Objective: To study the feasibility of a critical illness scoring system for predicting the prognosis of critically ill patients in the emergency medicine department.
Methods: 103 critically ill patients in the emergency medicine department from October 2005 to August 2006 were selected, and the Acute Physiology and Chronicity Score II (APACHEII), Acute Physiology and Chronicity Score III (APACHEIII), Simplified Acute Physiology Score II (SAPSII), and Probability of Mortality Model (MPM) scoring systems were applied to calculate the prognosis and prediction of death at the time of admission to the emergency room and at the 24-hour worst. The MPM scoring system was used to calculate the scores and predict mortality at the time of admission to the emergency room and at the 24-hour worst value, and to analyze whether the differences in prognosis at the time of admission to the emergency room and at the 24-hour worst value were statistically significant.
Results: There was no significant difference in the prognosis of the various scoring systems at the time of admission and at the 24-hour worst value (P>0.05). The scores and predicted mortality of various scoring systems differed significantly between the survival and death groups, and the higher the score at admission, the more severe the disease, and the higher the mortality.
Conclusion: Various scoring systems can be used to predict the prognosis of critically ill patients in emergency medicine. The selection of parameters scored at the time of admission had no significant effect on predicting prognosis, but the APACHEII scoring system at the time of admission was still preferred.
Keywords: critical illness; acute physiology and chronicity score II; acute physiology and chronicity score III; simplified acute physiology score II; probability of death model
, LIU Hong-jie, ZHAO Zi-ying. Emergency Department , Haidian Hospital of Beijing , Beijing 100080Abstract: Objective: to investigate the possibility Methods: all of the 103 patients who were The results of the study were presented in the following way: all of the 103 patients who were hospitalized to emergency rescue room from 10,2005 to 8,2006 were scored and their predicted morality were calculated with APACHEII, APACHEIII, SAPSII, MPMo, MPM, and MPA. MPMo.
Since Knaus proposed the critical care scoring system in 1981, scholars at home and abroad have introduced the fourth generation of scoring systems after more than 20 years of intensive research, but the most commonly used scoring systems are still the second and third generations, especially the APACHE scoring system, which is still the most widely used and authoritative scoring system. However, the timing of the selection of the parameters of the critical care scoring system is still under debate. In this paper, we selected 103 critically ill patients in emergency medicine from October 2005 to August 2006, and selected the parameters at the time of admission to the emergency room and at the worst value at 24 hours, and applied the APACHEII, APACHEIII, SAPSII, MPMo, and MPM24 scoring systems to analyze and compare the differences and accuracy of the two scoring methods in predicting the prognosis of critically ill patients in emergency medicine. The accuracy of the two scoring systems was analyzed to compare the differences and accuracy of the two scoring methods in predicting the prognosis of critically ill patients in the emergency department.
Data and Methods
All critically ill medical patients aged >14 years who were admitted to the emergency department from October 2005 to August 2006 were selected, excluding those who died within 24 hours, those who left the hospital spontaneously or abandoned treatment, those aged < span="">14 years, and those who were post-coronary artery bypass grafting, and were observed from the time they were admitted to the emergency department to the time they left the hospital, or were followed up to discharge if they were hospitalized during the observation period.
Methods
Separate data were collected at the time of patient admission to the resuscitation unit and at the 24-hour worst value, recording the general condition, primary diagnosis, and required parameters for each patient.
The scores of each scoring system and the risk of death and the required variables were recorded; the “Hundred Cities Clinical Medical Scoring Software 2005” was applied to each patient to score the worst value at the time of admission and at 24 hours and to calculate the risk of death.
A t-test was applied to compare the differences between the scores of surviving and deceased patients, and a u-test was used to predict the difference between the risk of death and the actual mortality rate.
Application of the 2 test to compare the accuracy of the four scores in evaluating the condition of critically ill patients in acute care and in predicting prognosis and mortality.
Statistical significance was found.
Results
A total of 103 patients were enrolled, including 44 males and 59 females, with a mean age of 67.4 ± 20.5 years, the oldest being 96 years and the youngest 18 years. All enrolled cases included 13 cases of severe infections, 5 cases of advanced tumors, 14 cases of cardiovascular diseases, 33 cases of respiratory diseases, 7 cases of digestive diseases, 6 cases of cerebrovascular accidents, 14 cases of poisoning, 6 cases of metabolic diseases, 1 case of asphyxia, and 1 case of drowning.
