Introduction

Patients undergoing neurosurgery are susceptible to rapid postoperative deterioration due to multiple factors including underlying pathologies, anesthetic effects, and surgical trauma, with severe cases potentially progressing to life-threatening conditions.1 The capacity of healthcare providers to accurately anticipate postoperative clinical trajectories is paramount for ensuring patient safety, minimizing emergent complications, and ultimately improving survival outcomes.2,3 This underscores the necessity for developing standardized, quantifiable objective assessment tools for patient evaluation.

Currently available clinical scoring systems for assessing illness severity include the Acute Physiology and Chronic Health Evaluation (APACHE), Early Warning Score (EWS), Modified Early Warning Score (MEWS), and Corrected Modified Early Warning Score (CM EWS).4 Among these, the Corrected MEWS has garnered increasing attention in clinical practice owing to its readily accessible parameters, operational simplicity, and reliable predictive validity.5,6 While predominantly utilized in emergency, transport, and resuscitation settings, the application of Corrected MEWS in specialized clinical domains remains exploratory.7–9

Notably, research specifically examining the utility of Corrected MEWS for postoperative neurosurgical assessment is limited, with no established consensus regarding optimal cutoff values in domestic clinical practice.10,11 This study therefore aims to systematically evaluate the predictive value of Corrected MEWS for clinical status and prognosis in post-neurosurgical patients.

Materials and MethodsStudy Population and Ethical Considerations

This retrospective cohort study enrolled postoperative neurosurgical patients admitted to the Department of Neurosurgery at the Army Medical Center of the PLA between December 2021 and April 2022.

Inclusion criteria:

Patients who were successfully transferred to the general ward following neurosurgical intervention.

Exclusion criteria:

ICU admission within 72 hours postoperatively Perioperative mortality Treatment discontinuation or withdrawal of care Discharge due to full recovery Incomplete clinical datasets Loss to follow-up within 90 postoperative days COVID-19-related mortality within 90 postoperative days

Following rigorous screening per the predefined criteria, 281 eligible patients were included in the final analysis, with an age range of 14–87 years (mean ± SD: 53.8 ± 13.2 years).

The study protocol received approval from the Institutional Review Board of our medical center (Ethics Approval No. 2023-116), ensuring compliance with ethical standards for human subject research.

Research Methods

Corrected MEWS scores were calculated based on seven physiological parameters recorded within 3 days post-surgery: axillary temperature, respiratory rate, heart rate, systolic blood pressure, level of consciousness (using the AVPU scale), oxygen saturation, and age (see Table 1 for scoring criteria). Each parameter was scored from 0 to 3 according to predefined thresholds, adapted from previously validated MEWS modifications.12 Patients were categorized into three risk groups based on total corrected MEWS: Group I (0–3 points), Group II (4–6 points), and Group III (≥7 points). Clinical outcomes at 90 days were determined via medical record review or follow-up interviews and classified as improved, unchanged, worsened, or deceased, based on documented functional status and survival. ROC curve analysis was conducted using 90-day mortality as the outcome variable.

Table 1 Corrected MEWS Scoring Criteria

Clinical Outcomes

Clinical outcomes at 90 days postoperatively were categorized as follows. Improved: Functional status showed improvement compared to the preoperative baseline, with no new neurological deficits. This corresponds to a Glasgow Outcome Scale (GOS) score of ≥4 or a modified Rankin Scale (mRS) score of ≤2. Unchanged: No significant clinical changes were observed, and the patient maintained a stable functional status. The GOS and mRS scores remained the same as at baseline. Deteriorated: The patient developed new or worsened neurological deficits or experienced increased functional dependence. This is reflected by a GOS score of ≤3 or an mRS score of ≥3. Death: The patient experienced all-cause mortality within 90 days following surgery.

Statistical Methods

Statistical analysis was performed using SPSS version 22.0. Continuous data were expressed as x±s, while categorical data were presented as counts or percentages. The Chi-square test was used for comparisons. Predictive probabilities were utilized to generate the ROC curve, with χ2 plotted on the x-axis and sensitivity on the y-axis. The area under the ROC curve (AUC) was calculated to assess predictive accuracy.

