Cardiac Intensive Care
Health-Related Quality of Life in Pediatric Cardiac Extracorporeal Life Support Survivors* Gonzalo Garcia Guerra, MD, MSc1; Charlene M. T. Robertson, MD1,2; Gwen Y. Alton, RN1; Ari R. Joffe, MD1; Elham Khodayari Moez, MSc2; Irina A Dinu, PhD2,3; David B. Ross, MD1,4; Ivan M. Rebeyka, MD1,4; Laurance Lequier, MD1; the Western Canadian Complex Pediatric Therapies Follow-Up Group
Objective: To assess the health-related quality of life of children who received cardiac extracorporeal life support. We hypothesized that extracorporeal life support survivors have lower health-related quality-of-life scores when compared with a healthy sample, with children with chronic conditions, and with children who had surgery for congenital heart disease and did not receive extracorporeal life support. Design: Prospective cohort study. Setting: Stollery Children’s Hospital and Complex Pediatric Therapies Follow-up Program clinics. Patients: Children less than or 5 years old with diagnosis of cardiac disease (congenital or acquired) who received extracorporeal life support at the Stollery Children’s Hospital from 1999 to 2009. Interventions: None. Measurements and Main Results: Health-related quality of life was assessed using the PedsQL 4.0 Generic Core Scales completed by the children’s parents at the time of follow-up. Forty-seven car*See also p. 775. 1 Department of Pediatrics, University of Alberta, Edmonton, AB, Canada. 2 Pediatric Rehabilitation Outcomes Evaluation and Research Unit, Glenrose Rehabilitation Hospital, Edmonton, AB, Canada. 3 School of Public Health, University of Alberta, Edmonton, AB, Canada. 4 Department of Surgery, University of Alberta, Edmonton, AB, Canada. This work was performed at University of Alberta, Edmonton, AB, Canada. Chairs of the Western Canadian Complex Pediatric Therapies Follow-up Group are Charlene M.T. Robertson, Reg Sauve, and Gwen Y. Alton. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ pccmjournal). Supported, in part, by the Registry and Follow-up of Complex Pediatric Therapies Project, Alberta Health and Wellness. Dr. Robertson’s institution received grant support from Alberta Health and Wellness. The remaining authors disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: [email protected] Copyright © 2014 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: 10.1097/PCC.0000000000000212
720
www.pccmjournal.org
diac extracorporeal life support survivors had their health-related quality of life assessed at a median age of 4 years. Compared with a healthy sample, children who received venoarterial extracorporeal life support have significantly lower PedsQL (64.9 vs 82.2; p < 0.0001). The PedsQL scores of children who received extracorporeal life support were also significantly lower than those of children with chronic health conditions (64.9 vs 73.1; p = 0.007). Compared with children with congenital heart disease who underwent cardiac surgery early in infancy and who did not receive extracorporeal life support, extracorporeal life support survivors had significantly lower PedsQL scores (64.9 vs 81.1; p < 0.0001). Multiple linear regression analysis found an independent association between both higher inotrope score in the first 24 hours of extracorporeal life support and longer hospital length of stay, with lower PedsQL scores. Conclusions: Pediatric cardiac extracorporeal life support survivors showed lower health-related quality of life than healthy children, children with chronic conditions, and children with congenital heart disease who did not receive extracorporeal life support. (Pediatr Crit Care Med 2014; 15:720–727) Key Words: critical care; extracorporeal membrane oxygenation; heart diseases; outcomes research; pediatrics; quality of life
S
ince Bartlett et al (1) started using extracorporeal membrane oxygenation (ECMO) to support children with congenital heart disease (CHD) in the 1970s, extracorporeal life support (ECLS) has evolved from an extraordinary, last resort, life-saving intervention to a standard of care in many pediatric tertiary centers with thousands of patients supported to date (1–3). Indications for ECLS now include cardiovascular failure due to heart failure, low cardiac output syndrome, inability to wean from cardiopulmonary bypass after cardiac surgery, sepsis, and sustained cardiac arrest, as well as respiratory failure due to acute respiratory distress syndrome, pneumonia, airway obstruction, etc. ECLS has also gone from an intervention using large complex equipment available only in the operating room, to more “simple” circuits and components October 2014 • Volume 15 • Number 8
Cardiac Intensive Care
used in the ICU, with more compact devices that can be used to transport patients from the operating room to other parts of the hospital or even between centers. It has been shown that ECLS can improve short-term outcome and survival of children with congenital or acquired heart disease that otherwise would have had significant adverse outcomes including death (4–7). Many centers have reported survival rates in pediatric cardiac ECLS above 40% (4, 8–10). However, ECLS is not exempt from complications, morbidity, and potential long-term adverse outcomes (11, 12). Hospital survival outcomes after ECMO have been reported in the Extracorporeal Life Support Organization registry for decades; however, long-term outcomes and health-related quality of life (HRQL) in survivors have not been part of this registry data or previous publications (3, 10). HRQL is a multidimensional construct that includes physical, mental, and social well-being dimensions and not merely the absence of disease. The literature consistently suggests that the clinical condition or severity of disease is not necessarily associated with perception of quality of life (12). Accordingly, HRQL refers to a person’s subjective, emotional evaluation and reaction to his/her health condition (13). To date, there have been few publications reporting HRQL in ECLS survivors (14–16). To address this issue, a study was undertaken to assess HRQL in children who received cardiac ECLS in our center. We hypothesized that cardiac ECLS survivors have lower HRQL scores when compared with a healthy sample, with children with chronic conditions, and with children who had surgery for congenital heart disease early in infancy and who did not receive ECLS.
MATERIALS AND METHODS This study is part of a prospective interprovincial inception cohort outcomes follow-up project conducted in four provinces in Western Canada (the Complex Pediatric Therapies Follow-up Program, CPTFP). Patients were identified at the time of ECLS and were followed prospectively. In this study, we included all children less than or 5 years old with diagnosis of cardiac disease (congenital or acquired) who received ECLS at the Stollery Children’s Hospital between January 1999 and December 2009. We excluded those patients who received ECLS for respiratory failure without cardiac disease and those children older than 6 years at the time of ECLS as these children are not prospectively followed up by the CPTFP. Patients included in this study received venoarterial ECLS with a centrifugal pump. Demographic and hospitalization variables that were previously agreed upon were collected prospectively (17). The following pre-ECLS and ECLS variables were collected retrospectively: left atrium vent, duration ECLS run, ECLS flows, fluid balance, and blood product administration. Inotrope score was calculated as follows: dopamine/dobutamine dose (μg/kg/min) + 100 × epinephrine/norepinephrine dose (μg/ kg/min) + 10 × milrinone dose (μg/kg/min) + 10,000 × vasopressine dose (U/Kg/min) (18). Long-term follow-up was discussed with parents or guardians once survival was probable, and with their written consent for participation, contact was Pediatric Critical Care Medicine
made with their respective follow-up clinics at the tertiary site of origin. The follow-up study and database have received institutional health research ethics board approval. History and physical measurements were obtained as have been described (17). The family socioeconomic status was determined using the Blishen Index (19). Maternal education was indicated by years of schooling. Outcomes assessments were completed at the tertiary site of origin. Single-ventricle physiology was defined as arterial oxygen saturation persistently below 90% due to the mixing of poorly oxygenated with oxygenated blood secondary to a right to left intracardiac or extracardiac shunt. Patient characteristics are described in Table 1. The HRQL Instrument HRQL was assessed using the PedsQL 4.0 Generic Core Scales completed by the children’s parents at the time of follow-up. This instrument invites parents to report on the quality of life of their child over the past 30 days as seen from the perspective of the child. The PedsQL 4.0 Generic Core Scales is a 23-item tool that encompasses the following domains: physical functioning (eight items), emotional functioning (five items), social functioning (five items), and school functioning (five items). Items are reverse scored and linearly transformed to a 0–100 scale; higher scores indicate better HRQL. A Physical Health Summary score (eight items) is the same as the Physical Functioning subscale. To create a Psychosocial Health Summary score, the mean is computed as the sum of the items divided by the number of items in the Emotional, Social, and School Functioning scales. The PedsQL 4.0 is a well-known instrument used to measure HRQL in children; its validity and reliability have been demonstrated and are acceptable for group comparisons (20, 21). Statistics Categorical variables are described as frequencies, and continuous variables are described as means and sds or median and interquartile range (IQR) as appropriate. We compared our cohort of patients with published values for a healthy sample, children with chronic conditions, and a cohort of children who had surgery for CHD with cardiopulmonary bypass (CPB) early in infancy and who did not receive ECLS (previously published by our center) (22, 23). In a post hoc analysis, we also compared our cohort of patients with a subgroup of children with single-ventricle physiology who underwent surgical palliation with the Norwood-Sano procedure, a high-risk subgroup from our previously published data (23). We used t test for independent samples to compare results across our cohort of patients and the other groups. We also calculated the percentage of patients who had a score difference from the healthy sample larger than the minimal clinically important difference (MCID). MCID is defined as the smallest difference in a score that the patients perceived to be beneficial and that would mandate, in the absence of troublesome side effects and excessive costs, a change in the patient’s management (21, 23). Stepwise multiple linear regression analysis was performed www.pccmjournal.org
721
Garcia Guerra et al
Table 1.
