worksheet for evidence-based review of science for emergency

C2010 Worksheet:Peds-024A 12-Oct-2009.doc of 22

WORKSHEET for Evidence-Based Review of Science for Emergency Cardiac Care Worksheet author(s) Date Submitted for review:

Oct 9, 2009

Clinical question.

In pediatric patients with ROSC after cardiac arrest (pre-hospital [OHCA] or in-hospital [IHCA]) who have signs of cardiovascular dysfunction (P),

does the use of any specific cardioactive drugs (I) as opposed to standard care (or different cardioactive drugs) (C), improve physiologic endpoints

(oxygen delivery, hemodynamics) or patient outcome (eg, survival to discharge or survival with favorable neurologic outcome) (O)?

Is this question addressing an intervention/therapy, prognosis or diagnosis? Therapy

State if this is a proposed new topic or revision of existing worksheet: Revision of existing worksheet

Conflict of interest specific to this question

Do any of the authors listed above have conflict of interest disclosures relevant to this worksheet? No

Search strategy (including electronic databases searched).

All searches build on the previous worksheet for this topic performed by this worksheet author for C2005, who used the same search

strategy (ie. Current searches limited to post-Jan 2004)

PubMed A. ("Cardiopulmonary Bypass"[Mesh] OR "Cardiac Surgical Procedures"[Mesh] OR "Myocardial Stunning"[Mesh])

AND

("Vasoconstrictor Agents"[Mesh] OR "Cardiotonic Agents"[Mesh] OR "Vasodilator Agents"[Mesh])

(no difference in hits when searched for all individual cardiotonics and pressors together as ‘OR’)

Limited to: All Infant: birth-23 months, Preschool Child: 2-5 years, Child: 6-12 years

70 hits

B. ("Heart Arrest"[Mesh] OR "Cardiopulmonary Resuscitation"[Mesh]) AND ("Vasoconstrictor Agents"[Mesh] OR "Cardiotonic Agents"[Mesh]);

no limits

173 hits

Combination of A) and B) thinned to 6 relevant citations

When strategy A) searched selecting adults, and children over 12 years age (and limiting for high quality papers, ie. RCTs and metanalyses),

114 hits with 10 selected for further review

EMBASE

1. vasoconstrictor Agent

2. Phosphodiesterase Inhibitor

3. milrinone

4. amrinone

5. hypertensive Agent

6. levosimendan

7. adrenalin

8. dobutamine

9. dopamine

10. inotropic Agent

11. heart arrest

12. Heart Surgery

13. Cardiopulmonary Bypass

14. myocardial stun

15. cardiopulmonary resuscitation

16. 11 or 12 or 13 or 14 or 15

17. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10

18. 16 and 17

81 hits: 2 hits supplemental to PubMed search (2007 Mayr paper, 2006 Egan paper)

All Cochrane databases 1. heart arrest

2. cardiopulmonary arrest

3. cardiopulmonary resuscitation

4. myocardial stun

5. cardiopulmonary bypass


6. cardiac surgery

7. heart surgery

8. 1 or 2 or 3 or 4 or 5 or 6 or 7

9. milrinone

10. amrinone

11. levosimendan

12. adrenalin

13. dobutamine

14. dopamine

15. norepinephrine

16. vasopressin

17. inotropic agent

18. cardiotonic

19. vasopressor

20. 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19

21. 8 and 20

47 hits; none relevant

Hand search of papers’ references: 2 hits (Vasquez and Nijhawan)

Web of Science of major articles (post-arrest): 0 supplemental hits

AHA Master library: 0 supplemental hits

The search strategies were run again on Oct 1, 2009 with no additional hits found.

State inclusion and exclusion criteria

• Only searched literature published since Jan 1, 2004

• Adult studies examining pre-CPB drug use or on-CPB were excluded from analysis as it was felt inappropriate to extrapolate findings from

these studies to the post-cardiac arrest setting

• Case reports and editorials excluded

• Number of articles/sources meeting criteria for further review: Since 2005 worksheet 13 LOE 5 I have included relevant ‘high impact’ studies from the 2005 worksheet 6 LOE 5


Summary of evidence

Evidence Supporting Clinical Question

Good

Hoffman 2003 E

Huang 1, 2005 AB

Huang 2, 2005 AB

Jorgensen, 2008 E

Lobato, 2005 E

Kern 1997E

Meyer 2002 E

Niemann 2003 E,A

Nijhawan, 1999 E

Stocker, 2007 E

Studer, 2005 ABE

Vasquez, 2004 E

Wang 2005 ABE

Ward 1982 E

Ward 1984 E

Fair

Alvarez, 2006 E

Duggall, 2005 E

Poor

Egan, 2006 E

Mayr, 2005 E

1 2 3 4 5

Level of evidence

A = Return of spontaneous circulation C = Survival to hospital discharge E = Other endpoint

B = Survival of event D = Intact neurological survival Italics = Animal studies


Evidence Neutral to Clinical question

Good

Fair

Poor

1 2 3 4 5

Level of evidence



Evidence Opposing Clinical Question

Good

Fair

Poor

1 2 3 4 5

Level of evidence




REVIEWER’S FINAL COMMENTS AND ASSESSMENT OF BENEFIT / RISK:

Animals and humans that are successfully resuscitated from cardiac arrest often have biventricular myocardial dysfunction (systolic and diastolic)

that is transient. Hemodynamics improve with the short term use of cardiotonic agents after successful resuscitation (Kern 1997, Meyer 2002,

Studer 2005). Whether this improves long-term survival, let alone neurologic prognosis, is far from clear. Controversy continues as to what the

best cardiotonic agent is to use in this setting, as well as what the desired hemodynamic target is (a specific cardiac output or blood pressure,

diastolic function, systolic function, reduced myocardial 02 consumption, etc). Some investigators have described a post-resuscitation vasodilatory

state (low systemic vascular resistance or SVR) that might respond to vasopressor support as opposed to primary inotropic support (Mayr, 2007).

Animal studies have not documented this post-resuscitation condition, and if anything (extrapolating from the post-CPB literature), a high SVR

state as opposed to a low SVR state is a more common outcome in survivors of cardiopulmonary arrest. Further investigation of this post-

resuscitative state is needed before a vasopressor-based treatment strategy can be recommended.

Most human studies of post-cardiac arrest myocardial dysfunction and its treatment have come from the cardiac surgical literature. Variations in

study design make it difficult to compare the results of many of these studies, let alone extrapolate their results to the post-cardiac arrest setting.

• These studies have examined the myocardial dysfunction that ensues after a period of cardioplegia and cardiac arrest. The use of a controlled

systemic circulation while on cardiopulmonary bypass (CPB) allows for most organ systems to continue to be perfused while the heart is

rendered ischemic. In the post-cardiac arrest setting, it is not only the heart, but all body organ systems that have sustained a hypoxic-ischemic

injury.

• Much of the cardiac surgical literature has studied the post-bypass response to inotropes in patients with ‘good’ pre-operative cardiac

function, and not patients in pre-cardiac arrest low cardiac output (typical of most patients post-cardiac arrest).

• Is it appropriate to compare the results of a study that examines starting an inotrope before CPB/ cardioplegia, to a study examining the use of

the same inotrope after separation from CPB?

• Is it appropriate to extrapolate the results of studies that examine loading a patient with a drug while they are still hemodynamically supported

on CPB (especially drugs that have significant effects on patient hemodynamics), to the ‘fragile’ post-cardiac arrest patients?

• Animal studies investigating myocardial dysfunction post-cardiac arrest have dealt predominantly with cardiac arrest following VF, and not

asphyxial arrest. It is unclear whether the myocardial dysfunction seen after non-asphyxial insults is the same as that seen after the

progressive hypoxemia/ ischemia culminating in asphyxial cardiac arrest (as is seen in the vast majority of pediatric cardiopulmonary arrests).

• only one pediatric animal study exists, and this is a CPB model of myocardial dysfunction. There is no published pediatric (animal or human)

study specifically investigating the role of inotropes after cardiac arrest (asphyxial or non-asphyxial).

The animal literature suggests that the use of drugs with inotropic, afterload reducing and luciotropic properties might benefit the patient (and

his/her heart) after cardiac arrest. Post-VF arrest studies in pigs suggest that dobutamine can fulfill all of these roles, but that doses higher than 5

mcg/ kg/ min may have an associated metabolic cost of increased tachycardia and myocardial 02 consumption that could exceed available

myocardial 02 delivery (Vasquez, 2004). Studies of the pediatric patient after CPB suggest that the early (post-injury) use of agents such as

milrinone can reduce the likelihood of the patient progressing into a low cardiac output state after the injury created by cardiotomoy and

cardioplegia/ myocardial ischemia (Hoffman, 2003). Human (adult post-CPB) literature suggests that either low dose epinephrine or milrinone

may achieve the same endpoints of improved cardiac output and reduced SVR, but that the efficacy of either of these agents as luciotropes may be

limited to a short period of time (ie. several hours) post-CPB (Lobato, 2005).

The majority of animal studies of post-cardiac arrest cardiotonic use (since the C2005 worksheet) have focused on leveosimendan, a calcium-

sensitizing agent with inotropic, luciotropic and vasodilatory effects. Porcine post-VF arrest studies suggest that in comparing leveosimendan (L)

and dobutamine (D), L has less associated tachycardia, equivalent inotropy and cardiac output, but improved luciotropy over D (Huang, 2005).

These same investigators have performed one of the few animal studies investigating longer term survival (72 hrs) in an animal (rat) VF model,

and found a survival benefit to both L and D over the control group, but an increased survival benefit specifically with the use of L.

A study in a piglet CPB model (Stocker, 2007) compared the ease of separation from CPB using either L (80 mcg/ kg load and 0.7 mcg/ kg/ min

infusion) or M (2 mcg/ kg/ min). Diastolic function, contractility and the degree of vasodilation were superior with the use of L as compared to M.

Varied dosing strategies for levosimendan have been investigated in adults post-CPB (Nijhawan, 1999, Alvarez 2006, Jorgensen 2008). Concern

remains that with the use of higher doses of L (36 mcg/kg bolus followed by infusion of 0.3 mcg/ kg/ min) that the benefits of increased cardiac

output and systemic vasodilation might be outweighed by drug-related tachycardia (baroreceptor-related?) and an associated myocardial metabolic

burden (ie. increased 02 consumption) (Nijhawan, 1999). The pediatric post-operative cardiac literature studying L use is limited, and addresses

multiple patient subgroups simultaneously (single ventricle, biventricular, R or L sided congenital heart defects, large age ranges within studies,

pre or post-CPB use), making it difficult to interpret this data, let alone extrapolate it to use in infants/ children after cardiac arrest in the non-CPB

setting (Egan, 2006).

Animal studies have investigated the role of agents such as propanolol in reducing myocardial 02 consumption in the post-VF arrest state. A

concern has been that the negative inotropic effects of this agent might counteract any benefit gleaned by improving the myocardial 02

delivery/consumption balance. An adult pig study (Wang, 2005) combined the use of L and propanolol in a VF cardiac arrest model. It found that

pre-defibrillation dosing with propanolol helped to reduce the number of shocks necessary to defibrillate, using a significantly lesser total energies

of shocks, and leading to significantly reduced post-resuscitation premature ventricular beats (PVB) and associated ST-changes. The treatment of

animals with levosimendan after successful resuscitation increased the animals’ ejection fraction beyond that seen with either placebo animals or

animals that were only treated intra-arrest with propanolol.

