Advertisement

Outcomes of neonates requiring prolonged stay in the intensive care unit after surgical repair of congenital heart disease

Open ArchivePublished:April 19, 2016DOI:https://doi.org/10.1016/j.jtcvs.2016.04.040

      Abstract

      Objective

      After neonatal cardiac surgery, a number of patients need a prolonged stay in the intensive care unit (ICU). Those patients require tremendous resources and strain the capacity of cardiac units. To date, little knowledge of early and late survival for this challenging population exists.

      Methods

      From 2002 to 2012, 108 neonates required a postoperative ICU stay >30 days. Multivariable regression analyses examined factors associated with hospital death and late survival. Comparison of late outcomes in hospital survivors was made between those who had prolonged ICU stay (n = 82) and contemporaneous neonates who did not (n = 1329).

      Results

      Hospital mortality occurred in 26 of 108 patients (24%). On multivariable analysis, factors associated with mortality were use of extracorporeal membrane oxygenation (odds ratio, 3.4 [95% confidence interval, 1.3-9.1], P = .014) and renal failure that required dialysis (odds ratio, 3.1 [95% confidence interval, 1.0-10.0], P = .056). Overall survival at 1 and 8 years was 57% and 51%. Comparison of late outcomes for hospital survivors showed that neonates who required prolonged postoperative stay in the ICU had significantly worse 8-year survival (69% vs 92%; P < .001) and that the effect of prolonged stay in the ICU on diminished survival was more pronounced in neonates with 2 ventricles (68% vs 95%, hazard ratio, 8.0 [95% confidence interval, 4.2-15.1], P < .001) than in those with single ventricle (66% vs 81%; hazard ratio, 2.0 [95% confidence interval, 1.1-3.5], P = .021). Overall, 77% of single-ventricle hospital survivors who required prolonged stay in the ICU progressed to Glenn, with 82% of them reaching or qualifying for subsequent Fontan.

      Conclusions

      Prolonged postoperative stay in the ICU is associated with high hospital and significant postdischarge mortality, mainly during the first year. In neonates with single ventricle, prolonged stay in the ICU was associated with high hospital and interstage mortality and usual progression subsequent to Glenn shunt. In contrast, prolonged stay in the ICU in neonates with 2 ventricles was associated with high hospital mortality and considerable decrease in late survival, suggesting a more pronounced deviation from expected survival in those patients.

      Key Words

      Abbreviations and Acronyms:

      CI (confidence interval), ECMO (extracorporeal membrane oxygenation), HR (hazard ratio), ICU (intensive care unit), IQR (interquartile range)
      Figure thumbnail fx1
      Effect of prolonged stay in the intensive care unit on postdischarge survival in single and 2 ventricle neonates.
      Prolonged postoperative stay in the intensive care unit increases postdischarge mortality, especially in neonates with 2 ventricle anomalies.
      Prolonged postoperative stay in the intensive care unit (ICU) is associated with high hospital and postdischarge mortality. In single-ventricle patients, prolonged stay in the ICU is associated with high hospital mortality and usual progression subsequent to Glenn shunt. Conversely, prolonged stay in the ICU in 2-ventricle patients is associated with high hospital mortality and considerable decrease in late survival.
      See Editorial Commentary page 727.
      The advances in operative management and perioperative care have allowed the successful treatment of neonates born with complex congenital heart disease. Subsequently, an increasing number of neonates who undergo cardiac surgery require prolonged stay in the intensive care unit (ICU). Chronic critical illness is defined in the pediatric and adult literature as duration of stay in the ICU longer than 28-30 days.
      • Lagercrantz E.
      • Lindblom D.
      • Sartipy U.
      Survival and quality of life in cardiac surgery patients with prolonged intensive care.
      • Bapat V.
      • Allen D.
      • Young C.
      • Roxburgh J.
      • Ibrahim M.
      Survival and quality of life after cardiac surgery complicated by prolonged intensive care.
      • Gaudino M.
      • Girola F.
      • Piscitelli M.
      • Martinelli L.
      • Anselmi A.
      • Della Vella C.
      • et al.
      Long-term survival and quality of life of patients with prolonged postoperative intensive care unit stay: unmasking an apparent success.
      • Grothusen C.
      • Attmann T.
      • Friedrich C.
      • Freitag-Wolf S.
      • Haake N.
      • Cremer J.
      • et al.
      Predictors for long-term outcome and quality of life of patients after cardiac surgery with prolonged intensive care unit stay.
      • Joskowiak D.
      • Kappert U.
      • Matschke K.
      • Tugtekin S.
      Prolonged intensive care unit stay of patients after cardiac surgery: initial clinical results and follow-up.
      Although it's natural that hospital mortality is increased in patients with chronic critical illness, little is known about the late outcomes of those patients subsequent to hospital discharge, especially in the pediatric cardiac population. In adult cardiac patients, the effect of prolonged postoperative stay in the ICU has been found to be associated with reduced late survival, low quality of life, and high rates of readmission.
      • Lagercrantz E.
      • Lindblom D.
      • Sartipy U.
      Survival and quality of life in cardiac surgery patients with prolonged intensive care.
      • Bapat V.
      • Allen D.
      • Young C.
      • Roxburgh J.
      • Ibrahim M.
      Survival and quality of life after cardiac surgery complicated by prolonged intensive care.
      • Gaudino M.
      • Girola F.
      • Piscitelli M.
      • Martinelli L.
      • Anselmi A.
      • Della Vella C.
      • et al.
      Long-term survival and quality of life of patients with prolonged postoperative intensive care unit stay: unmasking an apparent success.
      • Grothusen C.
      • Attmann T.
      • Friedrich C.
      • Freitag-Wolf S.
      • Haake N.
      • Cremer J.
      • et al.
      Predictors for long-term outcome and quality of life of patients after cardiac surgery with prolonged intensive care unit stay.
      • Joskowiak D.
      • Kappert U.
      • Matschke K.
      • Tugtekin S.
      Prolonged intensive care unit stay of patients after cardiac surgery: initial clinical results and follow-up.
      Neonates who need prolonged ICU care require tremendous resources and strain the capacity of cardiac units. Although previous studies have identified risk factors associated with prolonged stay in the ICU after cardiac surgery,
      • Gaies M.G.
      • Gurney J.G.
      • Yen A.H.
      • Napoli M.L.
      • Gajarski R.J.
      • Ohye R.G.
      • et al.
      Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass.
      • Ettema R.G.
      • Peelen L.M.
      • Schuurmans M.J.
      • Nierich A.P.
      • Kalkman C.J.
      • Moons K.G.
      Prediction models for prolonged intensive care unit stay after cardiac surgery: systematic review and validation study.
      • Messaoudi N.
      • De Cocker J.
      • Stockman B.A.
      • Bossaert L.L.
      • Rodrigus I.E.
      Is EuroSCORE useful in the prediction of extended intensive care unit stay after cardiac surgery?.
      • Pasquali S.K.
      • He X.
      • Jacobs M.L.
      • Shah S.S.
      • Peterson E.D.
      • Gaies M.G.
      • et al.
      Excess costs associated with complications and prolonged length of stay after congenital heart surgery.
      the impact of prolonged stay in the ICU on the long-term prognosis of pediatric cardiac population remains unclear.
      We aim in the current series to describe the characteristics of neonates who required prolonged ICU care after cardiac surgery at our institution, to examine early and late results and risk factors associated with survival, and finally to compare late outcomes with a control of contemporaneous neonates who survived hospitalization without requiring prolonged ICU stay.

