 and RR 02172). Serena Cottrell received a traveling scholarship from the Pediatric Intensive Care Society, UK.
|
References
|
|---|
- Newburger J.W., Jonas R.A., Wernovsky G., et al. A comparison of the perioperative neurological effects of hypothermic circulatory arrest versus low-flow cardiopulmonary bypass in infant heart surgery. N Engl J Med 1993;329:1057-1064.[Abstract/Free Full Text]
- Du Plessis A.J., Jonas R.A., Wypij D., et al. Perioperative effects of alpha-stat versus pH-stat strategies for deep hypothermic cardiopulmonary bypass in infants. J Thorac Cardiovasc Surg 1997;114:991-1001.[Abstract/Free Full Text]
- Jonas R. Optimal pH strategy for cardiopulmonary bypass in neonates, infants and children. Perfusion 1998;13:377-387.[Free Full Text]
- Bellinger DC, Jonas RA, Rappaport LA, et al. Developmental and neurological status of children after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. N Engl J Med 1995;332 No 9:549–555
- Bellinger D.C., Wypij D., Kuban K.C.K., et al. Developmental and neurological status of children at 4 years of age after heart surgery with hypothermic circulatory arrest or low flow cardiopulmonary bypass. Circulation 1999;100:526-532.[Abstract/Free Full Text]
- Hickey P.R. Neurological sequelae associated with deep hypothermic circulatory arrest. Ann Thorac Surg 1998;65:S65-70.[Abstract/Free Full Text]
- Bellinger D.C., Wypij D., du Plessis A.J., et al. Developmental and neurologic effects of alpha-stat versus pH-stat strategies for deep hypothermic cardiopulmonary bypass in infants. J Thorac Cardiovasc Surg 2001;121:374-383.
- Wernovsky G., Wypij D., Jonas R.A., et al. Postoperative course and haemodynamic profile after the arterial switch operation in neonates and infants: a comparison of low flow cardiopulmonary bypass versus circulatory arrest. Circulation 1995;92:2226-2235.[Abstract/Free Full Text]
- Newburger J.W., Wypij D. Methods and procedures. In: Jonas R.A., Newburger J.W., Volpe J.J., eds. Brain Injury and Pediatric Cardiac Surgery. Boston: Butterworth-Heinemann, 1996:289-310.
- Bayley N. Manual for the Bayley Scales of infant development. . New York: The Psychological Corporation, 1969.
- Bayley N. Manual for the Bayley Scales of infant development, 2nd Ed San Antonio, TX: The Psychological Corporation, 1993.
- Wechsler D. Wechsler Preschool and Primary Scale of Intelligence-Revised. . San Antonio, TX: The Psychological Corporation, 1989.
- Shum-Tim D., Nagashima M., Shinoka T., et al. Postischemic hyperthermia exacerbates neurological injury after deep hypothermic circulatory arrest. J Thorac Cardiovasc Surg 1998;116:780-792.[Abstract/Free Full Text]
- Baena R.C., Busto R., Dietrich W.D., Globus M.Y.T., Ginsberg M.D. Hyperthermia delayed by 24 hours aggravates neuronal damage in rat hippocampus following global ischaemia. Neurology 1997;48:768-773.[Abstract/Free Full Text]
- Chopp M., Chen H., Dereski M.O., Garcia J.H. Mild hypothermic intervention after graded ischemic stress in rats. Stroke 1991;22:37-43.[Abstract/Free Full Text]
- Carroll M., Beek O. Protection against hippocampal CA1 cell loss by post-ischemic hypothermia is dependent on delay of initiation and duration. Metab Brain Dis 1992;7:45-50.[Medline]
- Coimbra C., Wieloch T. Moderate hypothermia mitigates neuronal damage in the rat brain when initiated several hours following transient cerebral ischemia. Acta Neuropathol 1994;87:325-331.[Medline]
- Thoresen M., Bagenholm R., Loberg E.M., Apricena F., Kjellmer I. Posthypoxic cooling of neonatal rats provides protection against brain injury. Arch Dis Child 1996;74:F3-F9.
- Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346 No 8:557–63
- The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurological outcome after cardiac arrest. N Engl J Med 2002;346 no(8):549–56
- Grether J.K., Nelson K.B. Maternal infection and cerebral palsy in infants of normal birth weight. JAMA 1997;278:207-212.[Abstract/Free Full Text]
- Natale JE, Joseph JG, Helfaer MA, Shaffner DH. Early hyperthermia after traumatic brain injury in children; risk factors, influence on length of stay, and effect on short term neurological status. Crit Care Med 2000;28 no 7:2608–2615
- Takino M., Okada Y. Hyperthermia following cardiopulmonary resuscitation. Intensive Care Med 1991;17:419-442.[Medline]
- Reith J., Jorgensen H.S., Pedersen P.M., et al. Body temperature in acute stroke. relation to stroke severity, infarct size, mortality and outcome. Lancet 1996;347:422-425.[Medline]
- Albrecht R.F., Wass C.T., Lanier W.L. Occurrence of potentially detrimental temperature alterations in hospitalised patients at risk for brain injury. Mayo Clin Proc 1998;73:629-635.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
K. Watanabe, M. Matsui, J. Matsuzawa, C. Tanaka, K. Noguchi, N. Yoshimura, K. Hongo, M. Ishiguro, S. Wanatabe, K. Hirono, et al.
Impaired neuroanatomic development in infants with congenital heart disease.
J. Thorac. Cardiovasc. Surg.,
January 1, 2009;
137(1):
146 - 153.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Wernovsky
Improving neurologic and quality-of-life outcomes in children with congenital heart disease: Past, present, and future
J. Thorac. Cardiovasc. Surg.,
February 1, 2008;
135(2):
240 - 242.
[Full Text]
[PDF]
|
|
|
|
|
|
|
Intracardiac temperature monitoring in infants after cardiac surgery.
J. Thorac. Cardiovasc. Surg.,
March 1, 2006;
131(3):
614 - 620.
|
|
|
|
|
|
|
F. L. Hanley
Religion, politics...deep hypothermic circulatory arrest
J. Thorac. Cardiovasc. Surg.,
November 1, 2005;
130(5):
1236 - 1236.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