As stated in the C2005 worksheet, there is a need to study the use of cardiotonic support in children after cardiac arrest. For the time being,

conclusions must be extrapolated from the post-operative cardiac literature and animal models. Short term goals (such as hemodynamic


parameters, venous oxygen saturation, lactate concentration or other correlates of cardiac output) will continue to be targeted without knowing

what the best parameters are to use as therapeutic endpoints. Until definitive data exists suggesting improved outcomes with specific drug regimes,

the choice of cardiotonic agents to support the child post-resuscitation will continue to be physician and patient-dependent, weighing the relative

benefits/ side effects of the various cardiotonic agents, and the clinical setting that the practitioner is operating in.

Conclusion

DISCLAIMER: Potential possible wording for a Consensus on Science Statement. Final wording will differ due to other input and discussion.

CONSENSUS ON SCIENCE:

• Evidence from 2 LOE-5 studies in children (Hildebrand, 1988, 331*; Checchia, 2003, 131*), 2 LOE-5 studies in adults (Laurent, 2002*,

2110; Mayr, 2007, 35) and 2 LOE-5 animal studies (Kern, 1997, 2610, Meyer, 2002, 187) indicate that myocardial dysfunction and

(potentially) vascular instability are common after resuscitation from cardiac arrest. (Note: *studies were cited in the 2005 CoSTR statements

from the 2005 worksheet, and have not been reanalyzed in this worksheet)

• Evidence from 6 LOE-5 animal studies (Huang, 2005, 256;Huang, 2005, 487; Kern, 1997, 2610, Meyer, 2002, 187, Studer, 2005, 227;

Vasquez, 2004, 199) documents consistent improvement in hemodynamics when selected vasoactive drugs are given in the post–cardiac

arrest period.

• Evidence extrapolated from multiple adult (LOE-5) and pediatric studies (Hoffman, 2003, 996) of cardiovascular surgical patients with low

cardiac output documents consistent improvement in hemodynamics when vasoactive drugs are titrated in the period after cardiopulmonary

bypass

TREATMENT RECOMMENDATION (unchanged from C2005): Vasoactive drugs should be considered to improve hemodynamic status in the post–cardiac arrest phase. The choice, timing, and dose of specific

vasoactive drugs must be individualized and guided by available monitoring data.

After presentation to the Pediatric Task Force in Osaka in March 2009, the CoSTR statement was modified to the following

DISCLAIMER: Potential possible wording for a Consensus on Science Statement. Final wording will differ due to other input and discussion. CONSENSUS ON SCIENCE:

• Evidence from two (LOE-5) studies in children (Hildebrand, 1988, 331*; Checchia, 2003, 131*) and two (LOE-5) studies in adults

(Laurent, 2002, 2110; Mayr, 2007, 35) and two animal studies (Kern, 1997, 2610; Meyer, 2002, 187) documented that myocardial

dysfunction and vascular instability following resuscitation from cardiac arrest is common.

• Evidence from six (LOE-5) animal studies (Huang 1, 2005, 256 ; Huang 2, 2005, 487; Kern, 1997, 2610; Meyer 2002, 187; Studer, 2005,

227; Vasquez, 2004, 199) documented hemodynamic improvement when vasoactive drugs (dobutamine, milrinone, levosimendan) were

given in the post-cardiac arrest period.

• Evidence from multiple adult (LOE-5) and one (LOE-5) large pediatric study (Hoffman, 2003, 996) of patients with low cardiac output

syndrome following cardiac surgery documented consistent improvement in hemodynamics when selected vasoactive drugs were

prescribed.

• There are no studies evaluating the use of vasoactive agents after ROSC in children

TREATMENT RECOMMENDATION

Acknowledgements: Nil

Citation List

Alvarez, J., M. Bouzada, et al. (2006). "[Hemodynamic effects of levosimendan compared with dobutamine in patients with low cardiac output after cardiac

surgery]." Rev Esp Cardiol 59(4): 338-45

INTRODUCTION AND OBJECTIVES: Levosimendan is an inotropic agent that is effective in the treatment of heart failure. However, experience

with levosimendan in patients with reduced cardiac output following cardiopulmonary bypass is limited. The objective of this study was to compare

the short-term hemodynamic effects of levosimendan with those of dobutamine in managing low cardiac output after cardiac surgery. METHODS:

Forty-one patients who had low cardiac output after cardiopulmonary bypass were randomly assigned to dobutamine (n=20), 24-hour infusion of 7.5

microg/kg per min, or levosimendan (n=21), at a loading dose of 12 microg/kg followed by 24-hour infusion of 0.2 microg/kg per min. The

following parameters were determined during a 48-hour observation period: arterial, central venous, pulmonary arterial and pulmonary capillary

wedge pressure, cardiac index, heart rate, stroke volume, and systemic and pulmonary vascular resistance. RESULTS: Although both dobutamine

and levosimendan improved the cardiac index, the increase was significantly greater with levosimendan (2.4 [0.2] l/min per m2 vs 2.9 [0.3] l/min

per m2, respectively, at 24 h; P<.05). Moreover, levosimendan significantly reduced systemic and pulmonary vascular resistance, and significantly


decreased systemic arterial, pulmonary arterial, pulmonary capillary wedge, and central venous pressure. CONCLUSIONS: Both dobutamine and

levosimendan are effective in managing postoperative low cardiac output. However, levosimendan induces non-specific systemic, venous and

pulmonary vasodilation which can result in hypotension as a adverse event. In these patients, it is advisable to omit or reduce the loading dose.

An open randomized trial in adults post-CPB (n=41) with low cardiac output syndrome or LCOS (CI <2.2L/ min/ m2 and PCP> 15 mm Hg) comparing the

use the use of dobutamine (D) (7.5 mcg/ kg/ min) to Levosimendan (L)(loading dose of 12 mcg/ kg then infusion of 0.2 mcg/ kg/ min) started upon

ICU admission and continued for 24 hrs. Patients were excluded from the study for protocol violation, including changes in doses of L or D, need to

continue agents for >24 hrs for LCOS, need to add other inotropes or vasoactive agents.

Main findings

1) HR rose with the use of both agents, but became significantly higher in the L group by 24 hrs post-op (as the D group HR began to fall once D therapy was

stopped); L has metabolites that appear to linger for longer periods of time

2) CI rose to higher levels in the L group, but took up to 12 hrs to reach higher levels than D group, and remained higher past the discontinuation of study

drugs at 24 hrs (retention of metabolites in L group as above)

3) MAP was persistently higher in D group even out to 48 hrs (24 hrs past d/c of study drugs); right and left-sided cardiac filling pressures, ad SVR, remained

lower in L group out to 48 hrs (prolonged vasodilatory effect)

4) SV02 was higher in L group as of 6 hrs, and continued as such out to 48 hrs; D02 was higher in L group as of 6 hrs, and remained higher than D group

out to 24 hrs

Limitations

1) small study

2) although randomized, no blinding

3) no comment made as to S/A of therapy (eg. hypotension and need for fluid requirements in either group); they do allude to hypotension in their conclusions

of their abstract (ie. reduce the loading dose of L to reduce its effect), but do not give data to support their concerns

4) non-pediatric model

5) non-asphyxial model

6) non-cardiac arrest model

LOE: 5

Quality: Fair

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Duggal, B., U. Pratap, et al. (2005). "Milrinone and low cardiac output following cardiac surgery in infants: is there a direct myocardial effect?" Pediatr

Cardiol 26(5): 642-5

We assessed the effect of milrinone on myocardial function in pediatric patients with postoperative low cardiac output syndrome by index of

myocardial performance in a prospective, open-label, nonrandomized, consecutive study. Fifteen patients with low cardiac output syndrome

following cardiac surgical treatment were studied in the tertiary cardiothoracic pediatric intensive care unit between April 2001 and November 2003

(age range, 0.2-16 months; median, 7; weight, 2.7-11.8 kg; median, 5). Echocardiographic, Doppler-derived, time interval-based index of

myocardial performance (Tei index) was used to study cardiac function prior to and while on intravenous milrinone treatment for 18-24 hours.

Treatment with milrinone led to improvement in biventricular myocardial function [mean right ventricular index from 0.521 (SD-0.213) to 0.385

(SD-0.215), p = 0.003; mean left ventricular index from 0.636 (SD-0.209) to 0.5 (SD-0.171), p = 0.012). No difference was found in the values of

heart rate corrected right or left ventricular ejection time prior to and while on treatment with milrinone (right ventricle: mean, 1.23 (SD-0.42) and

1.14 (SD-0.48), p = 0.29; left ventricles: mean, 1.17 (SD-0.51) and 1.13 (SD-0.48), p = 0.66) Our data support the direct myocardial effect of

milrinone as part of the mechanism behind its already proven benefit in children with low cardiac output syndrome following cardiac surgery.

A case series (n=15) describing the echo-based measurement of ‘myocardial performance index’ in infants treated with open label milrinone post-cardiac

surgery for low cardiac output syndrome (clearly defined in the paper). Patients had serial echos post-operatively, at ~3 hrs and 18-24 hrs post-

operatively. All patients had milrinone started after the first echo (no loading dose, infusion between 0.3-0.6 mcg/ k/ min). Biventricular function

improved with milrinone treatment as measured by the R and L myocardial performance index’ demonstrate improved contractility, and not just a

vasodilatory effect.

Limitations of the paper include:

• Case series nature (absence of a control group)

• Wide range of infusion dosing

• Many concurrent factors are not accounted/ controlled for that could have varied between groups

o Cross-clamp and CPB times

o Use of intra-operative hypothermia

o Use of and amount of inotropes/ catecholamines in the post-operative phase

LOE: 5

Quality: Fair

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Egan, J. R., A. J. Clarke, et al. (2006). "Levosimendan for low cardiac output: a pediatric experience." J Intensive Care Med 21(3): 183-7


This was a retrospective observational study in a pediatric intensive care unit, in which 19 patients received levosimendan. There were no adverse

events attributable to levosimendan and no instances where the clinical condition worsened after administration. Arterial lactate levels decreased

significantly following levosimendan administration during cardiopulmonary bypass for anticipated low cardiac output. In those with established

low cardiac output, trends toward improved hemodynamics were seen, with heart rate reduction, an increase in mean blood pressure, a reduction in

arterial lactate, and reduced conventional inotrope use. Levosimendan was safely used in a small number of pediatric patients with established low

cardiac output state who demonstrated improved hemodynamics and tissue perfusion, with a tendency to reduced conventional inotrope usage, and

this warrants its evaluation as an inotrope in the pediatric population.

