Journal of Clinical Neuroscience 22 (2015) 111–115
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Use of APACHE II and SAPS II to predict mortality for hemorrhagic and ischemic stroke patients Byeong Hoo Moon, Sang Kyu Park ⇑, Dong Kyu Jang, Kyoung Sool Jang, Jong Tae Kim, Yong Min Han Department of Neurosurgery, Incheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea
a r t i c l e
i n f o
Article history: Received 12 February 2014 Accepted 10 May 2014
Keywords: APACHE II Hemorrhage Ischemia Mortality SAPS II Stroke
a b s t r a c t We studied the applicability of the Acute Physiology and Chronic Health Evaluation II (APACHE II) and Simplified Acute Physiology Score II (SAPS II) in patients admitted to the intensive care unit (ICU) with acute stroke and compared the results with the Glasgow Coma Scale (GCS) and National Institutes of Health Stroke Scale (NIHSS). We also conducted a comparative study of accuracy for predicting hemorrhagic and ischemic stroke mortality. Between January 2011 and December 2012, ischemic or hemorrhagic stroke patients admitted to the ICU were included in the study. APACHE II and SAPS IIpredicted mortalities were compared using a calibration curve, the Hosmer–Lemeshow goodness-of-fit test, and the receiver operating characteristic (ROC) curve, and the results were compared with the GCS and NIHSS. Overall 498 patients were included in this study. The observed mortality was 26.3%, whereas APACHE II and SAPS II-predicted mortalities were 35.12% and 35.34%, respectively. The mean GCS and NIHSS scores were 9.43 and 21.63, respectively. The calibration curve was close to the line of perfect prediction. The ROC curve showed a slightly better prediction of mortality for APACHE II in hemorrhagic stroke patients and SAPS II in ischemic stroke patients. The GCS and NIHSS were inferior in predicting mortality in both patient groups. Although both the APACHE II and SAPS II systems can be used to measure performance in the neurosurgical ICU setting, the accuracy of APACHE II in hemorrhagic stroke patients and SAPS II in ischemic stroke patients was superior. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Stroke is the second most common cause of death worldwide and the most frequent cause of permanent disability. While the mortality rate of stroke has markedly decreased in most countries with the introduction of intensified treatment protocols and advanced supportive medical therapy, mortality is still high in patients requiring intensive care treatment [1]. The admission of stroke patients to an intensive care unit (ICU) involves immense technological, professional, and financial resources, thus efforts to predict outcomes are essential. Most outcome scoring systems have been devised for general ICU populations and have not been validated for specific patient groups. Reliability and validity are important issues that allow confident use of a scoring system in ICU patients with different case mixes and baseline characteristics. There have been few reports regarding the use of severity scoring systems specifically in acute stroke patients to determine survival.
⇑ Corresponding author. Tel.: +82 32 280 5973; fax: +82 32 280 5991. E-mail address: [email protected] (S.K. Park). http://dx.doi.org/10.1016/j.jocn.2014.05.031 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.
The Glasgow Coma Scale (GCS) and National Institutes of Health Stroke Scale (NIHSS) have been shown to be predictors of prognosis for acute stroke patients in several studies [2–6]. However, there have been few reports in which the GCS or NIHSS have been compared with other scoring systems. The Acute Physiology and Chronic Health Evaluation II (APACHE II), Simplified Acute Physiology Score II (SAPS II), GCS and NIHSS were thus compared as predictors of mortality in acute stroke patients in the current study. The purpose of this study was therefore twofold, comparing APACHE II, SAPS II, GCS and NIHSS in predicting mortality in stroke patients in the ICU, and assessing the applicability of APACHE II and SAPS II in two specific disease categories (hemorrhagic and ischemic stroke).