      Patients and Methods

       Inclusion Criteria

      We examined outcomes of neonates who underwent cardiac surgery at Children's Healthcare of Atlanta, Emory University from 2002 to 2012. Premature neonates who received patent ductus arteriosus closure and children who were in heart failure supported by ventricular assist device were excluded. Patients were identified with the use of our institutional surgical database. Demographic, morphologic, clinical, operative, and hospital details were abstracted from the medical records for analysis. Approval of this study was obtained from our hospital's Institutional Review Board, and requirement for individual consent was waived for this observational study.

       Follow-up

      Time-related outcomes were determined from recent office visits documented in the electronic chart of Children's Healthcare of Atlanta system or from direct correspondence with pediatric cardiologists outside the system. Mean follow-up duration was 3.0 ± 3.45 years and was 95% complete.

       Statistical Analysis

      Statistical analyses were performed with SAS 9.3 (SAS Institute, Inc, Cary, NC), and statistical significance was assessed at the .05 level of significance. Data are presented as medians with interquartile ranges (IQRs) or frequencies and percentages as appropriate. Comparisons among neonates who required a prolonged stat in the ICU and who died before hospital discharge with those who survived hospital discharge were made by the use of χ2 tests, Fisher exact tests for categorical data, and Mann-Whitney U tests or Kolmogorov-Smirnov tests for continuous data. Similar analyses also were performed to compare characteristics of neonates who required a prolonged postoperative stay in the ICU with those who did not. On the basis of univariate analyses, variables significant at the 0.2 level were entered into a multivariable logistic regression model to identify predictors of mortality after surgery. Variable were retained in the multivariable logistic model if they were significant at the 0.1 level. The time-dependent outcomes (overall survival since surgery and since hospital discharge) were parametrically modelled with the HAZARD procedure in SAS. Parametric probability estimates for time-dependent outcomes uses models based on multiple, overlapping phases of risk (available for use with the SAS system at https://www.lerner.ccf.org/qhs/software/hazard). The HAZARD procedure uses maximum likelihood estimates to resolve risk distribution of time to event in up to 3 phases of risk (early, constant, and late). These parametric hazard models were created for overall mortality and mortality conditional on surviving hospital discharge after surgery. To examine the effect of prolonged ICU stay and single-ventricle anomalies on survival, models were stratified by prolonged ICU status and ventricle anomaly, and survival probabilities are reported for each group. The effect of prolonged stay in the ICU on subsequent progression to Glenn and Fontan after the initial surgery was examined with the use of χ2 tests.

      Results

       Patients' Characteristics, Morphologic, and Operative Details of the Prolonged ICU Stay Cohort