A retrospective cohort study describing the patient characteristics surrounding the use of levosimendan (L) in the intra- and post-CPB care of children having

congenital heart surgery. There were 19 patients described over a 32-month period that were treated with L. Hemodynamic data was collected at 3

time points: the hour prior to starting treatment with L, 4-hrs after starting L-infusion and 1-hr after the 24-hr infusion of L finished. The authors

retrospectively divided the patients into 2 groups; one group (n=6) where L had been started empirically before separation from CPB, as the

patients had been identified by the caregivers as being ‘high-risk’; the second group (n=13) had L started after separation from CPB as rescue

therapy (no set criteria). Of the patients treated, 1/6 of the ‘emprical treatment’ group did not have hemodynamic data collected fro the study, while

5/13 of the ‘rescue therapy’ group did not contribute hemodynamic data to the study. Drug dosing/ delivery was inconsistent (4/19 patients did not

receive a bolus dose of L).

The limited data shows that heart rate increased and then fell post-op, and that blood pressure fell and then rose, in the ‘rescue therapy’ group. Looking at

the ‘empiric group’, lactates rose and then fell post-op. Other hemodynamic changes are described, but these are just trends, and do not reach

statistical significance. The authors describe this as a safety study.

Weaknesses/ limitations of the study include:

• In the absence of a control group, these findings potentially just reflect the natural post-CPB changes of these variables, with no relevance to the

use of L

• The study conclusions are severely limited, even as a safety study, by the small numbers

• It is unclear how exactly the patient groups differ other than timing of dosing, as the criteria used to justify starting L are not specified (‘refractory

to conventional inotropes’)

• There is a significant fall-out of patients from the data collection that raise concern as to the validity of the data

• Relevance to post-asphyxial cardiac arrests such as occurs in most pediatric cardiac arrests?

• Many concurrent factors are not accounted/ controlled for that could have varied between groups

o Cross-clamp and CPB times

o Use of intra-operative hypothermia

o Use of and amount of inotropes/ catecholamines in the post-operative phase

LOE: 5

Quality: Poor

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Hoffman, T. M., G. Wernovsky, et al. (2003). "Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after

corrective surgery for congenital heart disease." Circulation 107(7): 996-1002.

BACKGROUND: Low cardiac output syndrome (LCOS), affecting up to 25% of neonates and young children after cardiac surgery, contributes to

postoperative morbidity and mortality. This study evaluated the efficacy and safety of prophylactic milrinone in pediatric patients at high risk for

developing LCOS. METHODS AND RESULTS: The study was a double-blind, placebo-controlled trial with 3 parallel groups (low dose, 25-

microg/kg bolus over 60 minutes followed by a 0.25- microg/kg per min infusion for 35 hours; high dose, 75- microg/kg bolus followed by a 0.75-

microg/kg per min infusion for 35 hours; or placebo). The composite end point of death or the development of LCOS was evaluated at 36 hours and

up to 30 days after randomization. Among 238 treated patients, 25.9%, 17.5%, and 11.7% in the placebo, low-dose milrinone, and high-dose

milrinone groups, respectively, developed LCOS in the first 36 hours after surgery. High-dose milrinone significantly reduced the risk the

development of LCOS compared with placebo, with a relative risk reduction of 55% (P=0.023) in 238 treated patients and 64% (P=0.007) in 227

patients without major protocol violations. There were 2 deaths, both after infusion of study drug. The use of high-dose milrinone reduced the risk

of the LCOS through the final visit by 48% (P=0.049). CONCLUSIONS: The use of high-dose milrinone after pediatric congenital heart surgery

reduces the risk of LCOS.

A multicenter prospective double blinded randomized placebo controlled trial investigating the response of children with biventricular cardiac repairs to

milrinone, that were within 90 minutes of PICU arrival post CPB, and at risk for a low cardiac output state. Milrinone at low dose (25 mcg/ kg

bolus over 60 min and infusion at 0.25 mcg/ kg/ min) and high dose (75 mcg/ kg bolus over 60 min followed by infusion at 0.75 mcg/ kg/ min) were

compared to the placebo arm. Progression into a low cardiac output syndrome (LCOS) was the primary endpoint, as defined by set clinical criteria

of worsening cardiac output, doubling of inotrope score, the addition of a new open label inotrope, or the use of ECLS.

There was a significant and progressive reduction in the incidence of post-operative LCOS when comparing placebo and high dose milrinone (although not in

comparing placebo to low dose milrinone). In assessing secondary study endpoints, there was no difference between groups in duration of assisted

ventilation or hospital stay. The incidence of prolonged hospital stay was highest with placebo, followed by high dose and then low dose milrinone

groups. While there was a significant drop in BPs and BPd immediately after the milrinone bolus (both milrinone groups relative to placebo), there

was no significant difference by the 12 hr mark of the study. HR was significantly higher in the treatment arms at 1, 12 and 24 hrs into the study.

There was no significant inter-group differences in hypotension, dysrhythmia or thrombocytopenia.


Limitations of the study are pointed out by the authors. Some patients that would have been eligible for study inclusion were empirically dosed in the OR with

milrinone outside of the study protocol (how many?). Low cardiac output was defined by clinical criteria, and not by objective measurement of

cardiac output (echo or thermodilution technique). Patients were both L and R sided lesions, so potentially may have responded differently to the

study drugs (although they were evenly distributed between study groups, subgroup differences in response may have been hidden in the larger

group analysis). All limitations considered, this is the closed study that exists looking at inotrope usage that would be “the gold standard study”.

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Huang, L., M. H. Weil, et al. (2005). "Levosimendan improves postresuscitation outcomes in a rat model of CPR." J Lab Clin Med 146(5): 256-61

In this study we sought to determine whether a calcium sensitizer, levosimendan, would have a more favorable effect on postresuscitation

myocardial function and, consequently, postresuscitation survival than beta-adrenergic dobutamine. The extreme decrease in survival before

hospital discharge of resuscitated victims is attributed, in part, to postresuscitation myocardial failure, and dobutamine has been recommended for

the management of postresuscitation myocardial failure. We studied a total of 15 animals. Ventricular fibrillation was induced in Sprague-Dawley

rats weighing 450 to 550 g. Cardiopulmonary resuscitation (CPR), including chest compressions and mechanical ventilation, was begun after 8

minutes of untreated cardiac arrest. Electrical defibrillation was attempted after 6 minutes of CPR. Each animal was resuscitated. Animals were

randomized to undergo treatment with levosimendan, dobutamine, or saline-solution placebo. These agents were administered 10 minutes after the

return of spontaneous circulation. Levosimendan was administered in a loading dose of 12 microg kg(-1) over a 10-minute period, followed by

infusion of 0.3 microg kg(-1) min(-1) over the next 230 minutes. Dobutamine was continuously infused at a dosage of 3 microg kg(-1) min(-1).

Saline-solution placebo was administered in the same volume and over the same amount of time as levosimendan. Levosimendan and dobutamine

produced comparable increases in cardiac output and rate of left-ventricular pressure increase. However, administration of levosimendan resulted in

lower heart rates and lesser increases in left ventricular diastolic pressure compared with both dobutamine and placebo. The duration of

postresuscitation survival was significantly greater with levosimendan (16 +/- 2 hours), intermediate with dobutamine (11 +/- 2 hours) and least

with saline-solution placebo (8 +/- 1 hour). Levosimendan and dobutamine both improved postresuscitation myocardial function. However,

levosimendan produced more favorable postresuscitation myocardial function and increased the duration of postresuscitation survival.

This is a RCT in an adult rat model, comparing the use of Levosimendan (L) to dobutmaine (D) to saline control in the post-resuscitative care of animals

resuscitated from VF-cardiac arrest, looking at hemodynamics and survival to 72 hrs as endpoints. Animals were randomized immediately pre-

study into 3 groups (there was no blinding of the investigators). After anesthesia and instrumentation, baseline hemodynamics were measured. An 8

min untreated VF-cardiac arrest followed, followed by 6 min of CPR. A strict resuscitation protocol was followed that did not include adrenergic

agents as part of it. Study drugs were started 10 minutes after ROSC; D (3 mcg/ kg/ min), L (12 mcg/ kg bolus over 10 min, followed by 0.3 mcg/ kg/

min), or saline control. After resuscitation, animals were followed by 72 hrs.

Baseline hemodynamics were comparable between L, D and saline control groups. All animals were successfully resuscitated after comparable resuscitation

times. Significantly greater heart rates were noted in the D group post resuscitation. Although there were no differences in blood pressure between

groups, cardiac index was higher in the D and L groups. Both D and L improved contractile and luciotropic function (preventing the drop in

cardiac output seen post-resuscitation in the control group), but only L produced a significant drop in LVEDP. There was a significantly higher

PaC02 and EtC02 in the dobutamine group, which was felt to be on the basis of increased metabolic activity created via the drug (although the

anaerobic threshold was apparently not crossed, as there was no intergroup differences in mixed venous 02 sats or lactates). Interestingly, PaC02

was consistently <30 in the post-resuscitation control groups, 30-35 in the L group and 35-40 in the D group. No comment is made suggesting

whether minute ventilation was reduced post-resuscitation. There was a significant difference in survival, with L and D being better than control,

but L being even better than D. Outcomes in the best survival group (L) were still poor (16 +/- 2 hrs)

Strengths of the study include a clearly laid out protocol.

Limitations include:

• The initiation of study drug was 10 minutes post ROSC (very soon after resuscitation); is this realistic in most clinical settings?


• Although there is not a statistically significant difference between the groups in regards PaC02, there is evidence of hyperventilation that would not

be in keeping with current clinical practice/ guidelines; did this in any way contribute to the poor outcomes?

LOE: 5

Quality: Good

Supportive

Labeled as Huang 1 in chart ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Huang, L., M. H. Weil, et al. (2005). "Comparison between dobutamine and levosimendan for management of postresuscitation myocardial dysfunction." Crit

Care Med 33(3): 487-91

OBJECTIVE: To investigate the effects of levosimendan, a nonadrenergic inotropic calcium sensitizer, in comparison with adrenergic dobutamine

for the management of postresuscitation myocardial dysfunction following resuscitation from prolonged cardiac arrest. DESIGN: Randomized

prospective animal study. SETTING: Animal research laboratory. SUBJECTS: Male Yorkshire-cross domestic pigs INTERVENTIONS:


Ventricular fibrillation was induced in male domestic pigs weighing between 35 and 40 kg. Cardiopulmonary resuscitation, including precordial

compression and mechanical ventilation, was started after 7 mins of untreated cardiac arrest. Electrical defibrillation was attempted after 5 mins of

cardiopulmonary resuscitation. Each animal was successfully resuscitated without pharmacologic intervention. Resuscitated animals were

randomized to treatment with levosimendan, dobutamine, or saline placebo. The inotropic agents or an equivalent volume of placebo diluents was

administered 10 mins after restoration of spontaneous circulation. Levosimendan was administered in a loading dose of 20 microg.kg over 10 mins

followed by a 220-min infusion of 0.4 microg.kg.min. Dobutamine was infused into the right atrium in an amount of 5 microg.kg.min. Treatment

was continued for a total of 230 mins. MEASUREMENTS AND MAIN RESULTS: Levosimendan and dobutamine produced comparable increases

in cardiac output. However, levosimendan produced significantly greater left ventricular ejection fraction and fractional area changes compared with

dobutamine and saline placebo. CONCLUSIONS: Levosimendan has the potential of improving postresuscitation myocardial function. It is likely to

serve as an alternative to dobutamine as an inotropic agent for management of postresuscitation myocardial dysfunction.

