 |
|
Original Articles Indian Pediatrics 2003; 40:303-309 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Follow Up of Survival and Quality of Life in Children After Intensive Care |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
M. Jayashree, S. C. Singhi, P. Malhi From the Division of Pediatric Intensive Care, Department of Pediatrics, Advanced Pediatric Centre, Postgraduate Institute of Medical Education and Research, Chandigarh, India. Correspondence to: Dr. Sunit Singhi, Professor and
Incharge, Pediatric Emergency and Intensive Care Unit, Department of
Pediatrics, PGIMER, Chandigarh 160 012, India. Manuscript received:March 13, 2002, Initial review
completed: April 30, 2002, Objective:To evaluate the health related quality of life of children after intensive care and to assess their long term survival. Design: Prospective. Setting: Tertiary Care Hospital Intensive Care Unit (ICU). Subjects and Methods: All patients admitted to ICU were enrolled prospectively over a period of 1 year. Children with ICU stay of less than 24 hrs, infants, readmission to ICU were however excluded. Survival was determined at the time of ICU discharge and 1 year later. Health status assessment was done with the help of the multiattribute health status classification (MAHSC), which has 6 domains; sensation, mobility, emotion, cognition, self care and pain. Assessment was done at two points of time - within 48 hours of admission to the ICU and 1 year after discharge. Results: 150 children (mean age 5.68 ± 3.6 years) with a mean duration of ICU stay (5.7 ± 5.5 days) were included in the study. The cumulative ICU mortality was 12.9%. Fifty-five (36.7%) had no overall health impairment (no affected domains) preceding the present illness. There was overall health impairment (³1 affected domain) preceding the present illness in 95 of the 150 patients (63.3%). In the domain specific health status mobility was affected in 74 (49%) followed by pain 61 (41.2%), self care 56 (38.8%), sensation 29 (20%), cognition 21 (14.8%) and emotion 14(9.5%). After 1 year, overall state of health had improved or was equal to the premorbid state in 87 (75%). In domain specific health, the proportion improving or remaining unchanged varied from 75% (emotional) to 80% (cognition), 85.3% (pain) and 88.7% (mobility). The overall state of health had worsened as compared to the premorbid state in 29 (25%) majority with neurological illnesses. Conclusion: Quality of life in three-quarters of the patients was preserved and one year survival was favorable. Worsening was noted primarily in-patients with neurological illnesses. Health related quality of life, Multi-attribute health status classification, Outcome, Pediatric intensive care.
Most of the available literature describes QOL of adult patients with few focussing on pediatric population. These few studies denote a favorable long-term survival of children admitted to ICU and a good functional outcome in terms of preservation of health(6,9,10). Similar studies are lacking in Indian set-up. We conducted this study with the aim of evaluating the health-related quality of life in children after intensive care and to assess their long-term survival. Subjects and Methods All patients admitted over a period of one year (May 1999-May 2000) to the Pediatric Intensive Care Unit (PICU) of Advanced Pediatric Centre, PGIMER, Chandigarh were prospectively enrolled in the study. Pediatric ICU has 8 beds and had admitted 347 patients during the study period. On admission to PICU, the demographic details were recorded with specific emphasis on primary diagnosis and indications for PICU referral. PRISM (III) scoring was done on Day 1 of admission to ascertain the severity of illness. Survival was determined at the time of ICU discharge; hospital discharge and one year after ICU discharge. Health Status Assessment The multi-attribute health status classification (MAHSC) was developed as a comprehensive generic health status measure for children. This classification has been used in pediatric patients in previous studies, originally in evaluating survivors of child-hood cancer(8). The MAHSC has been divided into 6 domains: sensation, mobility, emotion, cognition, self care and pain (Table I). Table I__The Multi-attribute Health Status Classification System (MAHSC).
Adopted from Feeny et al. (1992)(8).