      On the basis of the definition of chronic critical illness, we used ICU stay >30 days as our cutoff to describe patients who required prolonged postoperative ICU stay. Between 2002 and 2012, 171 children required prolonged postoperative ICU stay >30 days at our institution. This number represents ∼3% of all children who were admitted to our ICU after surgical repair of their congenital heart disease. We focused our current study on neonates. During this study interval, 1538 neonates underwent cardiac surgery at our institution; our prolonged ICU cohort comprised 108 neonates (∼7%) who required postoperative ICU stay >30 days.
      There were 58 male patients (54%), and median age at surgery was 5 days (IQR, 3-9). Mean weight was 3.0 ± 0.6 kg with 20 patients (19%) ≤2.5 kg at time of surgery. There were 22 patients (21%) who were born prematurely ≤36 weeks' gestation. Fifty patients (46%) had associated genetic syndromes and/or major extracardiac malformations, including heterotaxy syndrome (n = 16), DiGeorge syndrome (n = 12), CHARGE (ie, Coloboma of the eye, Heart defects, Atresia of the choanae, Retardation of growth and/or development, Genital and/or urinary abnormalities, and Ear abnormalities and deafness) syndrome (n = 4), VACTERL (ie, costo-Vertebral abnormalities, Anal atresia, Cardiac defects, Tracheal-esophageal abnormalities, Renal and radial abnormalities, nonradial Limb abnormalities) association (n = 3), Down syndrome (n = 2), Dandy Walker syndrome (n = 2), and other (n = 11), Table 1.
      Table 1Patient characteristics, anatomic, and operative details of neonates who required prolonged stay in the ICU after cardiac surgery with comparison between hospital survivors and nonsurvivors
      CharacteristicOverall (N = 108)Hospital survivalP value
      Yes (alive) (n = 82)No (dead) (n = 26)
      Male sex58 (54%)43 (52%)15 (58%).64
      Age at surgery (d), median (IQR)5 (3-9)5 (3-9)5 (2-10).49
      Premature ≤36 wk22 (21%)18 (23%)4 (15%).44
      Genetic syndromes and extracardiac malformations50 (46%)39 (48%)11 (42%).64
      Weight (kg), mean ± SD3.0 ± 0.63.0 ± 0.63.1 ± 0.5.51
      Weight ≤2.5 kg20 (19%)16 (20%)4 (15%).64
      Underlying cardiac anomaly.08
       Single ventricle63 (58%)44 (54%)19 (73%)
       Two ventricles45 (42%)38 (46%)7 (27%)
      STAT category 4 and 592 (85%)70 (85%)22 (85%).93
      Use of cardiopulmonary bypass90 (83%)67 (82%)23 (89%).42
      Cardiopulmonary bypass duration (min), median (IQR)139 (71-166)140 (81-169)127 (60-160).406
      Postoperative use of ECMO30 (28%)17 (21%)13 (50%).004
      Unplanned reoperation24 (22%)17 (21%)7 (27%).51
      Sepsis
      Data present for 106 of 108 patients.
      38 (36%)27 (34%)11 (42%).43
      Renal failure
      Data present for 106 of 108 patients.
      16 (15%)8 (10%)8 (31%).01
      Cerebrovascular accident
      Data present for 106 of 108 patients.
      7 (7%)5 (6%)2 (8%)1.00
      Necrotizing enterocolitis
      Data present for 106 of 108 patients.
      18 (17%)15 (19%)3 (12%).39
      Gastrointestinal bleeding
      Data present for 106 of 108 patients.
      4 (4%)2 (3%)2 (8%).25
      Chylothorax
      Data present for 106 of 108 patients.
      22 (21%)17 (21%)5 (19%).83
      Vocal cord paralysis
      Data present for 106 of 108 patients.
      4 (4%)4 (5%)0 (0%).25
      IQR, Interquartile range; SD, standard deviation; STAT, Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery Congenital Heart Surgery mortality categories; ECMO, extracorporeal membrane oxygenation.
      Data present for 106 of 108 patients.
      Sixty-three patients (58%) had underlying single-ventricle anomalies. At time of the index admission, 90 patients (83%) underwent open cardiac procedures that required the use of cardiopulmonary bypass, whereas the remaining 18 (17%) underwent closed cardiac procedures. After their cardiac procedure, 30 patients (28%) required extracorporeal membrane oxygenation (ECMO) support and 24 (22%) required unplanned cardiac reoperation (surgical revisions not including delayed sternal closure, mediastinal explorations, ECMO procedures, or procedure to treat noncardiac complications such as diaphragm plication or thoracic duct ligation), Table 1. Median postoperative ventilation, ICU, and hospital stay were 27.6 (IQR, 15.3-42.0), 44.7 (IQR, 36.0-60.0), and 54.0 (IQR, 44.0-76.0) days, respectively.

       Survival of the Prolonged ICU Stay Cohort

      Overall, hospital mortality occurred in 26 patients (24%). The differences in demographic, morphologic, and operative characteristics between hospital survivors and nonsurvivors are detailed in Table 1. On multivariable analysis, factors associated with hospital mortality were use of ECMO (odds ratio, 3.4 [95% confidence interval (CI), 1.3-9.1], P = .014) and renal failure requiring dialysis (odds ratio, 3.1 [95% CI, 1.0-10.0], P = .056).
      Subsequent to hospital discharge, there were 30 additional deaths (28%), with the majority of those additional deaths (n = 22, 73%) occurring within the first year after surgery and all occurring within 3 years from surgery. On importance, 28 of 30 late deaths were labeled cardiac, with 17 patients presenting with cardiac arrest at time of their death. Overall survival for the entire cohort was 57% at 1 year and 52% at 5 years after surgery, Figure 1. No factors were identified to be associated with the risk of death after hospital discharge, Table E1.
      Figure thumbnail gr1
      Figure 1A, Time-dependent survival and B, risk hazard of death over time after index surgical procedure in 108 neonates who required prolonged postoperative ICU stay. The solid lines in the parametric model represent parametric point estimates, and the dashed lines enclose the 95% confidence interval. Circles represent nonparametric estimates.

