P - ISSN : 2349-6592    |    E - ISSN : 2455-7099
Year : 2016 | Volume : 3 | Issue : 4 | Page : 16-19
Source of Funding:None Conflict of Interest:None
Objectives:To evaluate the sensitivity of the PRISM III score in patients admitted to PICU and to find a cut off PRISM III score.
Method:Four hundred children admitted to the PICU of a tertiary care hospital over a period of 18 months were enrolled in this cross-sectional observational study. PRISM III score was calculated using variables after admission in first 24 hours. The outcome was defined as survivors and deceased.
Results:Out of 400 patients, 70 died and 330 survived. The male to female ratio was 219:181 (55%:45%). Majority of patients were below or equal to 12 months of age (50%). The mean age was 32.9 months for non survivors, while 37.5 months for survivors. Gender, age, residence of the patient had no association with the outcome. Referred, incompletely immunized and children requiring ventilation within 1 hour of admission had poor outcome. Average PRISM III score was higher in non survivors (14.6) as compared to survivors (4.1). The mortality was significantly higher with high PRSIM III score. The area under the receiver operating characteristics curve was 0.96. The cut-off PRISM III score was 9 (kappa coefficient 0.747).
Conclusion:III score can be applied with a good degree of accuracy for severity assessment and mortality prediction to pediatric patients in PICU.
Key words:PICU, Mortality, PRISM III score, Outcome
A Pediatric intensive care unit (PICU) in a developing country has to provide the best possible care to the sick children taking into account the large patient load, scarcity of resources, lack of man power etc while ensuring a proper functioning. Evaluation of the outcomes requires use of accurate and easily applied methods. Scoring systems are aimed at quantifying case mix and using the resultant score to predict the outcome. Pediatric Risk of Mortality III (PRISM III) score is an updated version developed in 1996, has several improvements over the original PRISM score .
Materials and Methods:
This cohort study was carried out at a tertiary care teaching hospital having 120 bed Pediatric department and a 11 bedded PICU. Four hundred patients including directly referred cases were studied after hospital’s ethics committee approval over a period of 18 months (January 2014 to June 2015). Children between ages of 1 month to 12 years admitted to PICU were enrolled into the study after appropriate informed consent from the parents. The PRISM III score was evaluated as per the makers . The purpose of this study was to evaluate PRISM III score and to find a cut-off score for our PICU.
The clinical history was recorded at admission and the examination of heart rate, respiratory rate and pupillary reaction was made by pediatric resident doctor. Blood pressure (BP) was checked using standard NIBP monitor with appropriate cuff size(confirmed by sphygmomanometer), temperature using a standard mercury thermometer. Mental status was recorded according to Glasgow coma scale and modified Glasgow coma scale . White blood cell count, platelets, liver enzymes (SGOT, SGPT), bilirubin (total, direct), PT, PTT, glucose, blood urea, creatinine, sodium, potassium and calcium were measured by standard laboratory tests. Arterial blood gas analysis including bicarbonate, Paco2 and Pao2 was carried out in each patient. The most abnormal value of the variable is to be noted while entering the data during the first 24 h (PRISM III-24) after entry into PICU was considered while calculating the score. The children were followed up during hospital stay and the outcome was recorded as died or survived at the end of hospital stay. Data was collected in a predesigned proforma.Varaibles recorded were All the patients were given adequate treatment as per clinical scenario and hospital protocols.
The data was evaluated using Microsoft Excel 2010 and Software Package for Statistical Sciences (SPSS). The association between various variables for example, age, gender, residence, immunization status, referral, system involvement, PRISM III score and the outcome was tested using Chi square tests as appropriate. The aptness of the scoring system was assessed using Receiver Operating Characteristic (ROC) curve. The cut off PRISM III score was assessed using kappa coefficient.
A total of 400 children were enrolled, of which 330 survived. The observed mortality was 17.5%. There were 54.75% males in the study. Majority of our cases were infants (age < 12 months) (50%). Mean ages of survivors and non-survivors were 37.5 months and 32.9 months respectively. Most of the children (83.5%) resided in urban areas. The observed deaths were more in patients coming from rural area, however there was no statistical association between residence and outcome in PICU. 47% patients were referred from outside healthcare providers and had higher mortality(p=0.008). 233 children were completely immunized for age while the remaining cases were incompletely immunized, with significant association between immunization status and outcome (p value 0.000). 71 children required ventilation within 1 hour of admission with survival rate of 33.8%. Respiratory illnesses (36.25%), CNS disorder (21%), Multiple Organ Dysfunction Syndrome (MODS) (8.5%), Cardiovascular (5.5%), Hepatobiliary (3.25%) followed by others (25.5%) were conditions on admission. Mortality was highest in hepatobiliary disorder (76.9%) followed by MODS (73.5%). The average PRISM III score in non survivors (14.6) was higher than survivors (range 1 to 27). Mortality increased as the total PRISM III score increased. The area under the curve was 0.96 by ROC curve analysis and the cut-off score was 9 using kappa co-efficient in our study.
Table 1 summarizes the association between study variables and mortality in PICU. It can be noticed that age, gender, residence of the patients showed no significant association with the mortality. Referral, immunization status, ventilation in first hour of admission and PRISM III score showed association with mortality. Figure-1 shows increase in deaths as the total PRISM III score increases.
In this study of PRISM III score in a tertiary care PICU over 18 months ; high total PRISM III score was significantly associated with poor outcome.
PRISM III score was developed by Pollack et al in 1996 which describes the severity of illnesses according to physiological derangement .
The strength of this study lies in the number of children enrolled, prospective nature of the study and first time reporting of an association between immunization status and mortality in children admitted to PICU.