Functioning within each attribute is represented by four or five levels, varying from healthy to poor depending on absence or level of impairment. The state of health of a particular subject is described as a profile by a six element vector (X1X2X3X4X5X6) where X denotes the level (1-5) for domain 1. The level of each domain are meant to be interpreted as developmentally appropriate for the age. This system has not been validated for children less than 12 months of age. We recorded health status within 48 hours of ICU admission and 1 year after discharge from the ICU. The health status recorded within 48 hours of ICU admission, reflected the premorbid health status (i.e. before the present illness) and averaged over three months preceding the present illness. This was assessed by a questionnaire completed during a structured parental interview. Excluded from the analysis were children less than one year of age, survivors who stayed for less than 24 hrs in ICU, readmission and death within 24 hr of ICU stay. One year after discharge health status was re-assessed by an identical questionnaire completed during a structured interview at follow-up.
1. Overall health status: It was expressed as number of affected domains regardless of the degree of dysfunction.2. Domain specific health status: Degree of functional level within each of the domain. 3. Change in overall health status was obtained by comparing the number of affected domains before admission with that one year after discharge. 4. Change in domain specific health status: This was defined as the difference from preadmission to one year after discharge corresponding to each domain. A positive change was defined as an improved level and a negative change as a deteriorated level. Statistical Analysis Data are presented as mean values ± standard deviation. Comparison of two groups in different areas was performed by using Student t test. Significance was accepted as P <0.05 level. Correlation between PRISM and overall health status was done with the help of Chi square and Spearmann correlation. Results Overall 347 patients were seen during the study period. Forty five patients died in ICU (12.9%) and none during their subsequent hospital stay or in the first year following discharge. The cumulative ICU mortality was thus 12.9%. The following patients were excluded from the study: Patients younger than 1 year (n = 159), survivors who stayed less than 24 hrs in the ICU (n = 16) non survivors (n = 18), and ICU readmission (n = 4). There were 150 patients at the beginning of the study who were eligible for enrollment. The mean age of the study population was 5.7 ± 3.6 yr (range 1-12 yr) with a M:F ratio of 2.4:1. The mean duration of ICU stay was 5.7 ± 5.5 days (range 2-42 days). Twenty-two (14.7%) patients died in the study group. Fifty five (36.7%) had no overall health impairment (i.e., no affected domains) preceding the present illness. Ten of these died in the ICU. Thirty three survivors were in perfect health at the end of 1 year and in 12 of the 45 survivors with uncompromized health, a decline in overall health status was found. Overall health impairment (³1 affected domain) preceding the present illness was found in 95 (63.3%) patients. Twelve of these died in the ICU and 7 did not return for follow-up. Of the 76 patients available for follow-up at the end of one year, 32 patients (42%) improved to normal health, 20(26.3%) showed an improvement to better level as compared to admission and 17(22.3%) showed a decline in the health status. Table II and Fig. 1 show the one year outcome (overall health status) in relation to premorbid health status. Table II One Year Outcome in Relation to Premorbid Health Status.
Fig. 1. Health Status - Premorbid Admission vs one year In summary the overall health status had improved or remained unchanged in 87 (75%) at 1 year after discharge from PICU. In the domain specific health status, mobility was affected in 74(49%) followed by pain 61 (41.2%), self care 56 (38.8%), sensation 29 (20%) cognition 21 (14.8%) and emotion 14 (9.5%). One year post discharge, the domain specific health status is as shown in Fig. 2, comparing it to the premorbid health status. It had remained unchanged or improved in 75% with respect to emotional functions, 88.7% in mobility, 85.3% with respect to pain and 80% in cognition. Emotional domain was the worst affected in 28 (24.4%).
Fig. 2 Changes in Domains at one year The relation between overall health status and acute severity of illness on admission to PICU is shown in Table III. Seven (32%) of the non-survivors had an uncompromized health status at admission, 9 (40%) had one or two domains affected and 6 (27.2%) had more than two domains affected. In all the groups, majority of the patients who died were in the moderate to high risk mortality group as per the PRISM scoring. There was no definite correlation that could be drawn from the premorbid health status and mortality risk in the non-survivors. Table III Relation Between PRISM Score and Overall Health Status
No correlation could be found between the two (Chi square and Spearmann correlation).
