• 
• 

Three Part Question

In a [critically ill patient with severe burns], does [propranolol] [reduce the hypermetabolic state]?

Clinical Scenario

A 6 year old boy was admitted for 60% full thickness body surface area burns. His stay in pediatric intensive care unit (PICU) was complicated with recurrent episodes of sepsis hampering the progress of enteral nutrition. The team wondered if beta-blockers can be used in this patient to reduce his hypermetabolic state.

Search Strategy

Secondary source – Cochrane Database of Systematic Review, BestBETs
Primary source – PubMed, EMBASE
No time limitations.
We used the search terms of [‘adrenergic beta-antagonist’] AND [‘burns’].

Search Outcome

Using the above MeSH terms in PUBMED and EMBASE, a total of 39 articles and 27 articles were identified respectively. Four randomised controlled trials were relevant. The references of these papers were reviewed, along with the linked articles, but no further articles were found. Secondary search of The Cochrane Library and BestBETs did not yield any relevant review.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Herndon DN, Hart DW, Wolf SE, et al 2001 Texas	Patients aged 0.3-18 years old with burns more than 30% total body surface area burns and required 1 or more surgical interventions.	Prospective randomized trial 90 given propranolol 4mg/kg/day vs. 89 controls with no propranolol for 12 months.	Rate pressure product (95% confifence interval)	Decrease of 1706 mmHg x bpm (1200 - 2212) at 2 weeks to 1249 mmHg x bpm ( 722 - 1776) at 6 months in the propranolol group	Mode of administration of propranolol unclear. Randomisation, allocation concealment and blinding were not clear.
			Percent predicted resting energy expenditure (95% CI)	Decrease of 18% (8 - 27) in the propranolol group at 2 weeks to 13% (6 - 20) at 6 months in the propranolol group.
			Peripheral lean mass (95% CI)	Increase of 664g (22 - 1350) at 3 months to 956g (205 - 1706) at 6 months
			Central Mass (95% CI)	Decrease of 2112g (1089 - 3135) at 3 months to 3541g (2160 - 4922) at 12 months
			Odds of ≥ 5%TBMC/TBM in control and propranolol group	0.41 (0.19 - 0.89) at 6 months to 0.44 (0.20 - 0.94) at 12 months
			Mortality	No difference
			Length of stay	No difference
			Adverse events	2 cases of bradycardia, 1 case of hypoglycaemia, 1 case of cardiac arrhythmia, 2 cases of respiratory compromise and 4 deaths from sepsis was reported in the propranolol group
Herndon DN, Hart DW, Wolf SE, et al. 2001 Texas	Children less than 18 years old with more than 40 percent total body surface area burns	Prospective randomized trial 13 patients given oral propranolol (average 6.3mg/kg/day) to decrease heart rate by 20% from baseline value vs. 12 controls with no propranolol for 2-4 weeks	Resting energy expenditure (95% CI)	Decrease of 349 kcal/day (181 to 517) in propranolol group	Time of initiation of propranolol not clear. Randomisation carried out. Allocation concealment and blinding not clear. Small study and insufficient power. Baseline characteristics of two groups varied.
			Lean body mass (95% CI)	Increase of 5.6% (4 to 7) in propranolol group
			Mortality	Not mentioned
			Length of stay	Not mentioned
			Adverse Events	Propranolol was withheld temporarily from 3 patients at some time during therapy due to low mean arterial pressure.
Williams FN, Herndon DN, Kulp GA, et al. 2011 Texas	Patients with burns over 30% total body surface area	Prospective randomized trial 171 patients given oral propranolol, (average 4-6mg/kg/day) to decrease heart rate by 15% from baseline vs. 235 controls with no propranolol	Stroke volume (95% CI)	Increase of 18 (17.2 - 18.8) % for the propranolol group	Randomisation, allocation concealment and blinding were not clear. Duration of therapy not clear. No information on adverse events. M-mode echocardiogram was used which may underestimate cardiac output and stroke volume. No information on the number of patients with delayed/missed echocardiogram due to chest wounds.
			Rate pressue product (95% CI)	Decrease of 1500 (1331-1669) mmHg x bpm
			Duration of ICU stay	No difference
Jeschke MG, Norbury WB, Finnerty CC, et al 2007 Texas	Children younger than 18 years old with more than 40% total body surface area burns	Prospective randomized, placebo controlled trial . 102 patiens given oral propranolol (0.5-1.5mg/kg/dose 6 hourly) vs 143 controls with normal saline placebo for more than 3 days.	Length of ICU stay	No difference	Randomisation, allocation concealment and blinding were not clear. Time of initiation and duration of therapy not clear. No standard deviation provided for percent predicted Resting Energy Expenditure
			Mortality	No difference
			Infections and sepsis	No difference
			Percent predicted Resting Energy Expenditure	Decrease of approximately 20% ( p<0.05)

Comment(s)

