Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Reversal of catabolism by beta-blockade after severe burns.

Herndon DN, Hart DW, Wolf SE, Chinkes DL, Wolfe RR.

New Engl J Med 2001;345: 1223-1229. [abstract]

Reviewed by M Sharma, K Meert, Children's Hospital of Michigan, Detroit

Review posted April 21, 2003

---

I. What is being studied?:

The study objective:

To determine whether long term beta blockade with propranolol would decrease the resting energy expenditure and rate of muscle protein catabolism in children with severe burns.

The study design:

This study was designed as a prospective, randomized trial at a single center. It was conducted from January through December 1999.

The patients included:

The patient population was comprised of children less than 18 years of age with burns on more than 40% of their body surface area (BSA) transferred to the study center within one week after injury. Each of these patients underwent burn wound excision and grafting within 48 hours after admission.

The patients excluded:

Patients with a history of asthma were excluded.

The interventions compared:

In the propranolol group, patients were given propranolol through a nasogastric tube (0.33 mg/kg body weight every 4 hours) immediately after the second grafting operation. This second grafting operation was performed 6-10 days after the first excision-grafting operation (8-12 days after admission). Propranolol dose was adjusted to achieve a 20% decrease in heart rate compared with the 24-hour average heart rate immediately before drug treatment. Propranolol was continued for duration of 4 weeks or until discharge. Control group patients were not treated with propranolol.

The outcomes evaluated:

The main outcome variables were resting energy expenditure (REE) and net protein balance. Secondary outcome variables included fat free mass, serum hormone levels and lean body mass (see table 1 regarding methods and timings of measurements).

Table 1. Methods and timings of measurements of main and secondary outcome variables

Variable	Method of measurement	Before propranolol (baseline)	After 2 weeks of propranolol treatment	After 4 weeks of propranolol treatment	At discharge
REE	Metabolic cart	X	X
Net muscle-protein balance	Skeletal muscle 5-hour protein kinetics (stable-isotope) study	X	X
Fat free mass	Body composition by whole-body potassium-40 scanning	X		X
Serum potassium, serum glucose	Stat-5 analyzer	X	X
Insulin-like growth factor I (IGF I)	Ethanol extraction	X	X
Serum growth hormone, insulin and cortisol	Enzyme-linked immunosorbent assay/ enzyme immunoassay	X	X
Lean body mass	Body composition scanning by dual-image x-ray absorptiometry				X

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. Twenty-five patients were enrolled, of which 13 were randomized to the propranolol group, and 12 served as untreated controls. Randomization was performed with use of a random-number-generating scheme.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

One of the 13 patients in the propranolol group chose not to participate in any of the invasive studies (such as protein kinetics necessitating muscle biopsy), the reason for which is not reported. Measurements of REE, serum hormone levels were only available in 12 patients in the propranolol group and 12 patients in the control group. In one of the studies on a propranolol group patient undergoing protein kinetics a steady state was not reached, therefore this study was not included for analysis. Three patients (two in the control group and one in the propranolol treatment group) were fully healed and discharged before receiving 4 weeks of treatment; these subjects did not undergo a second (follow up) whole body potassium scanning. Thus measurements of fat free mass by whole body potassium scanning were available for comparison for 10 patients in the control group and 12 patients in the propranolol group. Further, 9 consecutive patients of the 25 enrolled were not able to undergo dual x-ray absorptiometry at the time of discharge because of technical difficulties with the scanner; this could be accomplished in 7 patients in the control group and 9 patients in the propranolol group.

Were patients analyzed in the groups to which they were randomized?

Yes. None of the patients crossed over from the treatment to control group or vice versa.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

No. Neither patients nor the health workers or the study personnel were blind to treatment.

4. Were the groups similar at the start of the trial?

Baseline characteristics for control and treatment groups are presented but statistical analysis is not noted in comparing some pertinent characteristics between the groups. It appears that the propranolol treatment group tended to have lower weight and body surface area (BSA) as compared to the control group. In addition, baseline insulin levels were twice as high in the treated compared to the control group.

5. Aside from the experimental intervention, were the groups treated equally?

Enteral nutrition was started for all patients and continued until the wounds were healed. The patients were fed the same commercial formula (Vivonex T.E.N., Sandoz) through a nasoduodenal tube to deliver 1500 Kcal/m2 of BSA burned plus another 1500 Kcal/m2 of total BSA. All patients underwent burn wound excision and grafting within 48 hours after admission. The patients remained in bed for five days after excision and grafting procedures, after which they were allowed to walk daily until the next excision-grafting procedure. All patients underwent sequential staged grafting procedures until the wounds were closed. However, the number of grafting procedures done on patients in control group versus treatment group has not been reported.