This is a prospective RCT comparing the use of levosimendan (L) 20 microg.kg over 10 mins followed by a 220-min infusion of 0.4 microg.kg.min or

dobutamine (D) 5 microg.kg.min, to saline control for post-resuscitation care of adult pigs with VF-induced cardiac arrest. Immediately prior to

study, animals were randomized to a study group by the sealed envelope method. After induction of VF, animals were left in VF without treatment

for 7 minutes, following which CPR was initiated. Defibrillation was attempted 5 minutes after starting resuscitation using a biphasic waveform

(150J) with 100% success. It is unclear whether any adrenergic agents were necessary to resuscitate the animals. Animals had an infusion of study

drug started 10 min after ROSC, and therapy was continued for a further 230 minutes.

Study groups were equivalent in baseline characteristics. There were no differences in the number of shocks necessary to resuscitate. Hemodynamics

(measured invasively) showed no differences in MAP in drug intervention groups (L and D) in comparison to control. Significantly lower PA

pressures, wedge pressures and right atrial pressures were observed in the L group relative to saline control from the initiation of drug therapy.

Significantly greater cardiac output was demonstrated for both L and D groups compared to placebo when measuring ejection fraction and

fractional area change by transesophageal echocardiogram, but greater as well for the L group compared to D group. Study conclusions were that

L has the potential to be an alternative to D as an inotropic agent for the resuscitation of post-resuscitation myocardial dysfunction.

A well designed animal study with some important limitations:

• Animals were free of heart disease (does this emulate the pediatric model even better than other models?)

• The initiation of study drug was 10 minutes post ROSC (very soon after resuscitation); is this realistic in most clinical settings?

• Short duration for study (for clinical relevance to patients): what would have happened beyond 4 hrs?


• Blood gases were analyzed post-resuscitation, but it is not clear (not reported) whether there were any differences in PaC02 between groups post-

resuscitation (inadvertent hyperventilation? Vent’n rates were apparently not reduced post-resuscitation)

Funding was partially industry-based (Abbott laboratories)

LOE: 5

Quality: Good

Supportive

Labeled as Huang 2 in chart

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Jorgensen K, Bech-Hanssen O, Houltz E, Ricksten S-E. Effects of Levosimendan on Left Ventricular Relaxation and Early Filling at Maintained

Preload and Afterload Conditions After Aortic Valve Replacement for Aortic Stenosis. Circulation. February 26, 2008 2008;117(8):1075-

1081.

Background We determined the effects of levosimendan, a calcium sensitizer, on left ventricular (LV) diastolic function in patients with LV

hypertrophy. Methods and Results In this prospective, randomized, blinded study, 23 patients received either levosimendan (0.1 and 0.2 {micro}g

{middle dot} kg1 {middle dot} min1; n=12) or placebo (n=11) after aortic valve replacement for aortic stenosis. The effects on LV performance,

dimensions, filling patterns, and isovolumic relaxation time, as well as systemic hemodynamics, were assessed by pulmonary artery thermodilution

catheterization and transesophageal 2-dimensional Doppler echocardiography. To circumvent the confounding effects of the levosimendan-induced

hemodynamic changes on Doppler echocardiographic indexes of LV early relaxation, heart rate and mean arterial and central venous pressures were

kept constant during levosimendan/placebo infusion by atrial pacing, vasopressor, and colloid infusions. In the levosimendan group, dose-dependent

increases in cardiac output (28%; P<0.001) and stroke volume (26%; P<0.001) and a decrease in systemic vascular resistance (22%; P<0.001) were

observed. There was a trend for an increase in LV ejection fraction (12%; P=0.058) with levosimendan. There were no significant differences in

systolic, diastolic arterial, or LV filling pressures or LV end-diastolic area between the 2 groups. Isovolumic relaxation time decreased (23%;

P<0.001), as did the deceleration slope of early diastolic filling (45%; P<0.01), whereas peak early diastolic filling velocity (16%, P<0.01) and peak

late diastolic filling velocity (15%; P<0.001) increased after levosimendan compared with placebo. Conclusion Levosimendan, in addition to its

inotropic effects, exerts a direct positive lusitropic effect in patients with LV hypertrophy as it shortens isovolumic relaxation time and improves LV

filling.

Summary

• There was a dose dependent increase in Cardiac output, stroke volume, and a drop in systemic vascular resistance, and a trend towards increased

ejection fraction (how much of the changes were a result of afterload reduction as opposed to inotropy?)

• There is improved diastolic relaxation (luciotropy) post CPB in this group of pts with the use of L

• Well designed study: only limitations of study design include:


o Blinding of echocardiographers/ investigators but not of anesthestists caring for the patients

Limitations of the study include:

• adult CPB model; non-arrested pts

• pts had normal ejection fraction pre-op

• duration of study is only 1 hr post-ischemic insult (separation from CPB): difficult to extrapolate to post-resuscitation care in meaningful way

LOE 5

Quality Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Kern, K. B., R. W. Hilwig, et al. (1997). "Postresuscitation left ventricular systolic and diastolic dysfunction. Treatment with dobutamine." Circulation

95(12): 2610-3.

BACKGROUND: Global left ventricular dysfunction after successful resuscitation is well documented and appears to be a major contributing factor

in limiting long-term survival after initial recovery from out-of-hospital sudden cardiac death. Treatment of such postresuscitation myocardial dysfunction has

not been examined previously. METHODS AND RESULTS: Systolic and diastolic parameters of left ventricular function were measured in 27 swine before

and after successful resuscitation from prolonged ventricular fibrillation cardiac arrest. Dobutamine infusions (10 micrograms.kg-1.min-1 in 14 animals or 5

micrograms.kg-1.min-1 in 5 animals) begun 15 minutes after resuscitation were compared with controls receiving no treatment (8 animals). The marked

deterioration in systolic and diastolic left ventricular function seen in the control group after resuscitation was ameliorated in the dobutamine-treated animals.

Left ventricular ejection fraction fell from a prearrest 58 +/- 3% to 25 +/- 3% at 5 hours after resuscitation in the control group but remained unchanged in the

dobutamine (10 micrograms.kg-1.min-1) group (52 +/- 1% prearrest and 55 +/- 3% at 5 hours after resuscitation). Measurement of the constant of isovolumic

relaxation of the left ventricle (tau) demonstrated a similar benefit of the dobutamine infusion for overcoming postresuscitation diastolic dysfunction. The tau

rose in the controls from 28 +/- 1 milliseconds (ms) prearrest to 41 +/- 3 ms at 5 hours after resuscitation whereas it remained constant in the dobutamine-

treated animals (31 +/- 1 ms prearrest and 31 +/- 5 ms at 5 hours after resuscitation). CONCLUSIONS: Dobutamine begun within 15 minutes of successful

resuscitation can successfully overcome the global systolic and diastolic left ventricular dysfunction resulting from prolonged cardiac arrest and

cardiopulmonary resuscitation.

A prospective ?non-randomized controlled investigation of the effect of dobutamine on post-resuscitation myocardial function after prolonged fibrillatory

cardiac arrest (15 minutes) in an adult swine model. After the fibrillatory period and subsequent ‘ACLS resuscitation’, a group of animals were placed on

dobutamine infusions (5 or 10 mcg/ kg/ min started 15 min after successful resuscitation) and their hemodynamics observed over the ensuing hour. The lower

dose dobutamine group was studied separately without a concurrent control group.

The deterioration in LV systolic and diastolic function was prevented by the use of post-resuscitation dobutamine (greater beneficial drug effects at 10 rather

than 5 mcg/ kg/ min). HR was significantly greater in the dobutamine treated animals(dose dependent) at all times (compared to controls), but stroke volumes

were increased, and wedge pressures significantly reduced in the dobutamine group.

Limitations of the model include the animal nature of the study group, but (regardless of the fibrillatory trigger) it would appear to mimic the hypoxic

ischemic model of the resuscitated child (although, was ventilation stopped at the time of the fibrillatory arrest?).

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Lobato, E. B., J. L. Willert, et al. (2005). "Effects of milrinone versus epinephrine on left ventricular relaxation after cardiopulmonary bypass following

myocardial revascularization: assessment by color m-mode and tissue Doppler." J Cardiothorac Vasc Anesth 19(3): 334-9

OBJECTIVE: The purpose of this study was to evaluate the left ventricular lusitropic effects of epinephrine versus milrinone after cardiopulmonary

bypass. DESIGN: Prospective randomized study. SETTING: Single institution, university teaching hospital. PARTICIPANTS: Adult patients

undergoing coronary artery bypass grafting under cardiopulmonary bypass. INTERVENTIONS: After separation from cardiopulmonary bypass,

patients were randomized to receive intravenous epinephrine by continuous infusion (0.03 microg/kg/min) or milrinone (50 microg/kg followed by

0.5 microg/kg/min). Transesophageal echocardiographic evaluation of left ventricular diastolic function, with emphasis on relaxation, was

performed before and after bypass and after the administration of either epinephrine or milrinone. MEASUREMENTS AND MAIN RESULTS:

Measurements included pulse-wave Doppler analysis of mitral inflow and pulmonary vein and left ventricular outflow tract velocities. Left

ventricular inflow velocity of propagation measured with color M-mode and tissue Doppler assessment of early mitral annulus velocity were used to

evaluate left ventricular relaxation. Values of velocity of propagation and mitral annulus velocity improved significantly after bypass, suggesting

improved relaxation. The administration of either epinephrine or milrinone did not result in further improvement in left ventricular relaxation.

CONCLUSIONS: After cardiopulmonary bypass, left ventricular relaxation was significantly improved. Neither epinephrine nor milrinone

exhibited favorable lusitropic effects after bypass.