       Comparison of Postdischarge Survival Between Neonates Who Required Prolonged Stay in the ICU and Those Who Did Not

      We studied the effect of ICU stay as a continuous variable on postdischarge survival in all neonates who survived initial hospitalization after cardiac surgery (n = 1411). Longer stay in the ICU was associated with decreased survival after hospital discharge (log 2 ICU stay: hazard ratio [HR], 1.7 [95% CI, 1.5-1.9], P < .001). After adjusting for other established risk factors for mortality after neonatal cardiac surgery (ie, single-ventricle anomaly, low weight [≤2.5 kg], prematurity ≤36 weeks, genetic syndromes and extracardiac anomalies, use of cardiopulmonary bypass, and the Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery Congenital Heart Surgery mortality categories), we found that longer stay in the ICU continued to be associated with decreased survival after hospital discharge in the adjusted analysis (log 2 ICU stay: HR, 1.5 [95% CI, 1.3-1.7], P < .001).
      We compared postdischarge survival of neonates who required prolonged ICU stay >30 days and those who did not to better assess the effect of prolonged ICU stay on late events. After exclusion of those who did not survive initial hospital admission, our comparison cohort was composed of 82 neonates who required prolonged ICU stay >30 days and 1329 neonates who did not require ICU stay >30 days. The differences in the demographic, morphologic, and operative variables between the 2 comparison groups are detailed in Table 2. Neonates who required prolonged stay in the ICU were more likely to be premature, have genetic syndromes, and/or extracardiac malformations and to have single-ventricle anomalies. In addition, they were more likely to have undergone unplanned cardiac reoperation or have required postoperative ECMO support. Those who required prolonged postoperative ICU stay >30 days had significantly worse later survival compared with those who did not: 1-year survival, 77% versus 94%, and 8-year survival, 69% versus 92%, respectively (P < .001), Figure 2. Prolonged stay >30 days in the ICU was associated with decreased survival after hospital discharge (HR, 4.5 [95% CI, 2.9-6.9], P < .001) on the unadjusted analysis and continued to be associated with decreased survival on the adjusted analysis (HR, 2.5 [95% CI, 1.6-4.0], P < .001).
      Table 2The differences in the demographic, morphologic, and operative variables between neonates who required prolonged stay in the ICU and those who did not after exclusion of hospital nonsurvivors
      CharacteristicICU stay ≤30 d (n = 1329)ICU stay >30 d (n = 82)P value
      Age (d), median (IQR)6 (3-10)5 (3-9).537
      Weight (kg), mean ± SD3.2 ± 0.63.0 ± 0.6.023
      Weight ≤2.5 kg178 (13%)16 (20%).118
      Male sex802 (60%)43 (52%).156
      Premature ≤36 wk174 (13%)18 (23%).020
      Genetic syndromes and extracardiac malformations228 (17%)39 (48%)<.001
      Underlying cardiac anomaly
       Single ventricle359 (27%)44 (54%)<.001
       Two ventricles970 (73%)38 (46%)
      Initial procedure
       Palliation534 (40%)55 (68%)<.001
       Repair794 (60%)26 (32%)
      Use of cardiopulmonary bypass834 (63%)67 (82%)<.001
      Unplanned reoperation55 (4%)17 (21%)<.001
      ECMO requirement22 (2%)17 (21%)<.001
      ICU, Intensive care unit; IQR, interquartile range; SD, standard deviations; ECMO, extracorporeal membrane oxygenation.
      Figure thumbnail gr2
      Figure 2A and B, Parametric model for survival after index surgical procedure in neonates who survived initial hospitalization (hospital survivors ONLY) stratified by the requirement for prolonged postoperative ICU stay (yes vs no). ICU, Intensive care unit.
      There was an interesting interaction between underlying cardiac anomaly (single-ventricle vs 2 ventricles) and prolonged ICU stay that is best depicted in Figure 3. Although prolonged stay in the ICU was associated with worse survival in both groups, the effect of prolonged stay in the ICU on diminished late survival was more pronounced in neonates with 2 ventricles (8-year survival 68% vs 95%, HR, 8.0 [95% CI, 4.2-15.1], P < .001) than in those with single ventricle (8-year survival 66% vs 81%, HR, 2.0 [95% CI, 1.1-3.5], P = .021).
      Figure thumbnail gr3
      Figure 3Parametric model for survival after index surgical procedure in neonates who survived initial hospitalization (hospital survivors ONLY) showing the interaction between underlying cardiac anomaly (single ventricle vs 2 ventricles) and the requirement for prolonged postoperative ICU stay (yes vs no). ICU, Intensive care unit.

       Progress of Neonates With Single Ventricle Who Required Prolonged Stay in the ICU After First-Stage Palliation

      Comparison between single-ventricle hospital survivors who required a prolonged stay in the ICU after first-stage palliation versus those who did not was performed and is shown in Figure 4. The prolonged ICU stay cohort included 43 neonates (11%), whereas the control group of single-ventricle neonates who did not require prolonged ICU stay included 358 neonates (89%). Neonates who required prolonged ICU stay were less likely to progress to subsequent Glenn shunt (77% vs 88%, P = .037) and more likely to die or require heart transplantation before Glenn shunt (23% vs 11%, P = .013). Subsequent to Glenn shunt, there was a trend in those who required prolonged stay in the ICU to be more likely to die or require heart transplantation before Fontan (18% vs 10%, P = .093).
      Figure thumbnail gr4
      Figure 4Flow chart demonstrating the progress to subsequent palliation stages in neonates with single ventricle after first-stage palliation surgery with comparison shown between neonates who required prolonged ICU stay after first-stage palliation versus not. Only hospital survivors are included. SV, Single ventricle; ICU, intensive care unit.

