The effect of milrinone on hemodynamic and gas exchange parameters in children

Rohit S. Loomba; Vincent Dorsey; Enrique G. Villarreal; Saul Flores

doi:10.1017/S1047951119002865

The effect of milrinone on hemodynamic and gas exchange parameters in children

Published online by Cambridge University Press: 17 December 2019

Rohit S. Loomba

Vincent Dorsey ,

Enrique G. Villarreal

and

Saul Flores

Show author details

Rohit S. Loomba: Affiliation:
Cardiology, Pediatrics, Advocate Children’s Heart Institute, Advocate Children’s Hospital, Oak Lawn, IL, USA Medicine, Chicago Medical School/Rosalind Franklin University of Medicine and Science, North Chicago, IL, USA
Vincent Dorsey: Affiliation:
Cardiology, Pediatrics, Advocate Children’s Heart Institute, Advocate Children’s Hospital, Oak Lawn, IL, USA
Enrique G. Villarreal*: Affiliation:
Critical Care and Cardiology, Pediatrics, Texas Children’s Hospital/Baylor College of Medicine, Houston, TX, USA Tecnologico de Monterrey, Escuela de Medicina y Ciencias de la Salud, Monterrey, Nuevo Leon, Mexico
Saul Flores: Affiliation:
Critical Care and Cardiology, Pediatrics, Texas Children’s Hospital/Baylor College of Medicine, Houston, TX, USA
*: Author for correspondence: E. G. Villarreal, MD, Cardiac Intensive Care Unit, Section of Critical Care and Cardiology, Texas Children’s Hospital, Research Scholar Baylor College of Medicine, Houston, TX, USA. Tel: +1 312 282 6935; Fax: +1 832 825 2969; E-mails: quique_villarreal93@hotmail.com; noyola@bcm.edu

Article contents

Abstract
Materials and methods
Results
Discussion
Conclusion
Supplementary material
Financial Support
Conflicts of Interest
References

Rights & Permissions

Abstract

Milrinone is a drug frequently used for hemodynamic support in children during critical illness. Although the hemodynamic changes induced by milrinone in children may appear similar to those of adults, the physiologic contributors of these changes remain vastly unknown. A systematic review was conducted to identify studies characterising the hemodynamic effects of milrinone in children during critical illness for hemodynamic support for various medical conditions. Studies were assessed for quality and those of satisfactory quality with pre- and post-operative hemodynamics for each patient were included in the final analyses. Those not limited to children and those not limited to patients with critical illness were excluded from the final analyses. A total of six studies with 791 patients were included in the final analyses. Milrinone infusion doses ranged from 0.3 to 0.75 mcg/kg/minute with the mean infusion dose being 0.5 mcg/kg/minute. Patients whom received milrinone infusion had greater cardiac output, greater left ventricle shortening fraction, lower right ventricular systolic pressure, and lower serum lactate levels. Systolic blood pressure mean arterial blood pressure and arterial oxygen concentration did not significantly change with administration of milrinone. These results were irrespective of milrinone infusion dose, infusion duration, and study size. Milrinone was found to have several beneficial hemodynamic effects in children during critical illness when used at usual clinical doses.

Keywords

Paediatrics milrinone intensive care units phosphodiesterase 3 inhibitors heart disease

Type: Original Article
Information: Cardiology in the Young , Volume 30 , Issue 1 , January 2020 , pp. 55 - 61

DOI: https://doi.org/10.1017/S1047951119002865 [Opens in a new window]
Copyright: © Cambridge University Press 2019

Milrinone was developed more than 30 years ago as WIN 47203, a non-glycosidic, non-adrenergic cardiotonic agent with a chemical structure analogous to that of amrinone.Reference Maskin, Sinoway, Chadwick, Sonnenblick and Le Jemtel¹ Milrinone is a potent cardiac bipyridine with inotropic and vasodilator properties carried out by selective inhibition of the cardiac adenosine 3’,5’-monophosphate (cAMP) phosphodiesterase III isozyme, with resultant increases in cardiac cAMP levels.Reference Alousi, Canter, Montenaro, Fort and Ferrari² The pharmacologic effects of milrinone include relaxation of the smooth muscle, enhanced myocardial contractility, and improved myocardial relaxation.Reference Silver, Harris and Canniff³^, Reference LeJemtel, Scortichini, Levitt and Sonnenblick⁴ The cardiovascular effects of milrinone were initially characterised in animal models and adult humans by increased cardiac output and decreased left ventricular end-diastolic pressure while significantly decreasing systemic vascular resistance.Reference Alousi, Canter, Montenaro, Fort and Ferrari²^, Reference Sinoway, Maskin, Chadwick, Forman, Sonnenblick and Le Jemtel⁵^, Reference Baim, McDowell and Cherniles⁶

Milrinone has been used for a number of years to mitigate the effects of low cardiac output particularly in patients after congenital heart surgery.Reference Hoffman, Wernovsky and Atz⁷ The hemodynamic changes induced by milrinone in children appear to be similar to those of adults. However, the physiologic contributors of these changes remain vastly unknown. Furthermore, recent trials in children and adults have demonstrated conflicting results related to patient outcomes, medication effectiveness, and the impact on hemodynamics.Reference Cavigelli-Brunner, Hug and Dave⁸^–Reference Ushio, Egi and Wakabayashi¹⁰ Therefore, the primary objective of this study was to determine the effect of milrinone on hemodynamic parameters in critically ill children by comparing hemodynamic data before and after the initiation of milrinone. A systematic review and subsequent meta-analyses were conducted.