A prospective randomized controlled trial comparing the effect of low dose epinephrine infusion (E: 0.03 mcg/ kg/ min; n=12) or milrinone infusion (M: 0.5

mcg/ kg/ min; n=13) in their effects (compared to control (C) receiving no inotropic support separating from CPB (n=11) on echocardiogram-

evaluated diastolic dysfunction in adults after elective CPB for CABG. Patients were intra-operatively (after CPB separation but before protamine

administration) divided into three groups after informed consent and computer-randomization. Nine patients were excluded form study intra-

operatively as they required inotropes in order to successfully separate from CPB. All patients routinely separated from CPB on nitroglycerin 1

mcg/ kg/ min. Echo images were generated pre-CPB, immediately post-separation from CPB, after the study was initiated (5 min post-epinephrine


infusion starting or after loading with milrinone), and after sternal closure (intra-operative). These images were evaluated by an

echocardiographer blinded to patient allocation

Patient group characteristics differ only in that the epinephrine group had a lesser number of patients with chronic hypertension, and a lesser number of

patients on beta-blocker therapy pre-operatively. The cross clamp and CPB times were comparable between groups. The hemodynamic data

showed the following:

• BP was lower post-CPB in the drug intervention groups compared to baseline, but not in comparison to concurrent control patients

• SVR dropped post-CPB in all three groups; the M and E groups had statistically lower SVR in comparison to control patients after drug dosing and

after sternal closure

• Both drug intervention drug groups had a significant increase in cardiac index in comparison to pre-CPB

Echo data however showed the following:

• There was a significant improvement in measured diastolic function in only one of the measured indices but this difference between drug group

resolved within 5 minutes of starting the study drugs; LV relaxation appeared to improve in all patients post-CPB, that was unchanged by the

addition of either study drug; these differences in luciotropy had returned to baseline in all groups by the time of sternal closure

o The only way to observe differences in measured diastolic function as noted above was with the pooling of the M and E group data and comparing

it to the control group. When evaluating individual group differences (ie. M vs E vs control), there were no differences


• Small numbers of patients

• Select patient population

o just looking at changes in luciotropy in patients with normal pre-op systolic function and mild diastolic function; pertinence to the patient with pre-

op moderate or severe dysfunction, let alone post-arrest systolic and diastolic dysfunciton?

o Patients examined only included those that separated from CPB without the need for inotrope support


LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Mayr, V., G. Luckner, et al. (2007). "Arginine vasopressin in advanced cardiovascular failure during the post-resuscitation phase after cardiac arrest."

Resuscitation 72(1): 35-44

Arginine vasopressin (AVP) has been employed successfully during cardiopulmonary resuscitation, but there exist only few data about the effects of

AVP infusion for cardiovascular failure during the post-cardiac arrest period. Cardiovascular failure is one of the main causes of death after

successful resuscitation from cardiac arrest. Although the "post-resuscitation syndrome" has been described as a "sepsis-like" syndrome, there is

little information about the haemodynamic response to AVP in advanced cardiovascular failure after cardiac arrest. In this retrospective study,

haemodynamic and laboratory variables in 23 patients with cardiovascular failure unresponsive to standard haemodynamic therapy during the post-

cardiac arrest period were obtained before, and 30 min, 1, 4, 12, 24, 48, and 72 h after initiation of a supplementary AVP infusion (4 IU/h). During

the observation period, AVP significantly increased mean arterial blood pressure (58 +/- 14 to 75 +/- 19 mmHg, p < 0.001), and decreased

noradrenaline (norepinephrine) (1.31 +/- 2.14 to 0.23 +/- 0.3 mug/kg/min, p = 0.03), adrenaline (epinephrine) (0.58 +/- 0.23 to 0.04 +/- 0.03

mug/kg/min, p = 0.001), and milrinone requirements (0.46 +/- 0.15 to 0.33 +/- 0.22 mug/kg/min, p < 0.001). Pulmonary capillary wedge pressure

changed significantly (p < 0.001); an initial increase being followed by a decrease below baseline values. While arterial lactate concentrations (95

+/- 64 to 21 +/- 18 mg/dL, p < 0.001) and pH (7.27 +/- 0.14 to 7.4 +/- 0.14, p < 0.001) improved significantly, total bilirubin concentrations (1.12

+/- 0.95 to 3.04 +/- 3.79 mg/dL, p = 0.001) increased after AVP. There were no differences in the haemodynamic or laboratory response to AVP

between survivors and non-survivors. In this study, advanced cardiovascular failure that was unresponsive to standard therapy could be reversed

successfully with supplementary AVP infusion in >90% of patients surviving cardiac arrest. copyright 2006 Elsevier Ireland Ltd. All rights

reserved.

A retrospective study of the use of vasopressin infusion for fluid and inotrope/ vasopressor resistant shock after resuscitation from adult cardiac arrest (no

controls). Concurrent with the use of vasopressin, there was an increase in blood pressure and a reduction in the dose of vasopressors and

inotropes necessary to maintain blood pressure (as per protocol). There were no differences in hemodynamics or laboratory response to the use of

vasopressin between survivors and non-survivors. The major limitation of the study is that based upon its case control nature, it is unclear which of

the physiological and laboratory value changes are on the basis of vasopressin or other concurrent factors.

LOE: 5

Quality: Poor

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Meyer, R. J., K. B. Kern, et al. (2002). "Post-resuscitation right ventricular dysfunction: delineation and treatment with dobutamine." Resuscitation 55(2):

187-91.

BACKGROUND: Left ventricular dysfunction after resuscitation from cardiac arrest has been well described. Treatment with dobutamine improves

post-resuscitation left ventricular function. Right ventricular function following resuscitation has not been investigated. The purposes of this study were to

examine right ventricular function following resuscitation and determine whether dobutamine would improve post-resuscitation right ventricular function.

METHODS AND RESULTS: Right ventricular function was measured in 28 swine (29+/-1 kg) before and after resuscitation from 15 min of untreated


ventricular fibrillation. Twelve animals received dobutamine at 10 mcg/kg/min while 16 animals served as untreated controls. Among controls, right

ventricular dysfunction post-resuscitation was demonstrated by a decrease in right ventricular ejection fraction and an increase in right ventricular end-

diastolic pressure. Among animals treated with dobutamine, there was a significant improvement in right ventricular function post-resuscitation compared to

untreated controls. CONCLUSIONS: This study establishes that right ventricular systolic and diastolic dysfunction does occur after prolonged cardiac arrest

from ventricular fibrillation. Dobutamine can ameliorate post-resuscitation right ventricular dysfunction.

A prospective randomized controlled non-blinded trial studying the effects of dobutamine on post-resuscitation RV function in an adult swine model. Animals

had a 15 min fibrillatory arrest, followed by an ‘ACLS resuscitation’. After successful resuscitation, study drug (10 mcg / kg/ min)was infused and

hemodynamics monitored for the ensuing 5 hrs.

RV ejection fraction and RV end-diastolic pressure were significantly improved over the depressed values measured in the control animals (no difference in

hemodynamics in dobutamine animals relative to pre- arrest baseline). HR was significantly elevated above controls at all times in the dobutamine

group.

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Niemann, J. T., D. Garner, et al. (2003). "Milrinone facilitates resuscitation from cardiac arrest and attenuates postresuscitation myocardial dysfunction."

Circulation 108(24): 3031-5.

BACKGROUND: Left ventricular (LV) dysfunction with a low cardiac index after successful CPR contributes to early death attributable to

multiorgan failure, and an effective treatment has not been identified. The purpose of this study was to investigate the use of milrinone, a selective

phosphodiesterase III inhibitor, as treatment for LV dysfunction after resuscitation. METHODS AND RESULTS: Ventricular fibrillation (VF) was induced

electrically in 32 swine. After 5 minutes of VF, CPR was initiated and animals were randomized to receive either saline (control group, n=16) as a bolus and

infusion or milrinone 50 microg/kg as a bolus and then 0.5 microg/kg per min for 60 minutes (treatment group, n=16). After 2 minutes of CPR (total VF time,

7 minutes), countershocks were given. Coronary perfusion pressures during CPR were similar for the groups (24+/-2 versus 21+/-4 mm Hg). All animals were

defibrillated; 6 of 16 control animals developed refractory postcountershock pulseless electrical activity compared with 0 of 16 treated animals (P=0.018). At

30 minutes after restoration of spontaneous circulation, stroke volume (16+/-3 versus 26+/-7 mL, P<0.01) and LV dp/dt (793+/-197 versus 1108+/-316 mm

Hg/s, P<0.02) were higher in the treatment group. Similar differences were observed 60 minutes after restoration of spontaneous circulation. Significant

differences in heart rates between groups were not observed, and peripheral vascular resistance was significantly greater in the control group 30 and 60

minutes after resuscitation. CONCLUSIONS: Milrinone facilitates resuscitation from prolonged VF and attenuates LV dysfunction after resuscitation without

worsening major determinants of myocardial oxygen demand.

A prospective randomized controlled non-blinded trial studying the effects of ‘intra-resuscitation’ milrinone on post-resuscitation hemodynamics/ outcome in

an adult swine model. Animals underwent a relatively short (5 min) fibrillatory arrest before resuscitation. The study group was bloused with

milrinone 50 mcg/ kg at the time that CPR was initiated, and an infusion of 0.5 mcg/ kg/ min started. Hemodynamics were monitored for the ensuing

60 min.

All study drug animals were successfully converted into a perfusing rhythm (as opposed to only 10/ 16 control animals). At 30 min after ROSC, stroke volume

and cardiac output was greater in the milrinone group (although still less than pre-arrest controls). There were no inter-group differences in HR.

SVR was significantly lower in the milrinone group, but without inter-group differences in MAP. By 60 min post-return to ROSC, there was no

statistically significant differences between the milrinone group and the pre-arrest baseline.

Limitations of the study include the relatively short period of cardiac arrest. As well, note should be made that the study investigated the effects of initiating

milrinone while performing CPR, not later after successful resuscitation.

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Nijhawan, N., A. C. Nicolosi, et al. (1999). "Levosimendan enhances cardiac performance after cardiopulmonary bypass: a prospective, randomized placebo-

controlled trial." J Cardiovasc Pharmacol 34(2): 219-28

Levosimendan is a new myofilament calcium (Ca2+) sensitizer that increases myocardial contractility by stabilizing the Ca2+-bound conformation

of troponin C. We tested the hypothesis that levosimendan enhances cardiac performance after cardiopulmonary bypass (CPB). Anesthesia was

induced and maintained with midazolam, sufentanil, and vecuronium in 18 patients randomly assigned to receive levosimendan (18 or 36 microg/kg

loading dose and 0.2 or 0.3 microg/kg/min infusion, respectively) or placebo 15 min before and continued for 6 h after CPB. Significant (p < 0.05)

increases in heart rate (HR) and decreases in systemic vascular resistance (SVR) occurred 15 min after CPB in patients receiving placebo. Later

increases in mean arterial pressure (MAP) and cardiac output (CO) and decreases in stroke volume (SV) and pulmonary vascular resistance also

were observed. HR was greater in patients receiving high- but not low-dose levosimendan versus placebo immediately after CPB. MAP also was

lower in patients treated with either dose of levosimendan compared with placebo after CPB. Levosimendan increased CO and decreased SVR (4.2

+/- 0.4 to 7.9 +/- 0.4 L/min and 1,150 +/- 99 to 512 +/- 42 dyn/s/cm5, respectively, 15 min after CPB; mean +/- SEM). CO and SV were higher and

SVR was lower in patients receiving levosimendan versus placebo after CPB. No differences in arterial oxygenation and perioperative arrhythmias

(Holter analysis) were observed between groups. The results indicate that levosimendan enhances cardiac performance after CPB in humans.


A RCT in adults with normal heart function, undergoing CPB for CABG or valve replacement, treated with 2 dosing strategies of levosimendan (L) compared

to placebo prior to separation from CPB. Hemodynamics were assessed over 6 hrs post-CPB. The study was double-blinded (preparation by

pharmacist, blinded anesthetist). Intraoperative and post-operative cardiopulmonary care was protocolized. Patients were randomized to receive

low dose L (18 mcg/kg load and infusion of 0.2 mcg/ kg/ min) or high dose L (36 mcg/ kg/ min load and then 0.3 mcg/ kg/ min infusion) throughout

the study. The baseline patient demographics were similar, with each group having 6 patients. There were no inter-group differences in cross-clamp

or CPB times.

