Jinhong Yang1，Lian fu Wang2，Li song Huang1，Yonghui Wang1
1.Department of Emergency, Aerospace Center Hospital, Haidian, Beijing, China100049
2.Inner Mongolia Aerospace Hospital, Hohhot City, Inner Mongolia Autonomous Region, China，010010
Communication:Lian fu Wang:wwin1977@163.com
[Abstract] Objective: To evaluate the value and efficacy of Modified Early Warning Score (MEWS) in predicting ICU admission for non-traumatic acute abdominal pain patients in emergency departments. Methods: A retrospective study was conducted with 126 non-traumatic acute abdominal pain patients admitted to emergency departments between June 2023 and February 2025, divided into admission and non-admission groups based on ICU admission status. Comparative analysis was performed on basic vital signs, biochemical indicators, consciousness status, Acute Physiological and Clinical Assessment of the Intensive Care Unit (APACHEII score), and MEWS scores between groups. Multivariate logistic regression was conducted for statistically significant differences to identify independent risk factors for ICU admission. ROC curves were used to assess predictive efficacy of different risk factors, followed by further analysis of patient outcomes under varying MEWS scores. Results: Both univariate and multivariate logistic regression analyses demonstrated that oxygen therapy requirement, APACHEII score, MEWS score, and NEWS score were significant predictors of ICU admission (P<0.05). Among these risk factors, MEWS score showed higher predictive sensitivity and specificity than APACHEII score, followed by NEWS score, then oxygen therapy requirement (P<0.05). No statistically significant differences were observed in observation status, specialty ward placement, or mortality rates across different MEWS scores (P>0.05).5). There were statistically significant differences in discharge rates and ICU admission rates under different MEWS scores (P<0.05). Conclusion: A relatively high proportion of non-traumatic acute abdominal pain patients in the emergency department require ICU admission, which is influenced by multiple factors. The MEWS score can improve the predictive efficacy for ICU admission in such cases, guiding clinical diagnosis and treatment.
[Keywords] Emergency department; non-traumatic acute abdominal pain; ICU admission; improved early warning score; risk factors; predictive efficacy
There are many causes of non-traumatic acute abdominal pain, including acute pancreatitis, aortic dissection, intestinal obstruction and visceral perforation, etc., and the clinical manifestations are often nonspecific[1]In elderly patients, the decreased compensatory capacity may mask the disease, while in young patients, the high tolerance may lead to the neglect of early symptoms, resulting in the traditional triage tools can not accurately identify high-risk patients[2]Although the APACHE II scoring system can assess the severity of non-traumatic acute abdominal pain in emergency departments, it requires multiple laboratory tests and complex calculations, which is not suitable for rapid triage in emergency departments.[3]The improved early warning score (MEWS) requires only five physiological parameters: heart rate, systolic blood pressure, consciousness state, body temperature and respiratory rate. It can be assessed within 10 minutes at the bedside, which is more in line with the working mode of emergency work[4]At present, MEWS is widely used in emergency care, pre-hospital care, ICU and specialized wards. In emergency patients with closed abdominal injury, MEWS can quickly assess the patients condition and guide surgical decisions, providing a basis for verification in non-traumatic acute abdominal pain patients in emergency departments.[5]This study aims to explore the value and efficacy of MEWS in predicting the need for ICU admission in patients with non-traumatic acute abdominal pain in the emergency department. The report is as follows.
1 Data and methods
1.1 General information
This retrospective study included 126 non-traumatic acute abdominal pain patients from the emergency department between June 2023 and February 2025. Patients were divided into ICU admission group (n=81) and non-admission group (n=45) based on their ICU admission status. Inclusion criteria: (1) Meet the diagnostic and treatment guidelines for acute abdominal pain as outlined in “Acute Abdominal Pain Diagnosis and Treatment”.[6]Diagnostic criteria for acute abdominal pain: (1) Age ≥18 years with acute abdominal pain presenting for medical consultation; (2) No history of trauma, cooperation with biochemical testing, and complete clinical documentation. Exclusion criteria: (1) Patients with psychiatric disorders requiring discharge within 24 hours of hospital admission or refusing treatment; (2) Those with multiple hospitalizations (≥2 admissions) or acute renal failure; (3) Individuals with previously diagnosed malignancies or autoimmune diseases.