III. What were the results?

1. How large was the treatment effect?

The treatment effect in this study was to look at the efficacy of rHuEPO. Patients treated with propranolol had a decrease in REE at 2 weeks after treatment compared to baseline by approximately 24%. Those in the control group had an increase REE by about 9% at two weeks compared to baseline. Propranolol group patients had a positive net balance of protein synthesis at 2 weeks after treatment (positive net balance = protein synthesis > breakdown); those in the control group had a negative net balance. For the propranolol group patients, the net balance of protein synthesis was negative at baseline (before treatment); this net balance became positive when measured at 2 weeks after propranolol treatment.

Table 2. Magnitude and precision of primary and some secondary outcome variables (± values are ± s.e.)

Variable	Control (n = 12)	Propranolol (n= 12)	Mean difference	p-value	95% confidence interval of the mean difference
Change in REE (from baseline to 2 weeks of propranolol or no treatment)	140 ± 67 kcal/day	-422 ± 197 kcal/day	562 kcal/day	0.001	148 to 976 kcal/day
Net balance of protein synthesis and breakdown (after 2 weeks of propranolol or no treatment)	-0.042 ± 0.016 _mol/ min/100 ml of leg volume	0.035 ± 0.011 _mol/min/100 ml of leg volume	-0.077 _mol/min/100 ml of leg volume	0.001	-0.047 to -0.107 _mol/min/100 ml of leg volume
Change in % fat free mass (after 4 weeks of propranolol or no treatment)*	-9% (n=10)	-1% (n=12)	-8%	0.003	No SE or SD reported
Change in serum potassium (from baseline to 2 weeks of propranolol or no treatment)	-0.1 ± 0.1 mg/dl	0.4 ± 0.2 mg/dl	-0.5 mg/dl	0.05	-0.94 to 0.06 mg/dl
Change in serum IGF-I (from baseline to 2 weeks of propranolol or no treatment)	54 ± 12 ng/ml	44 ± 13 ng/ml		0.56
Change in serum insulin (from baseline to 2 weeks of propranolol or no treatment)	13.6 ± 24.8 _IU/ml	-21.0 ± 17.3 _IU/ml		0.29
Change in serum cortisol (from baseline to 2 weeks of propranolol or no treatment)	-4.4 ± 2.9 _g/dl	-2.2 ± 2.1 _g/dl		0.58
Change in serum growth hormone (from baseline to 2 weeks of propranolol or no treatment)	0.5 ± 1.0 ng/ml	_1.1 ± 0.9 ng/ml		0.26
Lean body mass (at discharge)*	73.5 ± 1.5% (n=7)	79.1 ± 1.2% (n=9)	-5.6%	0.01	-8.6 to 19.8%

* Number less than 12 because of loss of patients to discharge or not able to obtain test

2. How precise was the estimate of the treatment effect?

Confidence intervals of the main outcome variables are as shown in Table 2.

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

The reduction in REE and improvement in net protein balance are impressive, but the confidence intervals are large, making it difficult to know the magnitude of beneficial effect in an individual patient. The authors did not enroll children with serious inhalation injury or critically ill children requiring mechanical ventilation; the use of propranolol in such children needs to be considered separately (1).

2. Were all clinically important outcomes considered?

The duration of hospital stay, time taken for healing (two patients in the control group healed before 4 weeks, and one patient in the propranolol group healed before 4 weeks) and the number of grafting procedures done on the control group versus propranolol group subjects have not been reported.

3. Are the likely treatment benefits worth the potential harms and costs?

Large doses (average maximal dose, 1 mg/kg body weight every four hours) of propranolol were required to significantly lower heart rate and protein catabolism (1). Decreased blood pressure necessitated temporarily withholding of the drug in about 23% patients. Propranolol is not an expensive drug, but its effects on blood pressure would require constant monitoring in the PICU. Decreased cardiac output, particularly in septic children can lead to hypoperfusion. In patients with asthma, severe bronchospasm may be induced. Propranolol blocks both beta1 and beta2 adrenergic receptors, and can mask symptoms of hypoglycemia in diabetics, and those of thyrotoxicosis in patients with hyperthyroidism. Other side effects of high doses of propranolol (e.g., hyperkalemia, hypoglycemia) could possibly become apparent with treatment of a larger number of patients.

References

1. Sheridan RL. A great constitutional disturbance (editorial). New Engl J Med 2001,345(17): 1271-1272.

Comments

Back to the EB Journal Club Index