Acta Neurol Belg DOI 10.1007/s13760-014-0420-x
ORIGINAL ARTICLE
Application of the APACHE II score to assess the condition of patients with critical neurological diseases Yi Bian • Ping Zhang • Yongjie Xiong • Feng Xu • Suiqiang Zhu • Zhouping Tang Zheng Xue
•
Received: 23 June 2014 / Accepted: 25 December 2014 Ó Belgian Neurological Society 2015
Abstract The Acute Physiology And Chronic Health Evaluation II (APACHE II) scoring system has been commonly used to assess the severity of patients’ diseases in general intensive care units (ICUs). However, few studies have investigated the application of this scoring system in patients in neurologic ICUs. In this study, the APACHE II scores of 102 patients in the neurologic ICU were calculated within the first 24 h. The actual mortality and predicted mortality were obtained based on these scores and analyzed statistically. The data indicated that cerebral hemorrhage, cerebral infarction and intracranial infection accounted for the top three causes for admission to the neurologic ICU, and these conditions were associated with high APACHE II scores and high predicted mortality. Additionally, the actual mortality rate was lower than the predicted rate after effective treatment. All patients were divided into groups according to their APACHE II scores, and we found that higher APACHE II scores were associated with higher actual mortality, especially for patients whose APACHE II scores were greater than 10. The APACHE II scores of the deceased patient group were higher than those of the surviving group, and
Y. Bian and P. Zhang contributed equally to this work. Y. Bian Department of Emergency, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, People’s Republic of China P. Zhang Y. Xiong F. Xu S. Zhu Z. Tang Z. Xue (&) Department of Neurology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, People’s Republic of China e-mail: [email protected]
this difference was statistically significant. In conclusion, our study found that the APACHE II scoring system may provide valuable information for predicting patient’s condition and prognosis in neurologic ICUs. Keywords Acute physiology and chronic health evaluation II Neurologic intensive care unit Mortality Prognosis
Introduction The Acute Physiology And Chronic Health Evaluation II (APACHE II) is the most widely used and authoritative system used to evaluate critical disease condition. The system assesses disease severity based on human acute physiological alterations and the patient’s health history and age. The APACHE II scoring system is commonly used in general intensive care units (ICUs) worldwide and can be used to objectively assess a patient’s condition, predict mortality, provide a scientific basis for the development of a therapeutic plan and improve healthcare quality and the rational utilization of medical resources. However, application of the APACHE II to assess the disease condition of patients with critical neurological illness is not commonly performed. Moreover, few studies have investigated the application of this system in patients in neurologic ICUs. In the present study, the APACHE II scoring system was used to evaluate patients admitted to the neurologic ICU of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology between May of 2013 and October of 2013. We also sought to investigate the predictive value of this system in patients with critical neurological conditions.
123
Acta Neurol Belg
Materials and methods General patient information One hundred and two patients treated in the neurologic ICU of Tongji Hospital of Tongji Medical College at Huazhong University of Science and Technology between May of 2013 and October of 2013 were recruited into this study. These patients included 73 men and 29 women, and the male to female ratio was 2.5:1. The age of disease onset ranged from 16 to 84 years with an average of 52.7 ± 16.6 years. Among these patients, there were 41 cases of cerebral hemorrhage (40.2 %), 27 cases of cerebral infarction (26.5 %), 23 cases of intracranial infection (22.5 %), 4 cases of subarachnoid hemorrhage (3.9 %), 2 cases of encephalopathy syndrome (2.0 %), 2 cases of space-occupying lesions on the brain stem (2.0 %), 2 cases of myasthenia gravis (2.0 %) and 1 case of peripheral neuropathy (1.0 %, Fig. 1). Among the 102 patients, 6 patients died (5.9 %) at ICU discharge, including 3 patients who died from cerebral hemorrhage, 2 patients from cerebral infarction and 1 patient from intracranial infection. After active treatment, 67 patients (65.7 %) improved markedly, met the criteria for discharge from the ICU and were transferred to a general ward for continuous treatment. 29 patients (28.4 %) did not show significant
improvement after treatment, and due to financial burden and other factors the families of these patients requested transfer to another hospital. At the 28-day follow-up, four of the patients who were transferred to other hospitals died, two of them died from cerebral hemorrhage and the other two from cerebral infarction. So, the total 28-day mortality rate was 9.8 %. Scoring criteria According to the APACHE II scoring criteria reported by Knaus et al. [1], each patient was evaluated within 24 h after admission to the ICU and the worst score was used for the evaluation. Prior to discharge from the ICU after treatment, the patients were evaluated again. The predicted risk of mortality (R) for each patient was calculated according to the following equation: ln (R/1 - R) = -3.517 ? (APACHEIIscore 9 0.146) ? 0.603 (only for patients who had undergone operations in the emergency room) ? score of the disease causing admission to the ICU. The predicted group risk of mortality was calculated by dividing the sum of the R value of each patient by the total number of patients. Statistical analysis Statistical analysis was performed using the software SPSS12.0. Measurement data were presented as the mean ± standard deviation, and the difference between means was analyzed with the Student’s t test. Count data were analyzed using descriptive statistics and the v2 test. P \ 0.05 indicates significant difference. Ethical standards statement and patient consents The study was approved by the hospital ethics committee and was therefore performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Specific national laws were also observed. Written informed consent was obtained from all patients or their legal guardians.
