Copyright ©The Author(s) 2016.
World J Clin Oncol. Feb 10, 2016; 7(1): 98-105
Published online Feb 10, 2016. doi: 10.5306/wjco.v7.i1.98
Table 1 Summaries of the studies of adult hematopoietic stem cell transplantation patients admitted to intensive care unit published between 2000-2015
Ref. (study period)No. of patients admitted to ICU [total N of HSCTs (%)], ICU admission risk factorsReasons for ICU admission (%)Interventions (%)OutcomesFactors evaluated for outcome predictionPredictors of outcome on mutlivariate analysisNotes
Boyaci et al[31] (2007-2010)48 patients (7 Auto and 41 AlloSCT)Respiratory failure 86%, sepsis/septic shock 75%, renal failure, liver failure, AMSMV 75%Mortality: 79% in hospitalAge, gender, underlying disease, remission status, HSCT type, HLA match, conditioning intensity, cause of ICU admission, GVHD, SOS, APACHE II, GCS, SOFA, # of organ failures, various vitals and lab values, VA, MVAPACHE II score and VA in ICU a/w higher mortality
Bayraktar et al[8] (2001-2010)389 AlloSCT patients a/to ICU within 100 d of HSCT [Of 3039 patients (13%)]Respiratory failure 61%, septic shock 12%, AMS 9%, arrhythmia 5%, non-GI, non-CNS bleeding 4%N/RMortality: 64% in hospitalAge > 55, underlying disease, year of HSCT was, HSCT period at ICU admission, graft source, HLA match status, donor relation, conditioning intensity, aGVHD at ICU admission, HCT-CI scoreHCT-CI ≥ 2 , ablative conditioning, aGVHD at ICU admission a/w higher mortality. ICU admission during conditioning regimen a/w lower mortalityHCT-CI score, a measure of pre-transplant comorbidities, can be calculated even prior to HSCT
van Vliet et al[17] (2004-2009)49 AlloSCT [Of 319 (15%)]Infectious complications 86%, respiratory failure 67% Ablative conditioning and unrelated donor grafting a/w increased risk for ICU admissionN/R1-yr OS: 15% Mortality: 33% in ICU, 53% in hospitalNRUnivariate analyses demonstrated improved 100-d survival between 2004-2005 to 2008-2009
Agarwal et al[30] (1998-2008)123 HSCT patients (73% AlloSCT)Mortality: 41% in ICU, 62% in hospital. OS @ 1yr: 24%Age, underlying disease, type of HSCT, GVHD, neutropenia, hospital admission-ICU interval, organ failures, sepsis type, APS, APACHE II, MVFungal infection and number of organ failures a/w higher ICU mortalityHard to explain why GVHD was a/w lower ICU mortality
Depuydt et al[33] (2000-2007)44 AlloSCTBacterial infections 32%, non-bacterial infections 30%, non-infectious causes 39%. Overall, pulmonary related causes 39%MV (73%), RRT (27%)Mortality: 61% in ICU, 75% in hospital, 80% @ 6 mAge, gender, bacterial infection, GVHD grade, HSCT-ICU interval, SOFABacterial infection as the cause of ICU admission a/w lower hospital mortalityImprovement in SOFA score by 5th d of ICU was sig better in patients with bacterial infections
Benz et al[9] (1998-2007)33 AlloSCT [Of 250 (13%)]Pulmonary complications 33%, sepsis 24%, neurological disorders 18%, cardiovascular problems 6%,MV 64%, VA 42%, RRT 27%OS @ 1yr: 28%NRSAPS II and SOFA scores did not reliably predict survival
ICU admission risk factors: aGVHD grade II-IV and HLA mismatch
Townsend et al[13] (1996-2007)164 AlloSCT (majority TCD) [Of 552 (30%)]. ICU admission risk factors: Ablative conditioningSepsis 67%, respiratory failure 55%MV 50%Survival: 32% in ICU OS @ 1yr: 19% overall, 61% in patients who survived ICUDonor type, conditioning intensity, reason for ICU admission, NIV, MV, VA, RRT, various labs, APACHE II, duration of ICU stay, duration of MVMV, raised BUN at admission and ablative conditioning a/w worse ICU survival
Trinkaus et al[15] (2001-2006)34 AutoSCT patients admitted within 100 d of SCT [Of 1013 (3.3%)]Sepsis 32%, respiratory failure 29%, cardiovascular failure 26%ICU mortality: 38%NR
