Effect of remifentanil infusion rate on stress response to the pre-bypass phase of paediatric cardiac surgery

N. K. Weale1, C. A. Rogers2, R. Cooper3, J. Nolan1 and A. R. Wolf*,1

1 University Department of Anaesthesia and 2 Bristol Heart Institute, Bristol Royal Infirmary, Upper Maudlin Street, Bristol BS2 8HW, UK. 3 Department of Anaesthesia, Gloucestershire Royal Hospital, Great Western Road, Gloucester GL1 3NN, UK

*Corresponding author. E-mail: awolfbch@aol.com

Accepted for publication: September 9, 2003


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 
Background. Opioids are used routinely to eliminate the stress response in the pre-bypass phase of paediatric cardiac surgery. Remifentanil is a unique opioid allowing a rapidly titratable effect. No data are available regarding a suitable remifentanil dose regimen for obtunding stress and cardiovascular responses to such surgery.

Methods. We recruited 49 infants and children under 5 yr old who were randomized to receive one of four remifentanil infusion rates (0.25, 1.0, 2.5, or 5.0 µg kg–1 min–1). Blood samples were obtained at induction, pre-surgery, 5 min after opening the chest, and immediately pre-bypass. Whole blood glucose was measured at all time points while cortisol and neuropeptide Y (NPY) were measured in the first and last samples. Heart rate and arterial pressure were also recorded.

Results. There was a significant increase in whole blood glucose 5 min after opening the chest and pre-bypass (P=0.009, P=0.002) in patients receiving remifentanil 0.25 µg kg–1 min–1, but not in those receiving higher doses. Increased remifentanil dosage was associated with reduced plasma cortisol during surgery (P<0.001). Baseline NPY showed considerable variation and there was no association between pre-bypass NPY and remifentanil dose. There was a significantly higher heart rate at the pre-bypass stage of surgery in the remifentanil 0.25 µg kg–1 min–1 group compared with higher doses (P=0.0006). Four out of five neonates with complex cardiac conditions showed severe bradycardia associated with remifentanil.

Conclusions. In infants and children under 5 yr, remifentanil infusions of 1.0 µg kg–1 min–1 and greater can suppress the glucose increase and tachycardia associated with the pre-bypass phase of cardiac surgery, while 0.25 µg kg–1 min–1 does not. Remifentanil should be used with caution in neonates with complex congenital heart disease.

Br J Anaesth 2004; 92: 187–94

Keywords: anaesthesia, paediatric; analgesics opioid, remifentanil; complications, congenital heart disease; protein, neuropeptide Y; stress, response


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 
High dose opioids have been advocated for elimination of the stress response during paediatric cardiac surgery.1 Significant intraoperative stress responses,2 postoperative complications, and increased mortality1 3 have been reported in human infants receiving inadequate analgesia. In piglets, raised concentrations of circulating catecholamines have been associated with post-ischaemic myocardial dysfunction.4 However, high dose opioids cause significant side-effects such as ventilatory depression5 and immunomodulation,6 7 and there have been few attempts to determine the relationship of opioid dose to stress reduction in this age group. We have shown previously that doses of fentanyl lower than that reported previously can obtund the stress responses to the pre-bypass phase of paediatric cardiac surgery with beneficial effects on haemodynamic stability.8

Remifentanil is a synthetic opioid, 40 times more potent than alfentanil,9 which undergoes rapid metabolism by plasma esterases to virtually inactive metabolites. It is a drug which is of particular interest in paediatric anaesthesia; its unique pharmacological properties could allow large doses of remifentanil to be used intra-operatively to obtund stress responses while minimizing postoperative side-effects and time to recovery. However, there are few data on dose requirements for remifentanil in infants and these are based on previous adult data.10 11 We are not aware of any published studies evaluating the relationship between dose and stress responses in children. We therefore undertook a prospective, blinded, randomized trial to determine a suitable remifentanil infusion rate that could effectively obtund some of the key markers of the stress response to the pre-bypass phase of cardiac surgery in infants and children.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 
After local ethical approval (United Bristol Health Care Ethical committee) and informed, written parental consent, we studied 51 infants and children under the age of 5 yr. Patients were recruited between July 1999 and May 2001. All children were undergoing elective cardiac surgery requiring cardiopulmonary bypass.

