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

Hemodynamic effects of i.v. milrinone lactate in pediatric patients with septic shock. A prospective, double-blinded, randomized, placebo-controlled, interventional study.

Barton P, Garcia J, Kouatli A, Kitchen L, Zorka A, Lindsay C, Lawless S, Giroir B.

Chest 1996; 109: 1302-1312. [abstract]

Reviewed by Michael Verive, MD, Hope Children's Hospital, Oak Lawn, IL

Review posted December 18, 1999

---

I. What is being studied?:

The study objective:

To determine the hemodynamic effects of intravenous milrinone lactate in pediatric patients with nonhyperdynamic septic shock. Specifically, the study hypothesized that milrinone would increase cardiac index (CI) by 20% and decrease systemic vascular resistance index (SVRI) by 20% during a 2 hour study period.

The study design:

Prospective, double-blind, randomized, placebo-controlled, interventional study.

The patients included:

Twelve pediatric patients aged 9 months to 15 years (original inclusion criteria 5 months to 18 years) requiring inotropic support and pulmonary artery catheter for nonhyperdynamic septic shock as defined by at least four clinical and four physiologic findings including:

Clinical:

• Peripheral cyanosis
• Cold, clammy skin
• Capillary refill > 3 seconds
• Thready pulses
• Shallow breathing or need for mechanical ventilation
• Tachycardia (infant HR > 160, child HR > 150)

Physiologic:

• Oliguria
• Normal or decreased CI ( 3.5-5.5 liters/min/m2)
• Normal or increased SVRI ( 800 dynes*sec*cm5/m2)
• Mixed venous saturations < 65% or > 85%
• Metabolic acidosis (serum bicarbonate < 18 mmol/L with or without respiratory compensation)
• Normal or high central venous pressure (CVP 6-10 mmHg)
• Normal or high pulmonary capillary wedge pressure (PCWP 8-12 mmHg)

The patients excluded:

Patients were excluded if they had known congenital heart disease, history of supraventricular or ventricular dysrrhythmias, history of hyperthyroidism or hypothyroidism, idiopathic (immune) thrombocytopenic purpura or hypersplenism, inadequate filling pressures, history of chronic renal failure or serum creatinine > 2 mg/dL unless receiving renal replacement therapy (CAVH or CVVH), and patients who are already receiving other phosphodiesterase inhibitors (amrinone or other methylxanthines).

The interventions compared:

Patients were randomized to receive either milrinone bolus + milrinone infusion (Group A), or placebo bolus + placebo infusion (Group B). After two hours, the patients in the milrinone group were crossed over to receive a placebo load and infusion in addition to their existing milrinone infusion. Patients in the placebo group crossed over to receive milrinone load and infusion in addition to their existing placebo infusion.

The outcomes evaluated:

In addition to measurements of CI and SVRI, oxygen delivery and consumption (DO2 and VO2), and other hemodynamic parameters (PVRI, SVI, RVSWI, LVSWI, heart rate, systolic blood pressure, mean arterial pressure, MPAP, PCWP, and Qs/Qt) were compared. Left ventricular compliance was also calculated by dividing LVSWI by left ventricular end diastolic volume index (LVEDVI, obtained via echocardiography) and compared between the two groups and in Group B after placebo and again after milrinone.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes, but the method was not specified.

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

Was followup complete?

Yes. No patients were dropped from either treatment arm through the study.

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

Yes, although the study design involved the intentional cross over of patients. Ultimately, they compared all the milrinone-treated patients (all 12 eventually received the drug) to the "pure" placebo patients, of which there were only 6 in the first phase of the study.

Secondary questions:

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

All patients, health workers, and study personnel were blinded to the treatment until after the second 2 hour study period, at which time an unblinded pharmacist informed the attending physician caring for the patients which infusions were milrinone.

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

A qualified yes. Although there were no significant differences in the hemodynamic parameters between the two groups at the start of the study, no data is available to determine the differences in age, gender, weight, duration of shock prior to study entry, heterogeneity of disease states, type and duration of inotropic support, or other therapeutic interventions which may have been given. Since the number of study patients was small, either a comparison of these parameters or a listing of which group the patients were randomized to should have been included to determine if there were any differences between groups prior to study entry. This is a moot point for the group of patients that used itself as control, but may have made a difference in the other six patients.