Demographic Variables of Cardiac Extracorporeal Life Support Survivors (n = 47)
Demographic/Pre-ECLS
Mean or n (sd or %)
Median (IQR)
Age at time of ECLS (mo)
10.4 (16.2)
3.2 (0.6–7.5)
Weight (kg)
12.8 (16.2)
6.0 (4.0–11.0)
Gender (male)
28 (59)
Single ventricle (yes)
13 (27)
Chromosomal abnormality (yes)
7 (15)
Indication for ECLS Failed wean from CPB
15 (32)
Low cardiac output syndrome
13 (28)
Hypoxia Cardiopulmonary resuscitation
3 (6) 16 (34)
Location of ECLS initiation Operating room
19 (40)
PICU
26 (56)
Other
2 (4)
Cannulation site Chest
28 (60)
Neck
18 (38)
Multiple
1 (2)
Inotrope score pre-ECLS
18.8 (19.6)
15.0 (0.0–27.0)
Days on mechanical ventilation pre-ECLS
6.25 (14.9)
1.0 (0.0–7.0)
7.5 (5.4)
6.4 (3–11.8)
Highest plasma lactate (mmol/L) pre-ECLS ECLS in relationship with cardiac surgery ECLS perioperative Not related
44 (94) 3 (6) 189.9 (114.2)
169.0 (112.0–242.0)
Aortic cross-clamp time (min) (n = 35)
64.6 (47.0)
60.0 (25.0–90.0)
Deep hypothermic circulatory arrest time (min) (n = 11)
22.3 (15.6)
21.0 (4.0–37.0)
CPB time (min) (n = 37)
Re-CPB (yes)
12 (35)
ECLS Time to lactate ≤ 2 mmol/L (hr)
18.2 (13.3)
Highest plasma lactate (mmol/L)
9.1 (5.6)
Left atrium vent (yes)
19 (40)
ECLS run (hr) Inotrope score first 24 hr
15.5 (8.0–26.0) 7.3 (4.6–13.0)
157.7 (128.2)
137.0 (91.0–168.0)
8.5 (5.8)
9.0 (5.0–12.5)
ECLS flow at 24 hr (mL/kg/min)
106.6 (27.7)
104.0 (93.0–120.0)
FB first 24 hr (mL/kg)
262.7 (221.6)
202.8 (153.4–336.0)
First day negative FB
4.0 (1.5)
Packed RBC transfusion first 48 hr (mL/kg) Seizures during ECLS (yes)
183.1 (115.9)
4.0 (3.0–5.0) 141.0 (105.0–254.0)
17 (34%) (Continued)
722
www.pccmjournal.org
October 2014 • Volume 15 • Number 8
Cardiac Intensive Care
Table 1.
(Continued). Demographic Variables (n = 47) Mean or n (sd or %)
Demographic/Pre-ECLS
Median (IQR)
Post-ECLS, mean (sd) All ventilation days
36.3 (44.3)
23.0 (14.0–43.0)
PICU days
39.9 (38.1)
26.0 (16.0–46.0)
Age health-related quality-of-life assessment (yr)
4.5 (1.0)
4.0 (4.0–5.0)
ECLS = extracorporeal life support, IQR = interquartile range, CPB = cardiopulmonary bypass, FB = fluid balance. Data are presented as n (%), mean ± sd, or median (IQR).
to explore for acute variables independently associated with HRQL. This analysis consisted of all variables from Table 1 that were found significant at p value of less than or equal to 0.10 in the univariate analysis, after screening for multicollinearity. Results are presented as effect sizes along with 95% CI and two-sided p values. We considered statistically significant those variables that have a p value of less than 0.05 in the multiple regression analysis. Statistical analyses were performed using SAS 9.1 (SAS Institute, Cary, NC).
RESULTS Description of the Cohort Ninety-eight patients received ECLS between 2000 and 2009, of which 39 patients (40%) were cannulated during ongoing extracorporeal cardiopulmonary resuscitation (ECPR). Fifty patients (51%) survived and 47 children (94% of survivors) had their HRQL assessed at a median (IQR) age of 4 years (4–5 yr). Of those children assessed, 28 (59%) were male, 13 (28%) had single-ventricle physiology, and 34 (72%) had biventricular physiology. Most of these children (n = 37; 79%) were placed on ECLS after surgery for CHD, and almost half cannulated either in the operating room (19; 40%) or in the PICU (26; 56%). Cannulation was done via the open sternum in 28 children (60%), and in 18 (38%), this was done via the neck. Seven children (15%) had chromosomal abnormalities. The median (IQR) age at the time of ECLS was 3.2 months (0.6–7.5 mo) and weight 4.8 kg (3.4– 6.9 kg). The most common indication for ECLS was failing to come off CPB in 15 patients (32%), followed by ECPR 16 (34%), low cardiac output syndrome 13 (28%), and hypoxia three (6%).
All the patients included in this study received venoarterial ECLS with a median (IQR) ECLS run of 137 hours (91–168 hr). Demographics and patient characteristics are shown in Table 1. HRQL Outcomes Compared with a healthy sample, children who received ECLS have statistically significantly lower HRQL score in all domains (Table 2): total PedsQL 4.0, 64.9 versus 82.2 (p < 0.0001); Physical Health Summary, 69.2 versus 84.0 (p = 0.0002); Psychosocial Health Summary, 65.9 versus 81.2 (p < 0.0001); Emotional Functioning, 64.6 versus 81.2 (p < 0.0001); Social Functioning, 64.9 versus 83.0 (p < 0.0001); and School Functioning, 58.0 versus 78.2 (p < 0.0001). More importantly, almost 50% of the children who received ECLS had PedsQL 4.0 scores 1 sd below the healthy sample’s mean, and approximate 60% had a score difference larger than the MCID (Table 3). The HRQL of children who received ECLS was also significantly lower than the HRQL of those children with chronic health conditions, except in the Psychosocial Summary score (Table 4). Scores for those that received ECLS compared with children with chronic conditions were as follows: Total PedsQL 4.0, 64.9 versus 73.1 (p = 0.007); Physical Health Summary, 69.2 versus 76.9 (p = 0.04); Psychosocial Health Summary, 65.9 versus 71.0 (p = 0.06); Emotional Functioning, 64.6 versus 71.0 (p = 0.03); Social Functioning, 64.9 versus 75.0 (p = 0.002); and School Functioning, 58.0 versus 65.5 (p = 0.02). When compared with a cohort of children with CHD who underwent cardiac surgery early in infancy and who did not receive ECLS, children who survived after ECLS had significantly lower HRQL scores in all domains (Table 5): PedsQL 4.0, 64.9 versus 81.1 (p < 0.0001); Physical Health
Mean PedsQL 4.0 Scores in Children From a Healthy Sample and Extracorporeal Life Support Survivors Table 2.