Materials and methods

Study identification

A systematic review was conducted to identify studies pertaining to the hemodynamic effects of milrinone in children. A new review protocol was used for this study as there was no existing protocol. Pubmed, EMBASE, and the Cochrane databases were queried to identify such studies. The following keywords were used in individually and in various combinations with one another: “milrinone,” “children,” “pediatrics,” “hemodynamics,” “blood pressure,” “cardiac output,” “pulmonary artery pressure,” and “vascular resistance” (Supplementary File S1). Both authors (RL, SF) reviewed the abstracts and titles of resulting studies. Full-text manuscripts of studies found to be pertinent to study questions were then obtained. These were reviewed for quality using the Newcastle–Ottawa Scale and assessed for risk of bias using the Cochrane Risk of Bias tool. Any discrepancies in these evaluations between the two authors were then reviewed by both authors and a consensus achieved. Studies of adequate quality and low risk of bias were then considered for inclusion.

Studies were included in the final pooled analyses if they met the following requirements: (1) utilised milrinone; (2) included children (under 18 years of age); (3) contained data for hemodynamic parameters before and after the initiation of milrinone; (4) published data; and (5) English language. Studies that met these criteria were then deemed appropriate for inclusion. Any discrepancies in appropriateness of inclusion between the two authors were reviewed by both authors and a consensus achieved.

Endpoints

Reported endpoints from the studies deemed appropriate for inclusion were outlined. The number of studies reporting each endpoint were then reviewed and a decision was made to include endpoints that had data from two or more studies. The following endpoints were identified for pooled analyses: systolic blood pressure (mmHg), mean arterial blood pressure (mmHg), echocardiographic Doppler-derived cardiac output (ml/kg/minute), left ventricular shortening fraction (%), right ventricular systolic pressure (mmHg), heart rate (beats per minute), arterial oxygen concentration (mmHg), arterial carbon dioxide concentration (mmHg), fraction of inspired oxygen (%), and serum lactate (mmol/L).

Data extraction

Data were extracted using an electronic data extraction tool. Data were extracted by both authors independently. As all variables were continuous, means and standard deviations for the pre- and post-milrinone data were recorded. The individually extracted data were then compared to identify any discrepancies. Discrepancies in the extracted data were then reviewed by the authors together and discrepancies resolved by consensus.

Bias analysis

Bias analysis was performed with specific attention paid to patient selection, intervention selection, endpoint inclusion, and result reporting.

Statistical analysis

Once data were extracted each endpoint underwent evaluation of heterogeneity. A Q-statistic and its resulting p value were calculated as was an I² statistic. If the p value for the Q-statistic was less than 0.05 or if the I² value was greater than 50%, heterogeneity was deemed significant. In the absence of significant heterogeneity, a fixed-effects model was used for the pooled analyses while a random-effects model was used if there was significant heterogeneity present. All endpoints were continuous in nature and thus pooled analyses were conducted to determine the mean difference and 95% confidence interval. Publication bias was assessed using an Egger test for endpoints with data from three or more studies. Meta-regression was not conducted due to the low number of pooled studies. Sensitivity analyses were conducted to determine the effect of milrinone dose, duration of milrinone infusion, and study size on the individual endpoints.

Results

Study characteristics

A total of six studies including 791 patients were included in the analyses (Fig 1).Reference James, Bee, Corcoran, McNamara, Franklin and El-Khuffash¹¹^–Reference Bianchi, Cheung, Phillipos, Aranha-Netto and Joynt¹⁶ Studies were published between the years 2013 and 2018. A majority, four, of the six studies included patients in the first month of life. The mean age of included patients was 8.5 months. Milrinone infusion doses ranged from 0.3 to 0.75 mcg/kg/minute with the mean infusion dose being 0.5 mcg/kg/minute. The duration of milrinone infusion until repeat hemodynamics ranged from 20 to 408 hours with a mean of 119 hours (Table 1).

Figure 1. Study identification methodology.

Table 1. Characteristics of selected studies

¹Sample size, ²micrograms, ³kilograms, ⁴congenital heart disease, ⁵Cardiology in the Young, ⁶persistent pulmonary hypertension of the newborn, ⁷Journal of Pediatrics, ⁸heart failure

Systolic blood pressure

A total of five studies with data from 774 patients were included in the pooled analysis of systolic blood pressure. Heterogeneity analysis resulted in a Q-statistic with a p value of less than 0.001 and an I² value of 90%, indicating the presence of significant heterogeneity. Thus, a random-effects model was used. Systolic blood pressure was not significantly different with milrinone infusion. The mean difference in systolic blood pressure was 3.46 mmHg with a 95% confidence interval −1.20 to 8.12 (p = 0.15) (Fig 2).

Figure 2. Forest plot demonstrating impact of milrinone on systolic blood pressure.

Sensitivity analyses demonstrated that milrinone infusion dose and duration did not impact the pooled result. Study size, however, did. With the removal of data from group A of the study from Lee et al, the change in systolic blood pressure was significantly different with a mean difference of 2.18 mmHg (95% confidence interval 0.36–4.00, p = 0.02).

Mean arterial blood pressure

A total of four studies with data from 52 patients were included in the pooled analysis of mean arterial blood pressure.Reference James, Bee, Corcoran, McNamara, Franklin and El-Khuffash¹¹^–Reference McNamara, Shivananda, Sahni, Freeman and Taddio¹³^, Reference Bianchi, Cheung, Phillipos, Aranha-Netto and Joynt¹⁶ Heterogeneity analysis resulted in a Q-statistic with a p value of 0.62 and an I² value of 0%, indicating the absence of significant heterogeneity. Thus, a fixed-effects model was used. Mean arterial blood pressure was not significantly different with milrinone infusion. The mean difference in mean arterial blood pressure was 1.23 mmHg with a 95% confidence interval −0.57 to 3.04 (p = 0.18) (Fig 3).