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

After study entry, the groups were treated equally in that no additional inotropic agents or ventilatory parameters were changed during the study. All placebo patients required reloading at the 1 hour evaluation point due to persistent clinical hypoperfusion and failure to achieve CI 20% above baseline. Nine of the milrinone study periods required reloading with milrinone, four due to a 20% or less increase in CI at 1 hour, and five due to clinical hypoperfusion. (This reloading protocol was built into the study a priori.)

III. What were the results?

1. How large was the treatment effect?

A summary of the significant changes [*p < 0.05] is given in the table below (presented as mean(SD)) for placebo (n=6) and milrinone treated patients (n=12):

Parameter/group	Baseline	0.5 hour	1 hour	2 hours
CI (L/min/m2)
placebo	3.2(0.6)	3.1(0.7)	3.3(0.8)	3.2(0.5)
milrinone	3.7(0.8)	4.8(1.2)*	5.1(1.7)*	5.5(1.6)*
SVRI (dynes*s*cm5/m2)
placebo	1549(282)	1521(307)	1447(301)	1582(309)
milrinone	1402(329)	1092(354)*	1098(419)*	982(390)*
SVI (mL/m2)
placebo	24.5(9.1)	23.5(8.9)	25.3(10.0)	24.2(8.1)
milrinone	24.0(6.0)	30.0(7.9)*	33.0(10.2)*	35.3(9.8)*
PVRI (dynes*s*cm5/m2)
placebo	301(187)	378(339)	337(320)	365(310)
milrinone	361(224)	275(182)*	282(220)*	247(144)*

No significant change in CI, SVRI, PVRI, SVI, RVSWI, LVSWI, HR, SBP, DBP, MAP, MPAP, PCWP, Qs/Qt, DO2, VO2, or LVSWI/LVEDVI during the initial period in the placebo treated patients, while there were significant (> 20%) increases in CI at 0.5, 1, and 2 hour study periods, and significant decreases (< 20%) in SVRI at 0.5 and 2 hour study periods in milrinone treated patients compared to placebo patients. Additionally, there were statistically significant differences in PVRI, SVI, RVSWI, LVSWI, SBP, MPAP, PCWP, and DO2 in the milrinone treated patients compared to control, with statistically significant differences in SVI, Qs/Qt, and DO2 between the milrinone periods compared to control patients during the initial placebo-only 2 hour period.

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

Although no confidence intervals were stated, there was significant precision in the estimate of the treatment effect based on a comprehensive table of hemodynamic parameters obtained through the study. Although the standard deviations of the physiologic measurements were somewhat large, there were statistically significant differences both over time within the milrinone-treated patients (ANOVA) and between the treated and placebo patients (unpaired t tests). If repeated with a similar group of patients we would expect to see similar results.

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

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

A qualified yes (see comments), since I deal with a large number of pediatric patients in septic shock. Since early pediatric septic shock is hyperdynamic, however, these results would only be applicable to these patients after the hyperdynamic state has resolved, and elevated SVRI with inadequate CI become apparent.

2. Were all clinically important outcomes considered?

A qualified no (see comments). The goal of this study was to compare changes in cardiac index and systemic vascular resistance index, not the ultimate clinical status (and improvement in clinical status) of the patient.

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

Yes, if the improvements in CI and SVRI persist and are coupled with improvements in multi-organ system function, clinical status, and reductions in morbidity and mortality. No adverse effects were observed during the study period, but the study was small and brief. Larger, longer-term studies are needed to better determine the risk-benefit ratio.

Reviewer's Comments

A few comments are in order. As a vasodilator, milrinone can affect both pulmonary and systemic vascular beds, so the changes in PVRI, RVSWI, PCWP and LVEDVI may reflect changes of pulmonary vascular resistance apart from milrinone's effect on the systemic vasculature.

Milrinone may inhibit the production or release of tumor necrosis factor-alpha, as seen with pentoxifylline, another phosphodiesterase inhibitor. The reduction of TNF-alpha induced myocardial depression by milrinone may be a factor in the hemodynamic improvements seen (1,2).