n
Healthy Sample, Mean (sd)
n
Extracorporeal Life Support, Mean (sd)
Mean Difference
Effect Size (95% CI)
p
Total score
8,713
82.2 (15.5)
47
64.9 (19.7)
17.3
1.11 (0.53–1.21)
< 0.0001
Physical health
8,696
84.0 (19.7)
47
69.2 (25.8)
14.8
0.75 (0.26–0.88)
0.0002
Psychosocial health
8,714
81.2 (15.3)
40
65.9 (16.9)
15.2
0.99 (0.68–1.30)
< 0.0001
Emotional functioning
8,692
81.2 (16.4)
47
64.6 (19.9)
16.5
1.00 (0.49–1.16)
< 0.0001
Social functioning
8,690
83.0 (19.6)
47
64.9 (23.9)
18.0
0.91 (0.42–1.07)
< 0.0001
School functioning
7,287
78.2 (19.6)
40
58.0 (23.2)
20.2
1.02 (0.71–1.33)
< 0.0001
Quality Scores
Pediatric Critical Care Medicine
www.pccmjournal.org
723
Garcia Guerra et al
Table 3. Mean PedsQL 4.0 Scores in Children From a Healthy Sample and Extracorporeal Life Support Survivors Mean Difference > MCID
< 1 sda
n
Extracorporeal Life Support, Mean (sd)
Mean Difference
MCID
n
%
Score
n
%
Total score
47
64.9 (19.7)
17.3
4.50
32
68
66.7
25
53
Physical health
47
69.2 (25.8)
14.8
6.92
26
55
65.0
16
34
Psychosocial health
40
65.9 (16.9)
15.2
5.49
28
70
65.9
17
42
Emotional functioning
47
64.6 (19.9)
16.5
7.79
28
60
64.8
22
47
Social functioning
47
64.9 (23.9)
18.0
8.98
28
60
63.4
22
47
School functioning
40
58.0 (23.2)
20.2
9.67
23
58
58.6
20
50
Quality Scores
MCID = minimal clinically important difference.
Table 4. Mean PedsQL 4.0 Scores in Children With Chronic Health Conditions and Extracorporeal Life Support Survivors n
Chronic Health Condition, Mean (sd)
n
Extracorporeal Life Support, Mean (sd)
Total score
831
73.1 (16.4)
47
64.9 (19.7)
8.1a
0.49 (0.11–0.72)
Physical health
830
76.9 (20.2)
47
69.2 (25.8)
7.7
0.37 (0.003–0.60) 0.048
Psychosocial health
830
71.0 (17.3)
40
65.9 (16.9)
5.0
0.29 (-0.02–0.61)
0.069
Emotional functioning 829
71.0 (19.7)
47
64.6 (19.9)
6.3
0.32 (0.03–0.61)
0.031
Social functioning
824
75.0 (21.7)
47
64.9 (23.9)
10.0
0.46 (0.16–0.75)
0.002
School functioning
756
65.5 (20.7)
40
58.0 (23.2)
7.5
0.35 (0.04–0.67)
0.026
Quality Scores
Mean Difference
a
a a
Effect Size (95% CI)
p
0.007
Table 5. Mean PedsQL 4.0 Scores in Children With Congenital Heart Disease and Extracorporeal Life Support Survivors n
Congenital Heart Disease, Mean (sd)
n
Extracorporeal Life Support, Mean (sd)
Mean Difference
Effect Size (95% CI)
p
Total score
130
81.1 (13.9)
47
64.9 (19.7)
16.1a
1.03 (0.50–1.24)
< 0.0001
Physical health
130
86.4 (15.3)
47
69.2 (25.8)
17.1
0.92 (0.37–1.08)
< 0.0001
Psychosocial health
130
77.5 (16.4)
40
65.9 (16.9)
11.5
0.69 (0.33–1.05)
0.0001
Emotional functioning
130
77.2 (16.9)
47
64.6 (19.9)
a
12.5
0.70 (0.36–1.04)
< 0.0001
Social functioning
130
82 (17.5)
47
64.9 (23.9)
17.0
a
0.87 (0.40–1.11)
< 0.0001
School functioning
130
74.9 (21.3)
40
58.0 (23.2)
16.8a
0.77 (0.41–1.13)
< 0.0001
Quality Scores
Summary, 69.2 versus 86.4 (p < 0.0001); Psychosocial Health Summary, 65.9 versus 77.5 (p = 0.0001); Emotional Functioning, 64.6 versus 77.2 (p 10 yr self-report). When compared with the general U.S. population, cardiac ECLS survivors had significantly lower scores in the Physical Summary, Physical Functioning, General Health Perception, and Parent Impact-Emotional domains. However, scores were not significantly different from the general population in other domains. Cardiac ECLS patients compared with patients with other cardiac disease (patients after D-transposition of the great arteries [D-TGA] repair, Kawasaki disease, and Fontan palliation) had similar HRQL scores except for Physical Summary (where cardiac ECLS patients had significant lower scores than patients with D-TGA and Kawasaki disease). Nevertheless, parents of cardiac ECLS survivors reported high prevalence of attention problems (34%), speech problems (34%), developmental delay or mental retardation (29%), hearing impairment or deafness (19%), orthopedic problems (15%), and epilepsy (10%) (13). This study differs from ours in several respects including the use of a different instrument to measure HRQL, an older cohort of patients, and follow-up assessment of only 44% of eligible patients. There are two studies not looking at HRQL but at Health Status (HS) after ECLS. Taylor et al (26) followed up 69 children using the Health State Utility Index and reported low survival (33%) but good HS among 71% of the survivors. However, they also found that 39% of these children survived with a disability. The survival among those children who received cardiac ECLS was lower (22%) with a similar prevalence of disabilities (33%). Mahle et al (11) performed a cost-utility analysis of 32 children who needed “salvage” cardiac ECLS. In their cohort, almost 10% of the children had “poor” HS at follow-up but overall they demonstrated that “salvage” cardiac ECLS was within the limits of accepted cost-utility (10). Although HS gives a broader concept of the outcome of these children, it does not necessarily correlate with HRQL (15). www.pccmjournal.org
725
Garcia Guerra et al
Our results are similar to those found by a Dutch follow-up program that evaluated the HRQL of 95 children who received ECLS for respiratory failure during the neonatal period (15). Those authors found that children who received ECLS had significant lower PedsQL 4.0 scores than a healthy sample in all domains except for Emotional Functioning. Similar to our results, 42% of the parents reported a PedsQL 4.0 score below 1 sd of the healthy sample. Approximately one third of survivors had some degree of cognitive and/or neuromotor disability. Similarly, in a cohort of 49 pediatric and neonatal ECLS patients (only 13 having cardiac ECLS), Wagner et al (16) reported a survival rate of 44% with 72% of the survivors having some degree of neurologic impairment and 36% having reduced HRQL (15). These studies all support our finding of lower HRQL in survivors of cardiac ECLS than in the general population, children with chronic disease, and children with CHD (14–16). There are few studies exploring the association between periECLS factors and HRQL in children. Madderom et al (15) found longer ECLS duration and the presence of chronic lung disease were negatively associated with PedsQL 4.0 in children who received ECLS in the neonatal period. Costello et al (14) performed univariate analysis of variables associated with HRQL in children who received cardiac ECLS; postcardiotomy failure, greater number of noncardiac surgeries, total PICU and hospital LOS, and noncardiac medical conditions were associated with lower PedsQL 4.0 Summary scores. Our study explored in more detail acute variables around the time of ECLS and their association with the HRQL. Higher inotrope score in the first 24 hours of ECLS was independently associated with lower PedsQL 4.0 score, Physical Summary, Psychosocial Summary, Emotional Functioning, and Social Functioning scores. Whether this represents inadequate ECLS flows is not clear but deserves further investigation, as this could be a modifiable variable. The median (IQR) ECLS flow at 24 hours was 104 mL/kg/min (93–120 mL/ kg/min), suggesting these patients were receiving on average a reasonable amount of support, although this might not be enough in some cases. Similar to Costello et al (14), we found that longer hospital LOS was associated with lower PedsQL 4.0, Physical Summary, and Emotional Functioning scores and longer PICU LOS was associated with lower School Functioning and Social Functioning. These associations probably represent a complicated hospital course rather than a direct effect of longer stay in PICU/hospital on HRQL. Having ECLS in a more recent era was associated with better School Functioning scores. This could be related to technical improvements in clinical management in recent years. Interestingly, despite 34% of our cohort having ECPR (ECMO cannulation during active chest compressions), we did not find an association between ECPR and HRQL. Our study has several strengths. We used PedsQL 4.0, a wellvalidated instrument, allowing us to compare our cardiac ECLS cohort with a healthy sample, children with chronic health conditions, and children with CHD. From those who survived 4 years after receiving cardiac ECLS, only three patients (6%) did not receive the questionnaires and were lost to follow-up. All of our patients received ECLS early in their childhood providing a relatively homogeneous population. This is one of the 726
www.pccmjournal.org
few studies looking for acute potentially modifiable variables associated with HRQL, an important step in trying to improve the long-term outcome of these children. This study also has some limitations. First, the study was conducted in a single center. Consequently, the results reflect the outcomes of a highly experienced referral ECLS center. Second, the occurrence of cerebrovascular events during ECLS was not recorded; these events could have significant impact on the long-term outcome of these children. Third, other predictor variables that we did not examine, such as glucose levels, may explain some of the HRQL variance. Fourth, HRQL was evaluated using the parents’ perspective. Several studies have shown significant variability between parent and child reports of HRQL (13, 27). At the time of follow-up, a significant proportion of these children were 4 years old, an age when only proxy reports can be obtained. Also, for our patients, the School Functioning scores represent difficulties at a preschool level. The population norms refer to children 4–8 years old and thus include children both during and after preschool. Reassessment of the same cohort in the future will allow comparison with children at higher levels of education and between child and parent. Fifth, children who required ECLS are likely to die if this support is not provided. Whether ECLS itself, or the severity of illness that leads to starting ECLS, are responsible for lower HRQL in this population is unknown. However, the outcomes we describe may help in counseling of parents, in prioritizing follow-up of these high-risk children, and in informing further study of this vulnerable population. HRQL outcomes contribute to a better understanding of the overall long-term outcome of these patients and families, describing how the survivors perform in society and experience daily living. A better understanding of mechanisms and determinants of HRQL in these children has potential important implications for clinical practice including modifying acute variables and clarity for treatment decision making. Further research examining key elements contributing to outcomes for these children is still needed. Strategies for the care of ECLS survivors should take into account not only the medical aspects of their disease and its treatment but also factors affecting the external, interpersonal, and personal spheres of these children’s lives. Early intervention programs could potentially improve the long-term outcome and HRQL of these children, as has been the case with other pediatric high-risk groups.