Diazepam pre-medication (0.5 mg kg–1) was administered to patients over 4 months old. Anaesthesia was induced with sevoflurane, air, and oxygen. A baseline venous blood sample (3 ml) was taken immediately after insertion of i.v. access for measurement of whole blood glucose, plasma neuropeptide Y (NPY), and cortisol. A remifentanil infusion was commenced at one of four infusion rates (0.25, 1.0, 2.5, or 5.0 µg kg–1 min–1) assigned randomly from a closed envelope technique. Infusions had been prepared previously by an unblinded anaesthetist who took no part in the subsequent patient management. Intubation was facilitated with pancuronium 0.1 mg kg–1 and patients were ventilated with air, oxygen, and isoflurane 0–0.2%. Arterial, central venous, and urinary catheters were placed before starting surgery. Arterial pressure and heart rate were recorded at 5 min intervals. Dextrose infusions were not used pre- or intra-operatively.

Further blood samples were taken immediately before surgery, 5 min after opening the chest and immediately before cardiopulmonary bypass. One of two senior cardiac surgeons performed all of the surgical procedures in this study and we did not recruit any patients undergoing repeat surgery. All samples were analysed for whole blood glucose and the pre-bypass sample also underwent analysis for plasma cortisol and NPY concentrations.

Whole blood glucose was measured using the glucose oxidase method (One Touch®, coefficient of variation <4% in measured range). Baseline and pre-bypass plasma cortisol was measured from residual plasma in 27 patients after NPY estimation. Measurement was by chemiluminescence (ACS: 180) with a sensitivity of 5.5 nmol litre–1 and inter-assay variability of 6.4% (range 4.5–7.6%).

Blood samples for NPY were immediately stored on ice and samples spun in a refrigerated centrifuge to provide plasma samples. These were stored at –40°C until analysis. Peptides were extracted from the plasma using octadecylsilane (C18) cartridges (SEP-PAK) and eluted from the cartridges in acetonitrite 60%. Acetonitrite was removed by evaporation using an Aquavac (Uniscience, Cambridge) and dried peptides were stored at –80°C until assayed. NPY was assayed using an immunoassay kit (Peninsular laboratories). Intra-assay variation was less than 5% and inter-assay variation less than 14%.

After institution of cardiopulmonary bypass, the remifentanil infusion was stopped and the subsequent management was according to standardized guidelines used in the Bristol Children’s Hospital Cardiac Unit. This includes adding morphine 0.5 mg kg–1 to the bypass pump.

Statistical analysis
A power calculation was performed before submission of the study for ethical approval using previous data on stress responses to cardiac surgery in children.8 12 We determined that a group size of 10 would be sufficient for an 80% power to detect a difference at the 5% significance level with an effect size of 0.6. However, because of the cardiovascular instability of the neonates with remifentanil, we extended the study to a total of 51 patients using additional randomized allocation to ensure that all groups had at least 10 patients over 1 month of age. However, the statistical design of this study required that all patients were subsequently included in the analysis on an intention to treat basis. The statistical software package used for this analysis was SAS version 8, SAS Institute Inc., Campus Drive, Cary, NC 27513. All analyses were carried out independently by a clinical trials unit statistician (C.A.R.), who was aware only of the trial design and the raw data.

Cortisol and NPY analysis (two time points)
Analysis of covariance was used to compare responses to the four doses of remifentanil. Both variables were transformed to the log scale before analysis, as the assumption of constant variance was not tenable for the measurements on the original scale. For the NPY response, in addition to the adjustment for baseline response, an adjustment for age was also investigated. A series of models were fitted and compared using F statistics and assessed graphically and using Bartlett’s,13 and Cook and Weisberg tests.14 The simplest model that provided an adequate fit for the data was selected.