1.2 Methodology
(1) Basic vital signs. Collect data on gender, age, respiratory rate, heart rate, body temperature, systolic blood pressure, diastolic blood pressure, mean arterial pressure (MAP), and blood oxygen saturation from both groups; (2) Relevant biochemical indicators. Use an automated biochemical analyzer to measure white blood cell count, serum creatinine, total bilirubin, and platelet count; (3) Scale scoring. ① Acute Physiological Status and Chronic Health Assessment II (APACHEII score) consists of three components: Acute Physiological Status Score (APS, 12 physiological indicators, 48 points), Age Score (6 points), and Chronic Health Status Score (CPS, 17 points). The total score ranges from 0 to 71 points, where higher scores indicate more severe condition, poorer prognosis, and higher mortality rates.[7]The ②MEWS scale mainly includes heart rate, systolic blood pressure, respiratory rate, body temperature and consciousness state. Each parameter is 3 points, totaling 15 points. Among them, mild disease is 0-4 points, severe disease is 5-8 points, and critical disease is ≥9 points[8]③ National Early Warning Score (NEWS). Includes respiratory rate, blood oxygen saturation, body temperature, systolic blood pressure, pulse and consciousness. Each item is worth 3 points, and the lower the score, the better.[9];(5) Factors affecting the admission of non-traumatic acute abdominal pain patients to ICU in the emergency department. The independent risk factors for the admission of non-traumatic acute abdominal pain patients to ICU in the emergency department were screened by multivariate Logistic regression analysis of statistically significant indicators.
1.3. Statistical analysis
Processed with SPSS 28.0 software, the count data row χ2The test was conducted using n (%) as the indicator, while the t-test for measurement data rows used () as the indicator. A single-factor and multi-factor analysis were performed to examine potential factors affecting ICU admission for non-traumatic acute abdominal pain patients in the emergency department. The receiver-operating characteristic (ROC) curve was utilized to evaluate predictive efficacy of different risk factors. Further analysis compared patient retention and prognosis under varying MEWS scores, with P<0.05 indicating statistically significant differences.
2 Results
2.1 Analysis of single factors affecting the admission of non-traumatic acute abdominal pain patients to ICU in emergency department
The univariate analysis demonstrated no statistically significant associations between ICU admission rates for non-traumatic acute abdominal pain patients in the emergency department and gender, age, respiratory rate, heart rate, body temperature, systolic blood pressure, diastolic blood pressure, mean arterial pressure (MAP), blood oxygen saturation, white blood cell count, serum creatinine, total bilirubin, or platelet count (P> 0.05). However, statistically significant differences were observed between these factors and the presence of oxygen therapy, APACHEII score, MEWS score, and NEWS score (P <0.05), as shown in Table 1.