CONCLUSION Parental reports of HRQL in children who received cardiac ECLS are significantly lower than those from a healthy population, children with chronic conditions, and children with CHD who had cardiovascular surgery early in infancy. High inotropic support in the first 24 hours of ECLS, longer PICU/hospital LOS, chromosomal abnormalities, longer CPB time, and having ECLS in an older area were associated with lower HRQL scores on some dimensions. Further research is needed to determine interventions and/or programs that could improve the HRQL of ECLS survivors. October 2014 • Volume 15 • Number 8
Cardiac Intensive Care
REFERENCES
1. Bartlett RH, Gazzaniga AB, Jefferies MR, et al: Extracorporeal membrane oxygenation (ECMO) cardiopulmonary support in infancy. Trans Am Soc Artif Intern Organs 1976; 22:80–93 2. Soeter JR, Mamiya RT, Sprague AY, et al: Prolonged extracorporeal oxygenation for cardiorespiratory failure after tetralogy correction. J Thorac Cardiovasc Surg 1973; 66:214–218 3. Paden ML, Conrad SA, Rycus PT, et al; ELSO Registry: Extracorporeal Life Support Organization Registry Report 2012. ASAIO J 2013; 59:202–210 4. Lequier L, Joffe AR, Robertson CM, et al; Western Canadian Complex Pediatric Therapies Program Follow-up Group: 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.e3 5. Aharon AS, Drinkwater DC Jr, Churchwell KB, et al: Extracorporeal membrane oxygenation in children after repair of congenital cardiac lesions. Ann Thorac Surg 2001; 72:2095–2101 6. Alsoufi B, Shen I, Karamlou T, et al: Extracorporeal life support in neonates, infants, and children after repair of congenital heart disease: Modern era results in a single institution. Ann Thorac Surg 2005; 80:15–21 7. Kumar TK, Zurakowski D, Dalton H, et al: Extracorporeal membrane oxygenation in postcardiotomy patients: Factors influencing outcome. J Thorac Cardiovasc Surg 2010; 140:330–336.e2 8. Chan T, Thiagarajan RR, Frank D, et al: Survival after extracorporeal cardiopulmonary resuscitation in infants and children with heart disease. J Thorac Cardiovasc Surg 2008; 136:984–992 9. Chrysostomou C, Morell VO, Kuch BA, et al: Short- and intermediate-term survival after extracorporeal membrane oxygenation in children with cardiac disease. J Thorac Cardiovasc Surg 2013; 146: 317–325 10. Joffe AR, Lequier L, Robertson CM: Pediatric outcomes after extracorporeal membrane oxygenation for cardiac disease and for cardiac arrest: A review. ASAIO J 2012; 58:297–310 11. Mahle WT, Forbess JM, Kirshbom PM, et al: Cost-utility analysis of salvage cardiac extracorporeal membrane oxygenation in children. J Thorac Cardiovasc Surg 2005; 129:1084–1090 12. Mussatto K, Tweddell J: Quality of life following surgery for congenital cardiac malformations in neonates and infants. Cardiol Young 2005; 15(Suppl 1):174–178 13. Spijkerboer AW, Utens EM, De Koning WB, et al: Health-related quality of life in children and adolescents after invasive treatment for congenital heart disease. Qual Life Res 2006; 15:663–673
Pediatric Critical Care Medicine
14. Costello JM, O’Brien M, Wypij D, et al: Quality of life of pediatric cardiac patients who previously required extracorporeal membrane oxygenation. Pediatr Crit Care Med 2012; 13:428–434 15. Madderom MJ, Gischler SJ, Duivenvoorden H, et al: Neonatal extracorporeal membrane oxygenation: Impaired health at 5 years of age. Pediatr Crit Care Med 2013; 14:183–193 16. Wagner K, Risnes I, Berntsen T, et al: Clinical and psychosocial follow-up study of children treated with extracorporeal membrane oxygenation. Ann Thorac Surg 2007; 84:1349–1355 17. Robertson CM, Sauve RS, Joffe AR, et al: The registry and followup of complex pediatric therapies program of Western Canada: A mechanism for service, audit, and research after life-saving therapies for young children. Cardiol Res Pract 2011; 2011:965740 18. Gaies MG, Gurney JG, Yen AH, 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 19. Blishen BR: The 1981 socioeconomic index for occupations in Canada. Can Rev Soc Anth 1987; 24:465–488 20. Varni JW, Limbers CA, Burwinkle TM: Parent proxy-report of their children’s health-related quality of life: An analysis of 13,878 parents’ reliability and validity across age subgroups using the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes 2007; 5:2 21. Varni JW, Seid M, Kurtin PS: PedsQL 4.0: Reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care 2001; 39:800–812 22. Varni JW, Burwinkle TM, Seid M, et al: The PedsQL 4.0 as a pediatric population health measure: Feasibility, reliability, and validity. Ambul Pediatr 2003; 3:329–341 23. Garcia Guerra G, Robertson CM, Alton GY, et al; Western Canadian Complex Pediatric Therapies Follow-up Group: Quality of life 4 years after complex heart surgery in infancy. J Thorac Cardiovasc Surg 2013; 145:482–488.e2 24. Jaeschke R, Singer J, Guyatt GH: Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989; 10:407–415 25. Ebrahim S, Singh S, Hutchison JS, et al: Adaptive behavior, functional outcomes, and quality of life outcomes of children requiring urgent ICU admission. Pediatr Crit Care Med 2013; 14:10–18 26. Taylor AK, Cousins R, Butt WW: The long-term outcome of children managed with extracorporeal life support: An institutional experience. Crit Care Resusc 2007; 9:172–177 27. Lambert LM, Minich LL, Newburger JW, et al; Pediatric Heart Network Investigators: Parent- versus child-reported functional health status after the Fontan procedure. Pediatrics 2009; 124:e942–e949
www.pccmjournal.org
727
Health-Related Quality of Life in Pediatric Cardiac Extracorporeal Life Support Survivors* Gonzalo Garcia Guerra, MD, MSc1; Charlene M. T. Robertson, MD1,2; Gwen Y. Alton, RN1; Ari R. Joffe, MD1; Elham Khodayari Moez, MSc2; Irina A Dinu, PhD2,3; David B. Ross, MD1,4; Ivan M. Rebeyka, MD1,4; Laurance Lequier, MD1; the Western Canadian Complex Pediatric Therapies Follow-Up Group
Objective: To assess the health-related quality of life of children who received cardiac extracorporeal life support. We hypothesized that extracorporeal life support survivors have lower health-related quality-of-life scores when compared with a healthy sample, with children with chronic conditions, and with children who had surgery for congenital heart disease and did not receive extracorporeal life support. Design: Prospective cohort study. Setting: Stollery Children’s Hospital and Complex Pediatric Therapies Follow-up Program clinics. Patients: Children less than or 5 years old with diagnosis of cardiac disease (congenital or acquired) who received extracorporeal life support at the Stollery Children’s Hospital from 1999 to 2009. Interventions: None. Measurements and Main Results: Health-related quality of life was assessed using the PedsQL 4.0 Generic Core Scales completed by the children’s parents at the time of follow-up. Forty-seven car*See also p. 775. 1 Department of Pediatrics, University of Alberta, Edmonton, AB, Canada. 2 Pediatric Rehabilitation Outcomes Evaluation and Research Unit, Glenrose Rehabilitation Hospital, Edmonton, AB, Canada. 3 School of Public Health, University of Alberta, Edmonton, AB, Canada. 4 Department of Surgery, University of Alberta, Edmonton, AB, Canada. This work was performed at University of Alberta, Edmonton, AB, Canada. Chairs of the Western Canadian Complex Pediatric Therapies Follow-up Group are Charlene M.T. Robertson, Reg Sauve, and Gwen Y. Alton. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ pccmjournal). Supported, in part, by the Registry and Follow-up of Complex Pediatric Therapies Project, Alberta Health and Wellness. Dr. Robertson’s institution received grant support from Alberta Health and Wellness. The remaining authors disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: [email protected] Copyright © 2014 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: 10.1097/PCC.0000000000000212
720
www.pccmjournal.org