ResultsAssociation Between Postoperative Corrected MEWS Scores and 90-Day Clinical Outcomes in Neurosurgical Patients

A significant positive correlation was observed between elevated Corrected MEWS scores within the first 72 postoperative hours and adverse clinical outcomes (P < 0.001). Higher scores demonstrated a strong predictive value for clinical deterioration and mortality, with pairwise comparisons across score strata revealing statistically significant differences in outcomes (P < 0.001, Table 2).

Table 2 Relationship Between Corrected MEWS Scores Within 3 Days Post-Surgery and Clinical Outcomes of Patients [n (%)]

Predictive Performance of Corrected MEWS for 90-Day Mortality

ROC curve analysis was performed using 90-day mortality as the primary endpoint. The analysis yielded an AUC of 0.944, indicating excellent predictive accuracy. The optimal cutoff value was determined to be 4.5, demonstrating high sensitivity (92.9%) and specificity (82.0%). The maximum Youden’s index of 0.749 further confirmed the robust discriminative ability of this threshold (P < 0.001, Figure 1).

Figure 1 ROC curve of MEWS score correction within 3 days post-neurosurgery predicting 90-day outcomes in neurosurgical patients.

These findings demonstrate that the Corrected MEWS scoring system serves as a highly significant prognostic tool (AUC > 0.9) for predicting 90-day mortality in post-neurosurgical patients.

The cutoff value of 4.5 for corrected MEWS demonstrates excellent discrimination for 90-day mortality. In clinical practice, this threshold could be considered as an early warning trigger for intensified monitoring or intervention post-neurosurgery, though prospective validation in broader cohorts is needed.

Discussion

Patient safety remains a paramount priority in healthcare delivery worldwide. Clinical evidence demonstrates that adverse events are frequently preceded by measurable physiological derangements that serve as critical warning signs.13 This recognition has driven the global medical community to develop and implement various early warning scoring systems, aiming to facilitate timely, efficient, and precise identification of at-risk patients.

The evolution of these scoring systems began with Morgan et al’s pioneering work on the EWS.10 While this system offered advantages in terms of rapid parameter acquisition and operational simplicity, its clinical implementation demonstrated significant inter-hospital variability. Subsequent refinements by Subbe led to the development of the Modified Early Warning Score (MEWS), which gained widespread clinical adoption due to its enhanced user-friendliness.14–16 However, comparative analyses have revealed limitations in MEWS’s diagnostic performance. Specifically, it demonstrates inferior sensitivity compared to established intensive care scoring systems (APACHE II and SAPS II) in detecting potentially critical illness.17,18

Furthermore, the system’s parameter selection lacks comprehensive coverage of critical physiological indicators and suffers from insufficient specificity for certain clinical scenarios.19 To improve specificity and sensitivity, Tang et al5 incorporated oxygen saturation into MEWS, forming the Corrected MEWS, which is now applied in emergency and critical care. Recent multicenter studies have extensively validated early warning systems in respiratory emergencies. A Spanish prospective study of 902 emergency department patients with acute respiratory distress found that while Modified Sequential Organ Failure Assessment (mSOFA) showed the best predictive performance for 2-day mortality (AUROC 0.911), MEWS still maintained clinical utility with an AUROC of 0.82–0.87 across different respiratory conditions.19 This suggests that although specialized scores may outperform MEWS in specific scenarios, its balanced performance across diverse conditions supports its use as a general screening tool. In diabetic emergencies, comparative studies suggest more nuanced performance differences among scoring systems. A retrospective analysis of diabetic patients presenting with sepsis found that while SOFA score showed superior predictive value for septic shock onset (AUROC 0.866 at 48 hours), MEWS still demonstrated clinically useful performance (AUROC 0.79) and maintained advantages in rapid assessment due to its simpler parameters.4 This highlights the importance of selecting scoring systems based on both predictive accuracy and clinical practicality. However, the system’s performance varies significantly by pathology. In toxic alcohol poisoning cases, a seven-year retrospective study demonstrated that while APACHE II and SOFA scores showed superior predictive value for adverse outcomes (AUC 0.97 and 0.94 respectively), MEWS exhibited limited discriminative ability (AUC=0.68) compared to these specialized scoring systems.17 This underscores the importance of using pathology-specific scoring tools for certain toxicological emergencies. Our findings align with this refinement: Corrected MEWS scores calculated within three days post-neurosurgery showed strong correlation with 90-day clinical outcomes. Specifically, patients with scores ≥7 had a mortality rate of 33.4%, compared to 0% in those with scores ≤3, demonstrating a clear predictive gradient.