The study showed that high dose L produced a greater tachycardia than low dose L or placebo (baroreceptor-related?). Both doses of L caused equivalent

drops in MAP and SVR, but rises in stroke volume and cardiac output. There was no greater incidence of dysrhythmias in either L group. Oxygen

delivery and consumption were both greater in the high dose L group. In summary, it was felt that L caused increased cardiac output via increases

in heart rate, reduced afterload and increased stroke volume. On balance, it was felt that the low dose L produced equivalent efficacy compared to

higher dosing.

Study strengths were that it was a well designed study (double blinded placebo-controlled, protocol-driven)

Study limitations include:

• small numbers of patients

• patients with poor LV function were exluded (relevance to post-cardiac arrest victims?)non-asphyxial arrest model in adults

• study drug was delivered prior to going onto and also while on CPB, not just post-CPB

Funding was partially industry-based (Orion Corporation)

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Stocker, C. F., L. S. Shekerdemian, et al. (2007). "Mechanisms of a reduced cardiac output and the effects of milrinone and levosimendan in a model of infant

cardiopulmonary bypass." Crit Care Med 35(1): 252-9

OBJECTIVES: A low cardiac output state is an important cause of morbidity after pediatric cardiopulmonary bypass. The objectives of our study

were to define the early precipitants of the reduced cardiac output and to investigate the effects on these of milrinone and levosimendan in a model

of pediatric cardiopulmonary bypass. DESIGN: Experimental study. SETTING:: Research laboratory at a university-affiliated, tertiary pediatric

center. SUBJECTS: Eighteen piglets. INTERVENTIONS: Piglets, instrumented with systemic, pulmonary arterial, and coronary sinus catheters,

pulmonary and circumflex arterial flow probes, and a left ventricular conductance-micromanometer-tipped catheter, underwent cardiopulmonary

bypass with aortic cross-clamp and cardioplegic arrest. At 120 mins, they were assigned to control, milrinone, or levosimendan groups and studied

for a further 120 mins. MEASUREMENTS AND MAIN RESULTS: In controls, between 120 and 240 mins, cardiac output decreased by 15%.

Systemic vascular resistance was unchanged, but pulmonary vascular resistance increased by 19%. Systemic arterial elastance increased by 17%,

indicating increased afterload. End-systolic elastance was unchanged, and coronary sinus oxygen tension decreased by 4.0 +/- 1.7 mm Hg. In

animals receiving milrinone cardiac output was preserved, and in animals receiving levosimendan cardiac output increased by 14%. Both drugs

prevented an increase in arterial elastance and pulmonary vascular resistance after cardiopulmonary bypass. Systemic vascular resistance decreased

by 31% after levosimendan, and end-systolic elastance increased by 48%, indicating improved contractility. Both agents prevented a decrease in

coronary sinus oxygen tension. CONCLUSIONS: Increased afterload, which is not matched by an equivalent elevation in contractility, contributes

to the reduced cardiac output early after pediatric cardiopulmonary bypass in this model. This increase is prevented by milrinone and levosimendan.

Both agents exert additional beneficial effects on pulmonary vascular resistance and myocardial oxygen balance, although levosimendan has greater

inotropic properties.

Levosimendan is a calcium sensitizing agent. It produces potent inotropic actions by sensitizing myocardial troponin C to calcium and exerts vasodilator

effects through stimulation of the adenosine triphosphate-sensitive potassium channels of systemic, pulmonary, and coronary vascular smooth

muscle cells. This was an animal based RCT examining and comparing the separate influences of milrinone (M) and levosimendan (L) in a post

cardiopulmonary bypass / aortic cross clamped pediatric anima l(pig) model. Animals were placed on CPB, and then underwent a 45 minute period

of aortic cross-clamp. After removal of the cross clamp, animals separated from CPB. All animals (including controls) separated from CPB with

Dopamine 5 mcg/ kg/ min, as well as either milrinone 2 mcg/ kg/ min (after a 100 mcg/ kg load) or levosimendan 80 mcg/ kg, followed by a 0.7 mcg/

kg/ min infusion. Both intervention groups had the drug boluses administered 120 min after removal of the cross clamp. Hemodynamics were

measured for the ensuing 2 hrs.

There were no significant changes in systemic or pulmonary blood pressure during the study period, comparing either group to control. The use of milrinone

prevented the post cross clamp deterioration in cardiac output seen in the control animals, but the levosimendan group actually had an increase in

cardiac output above baseline during the same study period. SVR dropped and remained lower during the study period in L relative to either the

milrinone or control group (with SVR I he M group actually not being significantly different than control). Systemic 02 delivery was maintained

throughout the study period in both M and L while it progressively dropped in the control group, but there was no intergroup difference in

myocardial 02 delivery (coronary sinus 02 was better maintained in both L and M groups). Measures of diastolic function were improved in both M

and L groups relative to controls, with L being even better than M. Measures of contractility were increased in L relative to either M or control

groups. Although PVR was lower immediately after loading with L (compared to M), these inter-group differences disappeared in later time points

Limitations of the study:

1. Milrinone dose chosen is higher than is used clinically in infants

2. It is unclear how long the animals (individually or the respective groups) remained on CPB after removal of the cross clamp, so it is unclear how


long each of the animals had been separated from CPB at the time that the various measurements were performed (consistency within and between

groups?). This has relevance as to whether all animals received the same degree of ischemic insult and are comparable.

3. The duration of study only extended to 4 hrs after the ischemic insult, not extending beyond to the nadir of post-arrest contractility. It is unclear

whether the noted findings would continue through to this time, and consequently how beneficial the observed effects would be when they were

needed most.

4. Relevance to post-asphyxial cardiac arrests such as occurs in most pediatric cardiac arrests?

5. There is apparently a dosing difference between some animal models (such as this one) and the doses used in human clinical practice. How

applicable some of this data is (specifically dosing) directly to the human clinical area is questionable.

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Studer, W., X. R. Wu, et al. (2005). "Influence of dobutamine on the variables of systemic haemodynamics, metabolism, and intestinal perfusion after

cardiopulmonary resuscitation in the rat." Resuscitation 64(2): 227-232

Background: Global left ventricular dysfunction after successful resuscitation from cardiac arrest may be treated successfully with dobutamine but

the effects on intestinal perfusion are unknown. Methods: In 24 male Sprague-Dawley rats ventricular fibrillation was induced. After 4 min of

untreated cardiac arrest, precordial chest compression was performed for 4 min; adrenaline (epinephrine) (90 mug kg(-1)) was injected, followed by

defibrillation. Return of spontaneous circulation was achieved in 18 animals, which were allocated to receive saline 0.9% (control group, n = 6),

dobutamine at 5 mug kg(-1) min(-1) (n = 6) or dobutamine at 10 mug kg(-1) min(-1) (n = 6). Measurements of haemodynamic variables and

intestinal tonometer P-CO2 were made before induction of ventricular fibrillation and 15, 30, 60, and 120 min postresuscitation. Results: At 120

min postresuscitation, mean aortic pressure was 82 +/- 20, 104 +/- 19, and 113 +/- 15 mmHg for the control group, the dobutamine (5 mug kg(-1)

min(-1)) group and the dobutamine (10 mug kg(-1) min(-1)) group (P < 0.05 for comparison of the dobutamine (10 mug kg(-1) min(-1)) group

versus the control group). Respective abdominal aortic blood flow was 107 +/- 16, 133 +/- 49, and 145 +/- 18 ml min(-1) kg(-1) (P < 0.05 for

comparison of the dobutamine (10 mug kg(-1) min(-1)) group versus the control group), and superior mesenteric artery blood flow was 25 +/- 9, 28

+/- 8, and 33 +/- 8 ml min(-1) kg(-1). Arterial lactate was significantly higher (P < 0.05) in the control group (2.3 +/- 0.6 mmol 1(-1)) than in the

dobutamine (5 mug kg(-1) min(-1)) group (1.6 +/- 0.3 mmol 1(-1)) and dobutamine (10 mug kg(-1) min(-1)) group (1.5 +/- 0.3 mmol 1(-1)).

Tonometrically derived P-CO2 gap was highly elevated at 15 min of postresuscitation and returned to prearrest level at 120 min postresuscitation in

all groups. Conclusions: Dobutamine enhances the recovery of global haemodynamic and metabolic variables early after cardiac arrest. (C) 2004

Elsevier Ireland Ltd. All rights reserved.

An RCT evaluating the effects of varied doses of dobutamine on post-resuscitation systemic and regional hemodynamics in a rat model of VF-induced cardiac

arrest. Interestingly, CPR was given for 30 s post-defibrillation, emulating the new resuscitation guidelines.

Main findings

1) Dobutamine (10 mcg/ kg/ min) was more effective in restoring aortic blood flow to pre-arrest levels by 30 min post-resuscitation (compared to control),

most likely through its chronotropic properties

2) superior mesenteric blood flow was maintained at pre-arrest levels throughout the study course in all groups (regardless of the use of inotropic support)

3) Lactates had returned to normal levels in the dobutamine treated animals by 120 minutes, and were sat lower than control at that time

Limitations

1) animal model

2) non-asphyxial arrest model

3) non-pediatric model

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Vasquez, A., K. B. Kern, et al. (2004). "Optimal dosing of dobutamine for treating post-resuscitation left ventricular dysfunction." Resuscitation 61(2): 199-

207

OBJECTIVES: This study was designed to determine the optimal dose of dobutamine in the treatment of post-resuscitation left ventricular

dysfunction. BACKGROUND: Global left ventricular dysfunction following successful resuscitation from prolonged, ventricular fibrillation cardiac

arrest, negatively impacts long-term survival. Dobutamine can overcome this global myocardial stunning. Previous data indicate a dose of 10

mcg/kgmin improves systolic and diastolic function, but markedly increases the heart rate. METHODS: Twenty swine (24 +/- 0.4 kg) were

randomized to one of four doses (0, 2, 5, and 7.5 mcg/kgmin) of dobutamine for the treatment of post-resuscitation myocardial dysfunction

following 12.5 min of untreated ventricular fibrillation cardiac arrest. Cardiac function was measured at pre-arrest baseline and serially for 6 h post-

resuscitation. Left ventricular function was evaluated by contrast ventriculograms, left ventricular pressures, +dP/dt, Tau, -dP/dt, and cardiac output.

Myocardial oxygen consumption and myocardial blood flow were measured to assess the functional significance of any dobutamine-mediated heart

rate responses. RESULTS: Left ventricular dysfunction was evident at 25 min and peaked 4 h post-resuscitation. Significant (P < 0.05)

improvements in ventricular systolic (EF, CO) and diastolic (LVEDP, Tau) function were evident within minutes of dobutamine initiation and

persisted at 6h for the 5 and 7.5 mcg/kgmin groups. Tachycardia manifested with all dobutamine doses, but only affected myocardial oxygen

consumption significantly (P < 0.05) at the highest dose (7.5 mcg/kgmin). CONCLUSIONS: Dobutamine at 5 mcg/kgmin appears optimal for


restoring systolic and diastolic function post-resuscitation without adversely affecting myocardial oxygen consumption.