Discussion Quality of life (QOL) is a multi-dimensional entity. It includes, but is not limited to, the social, physical and emotional functioning of the child and when indicated his/her family and it must be sensitive to the changes that occur throughout development. In the majority of circumstances, the impact of disease on QOL is measured and hence the term health related quality of life has often been used(11). The multi-attribute health status classi-fication system (MAHSC) was first evaluated and applied to study QOL in survivors of cancer(8,12). A major advantage of this system is that it provides a comprehensive assessment and hence identifies the sequelae which impair long term outcome. It helps in identifying the type and severity of sequelae. It can be used for multiple assessments in long term follow up studies. Thus, the system focuses attention on the full array of the dimensions of health status. There are however some disadvantages associated with this sytem. The data lacks way to report organ toxicity; it also does not include a separate component for prognosis. It has not been validated for infants, who represent a major portion of ICU admissions. Including this group would hence be particularly relevant for long term outcome. There was no difference between data collection on admission and one year later, as on both occasions, the parents responded to identical questionnaire in a structured inter-view. We found that overall health impair-ment before the present illness was present in nearly two third of the patients and one third were normal (i.e., no affected attributes). Eighty percent of those who were normal continued to remain so after one year and 20% had shown a decline. On the other hand 80% of those who showed health impairment at one year, majority were health impaired at admission. These findings emphasize that pre-morbid health status has an important bearing on the long term outcome. The domain specific health status also reflects a similar trend. Major deterioration both in overall and domain specific health status was observed in 29 (25%) patients. It is interesting to note that the group which deteriorated belonged to the neurological illnesses, which left behind sequelae. Emotional domain was the most commonly affected followed by cognition, self care, pain, mobility and sensation. The affection of cognition, mobility and self care as a part of neurological sequelae is well accepted. The emotional attribute needs to be looked into more carefully. There is a dearth of studies addressing this issue in pediatric care, but a few that are available have indicated a favourable outcome(6,10). Butt et al.(9) found a cummulative mortality of 14.3% and overall health impairment on follow up in 58%. Health status remained unchanged in 68.7%, deteriorated in 8.3% and improved in 8.7% compared to preadmission. This was a retrospective study and had not used any health measuring instrument. In a prospective study Gemke et al.(6) found that overall pre-admission health status was impaired in 68.5% and after one year, it had improved or equalled the pre-admission state in 72.6%(6). In another prospective study by Morrison et al.(10), it was found that 59.3% of children had scores indicating a normal QOL and 32.4% had a fair QOL requiring some intervention. In adults cumulative one to two year mortality varied from 23-69% with 10-68% survivors regaining their previous state of health(5,13). Another study found that only 31% to 77% of adults who are admitted to ICU survive and of this only 62% to 70% regain their previous state of health(14). In contrast to these there are some studies, the results of which suggest that intensive care is most cost effective for children than adults(15,16). We did not find a definite correlation between the premorbid health status and the mortality risk in our study. A similar pros-pective study in pediatric patients, however has demonstrated high mortality risk in-patients with un-compromised health status as compared to patients with impaired health status where mortality risk was low(6). We conclude that overall health status at one year after ICU discharge is significantly affected by the premorbid health status. In spite of large number of health impairment preceding the present illness, two thirds of the patients had improved at one year. Worsening was noted in relation to the neurological illness and their sequelae. The domain of emotion and pain needs more attention as a part of ICU care. Contributors: All the three authors were involved in the planning and designing of the study protocol. PM guided the data collection, JM drafted the manuscript and SS critically reviewed the manuscript and finalized the draft and shall act as guarantor for the paper. Funding: Postgraduate Institute of Medical Education and Research, Chandigarh. Competing interests: None stated.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
References | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