Results The average APACHE II score for patients with different disease types
Fig. 1 The disease types encountered in the neurologic ICU
123
The average APACHE II score of each group of patients with different disease types at the time of admission to the ICU and at the time of discharge from the ICU are shown in Table 1.
Acta Neurol Belg Table 1 APACHE II score according to disease type Diagnosis
Cerebral hemorrhage
Cerebral infarction
Intracranial infection
Subarachnoid hemorrhage
Encephalopathy syndrome
Space-occupying lesions on brain stem
Myasthenia gravis
Peripheral neuropathy
Number of cases (%)
41 (40.2)
27 (26.5)
23 (22.5)
4 (3.9)
2 (2.0)
2 (2.0)
2 (2.0)
1 (1.0)
Score at admission to ICU
14.4 ± 5.0
13.2 ± 6.2
12.4 ± 6.1
12.5 ± 7.7
18.5 ± 0.7
16.5 ± 6.4
8.5 ± 6.4
4
Score at discharge from ICU
10.0 ± 6.1
10.9 ± 5.6
5.5 ± 4.2
15.0 ± 10.2
11.0 ± 5.7
10.5 ± 2.1
4.5 ± 2.1
0
Predicted risk of mortality (%)
47.7
18.1
16.7
42.6
27.5
24.0
9.4
4.4
Mortality of ICU discharge (%)
7.3
7.4
4.3
0
0
0
0
0
28-day mortality rate (%)
12.2
14.8
4.3
0
0
0
0
0
Table 2 Patient prognosis at ICU discharge according to the APACHE II score at the time of admission to the ICU Score at admission to ICU
0–4
5–9
10–14
15–19
20–24
25–29
Number of cases (%)
4 (3.9)
26 (25.5)
25 (24.5)
32 (31.4)
13 (12.7)
2 (2.0)
Number of deaths at ICU discharge (%)
0
0
1 (4.0)
2 (6.3)
2 (15.4)
1 (50.0)
Number of improvements (%)
4 (100.0)
21 (80.8)
19 (76.0)
19 (59.4)
4 (30.8)
0
Number transferring to other hospitals (%)
0
5 (19.2)
5 (20.0)
11 (34.4)
7 (53.8)
1 (50.0)
Average days in ICU
14.0 ± 10.1
7.6 ± 3.8
8.5 ± 4.1
9.2 ± 7.6
9.1 ± 6.1
16.0 ± 18.4
Patient prognosis according to the APACHE II score at the time of admission to the ICU The range of the APACHE II score for the total of 102 patients was 4–29 with an average score of 13.5 ± 5.8. The patients were classified according to the APACHE II score at the time of admission to the ICU. Each five-point increment in the score was classified as one group, and patient prognosis was summarized according to this classification. Table 2 shows the average length of stay in ICU and the prognosis of each group of patients at ICU discharge, which was classified into death, improvement (patients improved and were discharged from the ICU) and transferring (patients were transferred to other hospitals for continuous treatment due to financial burden and other factors). The predicted risk of mortality according to the APACHE II score at the time of admission to the ICU Based on the equation reported by Knaus et al. [1] for calculating the predicted risk of mortality, the predicted
risk of mortality for the total of 102 patients was 30.6 %. The actual mortality at ICU discharge was 5.9 % and the actual 28-day mortality rate was 9.8 %. The predicted risk of mortality was calculated for each patient group classified according to each five-point increment in the APACHE II score. The predicted risk of mortality was also compared to the actual mortality (Table 3). The actual mortality was significantly lower than the predicted risk of mortality for all groups. Comparison between the surviving group and the deceased group Comparison of the patient general data at the end point of ICU discharge between the surviving group of 96 patients and the deceased group of 6 patients revealed that (Table 4) the average age and the gender ratio had no statistically significant difference. The average APACHE II score at the time of admission to the ICU was 13.1 ± 5.5 in the surviving group and 20.0 ± 6.4 in the deceased group, and this difference was significant (P \ 0.05). The average length of stay in the ICU was 8.9 ± 6.4 days in the
123
Acta Neurol Belg Table 3 The predicted risk of mortality according to the APACHE II score at the time of admission to the ICU and the actual mortality Score at admission to ICU
0–4
5–9
10–14
15–19
20–24
25–29
Predicted risk of mortality (%)
4.4
12.6
26.3
41.3
51.0
69.0
Mortality of ICU discharge (%)
0
0
4.0
6.3
15.4
50.0
28-day mortality rate (%)
0
0
8.0
9.4
30.8
50.0
Table 4 Comparison between the surviving and deceased groups at ICU discharge Total
Surviving group
Deceased group
Age (years)
52.7 ± 16.6
52.7 ± 16.7
53.8 ± 15.7
Gender ratio (male/female)
2.5:1
2.4:1
5:1
Score at admission to ICU
13.5 ± 5.8
13.1 ± 5.5
20.0 ± 6.4**
Predicted risk of mortality (%)
30.6 ± 20.7
29.6 ± 20.0
47.8 ± 25.5*
Average days in ICU
8.9 ± 6.2
8.9 ± 6.4
9.3 ± 3.4
Total
Surviving group
Deceased group
Age (years)
52.7 ± 16.6
52.7 ± 17.0
53.5 ± 13.1
Gender ratio (male/female)
2.5:1
2.4:1
4:1
Score at admission to ICU
13.5 ± 5.8
12.9 ± 5.4
19.1 ± 6.0**
Predicted risk of mortality (%)
30.6 ± 20.7
28.9 ± 19.6
46.5 ± 25.2*
Average days in ICU
8.9 ± 6.2
9.2 ± 6.4
6.7 ± 4.4
Comparison between the surviving group and deceased group at ICU discharge * As the standard deviation was large, significance test was not performed ** P \ 0.05
Table 5 Comparison between the surviving and deceased groups at 28 days
Comparison between the surviving group and deceased group at 28 days * As the standard deviation was large, significance test was not performed ** P \ 0.05
surviving group and 9.3 ± 3.4 days in the deceased group, with no significant difference. Then we compared the data at the end point of 28 days between the surviving group of 92 patients and the deceased group of 10 patients (Table 5). The average APACHE II score at the time of admission to the ICU of the two groups also showed significant difference (P \ 0.05). Patient prognosis according to the APACHE II score at the time of discharge from the ICU The APACHE II score at the time of discharge from the ICU for the 96 survivors was 0–27 with an average of 9.2 ± 6.1. Among the survivors, 67 patients (69.8 %) demonstrated improvement at the time of discharge from the ICU, with an average APACHE II score of 7.1 ± 4.4. Of these patients, 29 (30.2 %) were transferred to other hospitals for continuous treatment, with an average score of 14.1 ± 6.7.