Glucose, heart rate, and arterial pressure (four time points)
Mixed model analysis was used to examine the effect of remifentanil dose on these variables. Mixed models were chosen to allow the correlation between measurements from the same child to be modelled explicitly. A variety of correlation structures were examined and that giving the highest value for Akaike’s information criterion15 was selected in each case. Interactions between dose and stage of the operation were examined and where significant at the 5% level, the mean response was estimated separately for each stage. Model assumptions were assessed graphically.

Glucose measurements were transformed to the log scale prior to analysis, as the assumption of constant variance and normally distributed residuals was not tenable for measurements on the original scale.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 
Results are reported for 49 infants and children. Two patients were excluded from the analysis; one child aged 8 yr old who was recruited in error and one patient for whom the remifentanil infusion rate was not clearly documented.

All groups were similar in age and type of surgical procedure (Table 1). Baseline data are given in Table 2 and all raw data are included in Appendices I–III.


View this table:
[in this window]
[in a new window]
 
Table 1 Physical characteristics of the four patient groups
 

View this table:
[in this window]
[in a new window]
 
Table 2 Baseline data at induction. NY=neuropeptide
 
Glucose
The natural logarithm of glucose (ln (glucose)) varied significantly with both remifentanil dose (F3,44.5=5.77, P=0.002) and stage of the operation (F2.71.1=19.31, P<0.0001). The effect of remifentanil dose on ln (glucose) also differed significantly between the stages of the operation (F6,78.9=3.75, P=0.003). A difference in response was indicated in the pre-bypass sample between 0.25 µg kg–1 min–1 and all other remifentanil doses (P=0.002 after Bonferroni correction) (Fig. 1).



View larger version (17K):
[in this window]
[in a new window]
 
Fig 1 Mean ln (glucose) (SE) vs stage of surgery for each dose of remifentanil (ln=natural logarithm). *Significantly different from other remifentanil dose at the time point (P=0.002).

 
Cortisol
Initial analysis of the transformed ln (cortisol) revealed one markedly outlying observation in the 2.5 µg kg–1 min–1 group, which was subsequently omitted from the analysis. There was a strongly significant difference between remifentanil groups and pre-bypass ln (cortisol) (F1,24=43.8, P<0.0001), with a lower mean ln (cortisol) associated with higher remifentanil doses. The estimated decrease in the mean ln (cortisol) per unit increase in remifentanil dose after back transformation to the original scale equates to a ratio of 0.77 (95% CI 0.71–0.84). The natural log transformed values of cortisol with standard errors are presented in Figure 2.



View larger version (14K):
[in this window]
[in a new window]
 
Fig 2 Mean ln (cortisol) (SE) at induction and pre-bypass vs remifentanil dose (ln=natural logarithm).

 
An additional non-parametric analysis was also performed, which included the outlying variable. The resulting analysis with this outlier still indicated that cortisol responses were significantly reduced with increasing remifentanil doses (trend test, P<0.0001).

NPY
There was no significant statistical relationship between remifentanil dose and pre-bypass NPY (F1,42=0.14, P=0.17). There was evidence to suggest that both the baseline ln (NPY) and age were individually associated with the pre-bypass ln (NPY) (P<0.001 and P=0.005, respectively). Younger ages and high baseline ln (NPY) were predictive of a high pre-bypass ln (NPY). However, when considered together, the strongest association was baseline ln (NPY). The estimated difference in mean ln (NPY) is shown in Table 3.


View this table:
[in this window]
[in a new window]
 
Table 3 Estimated difference in mean ln (NPY) and standard error with each remifentanil infusion rate
 
Heart rate and systolic arterial pressure
Heart rate varied significantly with both remifentanil dose (F3,43.6=4.55, P=0.007) and stage of operation (F2,87.1=24.48, P<0.0001). The effect of remifentanil on heart rate also differed significantly between the stages of the operation (F6,87.1=3.22, P=0.007) (Fig. 3). Further pair-wise analysis at each sampling stage revealed a highly significant difference at the pre-bypass stage when comparing 0.25 µg kg–1 min–1 with all higher remifentanil doses. Arterial pressure varied significantly with the stage of the operation (F2,43=58.04, P<0.0001), but no relationship was found with remifentanil dose.