diac extracorporeal life support survivors had their health-related quality of life assessed at a median age of 4 years. Compared with a healthy sample, children who received venoarterial extracorporeal life support have significantly lower PedsQL (64.9 vs 82.2; p < 0.0001). The PedsQL scores of children who received extracorporeal life support were also significantly lower than those of children with chronic health conditions (64.9 vs 73.1; p = 0.007). Compared with children with congenital heart disease who underwent cardiac surgery early in infancy and who did not receive extracorporeal life support, extracorporeal life support survivors had significantly lower PedsQL scores (64.9 vs 81.1; p < 0.0001). Multiple linear regression analysis found an independent association between both higher inotrope score in the first 24 hours of extracorporeal life support and longer hospital length of stay, with lower PedsQL scores. Conclusions: Pediatric cardiac extracorporeal life support survivors showed lower health-related quality of life than healthy children, children with chronic conditions, and children with congenital heart disease who did not receive extracorporeal life support. (Pediatr Crit Care Med 2014; 15:720–727) Key Words: critical care; extracorporeal membrane oxygenation; heart diseases; outcomes research; pediatrics; quality of life
S
ince Bartlett et al (1) started using extracorporeal membrane oxygenation (ECMO) to support children with congenital heart disease (CHD) in the 1970s, extracorporeal life support (ECLS) has evolved from an extraordinary, last resort, life-saving intervention to a standard of care in many pediatric tertiary centers with thousands of patients supported to date (1–3). Indications for ECLS now include cardiovascular failure due to heart failure, low cardiac output syndrome, inability to wean from cardiopulmonary bypass after cardiac surgery, sepsis, and sustained cardiac arrest, as well as respiratory failure due to acute respiratory distress syndrome, pneumonia, airway obstruction, etc. ECLS has also gone from an intervention using large complex equipment available only in the operating room, to more “simple” circuits and components October 2014 • Volume 15 • Number 8
Cardiac Intensive Care
used in the ICU, with more compact devices that can be used to transport patients from the operating room to other parts of the hospital or even between centers. It has been shown that ECLS can improve short-term outcome and survival of children with congenital or acquired heart disease that otherwise would have had significant adverse outcomes including death (4–7). Many centers have reported survival rates in pediatric cardiac ECLS above 40% (4, 8–10). However, ECLS is not exempt from complications, morbidity, and potential long-term adverse outcomes (11, 12). Hospital survival outcomes after ECMO have been reported in the Extracorporeal Life Support Organization registry for decades; however, long-term outcomes and health-related quality of life (HRQL) in survivors have not been part of this registry data or previous publications (3, 10). HRQL is a multidimensional construct that includes physical, mental, and social well-being dimensions and not merely the absence of disease. The literature consistently suggests that the clinical condition or severity of disease is not necessarily associated with perception of quality of life (12). Accordingly, HRQL refers to a person’s subjective, emotional evaluation and reaction to his/her health condition (13). To date, there have been few publications reporting HRQL in ECLS survivors (14–16). To address this issue, a study was undertaken to assess HRQL in children who received cardiac ECLS in our center. We hypothesized that cardiac ECLS survivors have lower HRQL scores when compared with a healthy sample, with children with chronic conditions, and with children who had surgery for congenital heart disease early in infancy and who did not receive ECLS.
MATERIALS AND METHODS This study is part of a prospective interprovincial inception cohort outcomes follow-up project conducted in four provinces in Western Canada (the Complex Pediatric Therapies Follow-up Program, CPTFP). Patients were identified at the time of ECLS and were followed prospectively. In this study, we included all children less than or 5 years old with diagnosis of cardiac disease (congenital or acquired) who received ECLS at the Stollery Children’s Hospital between January 1999 and December 2009. We excluded those patients who received ECLS for respiratory failure without cardiac disease and those children older than 6 years at the time of ECLS as these children are not prospectively followed up by the CPTFP. Patients included in this study received venoarterial ECLS with a centrifugal pump. Demographic and hospitalization variables that were previously agreed upon were collected prospectively (17). The following pre-ECLS and ECLS variables were collected retrospectively: left atrium vent, duration ECLS run, ECLS flows, fluid balance, and blood product administration. Inotrope score was calculated as follows: dopamine/dobutamine dose (μg/kg/min) + 100 × epinephrine/norepinephrine dose (μg/ kg/min) + 10 × milrinone dose (μg/kg/min) + 10,000 × vasopressine dose (U/Kg/min) (18). Long-term follow-up was discussed with parents or guardians once survival was probable, and with their written consent for participation, contact was Pediatric Critical Care Medicine
made with their respective follow-up clinics at the tertiary site of origin. The follow-up study and database have received institutional health research ethics board approval. History and physical measurements were obtained as have been described (17). The family socioeconomic status was determined using the Blishen Index (19). Maternal education was indicated by years of schooling. Outcomes assessments were completed at the tertiary site of origin. Single-ventricle physiology was defined as arterial oxygen saturation persistently below 90% due to the mixing of poorly oxygenated with oxygenated blood secondary to a right to left intracardiac or extracardiac shunt. Patient characteristics are described in Table 1. The HRQL Instrument HRQL was assessed using the PedsQL 4.0 Generic Core Scales completed by the children’s parents at the time of follow-up. This instrument invites parents to report on the quality of life of their child over the past 30 days as seen from the perspective of the child. The PedsQL 4.0 Generic Core Scales is a 23-item tool that encompasses the following domains: physical functioning (eight items), emotional functioning (five items), social functioning (five items), and school functioning (five items). Items are reverse scored and linearly transformed to a 0–100 scale; higher scores indicate better HRQL. A Physical Health Summary score (eight items) is the same as the Physical Functioning subscale. To create a Psychosocial Health Summary score, the mean is computed as the sum of the items divided by the number of items in the Emotional, Social, and School Functioning scales. The PedsQL 4.0 is a well-known instrument used to measure HRQL in children; its validity and reliability have been demonstrated and are acceptable for group comparisons (20, 21). Statistics Categorical variables are described as frequencies, and continuous variables are described as means and sds or median and interquartile range (IQR) as appropriate. We compared our cohort of patients with published values for a healthy sample, children with chronic conditions, and a cohort of children who had surgery for CHD with cardiopulmonary bypass (CPB) early in infancy and who did not receive ECLS (previously published by our center) (22, 23). In a post hoc analysis, we also compared our cohort of patients with a subgroup of children with single-ventricle physiology who underwent surgical palliation with the Norwood-Sano procedure, a high-risk subgroup from our previously published data (23). We used t test for independent samples to compare results across our cohort of patients and the other groups. We also calculated the percentage of patients who had a score difference from the healthy sample larger than the minimal clinically important difference (MCID). MCID is defined as the smallest difference in a score that the patients perceived to be beneficial and that would mandate, in the absence of troublesome side effects and excessive costs, a change in the patient’s management (21, 23). Stepwise multiple linear regression analysis was performed www.pccmjournal.org
721
Garcia Guerra et al
Table 1.