123
Discussion The APACHE scoring system was initially proposed by Knaus et al. [2] in 1981. The APACHE II scoring system is used to predict the disease condition and prognosis of patients with critical diseases according to acute physiological alterations and the patient’s health history and age. The APACHE II scoring system has become the most authoritative and widely used international system for evaluating patients with critical diseases, and this method has also been commonly used worldwide to assess the condition of patients with severe diseases treated in general ICUs. This scoring system can also be used in specialized ICUs. Application of the APACHE II scoring system to evaluate the condition of patients with neurological diseases such as stroke and severe traumatic brain injury has been reported [3–6]. However, few studies have investigated the application of this scoring system in patients in neurologic ICUs. Our study found that the APACHE II
Acta Neurol Belg
scoring system may provide valuable information for predicting patient’s condition and prognosis, developing a therapeutic plan and promoting the rational utilization of medical resources in neurologic ICUs. The top three diseases responsible for admission to the neurologic ICU at our hospital were cerebral hemorrhage, cerebral infarction and intracranial infection. Patients with these three diseases demonstrated high APACHE II scores and high predicted risks of mortality at the time of admission to the ICU. After active treatment, the actual mortality of these patients was significantly lower than the predicted risk of mortality, suggesting that patients suffering from diseases with high APACHE II scores and a high predicted risk of mortality could substantially benefit from treatment in a neurologic ICU. In our study, patients were classified according to the APACHE II score obtained at the time of admission to the neurologic ICU, and the patients’ prognoses were monitored. The results showed that as the APACHE II score increased, the mortality rate increased and the proportion of patients showing improvement decreased. These results are similar to previous findings [7–9], suggesting that the APACHE II score and the predicted risk of mortality have predictive value for assessing the condition and prognosis of patients with critical neurological disease. The most critical factor influencing the prognosis of patients with severe brain diseases is the severity of the brain injury. Thus, the difference in the indicator used to assess the severity of brain damage, i.e., the Glasgow Coma Scale (GCS), is most significant among the patients in the neurologic ICU. It has been reported that the APACHE II score is more accurate than the GCS score for predicting late mortality of patients with acute severe brain diseases, although the APACHE II score may be less accurate than the GCS score to predict early mortality [10]. Comparisons between the surviving group and the deceased group at two different end points (at ICU discharge and 28-day follow-up) in this study both revealed that the APACHE II score at admission to ICU in the deceased group was significantly higher than that in the surviving group. These results indicate that a high APACHE II score and high predicted risk of mortality for patients in neurologic ICUs may predict a poor prognosis. The predicted risk of mortality for the total patient group reported in the present study is similar to reports by others [10], whereas the actual mortality in this study was lower. Among the patients in the neurologic ICU, GCS might be a most significant indicator in APACHE II scoring. Patients in the medium-score groups might be in middle or deep coma at the onset of intracranial pathology, so that they got a relatively high APACHE II score when they were admitted to ICU. After active treatment such as applying mannitol, some patients came out of coma during their stay
in ICU, then the APACHE II score declined and they presented better prognosis. This might explain the difference between predicted and actual mortality. In addition, the APACHE II score is based on the prediction of mortality risk in a group and might not accurately predict the mortality risk for an individual. Thus, insufficient sample sizes may lead to biased results. Our study demonstrated that the actual mortality of patients with an APACHE II score at admission greater than 10 was significantly higher than that of patients with a score less than 10. The threshold APACHE II score for determining whether a patient should be admitted to the ICU has not been standardized. Naved et al. [11] evaluated the APACHE II scores of patients at the second or seventh day in the ICU and found that the discharge rate from the hospital in patients with scores of 3–10 was high (90 %) and that the mortality of patients with scores of 31–40 was also high (84.6 %). Jiang et al. [12] showed that an APACHE II score less than 10 was associated with a low probability of hospital death; in addition, a score of 10–20 was associated with a mortality rate of approximately 50 %, and a score greater than 20 was