Moreover, the ROC curve analysis yielded an AUC of 0.944, with a cutoff of 4.5, 92.9% sensitivity, and 82.0% specificity, supporting its strong prognostic accuracy. These findings are consistent with previous studies,5,9,20 suggesting that Corrected MEWS is a robust early warning tool for postoperative neurosurgical patients.

Validation of MEWS in Uganda and Cape Town supports its global feasibility: in Uganda, MEWS ≥ 5 predicted 7-day mortality with OR = 5.82 (95% CI 2.42–13.99) and a Cape Town consensus refined cutoffs for resource-limited wards.21 In Cape Town, experts used a Delphi process to tailor a MEWS chart for general wards. They reached ≥ 70% consensus on seven vital sign thresholds—including respiratory and heart rates, systolic BP, temperature, and consciousness/SpO2 cutoffs.22

Compared to APACHE II or NEWS, Corrected MEWS offers advantages in terms of ease of calculation, faster parameter acquisition, and minimal data requirements, making it well-suited for resource-limited or high-turnover settings such as neurosurgical wards or emergency units. Age-specific limitations warrant particular attention. A comprehensive systematic review of 21 studies (n=11,183) evaluating early warning systems found that while these scores showed strong predictive value for 48-hour mortality (AUROC 0.88–0.93) and cardiac arrest (AUROC 0.74–0.86) in adult populations, the evidence regarding their impact on clinical outcomes and resource utilization remains inconclusive due to methodological limitations in most studies.23 This suggests that while MEWS demonstrates good predictive validity, its clinical implementation requires careful consideration of patient populations and local care protocols. In emergency department triage settings, a multicenter prospective study of 798 patients found that while MEWS showed utility in risk stratification, clinical judgment by physicians demonstrated superior accuracy for predicting hospital admission (sensitivity 91.9% vs 44.0% for MEWS ≥3, p<0.001), suggesting that scoring systems should complement rather than replace clinician assessment.24 However, APACHE II provides greater granularity for ICU-level decision-making, and NEWS may offer better discrimination in respiratory or septic cases. Thus, the choice of scoring system should be tailored to the clinical context.

Study Limitations and Future Directions

This study has limitations, including its single-center design, limited sample size, and lack of a comparison group using alternative scoring systems. These factors may limit generalizability. To address this, we propose a prospective, multicenter study comparing Corrected MEWS, NEWS, and APACHE II in neurosurgical populations. Such research could assess not only predictive performance but also operational feasibility (eg, time to calculate, training needs, inter-rater reliability) to inform clinical adoption.

Conclusion

In summary, the corrected MEWS demonstrated strong predictive value for 90-day prognosis in neurosurgical patients, with an AUC of 0.944, sensitivity of 92.9%, and specificity of 82.0% at the optimal cutoff of 4.5. Clinical outcomes were significantly stratified by MEWS scores: Group I (score 0–3) had a 0% mortality and only 1.1% deterioration rate, while Group III (score ≥7) exhibited a mortality rate of 33.4% and a deterioration rate of 28.6%. These findings suggest that corrected MEWS can serve as an effective early warning tool for identifying clinical deterioration and guiding timely interventions in the neurosurgical postoperative setting.

However, this study has limitations, including a relatively small sample size, its single-center design, and the lack of a control or comparator group. Further multicenter prospective studies are needed to validate these findings and explore the generalizability of corrected MEWS across broader clinical contexts.

Data Sharing Statement

All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.