The overall mortality was 17.5% which was comparable to a study from South India . Maximum cases were < 12 months of age. There was no association between age, gender of the patients and mortality which was similar to other Indian studies [5-7]. The observed deaths were more in patients coming from rural area which was also noted by a study from Punjab . This can be due to delay in receiving medical attention. In our study, there was statistically significant association between referral and mortality. This can be attributable to transport delay or poor pre-admission medical care. This was in contrast to Bellad et al, Choi et al and Hwang et al [9-11]. Higher deaths were observed in children who were incompletely or totally unimmunized for age. Ours is the first Indian study to compare immunization status and the outcome in PICU our finding that unimmunized children are more likely to die in PICU has been reported for the first time to the best of our knowledge .This may be because of the socioeconomic factors that prevent these vulnerable children from accessing immunization services and may reflect the health seeking behavior of the family too. Similar to an Indian study , it was found that mortality was statistically higher in children who were ventilated within 1 hour.
As the PRISM III score increased, the observed percentage of deaths increased. There was a significant association between the poor outcome and high PRISM III score. Similar observations were made in many Indian and foreign countries [4,10,13-15]. The average total PRISM III score was higher in non survivors. This was similar to the observation made by an Indian study .
The ROC curve analysis confirms that PRISM III score is highly sensitive in predicting the outcome as noted by many studies [4,10,13-15]. The cut-off score of 9 was similar to an Indian study . This information would help the attending PICU physician to assess and prognosticate the critical status of the patients arriving in the PICU objectively.
High total PRISM III score was significantly associated with poor outcome i.e. death in this study.As the total PRISM III score increases, the mortality in PICU increases. Thus PRISM III score was found to be a valid predictor of outcome in our PICU.The strength of this study lies in the number of children enrolled,prospective nature of the study and first time reporting of an association between immunization status and mortality in children admitted to PICU.
Mrs. Aruna Deshpande helped with statistical analysis.
1. Castello F V, Cassano A, Gregory P, Hammond J. The Pediatric Risk of Mortality (PRISM) Score and Injury Severity Score (ISS) for predicting resource utilization and outcome of intensive care in pediatric trauma. Crit Care Med 1999;27:985-8.
2. Pollack MM, Patel KM, Ruttimann UE. PRISM III: an updated Pediatric Risk of Mortality score. Crit Care Med 1996;24:743-52.
3. Hartman M, Cheifetz I. Approach to the Emergency Evaluation of a Child. In: Kliegman R, Stanton B, St. Geme III J, Schor N, editors. Nelson Textbook of Pediatrics. 19th ed. Philadelphia: Elsevier; 2011. p. 281.
4. Prashanth M.R., Basavaraj G.V. PRISM III score - An effective tool for the prediction of outcome in a pediatric intensive care unit. Karnataka Paediatr J 2012;265–7.
5. Thukral A, Lodha R, Irshad M, Arora NK. Performance of Pediatric Risk of Mortality (PRISM), Pediatric Index of Mortality (PIM), and PIM2 in a pediatric intensive care unit in a developing country. Pediatr Crit Care Med 2006;7:356-61.
6. Singhal D, Kumar N, Puliyel JM, Singh SK, Srinivas V. Prediction of mortality by application of PRISM score in intensive care unit. Indian Pediatr 2001;38:714-9.
7. Khilnani P, Sarma D, Singh R, Uttam R, Rajdev S, Makkar A, et al. Demographic profile and outcome analysis of a tertiary level pediatric intensive care unit. Indian J Pediatr 2004;71:587-91.
8. Kaur G, Vinayak N, Mittal K, Kaushik JS, Aamir M. Clinical outcome and predictors of mortality in children with sepsis, severe sepsis, and septic shock from Rohtak, Haryana: A prospective observational study. Indian J Crit Care Med 2014;18:437-41.
9. Bellad R, Rao S, Patil VD, Mahantshetti NS. Outcome of intensive care unit patients using Pediatric Risk of Mortality (PRISM) score. Indian Pediatr 2009;46:1091–2.
10. Choi KMS, Ng DKK, Wong SF, Kwok KL, Chow PY, Chan C, et al. Assessment of the Pediatric Index of Mortality (PIM) and the Pediatric Risk of Mortality (PRISM) III score for prediction of mortality in a paediatric intensive care unit in Hong Kong. Hong Kong Med J 2005;11:97–103.
11. Hwang HS, Lee NY, Han SB, Kwak GY, Lee SY, Chung SY, et al. Performance effectiveness of pediatric index of mortality 2 (PIM2) and pediatricrisk of mortality III (PRISM III) in pediatric patients with intensive care in single institution: Retrospective study. Korean J Pediatr 2008;51:1158-64.
12. Soniwal S. Prediction of outcome of paediatric patients by validating prism III scoring within 12 hours and prism III at 72-84 hours after initiation of the treatment. J Evol Med Dent Sci 2014;3:11533–55.
13. Karambelkar GR, Mane SV, Agarkhedkar SR, Karambelkar RP, Singhania S. The relevance of 24 hour PRISM III score in predicting mortality in pediatric intensive care unit. Int J Pharm 2013;3:214-9.
14. Akhtar N, Fadoo Z, Panju S, Haque A. Outcome and prognostic factors seen in pediatric oncology patients admitted in PICU of a developing country. Indian J Pediatr 2011;78:969-72.
15. Bilan N, Galehgolab BA, Emadaddin A, Shiva S. Risk of mortality in pediatric intensive care unit, assessed by PRISM-III. Pak J Biol Sci 2009;12:480–5.