associated with a high mortality of 80–100 %. In our department, we tend to use an APACHE II score of 10 as the threshold value for admission to the ICU. However, the rationality for using this threshold value requires further investigation. In addition, when a patient’s condition improves after active treatment, the APACHE II score should be dynamically monitored to determine whether the patient meets the criteria for discharge from the ICU. This strategy can facilitate the development of specific therapeutic plans, the efficient utilization of medical resources, a reduction in medical expenses and an improvement in therapeutic efficacy. To predict trends in disease development and improve the accuracy of prognosis prediction, most researchers prefer to assess the APACHE II score at multiple time points for long-term dynamic monitoring [8]. Evaluation of the APACHE II score when patients are at an early stage of disease in the emergency room or at the time of admission to the ICU may exclude the interference of treatment in prognosis prediction. Critical neurological diseases such as stroke are frequently dynamically progressing, which suggests that dynamic monitoring of the APACHE II score may more accurately predict prognosis. Monitoring dynamic changes in the APACHE II score at multiple time points can also predict the efficacies of various therapeutic approaches and provide guidance for therapeutic plan improvement. The APACHE II score represents a comprehensive system for evaluating severe diseases, and the raw data are predominantly collected from the general ICU. Thus, this scoring system might not be applicable for specialized ICUs. However, application of the APACHE II
123
Acta Neurol Belg
scoring system in neurologic ICUs could be evaluated by increasing the sample size and further determining the efficacy and specificity of this scoring system. In summary, the actual mortality of patients in neurologic ICU increased with increasing APACHE II score. However, since the difference between the predicted and actual mortality was very large, it seems difficult to formulate hard prognostic predictions only based on the APACHE II score. Multicenter researches with large sample size are needed for further investigation. Conflict of interest of interest.
The authors declare that they have no conflict
5.
6.
7.
8.
References 1. Knaus WA, Draper EA, Wanger DP, Zimmerman JE (1985) APACHE II: a severity of disease classification system. Crit Care Med 13:818–829 2. Knaus WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE (1981) APACHE-acute physiology and chronic health evaluation: a physiologically based classification system. Crit Care Med 9:591–597 3. Park SK, Chun HJ, Kim DW, Im TH, Hong HJ, Yi HJ (2009) Acute Physiology and Chronic Health Evaluation II and Simplified Acute Physiology Score II in predicting hospital mortality of neurosurgical intensive care unit patients. J Korean Med Sci 24:420–426 4. Dalgic A, Ergu¨ngo¨r FM, Becan T, Elhan A, Okay O, Yu¨ksel BC (2009) The revised Acute Physiology and Chronic Health
123
9.
10.
11.
12.
Evaluation System (APACHE II) is more effective than the Glasgow Coma Scale for prediction of mortality in head-injured patients with systemic trauma. Ulus Travma Acil Cerrahi Derg 15:453–458 Riachy M, Sfeir F, Sleilaty G, Hage-Chahine S, Dabar G, Bazerbachi T et al (2008) Prediction of the survival and functional ability of severe stroke patients after ICU therapeutic intervention. BMC Neurol 8:24 Zali AR, Seddighi AS, Seddighi A, Ashrafi F (2012) Comparison of the acute physiology and chronic health evaluation score (APACHE) II with GCS in predicting hospital mortality of neurosurgical intensive care unit patients. Glob J Health Sci 4:179–184 Kim TK, Yoon JR (2012) Comparison of the predictive power of the LODS and APACHE II scoring systems in a neurological intensive care unit. J Int Med Res 40:777–786 Su YY, Li X, Li SJ, Luo R, Ding JP, Wang L et al (2009) Predicting hospital mortality using APACHE II scores in neurocritically ill patients: a prospective study. J Neurol 256:1427– 1433 Cho DY, Wang YC (1997) Comparison of the APACHE III, APACHE II and Glasgow Coma Scale in acute head injury for prediction of mortality and functional outcome. Intensive Care Med 23:77–84 Tan Y, Chen S, Wang L (2010) Application of the APACHE II score to analyze 73 cases of acute severe brain diseases. Chin J Integr Med Cardio Cerebrovasc Dis 8:886–887 Naved SA, Siddiqui S, Khan FH (2011) APACHE-II score correlation with mortality and length of stay in an intensive care unit. J Coll Phys Surg Pak 21:4–8 Jiang X (2000) The clinical application and significance of scoring the severity of diseases. Chin Crit Care Med 4:195–197
ORIGINAL ARTICLE
Application of the APACHE II score to assess the condition of patients with critical neurological diseases Yi Bian • Ping Zhang • Yongjie Xiong • Feng Xu • Suiqiang Zhu • Zhouping Tang Zheng Xue
•
Received: 23 June 2014 / Accepted: 25 December 2014 Ó Belgian Neurological Society 2015