2. Materials and methods 2.1. Study population We included every patient admitted to the ICU of our cerebrovascular centers with acute cerebral ischemia (ischemic stroke) or intracerebral hemorrhage (hemorrhagic stroke) over a 2 year
112
B.H. Moon et al. / Journal of Clinical Neuroscience 22 (2015) 111–115
period. Exclusion criteria were acute strokes correlating with trauma, tumor, and other vascular malformations, including moyamoya disease or aneurysms. All patients admitted to our ICU between January 2011 and December 2012 were screened for eligibility. Decisions regarding admission to the ICU were made by the patient’s physician. The criteria for ICU admission were GCS score 612 or NIHSS P5, and one of the following: severe cerebral infarction, defined as an infarction greater than two-thirds of the middle cerebral artery territory for a supratentorial stroke; massive cerebellar infarction, defined as an infarction involving at least the territory of the superior cerebellar artery, anterior inferior cerebellar artery, or posterior inferior cerebellar artery; severe brain stem infarction; severe intracerebral hemorrhage defined as a hematoma volume of >25 ml for a supratentorial hemorrhage; a cerebellar hemorrhage >3 cm in diameter; severe brainstem hemorrhage resulting in respiratory failure requiring endotracheal intubation and/or mechanical ventilation; or neurologic diseases with severe medical co-morbidity, such as cardiac arrhythmia, pneumonia, or organ failure needing intensive use of life-support measures. All demographic and clinical variables, including the GCS and NIHSS scores, were collected for each patient. Survival status was recorded up to the time of hospital discharge. For patients who were re-admitted to the ICU during the study period, only the first admission was considered. Patients with an ICU length of stay 0.05) suggested good calibration, and a small p value indicated a poor calibration of the model. To assess the ability of the scoring systems to discriminate between survivors and non-survivors, receiver operating characteristic (ROC) curves were constructed for the prediction of mortality [10]. The ROC curve was achieved by plotting the sensitivities for all individual cut-off values versus the corresponding (1-specificity) values for the same cut-offs. The discrimination of a prognostic model is considered as good if the area under the curve (AUC) is >0.8, moderate if the AUC is 0.6–0.8, and poor if the AUC
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Use of APACHE II and SAPS II to predict mortality for hemorrhagic and ischemic stroke patients Byeong Hoo Moon, Sang Kyu Park ⇑, Dong Kyu Jang, Kyoung Sool Jang, Jong Tae Kim, Yong Min Han Department of Neurosurgery, Incheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea
a r t i c l e
i n f o
Article history: Received 12 February 2014 Accepted 10 May 2014
Keywords: APACHE II Hemorrhage Ischemia Mortality SAPS II Stroke
a b s t r a c t We studied the applicability of the Acute Physiology and Chronic Health Evaluation II (APACHE II) and Simplified Acute Physiology Score II (SAPS II) in patients admitted to the intensive care unit (ICU) with acute stroke and compared the results with the Glasgow Coma Scale (GCS) and National Institutes of Health Stroke Scale (NIHSS). We also conducted a comparative study of accuracy for predicting hemorrhagic and ischemic stroke mortality. Between January 2011 and December 2012, ischemic or hemorrhagic stroke patients admitted to the ICU were included in the study. APACHE II and SAPS IIpredicted mortalities were compared using a calibration curve, the Hosmer–Lemeshow goodness-of-fit test, and the receiver operating characteristic (ROC) curve, and the results were compared with the GCS and NIHSS. Overall 498 patients were included in this study. The observed mortality was 26.3%, whereas APACHE II and SAPS II-predicted mortalities were 35.12% and 35.34%, respectively. The mean GCS and NIHSS scores were 9.43 and 21.63, respectively. The calibration curve was close to the line of perfect prediction. The ROC curve showed a slightly better prediction of mortality for APACHE II in hemorrhagic stroke patients and SAPS II in ischemic stroke patients. The GCS and NIHSS were inferior in predicting mortality in both patient groups. Although both the APACHE II and SAPS II systems can be used to measure performance in the neurosurgical ICU setting, the accuracy of APACHE II in hemorrhagic stroke patients and SAPS II in ischemic stroke patients was superior. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Stroke is the second most common cause of death worldwide and the most frequent cause of permanent disability. While the mortality rate of stroke has markedly decreased in most countries with the introduction of intensified treatment protocols and advanced supportive medical therapy, mortality is still high in patients requiring intensive care treatment [1]. The admission of stroke patients to an intensive care unit (ICU) involves immense technological, professional, and financial resources, thus efforts to predict outcomes are essential. Most outcome scoring systems have been devised for general ICU populations and have not been validated for specific patient groups. Reliability and validity are important issues that allow confident use of a scoring system in ICU patients with different case mixes and baseline characteristics. There have been few reports regarding the use of severity scoring systems specifically in acute stroke patients to determine survival.
