Carmel Delzoppo

Three decades of pediatric intensive care: Who was admitted, what happened in intensive care, and what happened afterward

Objective: To describe the characteristics of children admitted to intensive care in 1982, 1995, and 2005–2006, and their long-term outcome. Setting: Pediatric intensive care unit in a university-affiliated childrens hospital. Design/Methods: Information for 2005–2006 admissions was obtained from pediatric intensive care unit database, and long-term outcome was ascertained through telephone interviews. Results were compared to previous cohorts from 1982 and 1995. Results: A total of 4010 children were admitted on 5250 occasions. Readmissions increased from 11% for 1982 to 31% in 2005 to 2006 (p < .001). In 2005–2006, fewer children were admitted after accidents (p < .001), or with croup (p < .001), or epiglottitis (p = .01), and 8% were treated with noninvasive ventilation compared to none in 1982 (p < .0001). Among children aged ≥1 month, pediatric intensive care unit length of stay remained constant. The risk of death predicted by the Pediatric Index of Mortality (PIM) remained constant (approximately 6%) between 1995 and 2005–2006.The proportion that died in the pediatric intensive care unit fell from 11.0% in 1982 to 4.8% in 2005–2006 (p < .001). Among children aged ≥1 month, proportion admitted with a preexisting moderate or severe disability was similar: 12.0% in 1982 and 14.6% in 2005–2006 (p = .11), but the proportion with a moderate or severe disability at follow-up increased from 8.4% in 1982 to 17.9% in 2005–2006 (p < .001). The proportion of children aged ≥1 month who either died in the pediatric intensive care unit or survived with disability did not improve: it was 19.4% in 1982 and 22.7% in 2005–2006. Conclusion: Over the last three decades, the length of stay in the pediatric intensive care unit and the severity of illness have not changed, but there has been a substantial reduction in pediatric intensive care unit mortality. However, the proportion of survivors with moderate or severe disability increased significantly. Some children who would have been allowed to die in 1982 and 1995 were kept alive in 2005–2006, but survived with disability. This trend has important implications for our patients and their families, and for the community as a whole.

Hemolysis in pediatric patients receiving centrifugal-pump extracorporeal membrane oxygenation: prevalence, risk factors, and outcomes.

Objectives:To explore the prevalence and risk factors for hemolysis in children receiving extracorporeal membrane oxygenation and examine the relationship between hemolysis and adverse outcomes. Design:Retrospective, single-center study. Setting:Tertiary PICU. Patients:Two hundred seven children receiving extracorporeal membrane oxygenation. Interventions:None. Measurements and Main Results:Plasma-free hemoglobin was tested daily and hemolysis was diagnosed based on peak plasma-free hemoglobin as mild (< 0.5 g/L), moderate (0.5–1.0 g/L), or severe (> 1.0 g/L). Gender, age, weight, diagnosis, oxygenator type, cannulation site, mean venous inlet pressure, mean pump speed, mean flow, and visible clots in the extracorporeal membrane oxygenation circuit were entered into the ordered logistic regression model to identify risk factors of hemolysis. Complications and clinical outcomes were compared across four hemolysis groups. Of the 207 patients, 69 patients (33.3%; 95% CI, 27.0–40.2%) did not have hemolysis, 98 patients (47.3%; 95% CI, 40.4–54.4%) had mild hemolysis, 26 patients (12.5%; 95% CI, 8.4–17.9%) had moderate hemolysis, and 14 patients (6.8%; 95% CI, 3.7–11.1%) had severe hemolysis with a median peak plasma-free hemoglobin of 1.51 g/L (1.18–2.05 g/L). The independent risk factors for hemolysis during extracorporeal membrane oxygenation were use of Quadrox D (odds ratio, 7.25; 95% CI, 3.10–16.95; p < 0.001) or Lilliput (odds ratio, 37.32; 95% CI, 8.95–155.56; p < 0.001) oxygenators, mean venous inlet pressure (odds ratio, 0.95; 95% CI, 0.91–0.98; p = 0.002), and mean pump speed (odds ratio, 2.89; 95% CI, 1.36–6.14; p = 0.006). Patients with hemolysis were more likely to experience a longer extracorporeal membrane oxygenation run and require more blood products. After controlling for age, weight, pediatric index of mortality 2, and diagnosis, patients with severe hemolysis were more likely to die in the ICU (odds ratio, 5.93; 95% CI, 1.64–21.43; p = 0.007) and in hospital (odds ratio, 6.34; 95% CI, 1.71–23.54; p = 0.006). Conclusions:Hemolysis during extracorporeal membrane oxygenation with centrifugal pumps was common and associated with a number of adverse outcomes. Risk factors for hemolysis included oxygenator types, mean venous inlet pressure, and mean pump speed. Further studies are warranted comparing pump types while controlling both physical and nonphysical confounders.