Abstract The Acute Physiology And Chronic Health Evaluation II (APACHE II) scoring system has been commonly used to assess the severity of patients’ diseases in general intensive care units (ICUs). However, few studies have investigated the application of this scoring system in patients in neurologic ICUs. In this study, the APACHE II scores of 102 patients in the neurologic ICU were calculated within the first 24 h. The actual mortality and predicted mortality were obtained based on these scores and analyzed statistically. The data indicated that cerebral hemorrhage, cerebral infarction and intracranial infection accounted for the top three causes for admission to the neurologic ICU, and these conditions were associated with high APACHE II scores and high predicted mortality. Additionally, the actual mortality rate was lower than the predicted rate after effective treatment. All patients were divided into groups according to their APACHE II scores, and we found that higher APACHE II scores were associated with higher actual mortality, especially for patients whose APACHE II scores were greater than 10. The APACHE II scores of the deceased patient group were higher than those of the surviving group, and
Y. Bian and P. Zhang contributed equally to this work. Y. Bian Department of Emergency, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, People’s Republic of China P. Zhang Y. Xiong F. Xu S. Zhu Z. Tang Z. Xue (&) Department of Neurology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, People’s Republic of China e-mail: [email protected]
this difference was statistically significant. In conclusion, our study found that the APACHE II scoring system may provide valuable information for predicting patient’s condition and prognosis in neurologic ICUs. Keywords Acute physiology and chronic health evaluation II Neurologic intensive care unit Mortality Prognosis
Introduction The Acute Physiology And Chronic Health Evaluation II (APACHE II) is the most widely used and authoritative system used to evaluate critical disease condition. The system assesses disease severity based on human acute physiological alterations and the patient’s health history and age. The APACHE II scoring system is commonly used in general intensive care units (ICUs) worldwide and can be used to objectively assess a patient’s condition, predict mortality, provide a scientific basis for the development of a therapeutic plan and improve healthcare quality and the rational utilization of medical resources. However, application of the APACHE II to assess the disease condition of patients with critical neurological illness is not commonly performed. Moreover, few studies have investigated the application of this system in patients in neurologic ICUs. In the present study, the APACHE II scoring system was used to evaluate patients admitted to the neurologic ICU of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology between May of 2013 and October of 2013. We also sought to investigate the predictive value of this system in patients with critical neurological conditions.
123
Acta Neurol Belg
Materials and methods General patient information One hundred and two patients treated in the neurologic ICU of Tongji Hospital of Tongji Medical College at Huazhong University of Science and Technology between May of 2013 and October of 2013 were recruited into this study. These patients included 73 men and 29 women, and the male to female ratio was 2.5:1. The age of disease onset ranged from 16 to 84 years with an average of 52.7 ± 16.6 years. Among these patients, there were 41 cases of cerebral hemorrhage (40.2 %), 27 cases of cerebral infarction (26.5 %), 23 cases of intracranial infection (22.5 %), 4 cases of subarachnoid hemorrhage (3.9 %), 2 cases of encephalopathy syndrome (2.0 %), 2 cases of space-occupying lesions on the brain stem (2.0 %), 2 cases of myasthenia gravis (2.0 %) and 1 case of peripheral neuropathy (1.0 %, Fig. 1). Among the 102 patients, 6 patients died (5.9 %) at ICU discharge, including 3 patients who died from cerebral hemorrhage, 2 patients from cerebral infarction and 1 patient from intracranial infection. After active treatment, 67 patients (65.7 %) improved markedly, met the criteria for discharge from the ICU and were transferred to a general ward for continuous treatment. 29 patients (28.4 %) did not show significant
improvement after treatment, and due to financial burden and other factors the families of these patients requested transfer to another hospital. At the 28-day follow-up, four of the patients who were transferred to other hospitals died, two of them died from cerebral hemorrhage and the other two from cerebral infarction. So, the total 28-day mortality rate was 9.8 %. Scoring criteria According to the APACHE II scoring criteria reported by Knaus et al. [1], each patient was evaluated within 24 h after admission to the ICU and the worst score was used for the evaluation. Prior to discharge from the ICU after treatment, the patients were evaluated again. The predicted risk of mortality (R) for each patient was calculated according to the following equation: ln (R/1 - R) = -3.517 ? (APACHEIIscore 9 0.146) ? 0.603 (only for patients who had undergone operations in the emergency room) ? score of the disease causing admission to the ICU. The predicted group risk of mortality was calculated by dividing the sum of the R value of each patient by the total number of patients. Statistical analysis Statistical analysis was performed using the software SPSS12.0. Measurement data were presented as the mean ± standard deviation, and the difference between means was analyzed with the Student’s t test. Count data were analyzed using descriptive statistics and the v2 test. P \ 0.05 indicates significant difference. Ethical standards statement and patient consents The study was approved by the hospital ethics committee and was therefore performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Specific national laws were also observed. Written informed consent was obtained from all patients or their legal guardians.