Demographic Variables of Cardiac Extracorporeal Life Support Survivors (n = 47)
Demographic/Pre-ECLS
Mean or n (sd or %)
Median (IQR)
Age at time of ECLS (mo)
10.4 (16.2)
3.2 (0.6–7.5)
Weight (kg)
12.8 (16.2)
6.0 (4.0–11.0)
Gender (male)
28 (59)
Single ventricle (yes)
13 (27)
Chromosomal abnormality (yes)
7 (15)
Indication for ECLS Failed wean from CPB
15 (32)
Low cardiac output syndrome
13 (28)
Hypoxia Cardiopulmonary resuscitation
3 (6) 16 (34)
Location of ECLS initiation Operating room
19 (40)
PICU
26 (56)
Other
2 (4)
Cannulation site Chest
28 (60)
Neck
18 (38)
Multiple
1 (2)
Inotrope score pre-ECLS
18.8 (19.6)
15.0 (0.0–27.0)
Days on mechanical ventilation pre-ECLS
6.25 (14.9)
1.0 (0.0–7.0)
7.5 (5.4)
6.4 (3–11.8)
Highest plasma lactate (mmol/L) pre-ECLS ECLS in relationship with cardiac surgery ECLS perioperative Not related
44 (94) 3 (6) 189.9 (114.2)
169.0 (112.0–242.0)
Aortic cross-clamp time (min) (n = 35)
64.6 (47.0)
60.0 (25.0–90.0)
Deep hypothermic circulatory arrest time (min) (n = 11)
22.3 (15.6)
21.0 (4.0–37.0)
CPB time (min) (n = 37)
Re-CPB (yes)
12 (35)
ECLS Time to lactate ≤ 2 mmol/L (hr)
18.2 (13.3)
Highest plasma lactate (mmol/L)
9.1 (5.6)
Left atrium vent (yes)
19 (40)
ECLS run (hr) Inotrope score first 24 hr
15.5 (8.0–26.0) 7.3 (4.6–13.0)
157.7 (128.2)
137.0 (91.0–168.0)
8.5 (5.8)
9.0 (5.0–12.5)
ECLS flow at 24 hr (mL/kg/min)
106.6 (27.7)
104.0 (93.0–120.0)
FB first 24 hr (mL/kg)
262.7 (221.6)
202.8 (153.4–336.0)
First day negative FB
4.0 (1.5)
Packed RBC transfusion first 48 hr (mL/kg) Seizures during ECLS (yes)
183.1 (115.9)
4.0 (3.0–5.0) 141.0 (105.0–254.0)
17 (34%) (Continued)
722
www.pccmjournal.org
October 2014 • Volume 15 • Number 8
Cardiac Intensive Care
Table 1.
(Continued). Demographic Variables (n = 47) Mean or n (sd or %)
Demographic/Pre-ECLS
Median (IQR)
Post-ECLS, mean (sd) All ventilation days
36.3 (44.3)
23.0 (14.0–43.0)
PICU days
39.9 (38.1)
26.0 (16.0–46.0)
Age health-related quality-of-life assessment (yr)
4.5 (1.0)
4.0 (4.0–5.0)
ECLS = extracorporeal life support, IQR = interquartile range, CPB = cardiopulmonary bypass, FB = fluid balance. Data are presented as n (%), mean ± sd, or median (IQR).
to explore for acute variables independently associated with HRQL. This analysis consisted of all variables from Table 1 that were found significant at p value of less than or equal to 0.10 in the univariate analysis, after screening for multicollinearity. Results are presented as effect sizes along with 95% CI and two-sided p values. We considered statistically significant those variables that have a p value of less than 0.05 in the multiple regression analysis. Statistical analyses were performed using SAS 9.1 (SAS Institute, Cary, NC).
RESULTS Description of the Cohort Ninety-eight patients received ECLS between 2000 and 2009, of which 39 patients (40%) were cannulated during ongoing extracorporeal cardiopulmonary resuscitation (ECPR). Fifty patients (51%) survived and 47 children (94% of survivors) had their HRQL assessed at a median (IQR) age of 4 years (4–5 yr). Of those children assessed, 28 (59%) were male, 13 (28%) had single-ventricle physiology, and 34 (72%) had biventricular physiology. Most of these children (n = 37; 79%) were placed on ECLS after surgery for CHD, and almost half cannulated either in the operating room (19; 40%) or in the PICU (26; 56%). Cannulation was done via the open sternum in 28 children (60%), and in 18 (38%), this was done via the neck. Seven children (15%) had chromosomal abnormalities. The median (IQR) age at the time of ECLS was 3.2 months (0.6–7.5 mo) and weight 4.8 kg (3.4– 6.9 kg). The most common indication for ECLS was failing to come off CPB in 15 patients (32%), followed by ECPR 16 (34%), low cardiac output syndrome 13 (28%), and hypoxia three (6%).
All the patients included in this study received venoarterial ECLS with a median (IQR) ECLS run of 137 hours (91–168 hr). Demographics and patient characteristics are shown in Table 1. HRQL Outcomes Compared with a healthy sample, children who received ECLS have statistically significantly lower HRQL score in all domains (Table 2): total PedsQL 4.0, 64.9 versus 82.2 (p < 0.0001); Physical Health Summary, 69.2 versus 84.0 (p = 0.0002); Psychosocial Health Summary, 65.9 versus 81.2 (p < 0.0001); Emotional Functioning, 64.6 versus 81.2 (p < 0.0001); Social Functioning, 64.9 versus 83.0 (p < 0.0001); and School Functioning, 58.0 versus 78.2 (p < 0.0001). More importantly, almost 50% of the children who received ECLS had PedsQL 4.0 scores 1 sd below the healthy sample’s mean, and approximate 60% had a score difference larger than the MCID (Table 3). The HRQL of children who received ECLS was also significantly lower than the HRQL of those children with chronic health conditions, except in the Psychosocial Summary score (Table 4). Scores for those that received ECLS compared with children with chronic conditions were as follows: Total PedsQL 4.0, 64.9 versus 73.1 (p = 0.007); Physical Health Summary, 69.2 versus 76.9 (p = 0.04); Psychosocial Health Summary, 65.9 versus 71.0 (p = 0.06); Emotional Functioning, 64.6 versus 71.0 (p = 0.03); Social Functioning, 64.9 versus 75.0 (p = 0.002); and School Functioning, 58.0 versus 65.5 (p = 0.02). When compared with a cohort of children with CHD who underwent cardiac surgery early in infancy and who did not receive ECLS, children who survived after ECLS had significantly lower HRQL scores in all domains (Table 5): PedsQL 4.0, 64.9 versus 81.1 (p < 0.0001); Physical Health
Mean PedsQL 4.0 Scores in Children From a Healthy Sample and Extracorporeal Life Support Survivors Table 2.