The patients were divided into survival group (n=70) and death group (n=33) according to the prognosis, and the actual mortality rate was 32%. The worst value parameters at the time of admission and 24 hours after admission were used to calculate the scores and morbidity and mortality rates of the patients in both groups using APACHEII, APACHEIII, and SAPS II scoring systems, respectively (see Tables 1 and 2), and a two-by-two comparison was performed, and it was found that there was no significant difference in the scores calculated by the various scoring systems at the time of admission and 24 hours at the worst value (P>0.05). This indicates that there was no significant relationship between predicted prognosis and the time point of value taking. The predicted mortality rate of SAPSII at the time of admission (32%) coincided with the actual morbidity and mortality rate (32%). , the
Table 1. Comparison of scoring scores for each scoring system at the two time points of value taking
24h worst value at admission 24h worst value at admission 24h worst value at admission
Full group 19.26±7.88 20.08±8.65 62.02±26.70 64.11±28.50 41.63±17.90 42.87±
Survival group 16.88±6.57 17.12±6.80★ 53.58±2296 53.36±24.45★ 37.52±14.73 37.93±14.80★
Death group 24.68±8.15 26.61±8.96▲ 80.35±25.76 84.81±27.52▲▲▲ 51.03±15.44 54.13±16.46▲▲▲
Note: ▲ Comparison between surviving and dead groups, ▲P < 0.05,< span="">▲▲P < 0.01< span="">; ★ Comparison between admission and 24-hour worst values, ★
Table 2. Comparison of morbidity and mortality rates for each scoring system at the two time periods for which values were taken
Actual mortality rate
At admission (%) 24h worst value At admission (%) 24h worst value (%) (%)
Whole group 27.72±25.14☆★ 39.45±25.39☆★ 32±25.14☆★ 35±31.45☆★ 26.01±21☆★ 29.01±21☆★ 32 32 Survival group 31.9±21.4 32.5±21.4★ 25.3±22.1 27.4±23.6★ 24.06±16.31 26.06±
Death group 49.2±58.7 53.8±27.7▲ 47±25.8 52±27.4▲ 39.14±25.01 39.14±
Note: Comparison between the survival and death groups, ▲P < 0.01< span="">; comparison between the worst value at admission and 24 hours, ★P > 0.05, comparison between expected and actual mortality, ☆
Composition of morbidity and mortality rates in different score bands for each scoring system, comparison between expected and actual mortality rates (see Tables 3, 4)
Using the APACHEII score at admission as a scoring band of 10 points, the expected and actual mortality rates for each score band are shown in Table 3. there was no significant difference between the actual and expected mortality values (p>0.05). The table shows that the higher the APACHEII score, the worse the prognosis and the higher the morbidity and mortality rate, and the morbidity and mortality rate was significantly higher for APACHEII scores >20.
Number of cases Number of deaths Expected mortality Actual mortality
Score (cases) (cases) (%) (%)
Total 103 33 27.72 32★
Table 3. relationship between expected mortality and actual mortality in each score band of APACHEII score at admission
Note: Comparison of expected mortality with actual mortality ★
Using the APACHEIII score segment at admission with 30 as a scoring band, there were 7 cases with 10-29 points and 0 deaths; 62 cases with 30-59 points and 6 deaths, with 9.7% mortality; 21 cases with 60-89 points and 16 deaths, with 76.2 mortality; and 13 cases with >90 points and 11 deaths, with 84.6% mortality. It is suggested that as the score increases, the mortality rate also increases significantly.
When the SAPS II score at admission was segmented (see Table 4), the expected mortality rate increased with the increase of the score, but there was no significant difference compared with the actual mortality rate. (P>0.05)
Table 4. relationship between SAPS II score score segments at admission and actual mortality
Number of cases Number of deaths Expected mortality Actual mortality
Score (cases) (cases) (%) (%)
Total 103 33 32 32☆
Note: ☆ Comparison of expected mortality rate and actual mortality rate
Because the MPM scoring system has no score and can only calculate the patient’s risk of morbidity and mortality, this study showed that although the MPMII0 (26.01%) at admission was not significantly different from the actual morbidity and mortality rate (32%), it was significantly lower than the actual mortality rate compared with other scoring systems.