Results The average APACHE II score for patients with different disease types
Fig. 1 The disease types encountered in the neurologic ICU
123
The average APACHE II score of each group of patients with different disease types at the time of admission to the ICU and at the time of discharge from the ICU are shown in Table 1.
Acta Neurol Belg Table 1 APACHE II score according to disease type Diagnosis
Cerebral hemorrhage
Cerebral infarction
Intracranial infection
Subarachnoid hemorrhage
Encephalopathy syndrome
Space-occupying lesions on brain stem
Myasthenia gravis
Peripheral neuropathy
Number of cases (%)
41 (40.2)
27 (26.5)
23 (22.5)
4 (3.9)
2 (2.0)
2 (2.0)
2 (2.0)
1 (1.0)
Score at admission to ICU
14.4 ± 5.0
13.2 ± 6.2
12.4 ± 6.1
12.5 ± 7.7
18.5 ± 0.7
16.5 ± 6.4
8.5 ± 6.4
4
Score at discharge from ICU
10.0 ± 6.1
10.9 ± 5.6
5.5 ± 4.2
15.0 ± 10.2
11.0 ± 5.7
10.5 ± 2.1
4.5 ± 2.1
0
Predicted risk of mortality (%)
47.7
18.1
16.7
42.6
27.5
24.0
9.4
4.4
Mortality of ICU discharge (%)
7.3
7.4
4.3
0
0
0
0
0
28-day mortality rate (%)
12.2
14.8
4.3
0
0
0
0
0
Table 2 Patient prognosis at ICU discharge according to the APACHE II score at the time of admission to the ICU Score at admission to ICU
0–4
5–9
10–14
15–19
20–24
25–29
Number of cases (%)
4 (3.9)
26 (25.5)
25 (24.5)
32 (31.4)
13 (12.7)
2 (2.0)
Number of deaths at ICU discharge (%)
0
0
1 (4.0)
2 (6.3)
2 (15.4)
1 (50.0)
Number of improvements (%)
4 (100.0)
21 (80.8)
19 (76.0)
19 (59.4)
4 (30.8)
0
Number transferring to other hospitals (%)
0
5 (19.2)
5 (20.0)
11 (34.4)
7 (53.8)
1 (50.0)
Average days in ICU
14.0 ± 10.1
7.6 ± 3.8
8.5 ± 4.1
9.2 ± 7.6
9.1 ± 6.1
16.0 ± 18.4
Patient prognosis according to the APACHE II score at the time of admission to the ICU The range of the APACHE II score for the total of 102 patients was 4–29 with an average score of 13.5 ± 5.8. The patients were classified according to the APACHE II score at the time of admission to the ICU. Each five-point increment in the score was classified as one group, and patient prognosis was summarized according to this classification. Table 2 shows the average length of stay in ICU and the prognosis of each group of patients at ICU discharge, which was classified into death, improvement (patients improved and were discharged from the ICU) and transferring (patients were transferred to other hospitals for continuous treatment due to financial burden and other factors). The predicted risk of mortality according to the APACHE II score at the time of admission to the ICU Based on the equation reported by Knaus et al. [1] for calculating the predicted risk of mortality, the predicted
risk of mortality for the total of 102 patients was 30.6 %. The actual mortality at ICU discharge was 5.9 % and the actual 28-day mortality rate was 9.8 %. The predicted risk of mortality was calculated for each patient group classified according to each five-point increment in the APACHE II score. The predicted risk of mortality was also compared to the actual mortality (Table 3). The actual mortality was significantly lower than the predicted risk of mortality for all groups. Comparison between the surviving group and the deceased group Comparison of the patient general data at the end point of ICU discharge between the surviving group of 96 patients and the deceased group of 6 patients revealed that (Table 4) the average age and the gender ratio had no statistically significant difference. The average APACHE II score at the time of admission to the ICU was 13.1 ± 5.5 in the surviving group and 20.0 ± 6.4 in the deceased group, and this difference was significant (P \ 0.05). The average length of stay in the ICU was 8.9 ± 6.4 days in the
123
Acta Neurol Belg Table 3 The predicted risk of mortality according to the APACHE II score at the time of admission to the ICU and the actual mortality Score at admission to ICU
0–4
5–9
10–14
15–19
20–24
25–29
Predicted risk of mortality (%)
4.4
12.6
26.3
41.3
51.0
69.0
Mortality of ICU discharge (%)
0
0
4.0
6.3
15.4
50.0
28-day mortality rate (%)
0
0
8.0
9.4
30.8
50.0
Table 4 Comparison between the surviving and deceased groups at ICU discharge Total
Surviving group
Deceased group
Age (years)
52.7 ± 16.6
52.7 ± 16.7
53.8 ± 15.7
Gender ratio (male/female)
2.5:1
2.4:1
5:1
Score at admission to ICU
13.5 ± 5.8
13.1 ± 5.5
20.0 ± 6.4**
Predicted risk of mortality (%)
30.6 ± 20.7
29.6 ± 20.0
47.8 ± 25.5*
Average days in ICU
8.9 ± 6.2
8.9 ± 6.4
9.3 ± 3.4
Total
Surviving group
Deceased group
Age (years)
52.7 ± 16.6
52.7 ± 17.0
53.5 ± 13.1
Gender ratio (male/female)
2.5:1
2.4:1
4:1
Score at admission to ICU
13.5 ± 5.8
12.9 ± 5.4
19.1 ± 6.0**
Predicted risk of mortality (%)
30.6 ± 20.7
28.9 ± 19.6
46.5 ± 25.2*
Average days in ICU
8.9 ± 6.2
9.2 ± 6.4
6.7 ± 4.4
Comparison between the surviving group and deceased group at ICU discharge * As the standard deviation was large, significance test was not performed ** P \ 0.05
Table 5 Comparison between the surviving and deceased groups at 28 days
Comparison between the surviving group and deceased group at 28 days * As the standard deviation was large, significance test was not performed ** P \ 0.05