      Discussion

      In our experience, approximately 3% of all pediatric cardiac surgical patients required prolonged ICU stay >30 days, with the majority being neonates; approximately 7% of all neonatal cardiac surgical patients required prolonged ICU stay >30 days. Several previous studies have identified factors associated with prolonged ICU stay after pediatric cardiac surgery. Those factors included patient characteristics such as neonatal age, low weight, prematurity, genetic syndromes, and extracardiac malformations.
      • Pagowska-Klimek I.
      • Pychynska-Pokorska M.
      • Krajewski W.
      • Moll J.J.
      Predictors of long intensive care unit stay following cardiac surgery in children.
      • Alsoufi B.
      • McCracken C.
      • Ehrlich A.
      • Mahle W.T.
      • Kogon B.
      • Border W.
      • et al.
      Single ventricle palliation in low weight patients is associated with worse early and midterm outcomes.
      • Kalfa D.
      • Krishnamurthy G.
      • Duchon J.
      • Najjar M.
      • Levasseur S.
      • Chai P.
      • et al.
      Outcomes of cardiac surgery in patients weighing <2.5 kg: affect of patient-dependent and -independent variables.
      • Alsoufi B.
      • Gillespie S.
      • Mahle W.
      • Deshpande S.
      • Kogon B.
      • Maher K.
      • et al.
      The impact of non-cardiac and genetic abnormalities on outcomes following neonatal congenital heart surgery.
      • Costello J.M.
      • Pasquali S.K.
      • Jacobs J.P.
      • He X.
      • Hill K.D.
      • Cooper D.S.
      • et al.
      Gestational age at birth and outcomes after neonatal cardiac surgery: an analysis of the Society of Thoracic Surgeons Congenital Heart Surgery Database.
      • Brown K.L.
      • Ridout D.A.
      • Goldman A.P.
      • Hoskote A.
      • Penny D.J.
      Risk factors for long intensive care unit stay after cardiopulmonary bypass in children.
      • Alsoufi B.
      • Manlhiot C.
      • Mahle W.T.
      • Kogon B.
      • Border W.L.
      • Cuadrado A.
      • et al.
      Low-weight infants are at increased mortality risk after palliative or corrective cardiac surgery.
      • Nathan M.
      • Karamichalis J.M.
      • Liu H.
      • Emani S.
      • Baird C.
      • Pigula F.
      • et al.
      Intraoperative adverse events can be compensated by technical performance in neonates and infants after cardiac surgery: a prospective study.
      Additional factors were related to underlying cardiac anomaly, such as the presence of single ventricle, and related to preoperative condition, need for preoperative mechanical ventilation, increased operative complexity, increased cardiopulmonary bypass time, and postoperative complications.
      • Pagowska-Klimek I.
      • Pychynska-Pokorska M.
      • Krajewski W.
      • Moll J.J.
      Predictors of long intensive care unit stay following cardiac surgery in children.
      • Alsoufi B.
      • McCracken C.
      • Ehrlich A.
      • Mahle W.T.
      • Kogon B.
      • Border W.
      • et al.
      Single ventricle palliation in low weight patients is associated with worse early and midterm outcomes.
      • Kalfa D.
      • Krishnamurthy G.
      • Duchon J.
      • Najjar M.
      • Levasseur S.
      • Chai P.
      • et al.
      Outcomes of cardiac surgery in patients weighing <2.5 kg: affect of patient-dependent and -independent variables.
      • Alsoufi B.
      • Gillespie S.
      • Mahle W.
      • Deshpande S.
      • Kogon B.
      • Maher K.
      • et al.
      The impact of non-cardiac and genetic abnormalities on outcomes following neonatal congenital heart surgery.
      • Costello J.M.
      • Pasquali S.K.
      • Jacobs J.P.
      • He X.
      • Hill K.D.
      • Cooper D.S.
      • et al.
      Gestational age at birth and outcomes after neonatal cardiac surgery: an analysis of the Society of Thoracic Surgeons Congenital Heart Surgery Database.
      • Brown K.L.
      • Ridout D.A.
      • Goldman A.P.
      • Hoskote A.
      • Penny D.J.
      Risk factors for long intensive care unit stay after cardiopulmonary bypass in children.
      • Alsoufi B.
      • Manlhiot C.
      • Mahle W.T.
      • Kogon B.
      • Border W.L.
      • Cuadrado A.
      • et al.
      Low-weight infants are at increased mortality risk after palliative or corrective cardiac surgery.
      • Nathan M.
      • Karamichalis J.M.
      • Liu H.
      • Emani S.
      • Baird C.
      • Pigula F.
      • et al.
      Intraoperative adverse events can be compensated by technical performance in neonates and infants after cardiac surgery: a prospective study.
      Our current findings are comparable with those from previous studies, and the prevalence of prematurity, low weight, genetic syndromes, extracardiac malformations, and single-ventricle anomalies was significantly greater in neonates who required prolonged stay in the ICU. Similarly, neonates who required prolonged ICU stay were more likely to have needed unplanned cardiac reoperation or ECMO support. Those findings underscore the effect of residual lesions that have been linked to prolonged lengths of stay, increase in ventilation time, and mortality.
      • Nathan M.
      • Karamichalis J.M.
      • Liu H.
      • Emani S.
      • Baird C.
      • Pigula F.
      • et al.
      Intraoperative adverse events can be compensated by technical performance in neonates and infants after cardiac surgery: a prospective study.
      Late outcomes beyond hospital discharge in pediatric cardiac surgery patients who required prolonged postoperative ICU stay have not been well explored. In a study that examined children with prolonged pediatric ICU stay via the Australia and New Zealand registry, children with cardiac anomalies comprised the second poorest long-term survival among the examined disease categories, with single-ventricle anomalies being an independent risk factor of death.
      • Namachivayam S.P.
      • Alexander J.
      • Slater A.
      • Millar J.
      • Erickson S.
      • Tibballs J.
      • et al.
      Five-year survival of children with chronic critical illness in Australia and New Zealand.
      In our surgical cohort, we demonstrated that there was an increase in mortality risk beyond hospital discharge with the majority of deaths occurring within the first year after the operation. Of importance, the vast majority of deaths were cardiac and many readmissions occurred after cardiac arrest with subsequent deterioration of status and patient demise. This finding highlights that the most important factor in determining late survival is the presence of residual cardiac lesions or myocardial dysfunction that will subject the patient to continual or recurrent deterioration of the cardiac status and consequently diminished survival.
      We have examined the echocardiograms of patients who suffered from late mortality subsequent to hospital discharge. Although there were several patients who had residual lesions such as ventricular dysfunction, atrioventricular valve failure, pulmonary artery stenosis, etc, the presence of residual lesions was not found to be significantly associated with death in our series. Our analysis likely is limited in that regard because of the lack of universal definition of what constitutes a significant residual lesion, in addition to the small sample size from the statistical standpoint. Nonetheless, prolonged stay in the ICU should prompt physicians to identify and treat any important residual lesion that might have contributed to the complicated postoperative course early on to increase both hospital and late survival. This finding also suggests the need for a more vigilant postdischarge follow-up and the possible need for early assessment for candidacy for heart transplantation in those with myocardial dysfunction or residual lesions that are not amenable to surgical repair. In addition to cardiac deaths, a few patients died as the result of lingering multiorgan problems that were acquired during their complicated initial hospitalization, such as neurologic and lung injury.