View larger version (18K):
[in this window]
[in a new window]
 
Fig 3 Mean heart rate (SE) vs stage of surgery for each remifentanil dose.

 
Complications
Of the 49 patients in the study, nine exhibited significant bradycardia or hypotension requiring intervention. Four of these were neonates with complex cardiac anatomy requiring urgent surgery. Three patients required fluid boluses alone, a further two received calcium gluconate and one was commenced on a dopamine infusion. Three of the five neonates with transposition of the great arteries were given epinephrine before bypass: two were receiving remifentanil 1 µg kg–1 min–1 and one 2.5 µg kg–1 min–1.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 
Median sternotomy and cannulation of the great vessels provide a useful and consistent model for studying stress responses to surgery in children, in terms of intensity and timing of the stimulus. The procedure is standardized and routine in children having their first operation and is a major procedure in terms of stimulation of both somatic and visceral afferents. Using this model we have previously been able to show that a pre-surgical dose of fentanyl 2 µg kg–1 could not suppress the increase in glucose and cortisol associated with this stimulus, while fentanyl doses of 25 µg kg–1 and above could. The current study, using the same study design, indicates that intraoperative remifentanil at infusion rates of 1 µg kg–1 min–1 and above can prevent glucose and heart rate responses to surgery but that 0.25 µg kg–1 min–1 is insufficient. Higher infusion rates of remifentanil are associated with progressively lower cortisol concentrations immediately pre-bypass. This is in contrast to the previous study with fentanyl, which showed that at 2 µg kg–1 there was poor suppression but that at 25 µg kg–1 and above there was significant suppression.8

Most hormonal and metabolic markers of stress require lengthy and sophisticated analysis in the laboratory. In contrast, accurate measurement of whole blood glucose can be achieved at the bedside or in the operating theatre at minimal cost. Blood glucose is usually tightly controlled in the unstressed child, and the increase associated with surgery can provide a useful ‘real time’ summary of stress response with a variety of analgesic techniques.16 The ease of blood glucose measurement makes it feasible to monitor this aspect of the stress response and to act on the results. Group analysis of our results indicate that most patients receiving remifentanil 1 µg kg–1 min–1 or more do not have an increase in blood glucose in the pre-bypass phase of surgery. However, four patients in the 1 and 2.5 µg kg–1 min–1 groups did show substantial responses (Appendix I). It is interesting to speculate whether increasing the remifentanil infusion in response to the increase in blood glucose could reverse the responses at an early stage. If so, this kind of approach could allow individual modulation of opioid dosage to balance the intensity of the surgical stimuli with its consequences in terms of heart rate and blood glucose response.

Remifentanil has become popular for infant and neonatal anaesthesia as it can provide potent analgesia intra-operatively whilst allowing rapid recovery and extubation.17 Many of the pharmacokinetic data for remifentanil are extrapolated from adult studies,10 11 although more specific paediatric data are emerging.18 Suggested dosing regimens are variable and may include an initial bolus of remifentanil at induction.19 We have shown that remifentanil 0.25 µg kg–1 min–1 is insufficient to eliminate the glucose response to the pre-bypass phase of cardiac surgery but the higher rate of 1 µg kg–1 min–1 appears sufficient for most individuals. The minimum time from induction of anaesthesia and commencement of the remifentanil infusion to the start of surgery in our study was consistently longer than 30 min. This allowed a steady state plasma concentration to be established, and we therefore chose not to use an initial bolus dose.

Remifentanil is known to cause cardiovascular depression at higher infusion rates and particularly following bolus administration.20 The cardiovascular depression observed with remifentanil was relatively minor in all treatment groups with the exception of the neonatal patients undergoing correction of either truncus arteriosus or transposition of the great arteries. We therefore recommend caution in the use of remifentanil, especially in higher doses, in this sub-group of patients.

