containing 100 mg of hydrocortisone sodium succinate pow-der. Each daily dose was diluted in 100 mL of isotonic sodiumchloride solution at the hospital’s pharmacy and administeredtostudygrouppatientsasacontinuousinfusion.Onvasopres-sor cessation or on day 8 after cardiac arrest, the daily hydro-cortisonesodiumsuccinatedosewasconsecutivelyreducedto200 mg and 100 mg and then discontinued (see the onlinesupplementalmaterial).Controlgrouppatientswithpostresus-citation shock received daily infusions of 100 mL of isotonicsodiumchloridesolutionplacebo.Isotonicsodiumchlorideso-lution infusion bags bore the patient codes.
DEFINITIONS
Circulatory failure
was defined as inability to maintain meanarterial pressure greater than 70 mm Hg without using vaso-pressors after volume loading.
10
Respiratory failure
was de-fined as a ratio of arterial oxygen partial pressure to inspiredoxygen fraction of 200 mm Hg or less.
Coagulation failure
wasdefined as a platelet count of 50
10
3
/µL or less.
Hepatic fail-ure
was defined as serum bilirubin concentration of 6 mg/dL(toconverttomicromolesperliter,multiplyby17.104)orless.
Renalfailure
wasdefinedasserumcreatininelevelof3.5mg/dL(toconverttomicromolesperliter,multiplyby88.4)orgreaterand/or requirement of renal replacement therapy.
Neurologic failure
was defined as a Glasgow Coma Score of 9 or less.After resuscitation, cardiac arrest–induced cardiac and mi-crocirculatory dysfunction lasts approximately 24 hours.
6
Postresuscitation shock was defined as sustained (
4 hours),newpostarrestcirculatoryfailureorpostarrestneedforatleasta 50% increase in any prearrest vasopressor/inotropic supporttargeted to maintain mean arterial pressure above 70 mm Hg.
DOCUMENTATION AND PATIENT FOLLOW-UP
AttemptsatCPRweredocumentedaccordingtotheUtsteinstyle.
9
Additional data comprised periarrest arterial pressure, gas ex-change,electrolyteandlactatelevels,vasopressor/inotropicsup-port, and intravenous fluids given. Daily follow-up was con-ducted by 4 blinded investigators (N.K., S.G., A.P., and A.S.).Follow-uptoday60aftercardiacarrestincludedmedication,or-gan or system failures, and ventilator-free days. Morbidity andcomplicationsthroughoutICU/CCUandhospitalstayandtimesto ICU/CCU and hospital discharge were also recorded. En-coded patient data were entered into a database by 2 investiga-tors (N.K. and S.G.) and independently cross-checked by an-other 2 investigators (A.P. and A.S.). Data were independentlyscrutinized by a steering committee.
PLASMA CYTOKINE CONCENTRATIONS
Venipuncturebloodsampleswereobtainedonday0(at6hoursafter randomization) from the last 35 surviving patients withpostresuscitation shock; additional blood samples were ob-tainedondays1,3,and7afterrandomization.Serumconcen-trationsoftumornecrosisfactor,interleukin(IL)-1
,IL-6,IL-8,and IL-10 were measured by an enzyme-linked immunosor-bent assay (Quantikine; R&D Systems Europe Ltd, Abingdon,England) according to manufacturer instructions.
STUDY END POINTS
Primaryendpointswerereturnofspontaneouscirculationfor15 minutes or longer and survival to hospital discharge, de-fined as presence of an attending physician discharge order tohomeortoarehabilitationfacility.Secondaryendpointswerearterial pressure during and 15 to 20 minutes after CPR, in-tensity of postarrest systemic inflammatory response, numberof organ failure–free days until completion of follow-up, andcerebralperformanceaccordingtotheGlasgow-Pittsburghscaleathospitaldischarge(seeonlinesupplementalmaterialforde-tails on determination of end points).
STATISTICAL ANALYSIS
Initial rhythm is asystole in 75% to 80% of the refractory car-diac arrests occurring in our hospital. Sample-size calculation(G
*
Power version 3.0.8; Heinrich Heine University, Düssel-dorf, Germany) was based on a possible, drug-related, overall3.1-fold improvement in survival to hospital discharge of thestudy group vs the control group. Survival improvement wasexpected mainly for patients with asystole.
9
Thus, our overallprediction was equivalent to an experimental treatment–induced 3.8-fold rise in the survival of patients with asystole.This corresponds to an improvement of 22.6% relative to a re-centlyreportedvasopressin-induced3.1-foldriseinsurvivalaf-terasystoliccardiacarrest.
9
Predictedoverallsurvivalofthecon-trol group was 5%.
2
Calculated
2
effect size was 0.34. For an
value of .05 and a power of 0.80, the estimated sample sizewas 68 (ie, 34 patients per group). The inclusion of 100 pa-tients resulted in a safety margin of 32 of 68 (47%).An intention-to-treat analysis was conducted with SPSS ver-sion12.0statisticalsoftware(SPSSInc,Chicago,Illinois).Dataarereportedasmean(SD),median(interquartilerange[IQR]),ornum-ber(percentage),unlessotherwisespecified.Distributionnormal-itywastestedbytheKolmogorov-Smirnovtest.Dichotomousandcategoricalvariableswerecomparedbythe
2
orFisherexacttest.Continuousvariableswerecomparedbya2-tailed,independent-samples
t
testortheMann-Whitneyexacttest.Inpostresuscitationshock,weusedlinearmixed-modelanaly-sis to determine the overall effects of group, time, and their in-teraction(group
time)onlog-transformedplasmacytokinecon-centrationsthroughoutthefirst7daysafterrandomization.Theeffectsofgroup,time,andgroup
timeon(1)averagedailycen-tral venous oxygen saturation and arterial blood lactate (mea-suredevery12hours),meanarterialpressure(recordedevery3hours), and infusion rates of vasopressors; (2) daily fluid bal-ance; and (3) hemoglobin concentration (measured every 24hours) were also analyzed for the first 10 days after randomiza-tion.Fixed-effectssignificancewasdeterminedbythe
F
test.Modelselectionwasbasedontheminimumvaluesof−2restrictedlog-likelihoodandAkaikeinformationcriteria.Between-groupcom-parisonsatindividual,consecutivetimepointswereconductedwith the independent-samples
t
test;
P
values were not cor-rected for multiple comparisons.SurvivalwasanalyzedbytheKaplan-Meiermethod,andsur-vival data were compared by (1) the Fisher exact test to deter-mine any nonrandom association between group and survivalto hospital discharge and (2) the log-rank test to test the nullhypothesis that the probability of death did not differ betweenthestudyandcontrolgroupsthroughoutpatientfollow-up.Uni-variate and multivariate backward-stepwise Cox regressionanalysis was used to identify independent predictors of deathandtodeterminetherespectiveproportionalhazardsandtheir95% confidence intervals. Variable entry and removal criteriawere0.05and0.10,respectively.Reported
P
valuesare2sided.Statistical significance was set at
P
.05.
RESULTS
From June 8, 2006, to March 16, 2007, there were 139potentially eligible patients with cardiac arrest. Overallsurvival to hospital discharge was 37 of 139 patients
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 1), JAN 12, 2009 WWW.ARCHINTERNMED.COM
17
©2009 American Medical Association. All rights reserved.
Leave a Comment