      We attempted to identify what postoperative complications or risk factors were associated with poor early or late outcomes. The 2 factors that were associated with hospital mortality were renal failure and need for ECMO. Our analysis might again be limited with this regard because of the overlap between the postoperative complications, with many patients suffering more than 1 complication, in addition to the small sample size from the statistical standpoint. Postoperative need for ECMO and prolonged duration of ECMO have been shown to be associated with prolonged stay in the ICU and early and late mortality.
      • Gupta P.
      • Robertson M.J.
      • Beam B.
      • Gossett J.M.
      • Schmitz M.L.
      • Carroll C.L.
      • et al.
      Relationship of ECMO duration with outcomes after pediatric cardiac surgery: a multi-institutional analysis.
      • Alsoufi B.
      • Al-Radi O.O.
      • Gruenwald C.
      • Lean L.
      • Williams W.G.
      • McCrindle B.W.
      • et al.
      Extra-corporeal life support following cardiac surgery in children: analysis of risk factors and survival in a single institution.
      • Lequier L.
      • Joffe A.R.
      • Robertson C.M.
      • Dinu I.A.
      • Wongswadiwat Y.
      • Anton N.R.
      • et al.
      Two-year survival, mental, and motor outcomes after cardiac extracorporeal life support at less than five years of age.
      • Chan T.
      • Thiagarajan R.R.
      • Frank D.
      • Bratton S.L.
      Survival after extracorporeal cardiopulmonary resuscitation in infants and children with heart disease.
      • Alsoufi B.
      • Al-Radi O.O.
      • Nazer R.I.
      • Gruenwald C.
      • Foreman C.
      • Williams W.G.
      • et al.
      Survival outcomes after rescue extracorporeal cardiopulmonary resuscitation in pediatric patients with refractory cardiac arrest.
      This continuous mortality risk may be caused by the prolonged effect of insults incurred by surgery and postoperative intensive care stay or may simply reflect the severity of those patients' conditions. Regardless of the cause, patients who required prolonged stay in the ICU should be considered as a high-risk group and would likely benefit from more rigorous work up to identify residual lesions and close follow-up and monitoring at home. Although many of those patients possessed established factors that would categorize them as high risk, prolonged stay in the ICU should be considered as an independent risk factor that should draw an additional caution.
      An interesting and important finding in our study was that despite the negative effect of prolonged stay in the ICU on late outcomes was evident in all patients, that effect was more prominent in patients with 2 ventricles than in those with single ventricle. Despite the greater hospital mortality in single-ventricle patients, neonates with 2 ventricles who required prolonged postoperative ICU stay had more attrition beyond hospital discharge and in fact their late survival was equal to that for neonates with single-ventricle anomalies. Taking into consideration that survival in neonates with 2 ventricles generally is superior to that with single ventricle, this interesting finding of comparable late outcomes suggests a more pronounced deviation from expected survival in children with 2 ventricles requiring prolonged postoperative ICU stay. This finding may reflect the fact that those patients with 2 ventricles might be sicker or have more residual lesions to require prolonged ICU stay because their expected requirement for ICU care should be shorter than that for single-ventricle patients. It is also plausible that single-ventricle patients may have experienced a more rigorous follow-up, given the need for subsequent surgical palliation, in which case would again suggest the need for more intensive monitoring of all those patients after hospital discharge.
      In children with underlying single-ventricle anomalies, prolonged stay in the ICU was associated with high hospital mortality and increased interstage mortality before Glenn shunt. Nonetheless, the progression pattern after Glenn did not seem to be significantly affected, with 82% of patients receiving or qualifying for the subsequent Fontan operation.
      In our study, we identified the effect of stay in the ICU as a continuous variable on late survival in all neonates who survived to hospital discharge after their initial neonatal cardiac surgery. We found that longer stay in the ICU was associated with worse survival in those patients. When we divided our patient population to 2 groups, we used the 30-day cutoff based on the work presented in previous studies in the pediatric and adult population that defined chronic critical illness as ICU stay more than 28-30 days. This cutoff has not been validated yet in the pediatric cardiac population. Our finding of worse outcomes in neonates with 2 ventricles suggest that what should be considered prolonged ICU stay for neonates with single ventricle might not necessarily be the same for those with 2 ventricles, who generally are expected to require shorter ICU stay. In studies on adult population who underwent cardiac surgery and experienced prolonged ICU stay, long-term survival and quality of life was poor, with 80% postdischarge mortality at 6 years and only 7% of the survivors regaining full function.
      • Gaudino M.
      • Girola F.
      • Piscitelli M.
      • Martinelli L.
      • Anselmi A.
      • Della Vella C.
      • et al.
      Long-term survival and quality of life of patients with prolonged postoperative intensive care unit stay: unmasking an apparent success.
      On this basis, the authors have argued that the expensive and exhausting efforts to sustain those patients' lives may not be justifiable. The considerations for resource allocation cannot be overlooked in the pediatric population, because those patients require tremendous resources.
      • Pasquali S.K.
      • He X.
      • Jacobs M.L.
      • Shah S.S.
      • Peterson E.D.
      • Gaies M.G.
      • et al.
      Excess costs associated with complications and prolonged length of stay after congenital heart surgery.
      Although caring for those patients is resource intensive, our data demonstrated greater late survival than that reported for adults.
      Furthermore, given the multitude of factors that affect survival in those patients, we were unable to necessarily identify a pattern predictive of futile outcomes, short of renal failure requiring dialysis, to warrant early termination of care. Nonetheless, we have identified that prolonged stay in the ICU should be regarded as a marker for increased risk of late mortality and therefore approaches to identify and possibly treat modifiable factors could potentially improve prognosis. Finally, the association between prolonged stay in the ICU and quality of life in pediatric cardiac patients after discharge is not known and needs to be further determined to better understand the tangible prognosis of those challenging patients and help making clinical and administrative decisions.