NPY is a 36 amino acid peptide, which is co-stored with norepinephrine in sympathetic nerve endings. It is released in response to prolonged or intense sympathetic stimulation,21 and causes systemic arterial vasoconstriction.22 23 In healthy children there is a significant decrease in NPY with age,24 and elevated levels of NPY have been demonstrated after aortic occlusion in children undergoing cardiopulmonary bypass.25 Unlike catecholamines, which have a half-life of only a few minutes at normal blood temperature and are difficult to measure accurately, NPY has a plasma half life of 20–30 min and is relatively easy to detect.22 We measured NPY in preference to catecholamines because of its ease of measurement and its potential use as a better summary marker of sympathetic activation compared with catecholamines, which show large minute by minute variation. However, our results are disappointing in that we failed to show any significant difference in this variable between the remifentanil infusion rates or stages of the operation. There may be several possible explanations. Our study was limited to the pre-bypass phase of surgery and this stimulus may be of insufficient intensity or duration to observe a reproducible increase in NPY. We also observed a 5–6-fold variation in baseline NPY (Appendix III), and while there was an effect of age this did not appear to fully explain the wide variation before surgery. Infants with cardiac failure have raised levels of other potent vasoconstrictors such as angiotensin II and vasopressin and the large variability in NPY may reflect a similar trend. However, NPY is ubiquitous and is found outside the nervous system where its function remains largely unknown.26 Further investigation is needed to understand the underlying sources and causes of raised NPY in the infant with congenital cardiac disease.

In conclusion, we suggest that a remifentanil infusion rate of 1 µg kg–1 min–1 is a suitable starting infusion rate for paediatric cardiac surgery. Further titration of the dose may proceed using whole blood glucose measurements and heart rate changes. This approach could allow ‘real time’ alteration of opiate effect and avoidance of unwanted cardiovascular depression secondary to opiate overdose.


    Appendices
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 


View this table:
[in this window]
[in a new window]
 
Appendix I Whole blood glucose concentrations, absolute values (mmol litre–1)
 

View this table:
[in this window]
[in a new window]
 
Appendix II Serum cortisol, absolute values (nmol litre–1)
 

View this table:
[in this window]
[in a new window]
 
Appendix III NPY, absolute values (pg ml–1)
 

    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Appendices
 References
 
1 Anand KJ, Hickey PR. Halothane-morphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992; 326: 1–9[Abstract]

2 Hickey PR, Hansen DD. High-dose fentanyl reduces intraoperative ventricular fibrillation in neonates with hypoplastic left heart syndrome. J Clin Anesth 1991; 3: 295–300[Medline]

3 Anand KJ, Hansen DD, Hickey PR. Hormonal-metabolic stress responses in neonates undergoing cardiac surgery. Anesthesiology 1990; 73: 661–70[ISI][Medline]

4 Caspi J, Coles JG, Benson LJ, et al. Effects of high plasma epinephrine and Ca2+ concentrations on neonatal myocardial function after ischemia. J Thorac Cardiovasc Surg 1993; 105: 59–67[Abstract]

5 Purcell-Jones G, Dormon F, Sumner E. The use of opioids in neonates. A retrospective study of 933 cases. Anaesthesia 1987; 42: 1316–20[ISI][Medline]

6 Webster NR. Opioids and the immune system. Br J Anaesth 1998; 81: 835–6[Free Full Text]

7 Shavit Y, Lewis JW, Terman GW, Gale RP, Liebeskind JC. Opioid peptides mediate the suppressive effect of stress on natural killer cell cytotoxicity. Science 1984; 223: 188–90[ISI][Medline]

8 Duncan HP, Cloote A, Weir PM, et al. Reducing stress responses in the pre-bypass phase of open heart surgery in infants and young children: a comparison of different fentanyl doses. Br J Anaesth 2000; 84: 556–64[Abstract]

9 Glass PS, Iselin-Chaves IA, Goodman D, Delong E, Hermann DJ. Determination of the potency of remifentanil compared with alfentanil using ventilatory depression as the measure of opioid effect. Anesthesiology 1999; 90: 1556–63[ISI][Medline]

10 Egan TD, Huizinga B, Gupta SK, et al. Remifentanil pharmacokinetics in obese versus lean patients. Anesthesiology 1998; 89: 562–73[ISI][Medline]