This study is one of a large number of studies that have evaluated the effect of a treatment vs. placebo in septic shock by comparing changes in physiologic parameters or circulating/tissue levels of inflammatory mediators. Studies have shown an association between pro-inflammatory cytokines (including TNF-alpha, platelet activating factor, IL-1 and IL-6), acute phase reactants (including, fibrinogen, c-reactive protein, complement activation products, erythropoietin, and thrombomodulin), and severity of pediatric sepsis and septic shock (3-7). Similarly, several studies have shown a correlation between physiologic parameters (cardiac index, systemic vascular resistance, oxygen extraction) and severity of septic shock (8,9). It would seem obvious that interventions that reduce the levels of pro-inflammatory mediators, or those interventions that antagonize the adverse hemodynamic changes should decrease morbidity and mortality in septic shock, as differences in physiologic variables have been demonstrated in survivors vs. nonsurvivors of pediatric septic shock (10).

However, studies that have targeted these surrogate endpoints (levels of inflammatory mediators, alterations in physiologic parameters) have not demonstrated a consistent improvement in survival, irrespective of the intervention (steroids, anti-cytokine or anti-endotoxin antibodies, plasma filtration, or other antiinflammatory therapies, inotropic support or increased oxygen consumption based on Swan-Ganz catheter measurements) (11-13).

Two observations are clear: first, as we understand more fully the complex interplay of pro- and anti-inflammatory mediators, the time course of hemodynamic alterations (14), and the response to clinical interventions in early and late sepsis and septic shock, our therapies will become more refined and effective. Second, large clinical trials are necessary to determine clinical and statistical significance of therapies for septic shock.

References

1. Wu CC, Liao MH, Chen SJ, Yen MH. Pentoxifylline improves circulatory failure and survival in murine models of endotoxaemia. Eur J Pharmacol 1999 May 28;373(1):41-9. [abstract]
2. Del Moral T, Goldberg RN, Urbon J, Suguihara C, Martinez O, Stein-Streilein J, Feuer WJ, Bancalari E. Effects of treatment with pentoxifylline on the cardiovascular manifestations of group B streptococcal sepsis in the piglet. Pediatr Res 1996 Sep;40(3):469-74. [abstract]
3. Hazelzet JA, van der Voort E, Lindemans J, ter Heerdt PG, Neijens HJ. Relation between cytokines and routine laboratory data in children with septic shock and purpura. Intensive Care Med 1994 May;20(5):371-4. [abstract]
4. Krafte-Jacobs B, Brilli R. Increased circulating thrombomodulin in children with septic shock. Crit Care Med 1998 May;26(5):933-8. [abstract]
5. Krafte-Jacobs B, Bock GH. Circulating erythropoietin and interleukin-6 concentrations increase in critically ill children with sepsis and septic shock. Crit Care Med. 1996 Sep;24(9):1455-9. [abstract]
6. Dirkes K, Harris BH, Connolly RJ, Schwaitzberg SD, Dinarello CA, Gelfand JA. Platelet activating factor-antagonist improves survival in experimental staphylococcal septicemia. J Pediatr Surg. 1994 Aug;29(8):1055-7. [abstract]
7. Ceneviva G, Paschall JA, Maffei F, Carcillo JA. Hemodynamic support in fluid-refractory pediatric septic shock. Pediatrics 1998 Aug;102(2):e19. [abstract] [full-text]
8. Carcillo JA, Pollack MM, Ruttimann UE, Fields AI. Sequential physiologic interactions in pediatric cardiogenic and septic shock. Crit Care Med 1989 Jan;17(1)12-6. [abstract]
9. Pollack MM, Fields AI, Ruttimann UE. Distributions of cardiopulmonary variables in pediatric survivors and nonsurvivors of septic shock. Crit Care Med 1985 Jun;13(6):454-9. [abstract]
10. Natanson C, Esposito CJ, Banks SM. The siren's songs of confirmatory sepsis trials: Selection bias and sampling error (editorial). Crit Care Med 1998 Dec;26(12):1927-1931. [citation]
11. Schetz M, Ferdinande P, Van den Berghe G, et al: Removal of pro-inflammatory cytokines with renal replacement therapy: Sense or nonsense? Intensive Care Med 1995;21:169-176. [citation]
12. Yu M, Levy MM, Smith P, Takiguchi SA, Miyasaki A, Myers SA. Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective, randomized, controlled study. Crit Care Med 1993 Jun;21(6):830-8. [abstract]
13. Martinez MA, Pena JM, Fernandez A, Jimenez M, Juarez S, Madero R, Vazquez JJ. Time course and prognostic significance of hemostatic changes in sepsis: relation to tumor necrosis factor-alpha. Crit Care Med 1999 Jul;27(7):1303-8. [abstract]

Comments

Back to the EB Journal Club Index