      Summary

      Inherent patient, anatomic, and postoperative risk factors are prevalent in neonates who require prolonged stay in the ICU after cardiac surgery. Prolonged postoperative stay in the ICU is associated with high hospital and significant postdischarge mortality, mainly during the first year. In neonates with underlying single-ventricle anomalies, prolonged stay in the ICU is associated with high hospital and interstage mortality and usual progression subsequent to Glenn shunt. On the other hand, prolonged ICU stay after surgery in patients with 2 ventricles anomalies is associated with high hospital mortality and considerable decrease in late survival, suggesting a more pronounced deviation from expected survival in children with 2-ventricle anomalies requiring prolonged postoperative ICU stay.

       Conflict of Interest Statement

      Authors have nothing to disclose with regard to commercial support.

      Appendix

      Table E1Univariate analysis of factors associated with survival subsequent to hospital discharge in neonates who required prolonged postoperative stay in the ICU
      CharacteristicUnivariate
      Hazard ratio (95% CI)P value
      Male sex1.58 (0.72-3.48).26
      Age at surgery0.97 (0.91-1.04).37
      Premature ≤ 36 wk1.40 (0.58-3.33).45
      Genetic syndromes and extracardiac malformations1.90 (0.86-4.19).11
      Weight ≤2.5 kg0.79 (0.27-2.28).66
      Two-ventricle anomaly0.97 (0.45-2.09).94
      Cardiopulmonary bypass use1.32 (0.45-3.82).61
      Postoperative ECMO use0.47 (0.14-1.56).22
      Unplanned reoperation1.01 (0.38-2.67).99
      STAT category 4 and 51.13 (0.73-1.74).57
      Sepsis1.53 (0.71-3.34).28
      Renal failure0.32 (0.04-2.37).27
      Cerebrovascular accident1.28 (0.30-5.40).74
      Necrotizing enterocolitis1.73 (0.70-4.32).24
      Gastrointestinal bleeding1.74 (0.24-12.84).59
      Chylothorax1.11 (0.45-2.76).83
      Vocal cord paralysis1.48 (0.35-6.27).59
      CI, Confidence interval; ECMO, extracorporeal membrane oxygenation; STAT, Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery Congenital Heart Surgery mortality categories.