11 Egan TD, Minto CF, Hermann DJ, Barr J, Muir KT, Shafer SL. Remifentanil versus alfentanil: comparative pharmacokinetics and pharmacodynamics in healthy adult male volunteers. Anesthesiology 1996; 84: 821–33[CrossRef][ISI][Medline]

12 Sun LS, Du F, Schechter WS, Quaegebeur JM, Vulliemoz Y. Plasma neuropeptide Y and catecholamines in pediatric patients undergoing cardiac operations. J Thorac Cardiovasc Surg 1997; 113: 278–84[Abstract/Free Full Text]

13 Bartlett MS. Properties of sufficiency and statistical tests. Proc Roy Soc 1937; 160 (Series A): 268–82

14 Cook RD, Weisberg S. Diagnostics for heteroscedacity in regression. Biometrika 1983; 70: 1–10[ISI]

15 Akaike H. A new look at the statistical model identification. IEEE Transact Autonomic Control 1974; AC-19: 716–23

16 Wolf AR, Doyle E, Thomas E. Modifying infant stress responses to major surgery: spinal vs extradural vs opioid analgesia. Paediatr Anaesth 1998; 8: 305–11[CrossRef][ISI][Medline]

17 Davis PJ, Galinkin J, McGowan FX, et al. A randomized multicenter study of remifentanil compared with halothane in neonates and infants undergoing pyloromyotomy. I. Emergence and recovery profiles. Anesth Analg 2001; 93: 1380–6[Abstract/Free Full Text]

18 Ross AK, Davis PJ, Dear Gd GL, et al. Pharmacokinetics of remifentanil in anesthetized pediatric patients undergoing elective surgery or diagnostic procedures. Anesth Analg 2001; 93: 1393–401[Abstract/Free Full Text]

19 Twersky RS, Jamerson B, Warner DS, Fleisher LA, Hogue S. Hemodynamics and emergence profile of remifentanil versus fentanyl prospectively compared in a large population of surgical patients. J Clin Anesth 2001; 13: 407–16[CrossRef][ISI][Medline]

20 Elliott P, O’Hare R, Bill KM, Phillips AS, Gibson FM, Mirakhur RK. Severe cardiovascular depression with remifentanil. Anesth Analg 2000; 91: 58–61[Abstract/Free Full Text]

21 Pernow J, Lundberg JM, Kaijser L. Alpha-adrenoceptor influence on plasma levels of neuropeptide Y-like immunoreactivity and catecholamines during rest and sympathoadrenal activation in humans. J Cardiovasc Pharmacol 1988; 12: 593–9[ISI][Medline]

22 Pernow J, Lundberg JM, Kaijser L. Vasoconstrictor effects in vivo and plasma disappearance rate of neuropeptide Y in man. Life Sci 1987; 40: 47–54[CrossRef][Medline]

23 Abrahamsson C. Neuropeptide Y-1 and Y-2 receptor mediated cardiovascular effects in the anaesthetised guinea pig, rat and rabbit. J Cardiovascular Physiol 2000; 36: 451–58

24 Kogner P, Bjork O, Theodorsson E. Plasma neuropeptide Y in healthy children: influence of age, anaesthesia and the establishment of an age-adjusted reference interval. Acta Paediatr 1994; 83: 423–7[ISI][Medline]

25 Sun LS, Du F, Schechter WS, Quaegebeur JM, Vulliemoz Y. Plasma neuropeptide Y and catecholamines in pediatric patients undergoing cardiac operations. J Thorac Cardiovasc Surg 1997; 113: 278–84[Abstract/Free Full Text]

26 Balasubramaniam A. Clinical potentials of neuropeptide Y family of hormones. Am J Surg 2002; 183: 430–4[CrossRef][ISI][Medline]





This Article
Abstract
Full Text (PDF)
E-Letters: Submit a response to the article
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Weale, N. K.
Articles by Wolf, A. R.
PubMed
PubMed Citation
Articles by Weale, N. K.
Articles by Wolf, A. R.