Table 1 Single factor analysis of ICU admission for non-traumatic acute abdominal pain patients in the emergency department
| General information | Number of cases | Inpatient group (n=81 cases) and | Non-resident group (n = 45) | χ2 | P price | |
| sex | man | 71 | 45（55.56） | 26（57.78） | 0.058 | 0.810 |
| woman | 55 | 36（44.44） | 19（42.22） | |||
| Whether oxygen is administered | yes | 64 | 56（69.14） | 8（17.78） | 32.313 | 0.000 |
| deny | 62 | 25（30.86） | 37（82.22） | |||
| Age (years) | 68.95±5.61 | 70.25±5.69 | 1.235 | 0.220 | ||
| Breathing rate (times per minute) | 21.50±3.24 | 22.08±3.28 | 0.955 | 0.342 | ||
| Heart rate (beats/min) | 90.49±5.61 | 91.32±5.66 | 0.791 | 0.431 | ||
| temperature （℃） | 36.45±2.14 | 36.52±2.17 | 0.174 | 0.862 | ||
| Systolic pressure (mmHg) | 130.69±7.93 | 129.48±7.90 | 0.823 | 0.413 | ||
| Systolic pressure (mmHg) | 76.78±6.81 | 77.11±6.86 | 0.259 | 0.796 | ||
| MAP（mmHg） | 93.24±5.34 | 94.11±5.38 | 0.872 | 0.386 | ||
| degree of blood oxygen saturation （%） | 92.51±5.71 | 93.24±5.74 | 0.685 | 0.495 | ||
| White blood cell count (×109/L） | 8.78±0.95 | 8.85±0.99 | 0.386 | 0.701 | ||
| Blood creatinine (μmol/L) | 175.36±15.63 | 176.11±15.69 | 0.258 | 0.797 | ||
| Total bilirubin (μmol/L) | 35.59±5.71 | 35.71±5.76 | 0.112 | 0.911 | ||
| Platelets (×109/L） | 225.28±32.51 | 226.72±32.56 | 0.238 | 0.812 | ||
| APACHEII score | 18.51±2.14 | 13.25±1.95 | 14.006 | 0.000 | ||
| MEWS score (points) | 7.45±1.14 | 4.73±0.86 | 15.093 | 0.000 | ||
| NEWS score (out of 10) | 10.59±1.69 | 8.32±1.13 | 8.999 | 0.000 | ||
2.2 Analysis of multiple factors affecting the admission of non-traumatic acute abdominal pain patients to ICU in emergency department
The results of multivariate Logistic regression showed that oxygen intake, APACHEII score, MEWS score and NEWS score were the influencing factors for ICU admission in non-traumatic acute abdominal pain patients in emergency department (P<0.05), as shown in Table 2.
Table 2 Analysis of multiple factors affecting the admission of non-traumatic acute abdominal pain patients to ICU in emergency department
| argument | β | S.E. | wald χ2 | P | O.R. | 95%CI | |
| lower limit | superior limit | ||||||
| Whether oxygen is administered | 4.280 | 2.039 | 4.405 | 0.036 | 72.205 | 51.327 | 94.271 |
| APACHEII score | 4.482 | 3.289 | 1.857 | 0.173 | 88.374 | 50.140 | 102.631 |
| MEWS grade | 4.341 | 4.110 | 6.027 | 0.014 | 10.563 | 6.496 | 15.473 |
| NEWS grade | 6.231 | 2.830 | 4.847 | 0.028 | 6.792 | 4.986 | 8.965 |
2.3 Predictive efficacy of different risk factors for ICU admission in non-traumatic acute abdominal pain in emergency department
The predictive sensitivity and specificity of different risk factors in ICU admission for non-traumatic acute abdominal pain in emergency department were MEWS score>APACHEII score>NEWS score>oxygen inhalation (P<0.05), as shown in Table 3 and Figure 1.
Table 3 Predictive efficacy of different risk factors for ICU admission in non-traumatic acute abdominal pain in emergency department
| Test variables | AUC | cutoff value | standard error | P price | 95% confidence interval | sensitivity | specificity | |
| lower limit | superior limit | |||||||
| Whether oxygen is administered | 0.733 | yes | 0.012 | 0.000 | 0.757 | 0.803 | 78.34 | 70.12 |
| APACHEII score | 0.783 | 16.49 minutes | 0.014 | 0.000 | 0.748 | 0.822 | 79.13 | 71.89 |
| MEWS grade | 0.872 | 6.00 points | 0.023 | 0.000 | 0.895 | 0.935 | 91.49 | 89.46 |
| NEWS grade | 0.753 | 9.00 minutes | 0.019 | 0.000 | 0.773 | 0.852 | 79.48 | 74.41 |
Figure 1 ROC curves for different risk factors in ICU admission for non-traumatic acute abdominal pain in emergency department
2.4 MEWS score and retention of non-traumatic acute abdominal pain patients in emergency department
There was no statistical difference in Mews score under different scores, observation ward and mortality (P>0.05); there was a statistical difference in Mews score under different scores, discharge rate and ICU admission rate (P<0.05), as shown in Table 3.