Neumann et al[18] (1999-2006)64 AlloSCT [Of 319 (20%)]Pulmonary complications 53%, Sepsis 22%, renal failure 9%, bleeding 3%, status epilepticus 3%ICU mortality: 66%Age, gender, underlying disease, remission status, conditioning intensity, HLA match status, GVHD, ICU admission indication, HSCT-ICU interval, SOFA, various labs, SOSSOFA ≥ 12 and BUN > 60 a/w higher ICU mortality
Gilli et al[10] (1995-2005)91 AlloSCT (29% < 18 yrs old) [Of 661 (14%)]Respiratory failure 41%, septic shock 31%, neurological events 12%MV 48%, RRT 5%, VA 58%Mortality: 58% in ICU, 70% @1mConditioning intensity, reason for ICU transfer APACHE II, SOFA, VA, RRT, IMVSOFA score a/w 30 d mortalityAPACHE II underestimated mortality
Naeem et al[14] (1998-2003)25 UCBT [Of 44 (57%)] ICU admission risk factors: Ablative conditioningPneumonia 52%, GI bleeding (12%), Sepsis 8%, renal failure 8%MV 48%ICU mortality: 72%NR
Pène et al[11] (1997-2003)209 AlloSCT [Of 1025 patients (20%)]Repiratory 67%, hemodynamic 23%, neurologic 18%, renal 17%, other 5%MV (58%), RRT (28%), VA (47%)Survival: 48% in ICU, 32% in hospital, 27% @ 6 m, 21% @ 1 yr MV patients: 18% in ICU, 16% in hospitalAge, gender, underlying disease, remission status, conditioning intensity, graft source, HSCT-ICU interval ≤ 30 d, corticosteroid Rx, serum bilirubin level, MV, VA, RRTCorticosteroid Rx, serum bilirubin level at ICU admission, MVNone of the 35 patients with admission LOD score > 10 survived the hospital stay
Scale et al[41] (1992-2002)504 patients (264 AlloSCT) who were admitted to ICU following the BMT hospitalization [Of 2653 (19%)]MV 51%, RRT 7%1-yr mortality: 67%NR
Kim et al[42] (1999-2001)18 AlloSCT [Of 210 (9%)]Respiratory failure 50%, renal failure 39%, septic shock 11%ICU mortality: 94%
Soubani et al[16] (1998-2001)85 HSCT patients (45 AlloSCT) [Of 745 (11%)]Respiratory 48%, Sepsis 23%, cardiac 19%, neurologic 6%, bleeding 2%MV in 60%Mortality: 39% in ICU, 59% in hospital, 72% @ 6 m CU mortality 63% among patients with MVAge, gender, smoking history, race, underlying disease, HSCT type, HLA match, HSCT-ICU interval, GVHD, various labsHigh lactate level, MV, > 2 MOFs during ICU stay a/w higher ICU mortality
Kew et al[12] (1992-2001)37 HSCT patients (28 AlloSCT) [Of 440 (9%)]Respiratory failure 65%, hemodynamic instability 57%,MV in 68%29 patients died within 1 yrPre-ICU patient characteristics, MV, VAVA a/w shorter OS
Afessa et al[32] (1996-2000)111 patients (62 Auto, 50 AlloSCT)Respiratory failure 40%, cardiac reasons 26%, sepsis 14%, CNS dysfunction 5%, GI bleeding 5%MV 55%Mortality: 33% in ICU, 46% in hospital 30-d mortality was 78% among AlloSCT patientsType of HSCT, graft source, post-transplant days @ ICU admission, GVHD, APACHE III, APACHE II, ARDS, MOF, sepsis, MV, VAHigher APACHE III score @ ICU admission, AlloSCT, MV, ARDS, MOF, sepsis, VA a/w higher 30-d mortalityAUC of receiver operating characteristic curve for APACHE III and hospital mortality was 0.704
Table 2 Possible reasons for improved outcomes in patients who are admitted to intensive care unit after hematopoietic stem cell transplantation
Improvements in HSCT
Reduced intensity conditioning
Better antimicrobial prophylaxis
Pre-emptive therapy of cytomegalovirus infections
Improved antifungal therapy
Improvements in intensive care
Early use of non-invasive ventilation
Early goal-directed therapy for septic shock
Better patient selection
Improved recognition of clinical deterioration and earlier ICU admission
Use of palliative care for patients with a slim chance of recovery
Table 3 Factors that were found to be associated with outcomes of intensive care among hematopoietic stem cell transplantation patients
Patient/disease related factors
Pre-transplant comorbidities
Transplant related factors
Type of HSCT (allogeneic vs autologous)
Conditioning regimen intensity
Graft-vs-host disease
Patient functional status at ICU admission
Serum bilirubin level
Serum lactate level
Blood urea nitrogen level
Type of infection (bacterial vs fungal)
Post-ICU admission factors
Mechanical ventilation
Vasopressor support