Long-stay children in intensive care: long-term functional outcome and quality of life from a 20-yr institutional study.

Objective: Long-stay patients (≥28 days) in pediatric intensive care units consume a disproportionate amount of resources, and very few studies have reported their outcome. We determined the long-term outcome of these children admitted to intensive care over a 20-yr period (January 1, 1989 to December 31, 2008). Setting: Pediatric intensive care unit in a university-affiliated tertiary pediatric hospital in Melbourne, Australia Methods: Demographic and clinical characteristics were compared after dividing patients into four groups depending on year of admission (1989–1993, 1994–1998, 1999–2003, and 2004–2008). Preadmission health status and long-term functional outcome were evaluated by a modified Glasgow outcome scale. Quality of life was assessed by using the Health Utilities Index Mark 1. Results: Over the 20-yr period, 233 long-stay patients had 269 long stay admission episodes to the pediatric intensive care unit, accounting for 1% (269 of 27,536) of all pediatric intensive care unit admissions and utilized 18.5% (15,740 of 85,032) of occupied bed days. Bed occupancy of long stay patients (as percentage of overall pediatric intensive care unit bed occupancy) increased from 8% in 1989 to 21% in 2008 (p = .001). Median age at admission was 4.2 months [interquartile range 0.38–41.5] and median length of stay was 40 days [interquartile range 32–57]. One hundred sixteen of 233 (49.8%) patients had died at the time of follow-up. Children who died were younger compared to survivors (median 3.4 months [interquartile range 0.38–41.5 vs. median 7.6 months, interquartile range 0.6–71.1, p = .026], had a higher proportion of comorbid illness (91% vs. 80%, p = .026), and 63% had a preexisting moderate or severe disability compared to 51% of survivors (p = .215). One hundred seventeen of 233 children survived and long-term functional outcome was favorable (normal, functionally normal, or mild disability) in 27% (63 of 233) and unfavorable (moderate or severe disability) for 17.2% (40 of 233). Outcome status was not known for 6% (14 of 233). Among survivors (n = 117), more than 50% (63 of 117) had favorable outcome. The quality of life in patients aged >2 yrs at follow up was good in 21% (40 of 222), moderate in 8% (16 of 222), poor quality in 68% (130 of 222, this includes deaths), and very poor in 3% (5 of 222). Conclusions: More than two-thirds of children who stay in intensive care for ≥28 days have an unfavorable outcome (moderate disability, severe disability, or death). Long-stay patients in pediatric intensive care utilized a large proportion of resources and this utilization has considerably increased with time. Service provision and policy making should expect worsening of these trends in the future; its effects on critical care bed availability and overall activity levels could be substantial.

Hemofiltration is not associated with increased mortality in children receiving extracorporeal membrane oxygenation.

Objectives: To investigate whether the use of continuous renal replacement therapy is independently associated with increased in-hospital mortality in children on extracorporeal membrane oxygenation. Design: Retrospective, 1:1 propensity-matched cohort study. Setting: Tertiary PICU. Patients: Eighty-six children on extracorporeal membrane oxygenation, 43 of whom also received hemofiltration. Interventions: None. Measurements and Main Results: Demographics, pre–extracorporeal membrane oxygenation hemodynamic data, fluid status, and biochemistry tests were collected, as well as duration of extracorporeal membrane oxygenation, blood product use, complications, and mortality. Forty-three children receiving extracorporeal membrane oxygenation and continuous renal replacement therapy were matched to a cohort of 43 children on extracorporeal membrane oxygenation not receiving continuous renal replacement therapy. The main indication for hemofiltration was fluid overload in 29 patients (67.4%), renal failure in nine patients (20.9%), and electrolyte abnormalities in five patients (11.6%). The median duration of hemofiltration was 108 hours (47–209 hr). Patients receiving hemofiltration had a longer duration of extracorporeal membrane oxygenation (127 hr [94–302 hr] vs 121 hr [67–182 hr]; p = 0.05) and received more platelet transfusions (0.91 mL/kg/hr [0.43–1.58 mL/kg/hr] vs 0.63 mL/kg/hr [0.30–0.79 mL/kg/hr]; p = 0.01). There were otherwise no differences in mechanical or patient-related complications between both groups. There was no difference in the proportion of patients who were successfully decannulated (81.4% vs 74.4%; p = 0.44), survived to ICU discharge (65.1% vs 55.8%; p = 0.38), or survived to hospital discharge (62.8% vs 48.8%; p = 0.19) in the controls versus the hemofiltration group. Conclusions: In-hospital mortality was similar between children on extracorporeal membrane oxygenation with and without hemofiltration although hemofiltration appeared to be associated with a slight increase in the duration of extracorporeal membrane oxygenation and more liberal platelet transfusions.