Table 3 Relationship between MEWS score and retention of non-traumatic acute abdominal pain patients in the emergency department [n(%)]
| MEWS grade | Number of cases | Discharge | Observation | Specialist wards | ICU | die |
| 0-4 points | 57 | 17（29.82） | 15（26.32） | 10（17.54） | 15（26.32） | 0（0.00） |
| 5-8 points | 51 | 5（9.80） | 7（13.73） | 13（25.49） | 26（50.98） | 0（0.00） |
| ≥9 points | 18 | 0（0.00） | 0（0.00） | 2（11.11） | 15（83.33） | 1（5.56） |
| / | 8.546 | 5.528 | 1.405 | 7.002 | 2.640 | |
| P | / | 0.014 | 0.063 | 0.495 | 0.030 | 0.267 |
- Discussion
Both APACHEII score and NEWS score are commonly used predictors for non-traumatic acute abdominal pain patients admitted to ICU in emergency department[10]However, the APACHEII scoring system requires collection of patients physiological parameters and biochemical indicators at admission, which is time-consuming and costly. This evaluation method necessitates multiple biochemical tests with high technical requirements, limiting its clinical application. While the NEWS scoring system demonstrates significant advantages in mortality risk assessment, it has limitations for specific disease categories. For instance, it cannot predict cerebral dysfunction in patients, exhibits low sensitivity, and lacks specificity.[11]The Mews score is simple and easy to operate, can identify potentially critically ill patients early, provide objective basis for medical decision-making, improve the accuracy and efficiency of emergency triage, and help reduce the occurrence of unexpected events in predicting whether adults with non-traumatic acute abdominal pain in emergency department need to be admitted to ICU.[12]。
The results of univariate and multivariate logistic regression in this study indicate that oxygen therapy use, APACHEII score, MEWS score, and NEWS score are significant predictors for ICU admission in non-traumatic acute abdominal pain patients (P<0.05). These findings demonstrate that a substantial proportion of such patients require ICU care, influenced by multiple factors, with the MEWS score showing enhanced predictive efficacy. The MEWS scoring system—comprising temperature, respiratory rate, systolic blood pressure, heart rate, and level of consciousness—is composed of clinically common physiological parameters. These easily measurable indicators can be assessed at bedside within short timeframes without requiring specialized equipment, effectively detecting deterioration through quantitative monitoring. Previous research confirms that MEWS scores ≥5 indicate increased risk of patient deterioration, necessitating immediate intervention to achieve “early detection, early prevention, and early treatment”.[13]At the same time, MEWS score provides a quantitative reference for whether a patient needs to be admitted to ICU by objectively scoring the patients condition and reducing the error of subjective judgment of medical staff.[14]In this study, different risk factors showed varying predictive sensitivity and specificity in ICU admission for non-traumatic acute abdominal pain patients. The ranking of prognostic scores was: MEWS> APACHEII> NEWS> oxygen therapy (P<0.05). Statistical differences were observed in discharge rates and ICU admission rates across different MEWS scores (P<0.05). These results indicate that MEWS scoring demonstrates high predictive accuracy for ICU admission in non-traumatic acute abdominal pain patients, with higher scores correlating to increased ICU admission rates. The analysis reveals that MEWS scoring significantly improves triage quality and reduces misdiagnosis rates during emergency department evaluations. By dynamically monitoring physiological parameter changes, it captures early warning signs like rapid breathing and altered mental status 6-24 hours before disease progression, providing critical evidence for timely ICU admission. Additionally, the MEWS scoring system establishes standardized “doctor call criteria” and streamlines medical communication protocols, ensuring prompt initiation of appropriate treatment plans.[15]。
In conclusion, a high proportion of non-traumatic acute abdominal pain patients in the emergency department need to be admitted to ICU, which is affected by many factors. However, MEWS score can improve the prediction efficacy of non-traumatic acute abdominal pain patients in the emergency department need to be admitted to ICU and guide clinical diagnosis and treatment.