⇑ Corresponding author. Tel.: +82 32 280 5973; fax: +82 32 280 5991. E-mail address: [email protected] (S.K. Park). http://dx.doi.org/10.1016/j.jocn.2014.05.031 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.
The Glasgow Coma Scale (GCS) and National Institutes of Health Stroke Scale (NIHSS) have been shown to be predictors of prognosis for acute stroke patients in several studies [2–6]. However, there have been few reports in which the GCS or NIHSS have been compared with other scoring systems. The Acute Physiology and Chronic Health Evaluation II (APACHE II), Simplified Acute Physiology Score II (SAPS II), GCS and NIHSS were thus compared as predictors of mortality in acute stroke patients in the current study. The purpose of this study was therefore twofold, comparing APACHE II, SAPS II, GCS and NIHSS in predicting mortality in stroke patients in the ICU, and assessing the applicability of APACHE II and SAPS II in two specific disease categories (hemorrhagic and ischemic stroke).
2. Materials and methods 2.1. Study population We included every patient admitted to the ICU of our cerebrovascular centers with acute cerebral ischemia (ischemic stroke) or intracerebral hemorrhage (hemorrhagic stroke) over a 2 year
112
B.H. Moon et al. / Journal of Clinical Neuroscience 22 (2015) 111–115
period. Exclusion criteria were acute strokes correlating with trauma, tumor, and other vascular malformations, including moyamoya disease or aneurysms. All patients admitted to our ICU between January 2011 and December 2012 were screened for eligibility. Decisions regarding admission to the ICU were made by the patient’s physician. The criteria for ICU admission were GCS score 612 or NIHSS P5, and one of the following: severe cerebral infarction, defined as an infarction greater than two-thirds of the middle cerebral artery territory for a supratentorial stroke; massive cerebellar infarction, defined as an infarction involving at least the territory of the superior cerebellar artery, anterior inferior cerebellar artery, or posterior inferior cerebellar artery; severe brain stem infarction; severe intracerebral hemorrhage defined as a hematoma volume of >25 ml for a supratentorial hemorrhage; a cerebellar hemorrhage >3 cm in diameter; severe brainstem hemorrhage resulting in respiratory failure requiring endotracheal intubation and/or mechanical ventilation; or neurologic diseases with severe medical co-morbidity, such as cardiac arrhythmia, pneumonia, or organ failure needing intensive use of life-support measures. All demographic and clinical variables, including the GCS and NIHSS scores, were collected for each patient. Survival status was recorded up to the time of hospital discharge. For patients who were re-admitted to the ICU during the study period, only the first admission was considered. Patients with an ICU length of stay 0.05) suggested good calibration, and a small p value indicated a poor calibration of the model. To assess the ability of the scoring systems to discriminate between survivors and non-survivors, receiver operating characteristic (ROC) curves were constructed for the prediction of mortality [10]. The ROC curve was achieved by plotting the sensitivities for all individual cut-off values versus the corresponding (1-specificity) values for the same cut-offs. The discrimination of a prognostic model is considered as good if the area under the curve (AUC) is >0.8, moderate if the AUC is 0.6–0.8, and poor if the AUC
