• 
• 

Three Part Question

In [children undergoing cardiac surgery] is the use of [prophylactic steroids] of any benefit in terms of [attenuated inflammatory response or clinical benefit]?

Clinical Scenario

You are about to perform an arterial switch on a 3kg girl. You ask your anaesthetist to give her 3mg of dexamethasone on induction of anaesthesia. This anaesthetist has just come from another institution where this was never done and regarded as even possibly being dangerous. He asks you why you give steroids to all your children undergoing prolonged cardiopulmonary bypass. You quote an animal study that you were involved with as a registrar but you cannot recall any convincing clinical trials, so you resolve to search the literature that evening

Search Strategy

Medline 1966-March 2004 using the OVID interface
[exp steroids/ OR steroid$.mp OR prednisolone.mp OR methylprednisolone.mp OR corticosteroid$.mp OR hydrocortisone.mp OR dexamethasone.mp] AND [exp cardiac surgical procedures/ OR cardiac surgery.mp OR exp cardiovascular surgical procedures/ OR cardiac operation$.mp OR exp cardiopulmonary bypass/ OR cardiopulmonary bypass.mp OR exp Transposition of Great Vessels/ OR exp Heart septal defects, atrial/ OR heart septal defects, ventricular/ OR VSD.mp] AND [Maximally sensitive paediatric search filter] AND [Maximally sensitive RCT filter] LIMIT to human

Search Outcome

From 302 papers of which 6 were prospective, randomised controlled trials (PRCTs) in children were found comparing steroids to either differing regimes of steroid or placebo.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Checchia et al, 2003, USA	28 children undergoing open-heart surgery with CPB for congenital heart defects Dexamethasone, 1 mg/kg iv (group II, n= 15) or placebo, 1 hr before initiation of cardiopulmonary bypass.	Double blind PRCT (level 2b)	Troponin I at 24 hours	Steroid group: mean 33.4 ± 20.0 ng/mL Placebo group: mean 86.9 ± 81.1 ng/ml p = 0.04	This is an extension of the Bronicki study (Bronicki et al. 2000) with 11 overlapping patients in both papers No sample size calculations presented
			Clinical parameters	1 death in Steroid group. No wound infections, no reoperations, no sternal dehiscences and no incidents of gastrointestinal bleeding in either group
Mott et al, 2003, USA	246 children undergoing congenital heart surgery Steroid group received pre-CPB intravenous methylprednisolone (1 mg/kg) plus four additional intravenous doses over 24 h	Double blind PRCT (Level 2b)	Uncomplicated Post Pericardotomy Syndrome (PPS) defined as temperature >100.5°F, pericardial friction rub, patient irritability, small pericardial effusion	Steroid group: 21/126 (17%) had PPS Placebo group: 18/120 (15%) p=0.73	Subjective diagnosis between complicated and un-complicated PPS Did not use an objective means to measure the effectiveness of immune suppression Very heterogeneous group of operations , and bypass times not given Study was powered to detect a drop of PPS from 15% to 5% with 80% certainty 14 patients excluded due to deviations from protocol. These patients should have been included on an intention to treat basis.
			Complicated PPS defined as non-complicated PPS plus hospital readmission or pericardiocentesis	Steroid group: 8/126 (6%) Placebo group: 1/120 (0.8%) p=0.05
Varan et al, 2002, Turkey	30 children undergoing congenital heart surgery High dose: (n=15) 30mg/kg Methyl-prednisolone iv Low dose: 2mg/kg methyl-prednisolone pre onset of CPB	Unblinded PRCT (Level 2b)	IL-6 at 2 hrs	High dose: 149 ±237 pg/ml Low dose: 187 ±196 p=NS	No sample size calculations given thus it is not clear if these negative findings are adequate to conclude that there are no significant differences between these treatments
			IL-8 at 2 hrs	High dose: 351 ±375 pg/ml Low dose 245 ±303 p=NS
			CRP at 24 hrs	High dose: 38.1 ±34 mg/dl Low dose 40.2 ±36.9 p=NS
			Clinical parameters high dose vs low dose	Mech ventilation 9.0 vs 10.0 (p=0.69). ICU stay (hrs) 65 vs 62 (p=0.72). Inotropes, blood loss, urine output no difference
Bronicki et al, 2000, USA	29 children undergoing surgery for congenital heart disease Steroid group: dexa-methasone (1 mg/kg intravenously) Controls received saline solution 1 hour prior to CPB	Double blind PRCT (Level 2b)	Biochemical markers	IL-6 was 8 times higher in the placebo group. TNF-alpha and complement C3a no significant differences	Small numbers Exact numbers for biochemical markers not given Group comparison: CPB time Steroid gp 111 ± 41 Placebo gp 135 ± 37 P=0.11 Aortic cross-clamp time (min): Steroid gp 64 ± 20 Placebo gp 78 ± 25 P=0.10 Although not statistically different the steroid group did seem to have shorter bypass times
			Days of mechanical ventilation	Steroid group 3 days Placebo group 5 days p=0.02
			Alveolar-arterial oxygen gradients during the first 24 hrs	Steroid group: 144 ± 108 mm Hg Placebo group: 214 ±118 mm Hg p=0.02
			Clinical variables	Steroid group: temp 37.2° ± 0.4°C Control group temp: 37.7°± 0.4°C p=0.007. Also significant differences in fluid requirement, renal function
Lindberg et al, 2003, Sweden	40 children over 10kg undergoing congenital heart surgery Steroid group dexa-methasone (1 mg/kg) on induction or normal saline	Double blind PRCT (Level 2b)	Bio-chemical markers	No difference between groups for Von Willebrand factor or S100B (a brain inflammatory marker)	Small numbers Exact figures for biochemical results not given
			CRP on day 1	Steroid group: 29 (11) mg/litre Placebo group: 42(19) mg/litre P=0.01
			Clinical findings	No significant difference in oxygenation, body temperature, fluid balance, leucocyte and platelet counts, days in the ICU or days on mechanical ventilation
Schroeder et al, 2003, USA	29 children undergoing cardiopulmonary bypass Group I: Pre-operative and intraoperative Methyl-prednisolone (MP) (30 mg/kg 4 hours before bypass and in bypass prime, n=14) Group II: Intraoperative MP only (30 mg/kg, n=15)	Double blind PRCT (level 2b)	Biochemical markers	Combined therapy group showed lower myocardial inflammatory mediator mRNA expression, MCP-1, IL-6 and ICAM-1 levels	Age group of group II was 4.4 months +/-3.9 months vs group I 2.1 months +/-2.4 months
			ICU stay	Group I 4.4 days Group II 6.1 days, P=0.07
			Clinical outcomes	Reduced fluid requirements, lower body temperature, and lower _A_VO2 for the first 24 hours after surgery (P_0.05)