A RCT in an adult pig VF-cardiac arrest model investigating the effects of varied dosages of dobutamine on post-resuscitation hemodynamics. The study

objective was to identify the optimal dosage of dobutamine needed to improve hemodynamics and cardiac performance, but minimizing the increase

in myocardial 02 consumption associated with increasing dose-related tachycardia. 22 pigs were instrumented after induction of anesthesia, and

then had an electrically-induced VF-cardiac arrest. After a 12.5 minute period of no treatment, the animals were resuscitated by protocol, with 20

of the 22 pigs being successfully revived. Epinephrine was potentially used as part of the protocol, but it is unclear how many of the animals

required it to reestablish ROSC. While the resuscitation protocol is described, it is unclear whether there were any intergroup differences when it

came to amount of resuscitation necessary to reestablish ROSC. Hemodynamics were assessed by invasive monitoring pre-arrest and for a 6 hr

period post-arrest, while being treated with dobutamine (randomly assigned (method not described) to 0, 2, 5, 7.5 mcg/ kg/ min); drug therapy was

initiated 30 minutes after successful resuscitation.

Ejection fraction (EF) (assessed by L ventriculogram) reached its nadir by 4 hrs in control animals; this decrease was modified by the use of dobutamine,

with EF reaching baseline levels by 4 hrs with 5 and 7.5 mcg/ kg/ min of drug, and actually becoming supra-baseline by 6 hrs with the use of 7.5

mcg/ kg/ min. Heart rate (HR) increases were seen with all dobutamine groups relative to control by 6 hrs, with the highest HR being achieved with

7.5 mcg/ kg/ min. Corresponding cardiac output (CO) measurements showed normalization in all dobutamine groups by 35 min post-resuscitation,

but sustained normal CO at 6 hrs occurred only with 5 mcg/ kg/ min, and supra-normal CO at 6 hrs occurred only with 7.5 mcg/ kg/ min. Control

animals showed an initial increase in HR post-resuscitation, but this returned to baseline by 2 hrs post-resuscitation. LVEDP remained elevated in

all study groups (including control) post-resuscitation. Isovolumetric relaxation (IVR) (assessed as Tau by ventriculogram) improved and was

sustained with doses of 5 and 7.5 mcg/ kg/ min, but only 7.5 mcg/ kg/ min increased the speed of IVR to faster than baseline (pre-arrest).

Myocardial 02 consumption increased with all dobutamine groups relative to control, but was significantly greater with 7.5 mcg/ kg/ min. LV stroke

work reached its nadir at 4 hrs post resuscitation, and was not modified by the use of dobutamine.

This eloquent animal study demonstrated that while cardiac output and ejection fraction can be optimized with 7.5 mcg/ kg/ min of dobutamine, the price to

be paid is a high myocardial 02 consumption (despite a high myocardial blood flow) and a lower coronary sinus p02 compared to lower drug

doses. The metabolic price paid is such that the optimal dobutmaine dose to improve hemodynamics/ contractility/ luciotropy and minimize

myocardial metabolic demand is with the use of 5 mcg/ kg/ min.


• the nature of the cardiac arrest (VF vs. asphyxial) for the purposes of extrapolation to the majority of pediatric cardiopulmonary arrests.

• unclear form the paper as to whether the investigators were blinded to group assignments.



• From the study it is unclear whether there are any differences between the groups in regards inter-group resuscitation times and doses of

epinephrine doses necessary to re-establish ROSC

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Wang, J., M. H. Weil, et al. (2005). "Levosimendan improves postresuscitation myocardial dysfunction after beta-adrenergic blockade." J Lab Clin Med

146(3): 179-83

In earlier studies, we found that a nonselective beta-adrenergic blocking agent, propranolol, facilitated cardiac resuscitation, reduced

postresuscitation myocardial ectopy, and improved postresuscitation survival. However, the potential adverse effects and specifically the negative

inotropic actions of propranolol prompted our further investigation of the potential value of a non-beta-adrenergic inotropic drug, levosimendan, in

conjunction with propranolol, for minimizing postresuscitation myocardial dysfunction after successful resuscitation from cardiac arrest. Ventricular

fibrillation was induced and untreated for 7 minutes in 15 domestic pigs, which were divided into propranolol, propranolol plus levosimendan, and

control groups. Propranolol was administered as a bolus dose of 0.1 mg/kg during cardiac arrest. Electrical defibrillation was attempted after 12

minutes of cardiac arrest including 5 minutes of precordial compression. Levosimendan was administered at 10 minutes after successful

resuscitation in a dose of 20 microg/kg and followed by infusion of 0.4 microg/kg/min over the ensuing 220 minutes. Propranolol reduced energies

or numbers of defibrillatory shocks and postresuscitation myocardial ectopy, and it improved postresuscitation myocardial dysfunction. When

levosimendan was added, postresuscitation myocardial contractile function was improved even more.

An animal model (adult pig) RCT investigating the role of propanolol (P) and levosimendan (L) in improving post-resuscitation myocardial function after VF-

induced cardiac arrest. The hope was that a non-adrenergic inotrope such as levosimendan (with no associated increase in myocardial 02

consumption) would mitigate the negative inotropic effect of propanolol (despite being shown in earlier studies to reduce post-resuscitation

myocardial dysfunction). Animals were randomized by sealed envelope to one of 3 study groups 15 min pre-study. After induction of anesthesia and

instrumentation, animals had baseline investigations performed and then had VF-induced electrically. After a 6 minute period of untreated VF,

animals received either IV propanolol (0.1 mg/ kg) or placebo, and then at 7 minutes were resuscitated by protocol (CPR, then biphasic

defibrillation at 12 minutes into the arrest). The 3 study groups were P + L, P + saline, or saline control. All animals had ROSC established. No

adrenergic agents were necessary in order to reestablish circulation. Animals 10 min after successful resuscitation then received either placebo or

levosimendan (20 mcg/ kg bolus over 10 min, followed by infusion of 0.4 mcg/ kg/ min over the following 220 min. Measurements were performed

over the ensuing 4 hrs. During that time, there were no significant differences in EtC02/ blood gases reported between groups, although no

comment is made as to whether the minute ventilation was reduced post-resuscitation.


Baseline hemodynamics did not differ between groups. All animals were resuscitated with a single biphasic shock and no epinephrine, so apparently the

resuscitation time between groups was equivalent. There were no significant differences between groups over the next 4 hrs in regards

hemodynamics, PetC02, blood gases or arterial lactates. P administered during CPR facilitated resuscitation with a significantly smaller number of

shocks, a significantly lesser total energies of shocks, with significantly lesser post-resuscitation premature ventricular beats (PVB) and ST-

changes. There were no differences between P and P + L groups in regards these. Post-resuscitation EF (ejection fraction) and fraction of area

changes by echo were greater in P relative to control, but even greater in P + L.

This was a well designed animal study. The study demonstrated benefit to the use of propanolol pre-defibrillation, but even greater benefits when coupled

with the use of L after ROSC. How relevant this is when P is not a standard of care is unclear. Only limitations obvious include:

• Blinding of investigators?

• Relevance to asphyxial cardiac arrest patients (making up the majority of pediatric cardiac arrests)



• study of intra-arrest agent (propanolol) and not just of post-resuscitation cardiotonic use

Funding was partially industry-based (Abbott laboratories)

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Ward, H. B., S. Einzig, et al. (1982). "Effects of dopamine and dobutamine on the myocardial and systemic circulation during and following cardiopulmonary

bypass in dogs." Pediatr Cardiol 3(3): 257-64.

The effect of dopamine (Dp) and dobutamine (Db) on myocardial and systemic blood flow (BF) distribution was compared in dogs being weaned

from cardiopulmonary bypass (CPB) after 20 min of normothermic global myocardial ischemia. Drug infusions (10 mcg/kg/min) were begun and

BF was measured (radiolabeled microspheres) prior to weaning and 60 min off CPB. On CPB: Pump flows, by design, were similar (100

ml/kg/min) in Dp, Db, and saline control (NS) dogs. Dp and Db significantly (p less than 0.05) increased myocardial BF. Dp did not alter renal,

visceral, and skeletal muscle BF and only increased BF to the cervical spinal cord and medulla. Db, however, significantly reduced renal (-49%),

splenci (-58%), pancreatic (-27%), and colonic (-47%) BF but increased perfusion in essentially all of the central nervous system. Off CPB: Cardiac

output during Dp and Db infusions was significantly greater than NS dogs (107 +/- 7 vs 152 +/- 12 vs 82 +/- 10 ml/kg/min resp.); the greater

increase for Db resulting from a larger stroke volume. Dp and Db significantly increased myocardial BF. Dp increased splenic (+77%), gastric

(+139%), and gallbladder (+125%) BF but had no effect on renal, hepatic, skeletal muscle, and intestinal BF. Db infusion maintained renal BF

similar to NS and elevated BF to most visceral organs. The results of this study show that the myocardium responded to inotropic stimulation

despite the previous ischemic insult but the BF changes varied among regional vascular beds with Dp and Db infusions during and following CPB.

Of the 2 drugs, Db showed the greater inotropic response off CPB, a similar increase in myocardial perfusion and greater visceral organ bloodflow.

A prospective randomized controlled double-blinded study of the effects of dopamine and dobutamine on hemodynamics on and after separation from CPB.

Adult dogs were placed on CPB and experienced a 20 minute normothermic period of myocardial ischemia. Before coming off CPB (after a 30 min

recovery on CPB), they were started on catecholamine infusions, and their hemodynamic effects noted on and then off CPB (to a period of 60 min

post separation from CPB).

Cardiac output on either catecholamine infusion was significantly greater than that of the controls, that of dobutamine coming from a significantly increased

stroke volume. For equivalent dosages, dobutamine produced a greater rise in cardiac output than dopamine.

Limitations of the study to the worksheet are the animal study nature of the work. The absence of cardioprotection is something that the human cardiac

literature cannot provide, and is more relevant to the post-resuscitation setting that this worksheet addresses .This study examines the hemodynamic

effects after drug resuscitation for myocardial ischemia, not circulatory arrest (which would have been more relevant to the pediatric post-

resuscitation setting).

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Ward, H. B., S. Einzig, et al. (1984). "Comparison of catecholamine effects on canine myocardial metabolism and regional blood flow during and after

cardiopulmonary bypass." J Thorac Cardiovasc Surg 87(3): 452-65.