Burns is an important public health issue. Globally, close to 11 million people require medical attention for burn injuries. [1] Burns, ranked fourth in all injuries, has a higher incidence than tuberculosis and HIV infections combined together. [1] Fortunately, although burns and fires account for over 300,000 deaths each year throughout the world, the vast majority of burns are not fatal. Morbidity and mortality due to fire and flames has declined worldwide in the past decades. Nonetheless, in low- and middle-income countries, fire-related burns still remain the leading cause of disability-adjusted life years lost. [1] In severe burns, plasma catecholamine levels increase as much as ten-fold [2]. This catecholamine overdrive state is the primary mediator of the hypermetabolic state in patients with burns. This hypermetabolic state is characterized by tachycardia, increased cardiac work, metabolism, lipolysis and protein catabolism. The resultant insulin resistance, increase in fracture risk, increase in liver-size, growth and development retardation, risk of cardiac dysfunction, impaired strength and muscle function and increased risk for infections can lead to increased morbidity and mortality of the patient with severe burns. [3] Because of the central role of adrenergic “overdrive”, β-adrenergic receptor antagonist, such as propranolol, mitigates the actions of plasma catecholamines and thus, reduces the hypermetabolic state in patients with acute burn injury. [3] This has led to conduct of a few randomised controlled trials examining the role of beta blockers in children with severe burns. The randomised controlled trials included [4, 5, 6, 7] in our review unfortunately have high risk of bias. Three did not state the method of randomisation clearly. [5, 6, 7] Allocation concealment and blinding of patients, caregivers and outcome assessors were also not explicitly described in all the studies. Baseline characteristics of propranolol and control groups were similar in all studies except for the study by Herndon et al [4]. In this particular study, the percentage of total body surface area burned, weight, age and sex distribution differed between intervention and control arms. Attrition rates were also not reported in two studies [6, 7]. All of the studies were not powered to detect differences in the primary outcomes. All four studies were conducted in Shriners Hospitals for Children in Texas with overlap in the time period that the studies were conducted, raising concerns about the possibility of duplication of results. There was also variation in the route of administration of propranolol (parenteral or oral), dosage, time of initiation of propranolol and duration of propranolol given. None of the studies examined mortality and duration of stay as primary outcome measures. Only one study reported no significant difference in mortality between the propranolol and control group [5]. The two studies that reported survival numbers did not find any statistical difference in mortality rates between children treated and not treated with propranolol. [5, 7] Duration of intensive care unit and hospital stay were examined in two [6,7] and one study respectively. [5] All three studies did not demonstrate any significant difference in length of stay between the two groups. Three studies [4, 5, 7] demonstrated that propranolol led to a significant reduction in resting energy expenditure in children with severe burns. [4] Peripheral lean body mass also showed a significant improvement in two studies [4,5]. In one study, the reduction in energy requirement and improvement in lean mass persisted up till six months of treatment with propranolol. [4]. However, both these effects were no longer demonstrated in children in whom propranolol was used up to 12 months. This is most likely due to the normalisation of the burn physiology by 12 months. Central mass, which is composed of organs such as the liver, spleen, kidneys and mesenteric fat, showed a significant decrease up till 12 months on propranolol. However, the clinical significance of the reduction in central mass, such as fatty liver, remains uncertain. Children on propranolol were also found to have a lower likelihood of experiencing a 5% loss of total bone mineral content up to 12 months on propranolol. Pertinent clinical outcomes related to bone density (e.g., bone fractures or final height) were however, not reported. Rate pressure product (RPP), a correlate of myocardial oxygen consumption, is the product of mean arterial pressure (MAP) and heart rate [beats per minute (bpm)]. RPP was found to be significantly decreased in two studies [5, 6] and the decrease was sustained for 6 months on propranolol. Stroke volume determined on echocardiogram was also found to be significantly increased. [6] However, the association of reduction of RPP and improvement in stroke volume, on long term adverse cardiac events has not been established. In terms of safety outcomes, propranolol was relatively well tolerated with no significant blood pressure changes although a small number of patients did require omission of one or more doses of propranolol due to low blood pressure. There were reported incidences of bradycardia, cardiac arrhythmia, respiratory compromise in patients on propranolol but none of the studies were not powered to detect differences in side effects. In summary, propranolol is a useful and safe therapeutic option for reducing the hypermetabolic state option in paediatric patients with severe burns. However, these studies were done in highly specialised burn units with availability of close monitoring for the side effects of propranolol. Our review highlights a concerning lack of data of the effect of beta blockers on the clinically important outcomes of reduction in length of ICU stay, mortality and the long term outcome of growth and cardiac events in children with severe burns.

Clinical Bottom Line

Propranolol reduces cardiac work, RPP, resting energy expenditure, central deposition of fat with improvement in lean muscle mass and skeleton loss in children with severe burns. However, the use of this β-adrenergic receptor antagonist has not been shown to decrease the length of hospitalisation or improve mortality in this group of children.

References

1. Herndon DN, Hart DW, Wolf SE, et al Long-Term Propranolol Use in Severely Burned Pediatric Patients: A Randomized Placebo-Controlled Study
2. Herndon DN, Hart DW, Wolf SE, et al. Reversal of catabolism by beta-blockade after severe burns
3. Williams FN, Herndon DN, Kulp GA, et al. Propranolol decreases cardiac work in a dose dependent manner in severely burned children
4. Jeschke MG, Norbury WB, Finnerty CC, et al Propranolol does not increase inflammation, sepsis, or infectious episodes in severely burned children