n
Healthy Sample, Mean (sd)
n
Extracorporeal Life Support, Mean (sd)
Mean Difference
Effect Size (95% CI)
p
Total score
8,713
82.2 (15.5)
47
64.9 (19.7)
17.3
1.11 (0.53–1.21)
< 0.0001
Physical health
8,696
84.0 (19.7)
47
69.2 (25.8)
14.8
0.75 (0.26–0.88)
0.0002
Psychosocial health
8,714
81.2 (15.3)
40
65.9 (16.9)
15.2
0.99 (0.68–1.30)
< 0.0001
Emotional functioning
8,692
81.2 (16.4)
47
64.6 (19.9)
16.5
1.00 (0.49–1.16)
< 0.0001
Social functioning
8,690
83.0 (19.6)
47
64.9 (23.9)
18.0
0.91 (0.42–1.07)
< 0.0001
School functioning
7,287
78.2 (19.6)
40
58.0 (23.2)
20.2
1.02 (0.71–1.33)
< 0.0001
Quality Scores
Pediatric Critical Care Medicine
www.pccmjournal.org
723
Garcia Guerra et al
Table 3. Mean PedsQL 4.0 Scores in Children From a Healthy Sample and Extracorporeal Life Support Survivors Mean Difference > MCID
< 1 sda
n
Extracorporeal Life Support, Mean (sd)
Mean Difference
MCID
n
%
Score
n
%
Total score
47
64.9 (19.7)
17.3
4.50
32
68
66.7
25
53
Physical health
47
69.2 (25.8)
14.8
6.92
26
55
65.0
16
34
Psychosocial health
40
65.9 (16.9)
15.2
5.49
28
70
65.9
17
42
Emotional functioning
47
64.6 (19.9)
16.5
7.79
28
60
64.8
22
47
Social functioning
47
64.9 (23.9)
18.0
8.98
28
60
63.4
22
47
School functioning
40
58.0 (23.2)
20.2
9.67
23
58
58.6
20
50
Quality Scores
MCID = minimal clinically important difference.
Table 4. Mean PedsQL 4.0 Scores in Children With Chronic Health Conditions and Extracorporeal Life Support Survivors n
Chronic Health Condition, Mean (sd)
n
Extracorporeal Life Support, Mean (sd)
Total score
831
73.1 (16.4)
47
64.9 (19.7)
8.1a
0.49 (0.11–0.72)
Physical health
830
76.9 (20.2)
47
69.2 (25.8)
7.7
0.37 (0.003–0.60) 0.048
Psychosocial health
830
71.0 (17.3)
40
65.9 (16.9)
5.0
0.29 (-0.02–0.61)
0.069
Emotional functioning 829
71.0 (19.7)
47
64.6 (19.9)
6.3
0.32 (0.03–0.61)
0.031
Social functioning
824
75.0 (21.7)
47
64.9 (23.9)
10.0
0.46 (0.16–0.75)
0.002
School functioning
756
65.5 (20.7)
40
58.0 (23.2)
7.5
0.35 (0.04–0.67)
0.026
Quality Scores
Mean Difference
a
a a
Effect Size (95% CI)
p
0.007
Table 5. Mean PedsQL 4.0 Scores in Children With Congenital Heart Disease and Extracorporeal Life Support Survivors n
Congenital Heart Disease, Mean (sd)
n
Extracorporeal Life Support, Mean (sd)
Mean Difference
Effect Size (95% CI)
p
Total score
130
81.1 (13.9)
47
64.9 (19.7)
16.1a
1.03 (0.50–1.24)
< 0.0001
Physical health
130
86.4 (15.3)
47
69.2 (25.8)
17.1
0.92 (0.37–1.08)
< 0.0001
Psychosocial health
130
77.5 (16.4)
40
65.9 (16.9)
11.5
0.69 (0.33–1.05)
0.0001
Emotional functioning
130
77.2 (16.9)
47
64.6 (19.9)
a
12.5
0.70 (0.36–1.04)
< 0.0001
Social functioning
130
82 (17.5)
47
64.9 (23.9)
17.0
a
0.87 (0.40–1.11)
< 0.0001
School functioning
130
74.9 (21.3)
40
58.0 (23.2)
16.8a
0.77 (0.41–1.13)
< 0.0001
Quality Scores
Summary, 69.2 versus 86.4 (p < 0.0001); Psychosocial Health Summary, 65.9 versus 77.5 (p = 0.0001); Emotional Functioning, 64.6 versus 77.2 (p 10 yr self-report). When compared with the general U.S. population, cardiac ECLS survivors had significantly lower scores in the Physical Summary, Physical Functioning, General Health Perception, and Parent Impact-Emotional domains. However, scores were not significantly different from the general population in other domains. Cardiac ECLS patients compared with patients with other cardiac disease (patients after D-transposition of the great arteries [D-TGA] repair, Kawasaki disease, and Fontan palliation) had similar HRQL scores except for Physical Summary (where cardiac ECLS patients had significant lower scores than patients with D-TGA and Kawasaki disease). Nevertheless, parents of cardiac ECLS survivors reported high prevalence of attention problems (34%), speech problems (34%), developmental delay or mental retardation (29%), hearing impairment or deafness (19%), orthopedic problems (15%), and epilepsy (10%) (13). This study differs from ours in several respects including the use of a different instrument to measure HRQL, an older cohort of patients, and follow-up assessment of only 44% of eligible patients. There are two studies not looking at HRQL but at Health Status (HS) after ECLS. Taylor et al (26) followed up 69 children using the Health State Utility Index and reported low survival (33%) but good HS among 71% of the survivors. However, they also found that 39% of these children survived with a disability. The survival among those children who received cardiac ECLS was lower (22%) with a similar prevalence of disabilities (33%). Mahle et al (11) performed a cost-utility analysis of 32 children who needed “salvage” cardiac ECLS. In their cohort, almost 10% of the children had “poor” HS at follow-up but overall they demonstrated that “salvage” cardiac ECLS was within the limits of accepted cost-utility (10). Although HS gives a broader concept of the outcome of these children, it does not necessarily correlate with HRQL (15). www.pccmjournal.org
725
Garcia Guerra et al
Our results are similar to those found by a Dutch follow-up program that evaluated the HRQL of 95 children who received ECLS for respiratory failure during the neonatal period (15). Those authors found that children who received ECLS had significant lower PedsQL 4.0 scores than a healthy sample in all domains except for Emotional Functioning. Similar to our results, 42% of the parents reported a PedsQL 4.0 score below 1 sd of the healthy sample. Approximately one third of survivors had some degree of cognitive and/or neuromotor disability. Similarly, in a cohort of 49 pediatric and neonatal ECLS patients (only 13 having cardiac ECLS), Wagner et al (16) reported a survival rate of 44% with 72% of the survivors having some degree of neurologic impairment and 36% having reduced HRQL (15). These studies all support our finding of lower HRQL in survivors of cardiac ECLS than in the general population, children with chronic disease, and children with CHD (14–16). There are few studies exploring the association between periECLS factors and HRQL in children. Madderom et al (15) found longer ECLS duration and the presence of chronic lung disease were negatively associated with PedsQL 4.0 in children who received ECLS in the neonatal period. Costello et al (14) performed univariate analysis of variables associated with HRQL in children who received cardiac ECLS; postcardiotomy failure, greater number of noncardiac surgeries, total PICU and hospital LOS, and noncardiac medical conditions were associated with lower PedsQL 4.0 Summary scores. Our study explored in more detail acute variables around the time of ECLS and their association with the HRQL. Higher inotrope score in the first 24 hours of ECLS was independently associated with lower PedsQL 4.0 score, Physical Summary, Psychosocial Summary, Emotional Functioning, and Social Functioning scores. Whether this represents inadequate ECLS flows is not clear but deserves further investigation, as this could be a modifiable variable. The median (IQR) ECLS flow at 24 hours was 104 mL/kg/min (93–120 mL/ kg/min), suggesting these patients were receiving on average a reasonable amount of support, although this might not be enough in some cases. Similar to Costello et al (14), we found that longer hospital LOS was associated with lower PedsQL 4.0, Physical Summary, and Emotional Functioning scores and longer PICU LOS was associated with lower School Functioning and Social Functioning. These associations probably represent a complicated hospital course rather than a direct effect of longer stay in PICU/hospital on HRQL. Having ECLS in a more recent era was associated with better School Functioning scores. This could be related to technical improvements in clinical management in recent years. Interestingly, despite 34% of our cohort having ECPR (ECMO cannulation during active chest compressions), we did not find an association between ECPR and HRQL. Our study has several strengths. We used PedsQL 4.0, a wellvalidated instrument, allowing us to compare our cardiac ECLS cohort with a healthy sample, children with chronic health conditions, and children with CHD. From those who survived 4 years after receiving cardiac ECLS, only three patients (6%) did not receive the questionnaires and were lost to follow-up. All of our patients received ECLS early in their childhood providing a relatively homogeneous population. This is one of the 726
www.pccmjournal.org
few studies looking for acute potentially modifiable variables associated with HRQL, an important step in trying to improve the long-term outcome of these children. This study also has some limitations. First, the study was conducted in a single center. Consequently, the results reflect the outcomes of a highly experienced referral ECLS center. Second, the occurrence of cerebrovascular events during ECLS was not recorded; these events could have significant impact on the long-term outcome of these children. Third, other predictor variables that we did not examine, such as glucose levels, may explain some of the HRQL variance. Fourth, HRQL was evaluated using the parents’ perspective. Several studies have shown significant variability between parent and child reports of HRQL (13, 27). At the time of follow-up, a significant proportion of these children were 4 years old, an age when only proxy reports can be obtained. Also, for our patients, the School Functioning scores represent difficulties at a preschool level. The population norms refer to children 4–8 years old and thus include children both during and after preschool. Reassessment of the same cohort in the future will allow comparison with children at higher levels of education and between child and parent. Fifth, children who required ECLS are likely to die if this support is not provided. Whether ECLS itself, or the severity of illness that leads to starting ECLS, are responsible for lower HRQL in this population is unknown. However, the outcomes we describe may help in counseling of parents, in prioritizing follow-up of these high-risk children, and in informing further study of this vulnerable population. HRQL outcomes contribute to a better understanding of the overall long-term outcome of these patients and families, describing how the survivors perform in society and experience daily living. A better understanding of mechanisms and determinants of HRQL in these children has potential important implications for clinical practice including modifying acute variables and clarity for treatment decision making. Further research examining key elements contributing to outcomes for these children is still needed. Strategies for the care of ECLS survivors should take into account not only the medical aspects of their disease and its treatment but also factors affecting the external, interpersonal, and personal spheres of these children’s lives. Early intervention programs could potentially improve the long-term outcome and HRQL of these children, as has been the case with other pediatric high-risk groups.
