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

Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease.

Hoffman TM, Wernovsky G, Atz AM, et al.

Circulation. 2003 Feb 25;107(7):996-1002. [abstract]

Reviewed by Al Torres, MD, MS, University of Illinois College of Medicine at Peoria

Review posted August 29, 2003

---

I. What is being studied?:

The study objective:

To evaluate the efficacy and safety of the prophylactic use of milrinone in pediatric patients at high risk of developing low cardiac output syndrome (LCOS) after cardiac surgery.

The study design:

The PRIMACORP trial (Prophylactic Intravenous use of Milrinone After Cardiac OpeRation in Pediatrics) was a multicenter, randomized, double-blind, placebo-controlled trial using 3 parallel treatment groups of pediatric patients undergoing cardiac surgery.

The patients included:

238 patients, median age of 3 months (range of 2 days to 6.9 years) who underwent biventricular repair of certain cardiac lesions involving cardiopulmonary bypass, received the study medication. 242 patients were enrolled. 238 received study medication. 11 had major protocol violations leaving 227 patients (median age 3 months, range 2 days-6.9 years) in the "per protocol" population. 13 patients in the "per protocol" population received open label milrinone and their data were not included in the analysis of the secondary endpoint.

The patients excluded:

Patients were excluded if their body weight was < 2 kg, premature (birth < 36 weeks postconceptual age), renal dysfunction (creatinine > 1.5 mg/dl 48 hours before surgery), and LCOS or hypotension on arrival to the PICU.

The interventions compared:

Patients were randomly assigned to in a 1:1:1 ratio within 90 minutes of arriving in the PICU to receive either 1), low-dose IV milrinone (25 mcg/kg bolus over 60 minutes followed by a 0.25 mcg/kg per min infusion for 35 hours); 2), high-dose IV milrinone (75 mcg/kg bolus followed by an infusion of 0.75 mcg/kg per min infusion for 35 hours); or 3), placebo.

The outcomes evaluated:

The primary endpoint was a composite variable consisting of death or the development of LCOS requiring additional pharmacological support or other support administered within the first 36 hours after cardiac surgery. LCOS was defined as clinical signs or symptoms, e.g., tachycardia, oliguria, poor perfusion, cardiac arrest, with or without a widened arterial-mixed venous oxygen saturation difference or metabolic acidosis. Additional pharmacological support or other support was defined as mechanical support of the circulation (e.g., extracorporeal life support), an increase of > 100% of pharmacological support relative to baseline, the administration of a new, open-label inotropic agent, or other interventions specifically to treat LCOS. Secondary endpoints included evaluation of the composite variable (death or development of LCOS) in the interval between 36 hours after initiation of the study drug and the final visit (up to 30 days after randomization), duration of mechanical ventilation, length of hospital stay, creatinine clearance, and total urine output.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes, but the randomization method used was not described in the trial design paper published previously (1). The randomization scheme used was effective in creating three similar groups (see # 4 below).

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

Was followup complete?

Patients were followed for evaluation of the secondary outcomes up until their final visit, which was up to 30 days post-randomization. 209 of the 242 patients enrolled were followed to the final visit. All patients were followed up for at least 14 days after randomization.

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

Patients were analyzed in the groups to which they were randomized as far as the primary outcome variable is concerned. For the primary outcome, an intention to treat analysis (n= 238, including 11 patients with major protocol violations) and efficacy analysis (n=227, excluding patients with major protocol violations) was performed. For the secondary outcomes, the analysis excluded the 11 patients with major protocol violations, the 13 patients that received open label milrinone after study drug infusion and the 5 lost to follow-up after hospital discharge.

Secondary questions:

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

Maybe. The authors state that the trial was double-blinded but did not elaborate. Since there were two arms of the trail where IV milrinone was infused at different rates, a low- and high-dose, it would have been reassuring if the investigators had explained that the concentrations of the IV milrinone was adjusted so that the bedside staff would use the same delivery rate regardless of study medication assigned. The authors do state that the predictable hemodynamic changes during the institution of milrinone may have resulted in some bias. However, an independent blinded committee to adjudicate the primary endpoint was used in all cases and should have eliminated this bias.

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

Yes. Although other co-morbidities which influence outcomes (e.g., chromosomal abnormalities) were not addressed, it is reasonable to assume that they were equally distributed between all 3 groups because all the other variables examined were.

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

Yes. The use of treatment protocols was not mentioned in the paper. The authors included treatment interventions specifically directed at correcting LCOS as part of the primary outcome measure. A comparison of the volume expanders administered to the three groups would have reassured this reviewer that the groups were treated equally.

III. What were the results?

1. How large was the treatment effect?

No patients died during administration of study drug. The use of high-dose milrinone significantly reduced the risk of the development of LCOS compared with that of placebo in all 238 treated patients (p = 0.023, relative risk reduction, 55%) and in the 227 per-protocol population (p = 0.007, relative risk reduction 64%). There was no significant difference between the placebo group and the low-dose milrinone group in the development of LCOS (p = 0.183, relative risk reduction 34%). Six additional patients developed LCOS after the discontinuation of study drug infusion. There was a 48% relative risk reduction (p = 0.049) in LCOS/death by the final visit in the high-dose milrinone group compared to the placebo group.

The mean (geometric) duration of mechanical ventilation and hospital stay was similar in all three treatment groups (1.6 to 1.7 days, p = 0.964; 8.6 to 10.2 days, p = 0.159, respectively).

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

The 95% CI for the 64% relative risk reduction in development of LCOS in the high-dose milrinone (per protocol) group compared to the placebo group was 21 to 84%, i.e., as few as 21% or as great 84% a relative risk reduction could occur in the high-dose milrinone group in 95 out of 100 similar trials. The absolute risk reduction for developing LCOS in the high-dose milrinone group was 17%, 95% CI, 5 Ð30%. And finally, the number of patients needed to treat with high-dose milrinone to avoid developing LCOS in one patient was six (95% CI, 3 to 20).

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

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

Yes, as long as the population of cardiac surgery candidates we serve are similar in diagnosis and severity. The authors make it clear that the patients studied were only those who underwent biventricular repair. This limitation excludes many high-risk cardiac patients such as the child with hypoplastic left heart syndrome undergoing staged repair. The authors also explain that other high-risk patients, those treated with milrinone in the operating room immediately after cardiac surgery, were also not included.

2. Were all clinically important outcomes considered?

Yes. Although the investigators did not use objective measures of LCOS such as thermodilution measurements of cardiac index, since this is rarely measured in pediatric patients. The combination of clinical diagnosis plus therapeutic interventions confirmed by an independent committee was satisfactory.

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

Probably. The incidence of serious adverse events overall and by organ system was similar in the treatment groups. Thrombocytopenia, arrhythmias, and hypotension were infrequent. The investigators would have made a stronger argument for high-dose milrinone administration in this population of patients if they had compared costs in the three groups. Avoidance of long-term complications of LCOS (e.g., chronic renal failure, neurological impairment) should be incorporated into this comparison. However, these long term complications are rare following pediatric cardiac surgery and were not included in this study.

References

1. Hoffman TM, Wernovsky G, Atz AM, et al. Prophylactic intravenous use of milrinone after cardiac operation in pediatrics (PRIMACORP) study. Am Heart J 2002; 143:15-21 [abstract]

Comments

Back to the EB Journal Club Index