surviving group and 9.3 ± 3.4 days in the deceased group, with no significant difference. Then we compared the data at the end point of 28 days between the surviving group of 92 patients and the deceased group of 10 patients (Table 5). The average APACHE II score at the time of admission to the ICU of the two groups also showed significant difference (P \ 0.05). Patient prognosis according to the APACHE II score at the time of discharge from the ICU The APACHE II score at the time of discharge from the ICU for the 96 survivors was 0–27 with an average of 9.2 ± 6.1. Among the survivors, 67 patients (69.8 %) demonstrated improvement at the time of discharge from the ICU, with an average APACHE II score of 7.1 ± 4.4. Of these patients, 29 (30.2 %) were transferred to other hospitals for continuous treatment, with an average score of 14.1 ± 6.7.
123
Discussion The APACHE scoring system was initially proposed by Knaus et al. [2] in 1981. The APACHE II scoring system is used to predict the disease condition and prognosis of patients with critical diseases according to acute physiological alterations and the patient’s health history and age. The APACHE II scoring system has become the most authoritative and widely used international system for evaluating patients with critical diseases, and this method has also been commonly used worldwide to assess the condition of patients with severe diseases treated in general ICUs. This scoring system can also be used in specialized ICUs. Application of the APACHE II scoring system to evaluate the condition of patients with neurological diseases such as stroke and severe traumatic brain injury has been reported [3–6]. However, few studies have investigated the application of this scoring system in patients in neurologic ICUs. Our study found that the APACHE II
Acta Neurol Belg
scoring system may provide valuable information for predicting patient’s condition and prognosis, developing a therapeutic plan and promoting the rational utilization of medical resources in neurologic ICUs. The top three diseases responsible for admission to the neurologic ICU at our hospital were cerebral hemorrhage, cerebral infarction and intracranial infection. Patients with these three diseases demonstrated high APACHE II scores and high predicted risks of mortality at the time of admission to the ICU. After active treatment, the actual mortality of these patients was significantly lower than the predicted risk of mortality, suggesting that patients suffering from diseases with high APACHE II scores and a high predicted risk of mortality could substantially benefit from treatment in a neurologic ICU. In our study, patients were classified according to the APACHE II score obtained at the time of admission to the neurologic ICU, and the patients’ prognoses were monitored. The results showed that as the APACHE II score increased, the mortality rate increased and the proportion of patients showing improvement decreased. These results are similar to previous findings [7–9], suggesting that the APACHE II score and the predicted risk of mortality have predictive value for assessing the condition and prognosis of patients with critical neurological disease. The most critical factor influencing the prognosis of patients with severe brain diseases is the severity of the brain injury. Thus, the difference in the indicator used to assess the severity of brain damage, i.e., the Glasgow Coma Scale (GCS), is most significant among the patients in the neurologic ICU. It has been reported that the APACHE II score is more accurate than the GCS score for predicting late mortality of patients with acute severe brain diseases, although the APACHE II score may be less accurate than the GCS score to predict early mortality [10]. Comparisons between the surviving group and the deceased group at two different end points (at ICU discharge and 28-day follow-up) in this study both revealed that the APACHE II score at admission to ICU in the deceased group was significantly higher than that in the surviving group. These results indicate that a high APACHE II score and high predicted risk of mortality for patients in neurologic ICUs may predict a poor prognosis. The predicted risk of mortality for the total patient group reported in the present study is similar to reports by others [10], whereas the actual mortality in this study was lower. Among the patients in the neurologic ICU, GCS might be a most significant indicator in APACHE II scoring. Patients in the medium-score groups might be in middle or deep coma at the onset of intracranial pathology, so that they got a relatively high APACHE II score when they were admitted to ICU. After active treatment such as applying mannitol, some patients came out of coma during their stay