Prediction of Multidimensional Fatigue After Childhood Brain Injury.

Objectives: To determine (1) the presence of fatigue symptoms and predictors of fatigue after childhood brain injury and examine (2) the feasibility, reliability, and validity of a multidimensional fatigue measure (PedsQL Multidimensional Fatigue Scale [MFS]) obtained from parent and child perspectives. Setting: Emergency and intensive care units of a hospital in Melbourne, Australia. Participants: Thirty-five families (34 parent-proxies and 32 children) aged 8 to 18 years (mean child age = 13.29 years) with traumatic brain injury (TBI) of all severities (27 mild, 5 moderate, and 3 severe) admitted to the Royal Childrens Hospital. Design: Longitudinal prospective study. Fatigue data collected at 6-week follow-up (mean = 6.9 weeks). Main Outcome Measures: Postinjury child- and parent-rated fatigue (PedsQL MFS), mood, sleep, and pain based on questionnaire report: TBI severity (mild vs moderate/severe TBI). Results: A score greater than 2 standard deviations below healthy control data indicated the presence of abnormal fatigue, rates of which were higher compared with normative data for both parent and child reports (47% and 29%). Fatigue was predicted by postinjury depression and sleep disturbance for parent, but not child ratings. Fatigue, as rated by children, was not significantly predicted by TBI severity or other symptoms. The PedsQL MFS demonstrated acceptable measurement properties in child TBI participants, evidenced by good feasibility and reliability (Cronbach &agr; values >0.90). Interrater reliability between parent and child reports was poor to moderate. Conclusions: Results underscore the need to assess fatigue and associated sleep-wake disturbance and depression after child TBI from both parent and child perspectives.

Prevalence of dysglycemia and association with outcomes in pediatric extracorporeal membrane oxygenation.

Objectives: To evaluate the relationship between glucose derangement, insulin administration, and mortality among children on extracorporeal membrane oxygenation. Design: Retrospective cohort. Setting: Tertiary PICU. Patients: Two hundred nine children receiving extracorporeal membrane oxygenation, including 97 neonates. Interventions: None. Measurements and Main Results: Hyperglycemia and severe hyperglycemia were defined as a single blood glucose level greater than 15 mmol/L (270 mg/dL) and greater than 20 mmol/L (360 mg/dL), respectively. Hypoglycemia and severe hypoglycemia were defined as any single glucose level less than 3.3 mmol/L (60 mg/dL) and less than 2.2 mmol/L (40 mg/dL), respectively. A total of 15,912 glucose values were recorded. The median number of glucose values was 59 per patient, corresponding to a mean 0.53 ± 0.12 tests per hour. Sixty-nine patients (33.0%) without dysglycemia and who received no insulin were defined as the control group. Eighty-nine (42.6%) and 26 (12.4%) patients developed hyperglycemia and severe hyperglycemia, respectively. Sixty-three (30.1%) and 17 (8.1%) patients developed hypoglycemia and severe hypoglycemia, respectively. Sixty-one patients (29.2%) received IV insulin during extracorporeal membrane oxygenation. Both hyperglycemia and hypoglycemia were associated with increased mortality on extracorporeal membrane oxygenation (46% and 48%, respectively, vs 29% of controls; p = 0.03). However, after adjusting for severity of illness and extracorporeal membrane oxygenation complications, abnormal glucose levels were not independently related to mortality. Conclusions: Dysglycemia in children on extracorporeal membrane oxygenation was common but not independently associated with increased mortality. The optimal glucose range for this high-risk population requires further investigation.