CONCLUSION Parental reports of HRQL in children who received cardiac ECLS are significantly lower than those from a healthy population, children with chronic conditions, and children with CHD who had cardiovascular surgery early in infancy. High inotropic support in the first 24 hours of ECLS, longer PICU/hospital LOS, chromosomal abnormalities, longer CPB time, and having ECLS in an older area were associated with lower HRQL scores on some dimensions. Further research is needed to determine interventions and/or programs that could improve the HRQL of ECLS survivors. October 2014 • Volume 15 • Number 8
Cardiac Intensive Care
REFERENCES
1. Bartlett RH, Gazzaniga AB, Jefferies MR, et al: Extracorporeal membrane oxygenation (ECMO) cardiopulmonary support in infancy. Trans Am Soc Artif Intern Organs 1976; 22:80–93 2. Soeter JR, Mamiya RT, Sprague AY, et al: Prolonged extracorporeal oxygenation for cardiorespiratory failure after tetralogy correction. J Thorac Cardiovasc Surg 1973; 66:214–218 3. Paden ML, Conrad SA, Rycus PT, et al; ELSO Registry: Extracorporeal Life Support Organization Registry Report 2012. ASAIO J 2013; 59:202–210 4. Lequier L, Joffe AR, Robertson CM, et al; Western Canadian Complex Pediatric Therapies Program Follow-up Group: 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.e3 5. Aharon AS, Drinkwater DC Jr, Churchwell KB, et al: Extracorporeal membrane oxygenation in children after repair of congenital cardiac lesions. Ann Thorac Surg 2001; 72:2095–2101 6. Alsoufi B, Shen I, Karamlou T, et al: Extracorporeal life support in neonates, infants, and children after repair of congenital heart disease: Modern era results in a single institution. Ann Thorac Surg 2005; 80:15–21 7. Kumar TK, Zurakowski D, Dalton H, et al: Extracorporeal membrane oxygenation in postcardiotomy patients: Factors influencing outcome. J Thorac Cardiovasc Surg 2010; 140:330–336.e2 8. Chan T, Thiagarajan RR, Frank D, et al: Survival after extracorporeal cardiopulmonary resuscitation in infants and children with heart disease. J Thorac Cardiovasc Surg 2008; 136:984–992 9. Chrysostomou C, Morell VO, Kuch BA, et al: Short- and intermediate-term survival after extracorporeal membrane oxygenation in children with cardiac disease. J Thorac Cardiovasc Surg 2013; 146: 317–325 10. Joffe AR, Lequier L, Robertson CM: Pediatric outcomes after extracorporeal membrane oxygenation for cardiac disease and for cardiac arrest: A review. ASAIO J 2012; 58:297–310 11. Mahle WT, Forbess JM, Kirshbom PM, et al: Cost-utility analysis of salvage cardiac extracorporeal membrane oxygenation in children. J Thorac Cardiovasc Surg 2005; 129:1084–1090 12. Mussatto K, Tweddell J: Quality of life following surgery for congenital cardiac malformations in neonates and infants. Cardiol Young 2005; 15(Suppl 1):174–178 13. Spijkerboer AW, Utens EM, De Koning WB, et al: Health-related quality of life in children and adolescents after invasive treatment for congenital heart disease. Qual Life Res 2006; 15:663–673
Pediatric Critical Care Medicine
14. Costello JM, O’Brien M, Wypij D, et al: Quality of life of pediatric cardiac patients who previously required extracorporeal membrane oxygenation. Pediatr Crit Care Med 2012; 13:428–434 15. Madderom MJ, Gischler SJ, Duivenvoorden H, et al: Neonatal extracorporeal membrane oxygenation: Impaired health at 5 years of age. Pediatr Crit Care Med 2013; 14:183–193 16. Wagner K, Risnes I, Berntsen T, et al: Clinical and psychosocial follow-up study of children treated with extracorporeal membrane oxygenation. Ann Thorac Surg 2007; 84:1349–1355 17. Robertson CM, Sauve RS, Joffe AR, et al: The registry and followup of complex pediatric therapies program of Western Canada: A mechanism for service, audit, and research after life-saving therapies for young children. Cardiol Res Pract 2011; 2011:965740 18. Gaies MG, Gurney JG, Yen AH, 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 19. Blishen BR: The 1981 socioeconomic index for occupations in Canada. Can Rev Soc Anth 1987; 24:465–488 20. Varni JW, Limbers CA, Burwinkle TM: Parent proxy-report of their children’s health-related quality of life: An analysis of 13,878 parents’ reliability and validity across age subgroups using the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes 2007; 5:2 21. Varni JW, Seid M, Kurtin PS: PedsQL 4.0: Reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care 2001; 39:800–812 22. Varni JW, Burwinkle TM, Seid M, et al: The PedsQL 4.0 as a pediatric population health measure: Feasibility, reliability, and validity. Ambul Pediatr 2003; 3:329–341 23. Garcia Guerra G, Robertson CM, Alton GY, et al; Western Canadian Complex Pediatric Therapies Follow-up Group: Quality of life 4 years after complex heart surgery in infancy. J Thorac Cardiovasc Surg 2013; 145:482–488.e2 24. Jaeschke R, Singer J, Guyatt GH: Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989; 10:407–415 25. Ebrahim S, Singh S, Hutchison JS, et al: Adaptive behavior, functional outcomes, and quality of life outcomes of children requiring urgent ICU admission. Pediatr Crit Care Med 2013; 14:10–18 26. Taylor AK, Cousins R, Butt WW: The long-term outcome of children managed with extracorporeal life support: An institutional experience. Crit Care Resusc 2007; 9:172–177 27. Lambert LM, Minich LL, Newburger JW, et al; Pediatric Heart Network Investigators: Parent- versus child-reported functional health status after the Fontan procedure. Pediatrics 2009; 124:e942–e949
www.pccmjournal.org
727