in ICU, then the APACHE II score declined and they presented better prognosis. This might explain the difference between predicted and actual mortality. In addition, the APACHE II score is based on the prediction of mortality risk in a group and might not accurately predict the mortality risk for an individual. Thus, insufficient sample sizes may lead to biased results. Our study demonstrated that the actual mortality of patients with an APACHE II score at admission greater than 10 was significantly higher than that of patients with a score less than 10. The threshold APACHE II score for determining whether a patient should be admitted to the ICU has not been standardized. Naved et al. [11] evaluated the APACHE II scores of patients at the second or seventh day in the ICU and found that the discharge rate from the hospital in patients with scores of 3–10 was high (90 %) and that the mortality of patients with scores of 31–40 was also high (84.6 %). Jiang et al. [12] showed that an APACHE II score less than 10 was associated with a low probability of hospital death; in addition, a score of 10–20 was associated with a mortality rate of approximately 50 %, and a score greater than 20 was associated with a high mortality of 80–100 %. In our department, we tend to use an APACHE II score of 10 as the threshold value for admission to the ICU. However, the rationality for using this threshold value requires further investigation. In addition, when a patient’s condition improves after active treatment, the APACHE II score should be dynamically monitored to determine whether the patient meets the criteria for discharge from the ICU. This strategy can facilitate the development of specific therapeutic plans, the efficient utilization of medical resources, a reduction in medical expenses and an improvement in therapeutic efficacy. To predict trends in disease development and improve the accuracy of prognosis prediction, most researchers prefer to assess the APACHE II score at multiple time points for long-term dynamic monitoring [8]. Evaluation of the APACHE II score when patients are at an early stage of disease in the emergency room or at the time of admission to the ICU may exclude the interference of treatment in prognosis prediction. Critical neurological diseases such as stroke are frequently dynamically progressing, which suggests that dynamic monitoring of the APACHE II score may more accurately predict prognosis. Monitoring dynamic changes in the APACHE II score at multiple time points can also predict the efficacies of various therapeutic approaches and provide guidance for therapeutic plan improvement. The APACHE II score represents a comprehensive system for evaluating severe diseases, and the raw data are predominantly collected from the general ICU. Thus, this scoring system might not be applicable for specialized ICUs. However, application of the APACHE II
123
Acta Neurol Belg
scoring system in neurologic ICUs could be evaluated by increasing the sample size and further determining the efficacy and specificity of this scoring system. In summary, the actual mortality of patients in neurologic ICU increased with increasing APACHE II score. However, since the difference between the predicted and actual mortality was very large, it seems difficult to formulate hard prognostic predictions only based on the APACHE II score. Multicenter researches with large sample size are needed for further investigation. Conflict of interest of interest.
The authors declare that they have no conflict
5.
6.
7.
8.
References 1. Knaus WA, Draper EA, Wanger DP, Zimmerman JE (1985) APACHE II: a severity of disease classification system. Crit Care Med 13:818–829 2. Knaus WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE (1981) APACHE-acute physiology and chronic health evaluation: a physiologically based classification system. Crit Care Med 9:591–597 3. Park SK, Chun HJ, Kim DW, Im TH, Hong HJ, Yi HJ (2009) Acute Physiology and Chronic Health Evaluation II and Simplified Acute Physiology Score II in predicting hospital mortality of neurosurgical intensive care unit patients. J Korean Med Sci 24:420–426 4. Dalgic A, Ergu¨ngo¨r FM, Becan T, Elhan A, Okay O, Yu¨ksel BC (2009) The revised Acute Physiology and Chronic Health
123
9.
10.
11.
12.
Evaluation System (APACHE II) is more effective than the Glasgow Coma Scale for prediction of mortality in head-injured patients with systemic trauma. Ulus Travma Acil Cerrahi Derg 15:453–458 Riachy M, Sfeir F, Sleilaty G, Hage-Chahine S, Dabar G, Bazerbachi T et al (2008) Prediction of the survival and functional ability of severe stroke patients after ICU therapeutic intervention. BMC Neurol 8:24 Zali AR, Seddighi AS, Seddighi A, Ashrafi F (2012) Comparison of the acute physiology and chronic health evaluation score (APACHE) II with GCS in predicting hospital mortality of neurosurgical intensive care unit patients. Glob J Health Sci 4:179–184 Kim TK, Yoon JR (2012) Comparison of the predictive power of the LODS and APACHE II scoring systems in a neurological intensive care unit. J Int Med Res 40:777–786 Su YY, Li X, Li SJ, Luo R, Ding JP, Wang L et al (2009) Predicting hospital mortality using APACHE II scores in neurocritically ill patients: a prospective study. J Neurol 256:1427– 1433 Cho DY, Wang YC (1997) Comparison of the APACHE III, APACHE II and Glasgow Coma Scale in acute head injury for prediction of mortality and functional outcome. Intensive Care Med 23:77–84 Tan Y, Chen S, Wang L (2010) Application of the APACHE II score to analyze 73 cases of acute severe brain diseases. Chin J Integr Med Cardio Cerebrovasc Dis 8:886–887 Naved SA, Siddiqui S, Khan FH (2011) APACHE-II score correlation with mortality and length of stay in an intensive care unit. J Coll Phys Surg Pak 21:4–8 Jiang X (2000) The clinical application and significance of scoring the severity of diseases. Chin Crit Care Med 4:195–197