reference documentation
[1]Liang X .Application and Effectiveness of Modified Early Warning Score (MEWS) in Emergency Triage[J].Journal of Clinical and Nursing Research，2024，8（12）:337-342.
[2]Papadomanolakis-Pakis，Nicholas，Munch P V，et al.Prognostic clinical prediction models for acute post‐surgical pain in adults: a systematic review[J].Anaesthesia, 2024，79（12）：1335-1347.
[3]Octavius G S，Muljono M P，Budiputri C L.Serum pentraxin-3 in diagnosing acute appendicitis in patients with acute abdominal pain: A systematic review and meta-analysis of diagnostic test accuracy[J].Surgery，2023，173（5）:1122-1128.
[4]Tessitore A，Zupin L，Celsi F，et al.Ketorolac use in the emergency department in children: a systematic review and meta-analysis[J].European Journal of Pediatrics，2025，184（5）：296.
[5]Farooq R，Perin G，Balasubramanian S P.Inefficient admissions for abdominal pain under an acute general surgical unit[J].World Journal of Surgery，2024，48（3）：2401-2408.
[6] Sun Changyi and Zhang Guangwu. Diagnosis and Treatment of Acute Abdominal Pain [M]. Golden Shield Publishing House, 2001, pp. 45-53.
[7]Zhao Y，Xiao X.Efficacy of ultrasound-guided stellate ganglion block in relieving acute postoperative pain: a systematic review and meta-analysis[J].Journal of International Medical Research，2024，52（5）：3000605241252237.
[8]Patel P S，Booth M.Synchronous acute internal hernia and marginal ulceration with bleeding causing acute abdominal pain[J].ANZ Journal of Surgery，2025，95（3）：594.
[9]Stieler M，Carter G，Pockney CP.Somatic symptom severity, depression and anxiety associations with pancreatitis and undifferentiated abdominal pain in surgical inpatients[J].ANZ journal of surgery，2024，94（4）:634-639.
[10]Wu Q，Zhou Y，Sun S，et al.Clinical analysis of acute postoperative pain after total laparoscopic hysterectomy for adenomyosis and uterine fibroids-a prospective observational study[J].Annals of Medicine，2023，55（2）:2281510.
[11]Jnsson L R，Foss N，Orbk J，et al.Early intensive mobilization after acute high-risk abdominal surgery: a nonrandomized prospective feasibility trial[J].Canadian Journal of Surgery，2023，66：E236-E245.
[12]Wang T，Xiao P，Xue J，et al.Abdominal pain in an adolescent girl with acute phlegmonous gastritis: a case report[J].Journal of International Medical Research, 2023，51（11）：3000605231202136.
[13] Chai C，Peng S Z，Zhang R，et al.Advancing Emergency Department Triage Prediction With Machine Learning to Optimize Triage for Abdominal Pain Surgery Patients[J].Surgical Innovation，2024，31（6）：583-597.
[14]Liao C Y，Allen C W，Hill C，et al.Man With Acute Abdominal Pain[J].Annals of Emergency Medicine，2025，85（3）:276-277.
[15]Jane A，Kazemi Y，Sehgal A.Acute Abdominal Pain in a 15-Year-Old Male Patient:[J].Clinical Pediatrics，2024，63（5）:725-728.