. Discussion
There is no large-scale clinical validation of whether this is the best method of scoring the parameters of the three scoring systems, except for MPM0, which is often taken as the worst value 24 hours after admission to the ICU, and Knaus, in proposing the APACHE scoring system, has argued that it should be more accurate in determining the condition and prognosis if each parameter is taken as the value of the patient’s initial admission without therapeutic intervention [1]. And because it is difficult for emergency physicians to observe the 24-hour worst value for critical illness, there is no specific condition evaluation system for critically ill patients in emergency medicine, so if the above scoring system makes adjustments in the time of taking values, the critical illness scoring system can also be applied to critically ill patients in emergency medicine. It can be more convenient for emergency and ICU departments. In this paper, we selected the worst value at the time of admission to the emergency care unit and 24 hours to score and calculate the morbidity and mortality rate, and observed whether there was any statistically significant difference between the two scoring methods in evaluating the severity of illness and predicting the morbidity and mortality rate.
Since the MPMII scoring system does not have a score to predict the severity of the patient’s illness, this paper applied APACHEII, APACHEIII, and SAPSII to evaluate the severity of the patient’s illness, and there was no statistically significant difference between the scores of the three scoring systems at the time of admission and the worst value at 24 hours of admission, while the difference between the surviving and dead groups was significant, so the three scoring systems can be used to evaluate the severity of the patient’s illness at the time of emergency All of them can be used to evaluate the severity of the condition of patients in emergency medicine at admission. Since APACHEIII is only a score and cannot directly calculate the expected rate of death, the three scoring systems APACHEII, SAPSII, and MPMII were applied to calculate the expected rate of death at admission and at the worst value at 24 hours, and no significant differences were found in the comparison with the actual mortality rate, with SAPSII (32%) matching the actual mortality rate (32%) and MPM0 (26.01%) was slightly lower than the actual mortality rate compared to other scoring systems, which was slightly different from that reported by Yang Trail [2], and the reason for this may be related to the type of disease and treatment setting of the patients enrolled.
In Table 3, it was found that for each 1O-point increase in APACHE II score at admission, the morbidity and mortality rate increased accordingly. below 10 points of APACHEII, the morbidity and mortality rate was 0, and for APACHEI1 scores >20, the morbidity and mortality rate increased significantly. the expected mortality rate of APACHEII in the low score segment 0-10 (15.9%) was higher than the actual morbidity and mortality rate (0%), and the expected mortality in the high score segment The estimated risk of death for APACHEI1 (27.72%) was slightly lower than the actual mortality rate (32%), but the difference was not significant (P>0.05), suggesting that clinicians should pay attention to adjusting the estimated values when evaluating the condition and prognosis of patients in the actual application of this scoring system, and should make adjustments to the estimated values for patients in the higher scoring bands. The estimated value of the risk of death should be increased.
The APACHEIII score was positively correlated with the morbidity and mortality rate, and the APACHEIIIA score was used as a cut-off score of 6O, with every 3O points as a scoring band. The mortality rate at APACHEIII score << span="">60 is 0-9.7%, while the morbidity and mortality rate at APACHEIII score >60 is >70%, which is consistent with domestic and international scholarly reports [4][5]. It indicates that the APACHEIII scoring system can also be used for the assessment of acute critical illness and can be used to classify the severity of a patient’s illness with a 60-point cut-off.
The scoring system is a simplified acute physiologic score, which was developed in 1984 by Frenchman Le. Gah, TR et al. who simplified the APS scoring part of APACHE and first proposed the SAPSI, which was later updated and supplemented and then proposed the SAPS II in 1993. It has been argued that SAPS II is less accurate than APACHEII and MPM0 in predicting patients’ risk of illness and death, and that SAPS II can be used to determine the severity of a patient’s condition but not to accurately predict his or her prognosis [6], but the study in this paper shows that the application of SAPS II to predict mortality is close to the actual mortality rate, which is consistent with the study of Liang [7].
MPM0, 24 can predict mortality at the time of admission to ICU and at 24 hours, respectively, and MPM0 is the only system for evaluating in-hospital mortality immediately after admission, and this paper shows that MPM0 predicts mortality in critically ill patients in acute care with This paper shows that there is no significant difference between the MPM0 prediction of emergency critical care mortality and actual mortality, which is consistent with other literature [2], but the predicted value is slightly lower when compared with other scoring systems. Although MPM0 is convenient and quick to calculate, and excludes the influence of various therapeutic factors to the maximum extent, the most significant physiological variables in this scoring system are coma or profound rigor mortis and acute renal failure, and the probability of in-hospital death in these patients is 4.4 times higher than that in patients without coma or acute renal failure. All of these may affect the accuracy of predicting prognosis.
In conclusion, the four scoring systems selected at the time of admission to predict morbidity and mortality are suitable for critically ill patients in emergency medicine, and APACHEII is preferred, but the prognosis is better if the four scoring systems are combined to predict prognosis