The acute metabolic and hemodynamic effects of dopamine, dobutamine (both at 10 micrograms . kg-1 . min), and isoproterenol (at 0.05 or 0.1

micrograms . kg-1. min) were determined in dogs following 20 minutes of normothermic global myocardial ischemia. The catecholamines were

started 10 minutes before cardiopulmonary bypass (CPB) was discontinued and were continued for 1 hour after bypass. Regional myocardial and

systemic blood flow distribution was measured by means of the radioactive microsphere technique. On bypass all catecholamines sharply increased

heart rate, myocardial oxygen consumption, and left ventricular blood flow (p less than 0.01). Because the hearts were unloaded, these data suggest

that velocity of contraction is an important component of myocardial oxygen consumption. Although these drugs did not lower myocardial

adenosine triphosphate (ATP) and creatine phosphate (CP) levels, the significant rise in oxygen consumption suggested that inotropic treatment on


bypass may not be beneficial. Furthermore, renal blood flow was diminished in dobutamine-treated dogs (p less than 0.01) and tended to decrease

with isoproterenol infusion. No change was seen with dopamine infusion. After bypass, dobutamine treatment increased cardiac output (p less than

0.01) and stroke volume (p = 0.017) with no change in heart rate, myocardial oxygen consumption, high-energy phosphate levels, and total or

transmural distribution of left ventricular blood flow. Dopamine infusion did not change cardiac output but did increase oxygen consumption (p less

than 0.01). Isoproterenol showed a slight inotropic effect, but frequent ventricular arrhythmias were present during weaning from bypass. In all

treatment groups, blood flow in the other systemic beds (cerebral, gastrointestinal, and renal) was similar to that in control dogs. These data suggest

that dobutamine is the most efficient of the drugs tested for support of the heart following global myocardial ischemia but, when given during

bypass, it appears to decrease renal blood flow.

A prospective ?randomized controlled trial comparing the hemodynamic effects of dopamine, dobutamine or isoproterenol on adult dogs separated from

CPB. Animals were exposed to a 20 min normothermic cardiac arrest on CPB. After a 30 min period of stabilization on CPB after removal of the

aortic cross-clamp, the dogs separated from CPB, and 10 min later were started on a control or catecholamine infusion.

Dobutamine (10 mcg/ kg/ min) increased cardiac output and stroke volume (and decreased SVR) without changing HR, MAP or myocardial 02 consumption.

Dopamine(10 mcg/ kg/ min) did not increase cardiac output/ MAP/ HR/ stroke volume, but did increase myocardial 02 consumption. Isoproterenol

(0.05 mcg/ kg/ min) did show a slight inotropic effect, but at the expense of frequent ventricular tachydysrhythmias.

Limitations of the study to the worksheet are the animal study nature of the work. The absence of cardioprotection is something that the human cardiac

literature cannot provide, and is more relevant to the post-resuscitation setting that this worksheet addresses .This study examines the hemodynamic

effects after drug resuscitation for myocardial ischemia, not circulatory arrest (which would have been more relevant to the pediatric post-

resuscitation setting).

LOE: 5

Quality: Good

Supportive

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Summary of Articles cited in C2005 Worksheet (highlighted articles included in C2010 worksheet as well)

First author Pub

Year Age Species Model

inotrope

initiated

study

duration

drug(s)

studied

Low

CO

pts

includ

ed ?

Comm

ents

Abdallah 2003 Pediatric Human CPB on CPB 24 hrs M vs E

No 29 pts

with

TOF

Allen 1997 Adult Dog CPB on CPB 1 hr

Angelos 2002 Langerhans

prep Dog Db vs P

ex-vivo

model

Badner 1994 Adult Human CPB on CPB 4 hrs A vs P Yes 30 pts

Bailey 1997 Pediatric Human CPB

After

separation

from CPB

1 hr SNP vs A

No 10

infants,

mixed

CV

lesions,

pts act

as their

own

control

Booker 1995 Pediatric Human CPB on CPB 12 hrs Db vs D

No 19 pts,

mixed

ages and

CV

lesions,

pts acted

as their

own

controls;

PVR

rise

post-


First author Pub


inotrope

initiated

study

duration

drug(s)

studied

Low

CO

pts

includ

ed ?

Comm

ents

CPB

that was

attribute

d to

drugs

was

potential

ly just

due to

post-

CPB

changes

Butterworth 1993 Adult Human CPB on CPB 40 min post

CPB A vs P

Yes 39 pts

Chang 1995 Pediatric Human CPB 6 hrs post

CPB 1 hr

M

(pharmacokinetic

s)

Yes 10

neonates

, mixed

CV

lesions

D’Hollander 1982 Adult Human CPB

After

separation

from CPB

< 1hr D vs Db

No 34 pts

act as

their

own

control

(dobut

vs dop)

De Hert 1995 Adult Human CPB on CPB 4 hrs M (low vs hi

dose)

Yes 20 pts;

no

placebo

control

Doolan 1997 Adult Human CPB on CPB 12 hrs M vs P Yes 32 pts

Dupuis 1992 Adult Human CPB on CPB 2 hrs post

CPB M vs Db

Yes 30 pts;

no

placebo

control

Feneck 1992 (X2) Adult Human CPB

After

separation

from CPB

16 hrs M (varied dosing)

Yes 99 pts;

non-

randomi

zed,

non-

blinded

Feneck 2001 Adult Human CPB 2 hrs post

CPB 4 hrs M vs Db

Yes 120 pts;

no

placebo

control

Gray 1981 Adult Human CPB 2-18 hrs

post CPB 2 hrs

Db vs D vs

NE/Phentol

Yes 9 pts

Hoffman 2003 Pediatric Human CPB

90 min

post-ICU

admission

>24 hrs M vs P

Yes

227 pts

Holloway 1975 Adults Human CPB ICU POD-1 4 hrs Iso vs D

No 12 pts ;

cross-

over


First author Pub


inotrope

initiated

study

duration

drug(s)

studied

Low

CO

pts

includ

ed ?

Comm

ents

Innes 1994 Pediatric Human CPB on CPB 6 hrs Pnenox + Db vs

Enoximone

yes 28 pts;

mixed

age and

CV

lesions;

no

control

group

Jenkins 1997 Adults Human CPB on CPB 3 hrs A vs Db Yes 20 pts

Kern 1997 Adult Pig VF arrest 15 min post

ROSC 1 hr Db vs P

Jaccard 1984 Pediatric Human CPB

3-4 hrs post

sternal

closure

2 hrs Iso vs Db

yes 12 TOF

pts, 22

mos to

15 yrs

Kersting 1976 Adult Human CPB 4 hrs post-

op 2 hrs Iso vs Db

Yes 11 pts

Kiefer-Jensen 1995 Adult Human CPB ICU; when? 1 hr PGI vs SNP

Yes 10 pts,

cross-

over

Kikura 2002 Adult Human CPB On CPB 24 hrs A vs M vs P Yes 37 pts

Kunimoto 1999 Adult Human CPB ICU POD 1 6 hr

A varied dosing

in hi vs low

inotrope groups

No 45 pts;

no

placebo

group

Latinen 1997 Pediatric Human CPB On CPB 4-18 hrs A vs D/NTG

Yes 32

infants

with AV

canal

Latinen 1999 Pediatric Human CPB On CPB 18 hrs A vs D/NTG

Yes 35

infants

with

TGA

Lewis 2000 Adult Human CPB On CPB 18 hrs A vs P Yes 234 pts

Lobato 1998 Adult Human CPB On CPB 3 hrs M vs P Yes 21 pts

Lobato 2000 Adult Human CPB

After

separation

from CPB

<1 hr M vs Epi

Yes 20 pts;

No

placebo

group

Meyer 2002 Adult Pig VF arrest 15 Min post

ROSC

5 hrs post-

ROSC

Db vs untreated

controls

Niemann 2003 Adult Pig VF arrest Intra arrest 60 min M vs P

O Yrvati 1995 Adult Human CPB

After

separation

from CPB

24 hrs Db vs A or Db

‘controls’

No

80 pts

Ramamoorthy 1998 Pediatric Human CPB On CPB 7 hrs

Pharmacokinetics

of varied M

dosing

Yes

18 pts

Rathmell 1998 Adult Human CPB

After

separation

from CPB

10 min A vs M

Yes

44 pts

Romsom 1999 Adult Human CPB After 30 min Db vs untreated No 100 pts


First author Pub


inotrope

initiated

study

duration

drug(s)

studied

Low

CO

pts

includ

ed ?

Comm

ents

separation

from CPB

controls

Royster 1993 Adult Human CPB

After

separation

from CPB

~30 min P vs E vs A vs E +

A

No

40 pts

Sato 1982 Adult Human CPB

After

separation

from CPB

4 hrs E vs D vs Db

(crossover)

No

10 pts

Shibata 2001 Adult Human CPB ICU

admission 3 hrs M vs P

No 20 pts

Steen 1978 Adult Human CPB

After

separation

from CPB

1 hr D vs Db vs E

(crossover)

Yes

34 pts

Ward 1982 Adult Dogs CPB On CPB 60 min D vs Db vs P

Yes After 20

min

normoth

ermic

myocard

ial

ischemia

on CPB

Ward 1984 Adult Dogs CPB On CPB 60 min D vs Db vs Iso vs

P

Yes After 20

min

normoth

ermic

myocard

ial

ischemia

on CPB

Yamada 2000 Adult Humans CPB On CPB 60 min M vs P Yes 48 pts

Summary of Articles cited in C2010 Worksheet

First

author

Pub


inotrope

initiated

study

duration

drug(s)

studied

Low

CO

pts

includ

ed ?

Comments

Alvarez 2006 Adult Human CPB ICU

admission 48 hrs L vs Db

Yes 41 pts; no

placebo

control

Duggal 2005 Pediatric Human CPB ICU

admission 24 hrs M

Yes 15 pts; pts

their own

controls over

time

Egan 2006 Pediatric Human CPB

Pre/ on

CPB/ post-

CPB

24 hrs L

Yes 19 pts; pts

their own

controls over

time


Huang 2005 Adult Rat

VF

cardiac

arrest

10 min

post-ROSC 4 hrs L vs Db vs P

Huang 2005 Adult Pig

VF

cardiac

arrest

10 min

post-ROSC 72 hrs L vs Db vs P

Jorgensen 2008 Adult Human CPB

After

separation

from CPB

1 hr L (varied

dosing) vs P

No

23 pts

Lobato 2005 Adult Human CPB

After

separation

from CPB

After sternal

closure Epi vs. M vs. P

No

45 pts

Mayr 2007 Adult Human VF or

asphyxial Post-ROSC 72 hrs

V vs E vs NE

vs M

Yes Observational

study of 23

adults post

ROSC

Nijhawan 1999 Adult Human CPB On CPB 6 hrs L vs P No

18 pts

Stocker 2007 Pediatric Pig CPB

2 hrs after

removal of

AoXcl

2 hrs L vs M vs P

All

measurements

start 2 hrs

after removal

of AoXcl

Studer 2005 Adult Rat

VF

cardiac

arrest

15 min

post-ROSC 2 hrs

Db (2 doses) vs

P

Vasquez 2004 Adult Pig

VF

cardiac

arrest

30 min

post-ROSC 6 hrs

Graded Db

dosing vs P

Wang 2005 Adult Pig

VF

cardiac

arrest

Pranolol

intra-arrest

and L 10

min post-

ROSC

4 hrs

P vs prop

(intra-arret) vs

Prop (intra-

arrest) and L

worksheet for evidence-based review of science for emergency

Documents

c2010 worksheet

hospital discharge e

e wang

e lobato

e kern

e alvarez

e fairduggall

e stocker