Comment(s)

Several randomized controlled trials were found examining various regimes of steroid therapy in children undergoing congenital cardiac surgery, although their patient numbers were small and all studies were from single centres. Checchia et al performed a 28 patient RCT with Troponin I as the primary outcome measure. They found a significantly increased TnI in the placebo group, and inferred that therefore myocardial damage was attenuated by steroids. Mott et al performed the largest double blind RCT with 246 children randomized. They used Post Pericardotomy Syndrome (PPS) as their primary outcome measure and found that there was no difference in uncomplicated PPS. However they found that 8 patients treated with methylprednisolone had complicated PPS compared to only 1 in the control group and concluded that steroids may actually be harmful. Unfortunately this group did not report any biochemical or other clinical markers. Bronicki et al conducted a double blind RCT in 29 children. They provided the most comprehensive findings in support of dexamethasone at 1mg/kg. They found a significant improvement in ventilation, ICU stay, arterial-alveolar oxygenation, post-operative temperature, and reductions in levels of Interleukin-6. No significant difference was found in TNF-á or complement C3a. It should be noted that this is also a small study and that the steroid group had a mean CPB time of 21 min less than the control group and a crossclamp time 14 min shorter than the control group. Lindberg et al performed an RCT in 40 children comparing dexamethasone 1mg/kg to placebo. Although CRP was found to be significantly lower on day 1, no other differences were found in other biochemical markers or with any clinical markers, including days in ICU or ventilation time. Two studies compared difference steroid regimes: Varan et al performed a study to see if there was a difference between giving methylprednisolone at 30mg/kg or 2mg/kg. They found no significant differences in clinical parameters, or in IL-6, IL-8, or CRP. Unfortunately this was a very small study with no sample size calculations and thus this study is likely to have been too small to confidently exclude a difference in these treatments. Schroeder et al recently performed a 29 patient PRCT to compare pre-operative and intra-operative steroids to intra-operative steroids alone. They found that adding a pre-operative dose resulted in lower fluid requirement, lower body temperature , lower inflammatory marker expression and an almost significant lower ICU stay. It must be noted that the mean age of the group who did not receive pre-operative steroids was half of those who received pre-operative steroids. The studies summarised here do provide some evidence for a benefit in giving steroids but 2 studies reported negative findingd and the largest study found an increase in complicated post-pericadotomy syndrome with steroids. We were surprised not to find any large prospective or retrospective cohort studies that might have shed further light on the possible changes in clinical course secondary to steroids and we conclude that there is certainly a great need for further clinical studies in this area, which would optimally be by multi-centre PRCT but even a retrospective cohort study would be highly informative.

Clinical Bottom Line

Steroids may reduce Troponin I release, CRP and reduce Interleukin-6. In addition 2 studies, each in only 30 patients, found some evidence for improvements in clinical parameters such as ICU stay and fluid requirement. These findings need confirmation prior to any firm recommendations as to the benefits of steroids

References

1. Checchia PA, Backer CL, Bronicki RA, Baden HP, Crawford SE, Green TP, Mavroudis C. Dexamethasone reduces postoperative troponin levels in children undergoing cardiopulmonary bypass.[see comment]. Crit Care Med 2003;31(6):1742-5.
2. Mott AR, Fraser CD Jr, Kusnoor AV et al. The effect of short-term prophylactic methylprednisolone on the incidence and severity of postpericardiotomy syndrome in children undergoing cardiac surgery with cardiopulmonary bypass. J Am Coll Cardiol 2001;37(6):1700-6.
3. Varan B, Tokel K, Mercan S, Donmez A, Aslamaci S. Systemic inflammatory response related to cardiopulmonary bypass and its modification by methyl prednisolone: high dose versus low dose. Pediatr Cardiol 2002;23(4):437-41.
4. Bronicki RA, Backer CL, Baden HP, Mavroudis C, Crawford SE, Green TP. Dexamethasone reduces the inflammatory response to cardiopulmonary bypass in children. Ann Thorac Surg 2000;69(5):1490-5.
5. Lindberg L, Forsell C, Jogi P, Olsson AK. Effects of dexamethasone on clinical course, C-reactive protein, S100B protein and von Willebrand factor antigen after paediatric cardiac surgery. Br J Anaesth 2003;90(6):728-32.
6. Schroeder VA, Pearl JM, Schwartz SM, Shanley TP, Manning PB, Nelson DP. Combined steroid treatment for congenital heart surgery improves oxygen delivery and reduces postbypass inflammatory mediator expression. Circulation 2003;107(22):2823-8.