Evidence of cortisol suppression in neonates after major cardiac surgery: is supplementation necessary?

Introduction:  Infants and children undergoing major cardiac surgery mount a substantial stress response. Data from a previous investigation suggested that after cardiac surgery infants under 90 days had prolonged suppression of cortisol production (1) that might not be explained as simply due to the use of dexamethasone (used routinely for cardiopulmonary bypass). This study was designed to determine the origins of this suppressed cortisol response. Methods:  Local ethics committee approval was granted for the study. Ten neonates due to undergo cardiac surgery using cardiopulmonary bypass were recruited after informed parental consent. A standard anaesthetic technique was used. Baseline blood samples were taken following induction of anaesthesia and arterial line insertion: a dose of dexamethasone (0.5 mg·kg-1) was then administered as per usual protocol. Blood was taken at release of the cross clamp (ROCC) and then 2, 6, 24, 48 and 72 h later. Blood was not taken at 72 h if the patient no longer had invasive monitoring lines in place. At each time point concentrations of cortisol, adrenocorticotrophic hormone (ACTH) and dexamethasone were measured. Table 1. Results:  Time ACTH (ng·l-1) Cortisol (nmol·l-1) Dexamethasone (nmol·l-1) Baseline 68.7 (38.6–123.0) 647.0 (510.0–1038) 0.8 (0.8–1.0) ROCC 18.3 (11.3–28.7) 439.0 (340.0–556.0) 173.5 (153.3–186) ROCC + 2 11.2 (9.0–14.9) 201.0 (134.0–416.0) 147.8 (114.6–176) ROCC + 6 9.0 (9.0–10.4) 87.5 (49.0–94.0) 60.0 (50.6–103.3) ROCC + 24 9.0 (9.0–13.8) 98.0 (29.0–244.0) 5.1 (2.4–12.6) ROCC + 48 12.6 (9.0–16.4) 302.0 (173.0–451.0) 1.3 (1–3.9) ROCC + 72 12.4 (9.0–26.3) 494.0 (176.0–686.0) 1.7 (1–2.3) Note: Figures above are given as median (interquartile range) for the 10 patients in the study. Statistical analysis was performed after the data had been natural-log transformed. Plasma cortisol concentrations reached a nadir 6 h post-ROCC. At 24 h post-ROCC cortisol was significantly below baseline (P < 0.05), and was undetectable in four patients (<30 nmol·l-1). 48 h post-ROCC; eight out of 10 patients still had a cortisol level below baseline. Plasma ACTH concentrations were also suppressed up to 48 h post-ROCC (P < 0.05). However, plasma dexamethasone concentrations had returned to background (baseline) values by 48 h post-ROCC, and therefore were unlikely to still exert pharmacological effects. Furthermore, there was no significant correlation between plasma dexamethasone and cortisol concentrations at either 24 or 48 h post-ROCC. Discussion:  The data suggest that between 24 and 48 h post-ROCC the patients could be at risk of relative adrenocortical insufficiency as dexamethasone had been substantially cleared from the plasma and cortisol continued to remain below baseline in the majority of patients. The independence of dexamethasone and cortisol concentrations between 24 and 48 h is interesting and would indicate that other regulators are responsible for the prolonged pituitary adrenocortical suppression. Conclusions:  Neonates may be at risk of prolonged reduced cortisol secretion after major cardiac surgery via suppression of pituitary ACTH release. The results cannot be dismissed as simple feedback inhibition by prolonged elimination of dexamethasone. Neonates may benefit from supplemental corticosteroids after major cardiac surgery by measurement of plasma cortisol concentrations postoperatively and providing steroid supplementation if required. Acknowledgements:  We are grateful to the APA for funding this study, to the biochemistry laboratory at Guy's and St Thomas’ Trust for performing the assays, and to the parents of the patients for allowing them to be included in this study. Reference  1 Humphreys N et al.‘Plasma cortisol concentrations in children in PICU after cardiac surgery: evidence of prolonged suppression in infants under 90 days old’. Abstract presented at PICS ASM, September 2003.