      References

        • Lagercrantz E.
        • Lindblom D.
        • Sartipy U.
        Survival and quality of life in cardiac surgery patients with prolonged intensive care.
        Ann Thorac Surg. 2010; 89: 490-495
        • Bapat V.
        • Allen D.
        • Young C.
        • Roxburgh J.
        • Ibrahim M.
        Survival and quality of life after cardiac surgery complicated by prolonged intensive care.
        J Card Surg. 2005; 20: 212-217
        • Gaudino M.
        • Girola F.
        • Piscitelli M.
        • Martinelli L.
        • Anselmi A.
        • Della Vella C.
        • et al.
        Long-term survival and quality of life of patients with prolonged postoperative intensive care unit stay: unmasking an apparent success.
        J Thorac Cardiovasc Surg. 2007; 134: 465-469
        • Grothusen C.
        • Attmann T.
        • Friedrich C.
        • Freitag-Wolf S.
        • Haake N.
        • Cremer J.
        • et al.
        Predictors for long-term outcome and quality of life of patients after cardiac surgery with prolonged intensive care unit stay.
        Interv Med Appl Sci. 2013; 5: 3-9
        • Joskowiak D.
        • Kappert U.
        • Matschke K.
        • Tugtekin S.
        Prolonged intensive care unit stay of patients after cardiac surgery: initial clinical results and follow-up.
        Thorac Cardiovasc Surg. 2013; 61: 701-707
        • Gaies M.G.
        • Gurney J.G.
        • Yen A.H.
        • Napoli M.L.
        • Gajarski R.J.
        • Ohye R.G.
        • et al.
        Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass.
        Pediatr Crit Care Med. 2010; 11: 234-238
        • Ettema R.G.
        • Peelen L.M.
        • Schuurmans M.J.
        • Nierich A.P.
        • Kalkman C.J.
        • Moons K.G.
        Prediction models for prolonged intensive care unit stay after cardiac surgery: systematic review and validation study.
        Circulation. 2010; 122: 682-689
        • Messaoudi N.
        • De Cocker J.
        • Stockman B.A.
        • Bossaert L.L.
        • Rodrigus I.E.
        Is EuroSCORE useful in the prediction of extended intensive care unit stay after cardiac surgery?.
        Eur J Cardiothorac Surg. 2009; 36: 35-39
        • Pasquali S.K.
        • He X.
        • Jacobs M.L.
        • Shah S.S.
        • Peterson E.D.
        • Gaies M.G.
        • et al.
        Excess costs associated with complications and prolonged length of stay after congenital heart surgery.
        Ann Thorac Surg. 2014; 98: 1660-1666
        • Pagowska-Klimek I.
        • Pychynska-Pokorska M.
        • Krajewski W.
        • Moll J.J.
        Predictors of long intensive care unit stay following cardiac surgery in children.
        Eur J Cardiothorac Surg. 2011; 40: 179-184
        • Alsoufi B.
        • McCracken C.
        • Ehrlich A.
        • Mahle W.T.
        • Kogon B.
        • Border W.
        • et al.
        Single ventricle palliation in low weight patients is associated with worse early and midterm outcomes.
        Ann Thorac Surg. 2015; 99: 668-676
        • Kalfa D.
        • Krishnamurthy G.
        • Duchon J.
        • Najjar M.
        • Levasseur S.
        • Chai P.
        • et al.
        Outcomes of cardiac surgery in patients weighing <2.5 kg: affect of patient-dependent and -independent variables.
        J Thorac Cardiovasc Surg. 2014; 148: 2499-2506
        • Alsoufi B.
        • Gillespie S.
        • Mahle W.
        • Deshpande S.
        • Kogon B.
        • Maher K.
        • et al.
        The impact of non-cardiac and genetic abnormalities on outcomes following neonatal congenital heart surgery.
        Semin Thorac Cardiovasc Surg. November 17, 2015; ([Epub ahead of print])
        • Costello J.M.
        • Pasquali S.K.
        • Jacobs J.P.
        • He X.
        • Hill K.D.
        • Cooper D.S.
        • et al.
        Gestational age at birth and outcomes after neonatal cardiac surgery: an analysis of the Society of Thoracic Surgeons Congenital Heart Surgery Database.
        Circulation. 2014; 129: 2511-2517
        • Brown K.L.
        • Ridout D.A.
        • Goldman A.P.
        • Hoskote A.
        • Penny D.J.
        Risk factors for long intensive care unit stay after cardiopulmonary bypass in children.
        Crit Care Med. 2003; 31: 28-33
        • Alsoufi B.
        • Manlhiot C.
        • Mahle W.T.
        • Kogon B.
        • Border W.L.
        • Cuadrado A.
        • et al.
        Low-weight infants are at increased mortality risk after palliative or corrective cardiac surgery.
        J Thorac Cardiovasc Surg. 2014; 148: 2508-2514.e1
        • Nathan M.
        • Karamichalis J.M.
        • Liu H.
        • Emani S.
        • Baird C.
        • Pigula F.
        • et al.
        Intraoperative adverse events can be compensated by technical performance in neonates and infants after cardiac surgery: a prospective study.
        J Thorac Cardiovasc Surg. 2011; 142 (1107 e1-5): 1098-1107
        • Namachivayam S.P.
        • Alexander J.
        • Slater A.
        • Millar J.
        • Erickson S.
        • Tibballs J.
        • et al.
        Five-year survival of children with chronic critical illness in Australia and New Zealand.
        Crit Care Med. 2015; 43: 1978-1985
        • Gupta P.
        • Robertson M.J.
        • Beam B.
        • Gossett J.M.
        • Schmitz M.L.
        • Carroll C.L.
        • et al.
        Relationship of ECMO duration with outcomes after pediatric cardiac surgery: a multi-institutional analysis.
        Minerva Anestesiol. 2015; 81: 619-627
        • Alsoufi B.
        • Al-Radi O.O.
        • Gruenwald C.
        • Lean L.
        • Williams W.G.
        • McCrindle B.W.
        • et al.
        Extra-corporeal life support following cardiac surgery in children: analysis of risk factors and survival in a single institution.
        Eur J Cardiothorac Surg. 2009; 35: 1004-1011
        • Lequier L.
        • Joffe A.R.
        • Robertson C.M.
        • Dinu I.A.
        • Wongswadiwat Y.
        • Anton N.R.
        • et al.
        Two-year survival, mental, and motor outcomes after cardiac extracorporeal life support at less than five years of age.
        J Thorac Cardiovasc Surg. 2008; 136: 976-983
        • Chan T.
        • Thiagarajan R.R.
        • Frank D.
        • Bratton S.L.
        Survival after extracorporeal cardiopulmonary resuscitation in infants and children with heart disease.
        J Thorac Cardiovasc Surg. 2008; 136: 984-992
        • Alsoufi B.
        • Al-Radi O.O.
        • Nazer R.I.
        • Gruenwald C.
        • Foreman C.
        • Williams W.G.
        • et al.
        Survival outcomes after rescue extracorporeal cardiopulmonary resuscitation in pediatric patients with refractory cardiac arrest.
        J Thorac Cardiovasc Surg. 2007; 134: 952-959