Ethics Approval and Consent to Participate

This study was conducted with approval from the Ethics Committee of Chinese People’s Liberation Army Special Medical Center [Medical research ethics (2023)-116]. This study was conducted in accordance with the declaration of Helsinki. Written informed consent was obtained from all participants.

Acknowledgments

We would like to acknowledge the hard and dedicated work of all the staff that implemented the intervention and evaluation components of the study.

Funding

2024 Chongqing Nan’an District Science and Health Joint Medical Research Project (Key Project), funding number: 2024-06.

Disclosure

The authors declare that they have no competing interests in this work.

References

1. Alam N, Hobbelink EL, van Tienhoven AJ, et al. The impact of the use of the early warning score (EWS) on patient outcomes: a systematic review. Resuscitation. 2014;85(5):587–594. doi:10.1016/j.resuscitation.2014.01.013

2. Smith GB, Redfern OC, Pimentel MA, et al. The National Early Warning Score 2 (NEWS2). Clin Med. 2019;19(3):260. PMID: 31092526; PMCID: PMC6542226. doi:10.7861/clinmedicine.19-3-260

3. Zhang YJ, Liu XY, Xu WX, Yang YP. Reevaluation of prognostic and severity indicators for COVID-19 patients in the emergency department. Ann Med. 2024;56(1):2417178. PMID: 39435611; PMCID: PMC11497573. doi:10.1080/07853890.2024.2417178

4. Liengswangwong W, Siriwannabhorn R, Leela-Amornsin S, et al. Comparison of Modified Early Warning Score (MEWS), Simplified Acute Physiology Score II (SAPS II), Sequential Organ Failure Assessment (SOFA), and Acute Physiology and Chronic Health Evaluation II (APACHE II) for early prediction of septic shock in diabetic patients in Emergency Departments. BMC Emerg Med. 2024;24(1):161. PMID: 39232644; PMCID: PMC11376032. doi:10.1186/s12873-024-01078-8

5. Tang WJ, Zhang SQ, Huang SW, et al. A comparative study on the application of modified early warning score system and modified early warning score system in pre-hospital emergency care. Chin Gen Med. 2011;14(05):526–529.

6. Yu Z, Xu F, Chen D. Predictive value of Modified Early Warning Score (MEWS) and Revised Trauma Score (RTS) for the short-term prognosis of emergency trauma patients: a retrospective study. BMJ Open. 2021;11(3):e041882. doi:10.1136/bmjopen-2020-041882

7. Hincapié-Osorno C, van Wijk RJ, Postma DF, et al. Validation of MEWS, NEWS, NEWS-2 and qSOFA for different infection foci at the emergency department, the acutelines cohort. Eur J Clin Microbiol Infect Dis. 2024;43:2441–2452. PMID: 39414696. doi:10.1007/s10096-024-04961-1

8. Veldhuis LI, van der Weide L, Nanayakkara P, Ludikhuize J. The accuracy of predicting hospital admission by emergency medical service and emergency department personnel compared to the prehospital MEWS: a prospective multicenter study. BMC Emerg Med. 2024;24(1):111. PMID: 38982356; PMCID: PMC11234550. doi:10.1186/s12873-024-01031-9

9. Butkiewicz S, Zaczyński A, Pańkowski I, et al. Retrospective study to identify risk factors for severe disease and mortality using the modified early warning score in 5127 patients with COVID-19 admitted to an emergency department in poland between March 2020 and April 2021. Med Sci Monit. 2022;28:e938647. PMID: 36495006; PMCID: PMC9749323. doi:10.12659/MSM.938647

10. Morgan R, Williams F, Wright M. An early warning scoring system for detecting developing critical illness. Clin Interns Care. 1997;8(1):100–101.

11. Ludikhuize J, Borgert M, Binnekade J, et al. Standardized measurement of the modified early warning score results in enhanced implementation of a rapid response system: a quasi-experimental study. Resuscitation. 2014;85(5):676–682. doi:10.1016/j.resuscitation.2014.02.009

12. Rohith N, Narayanaswamy S, Hegde S. Clinical study of a new modified early warning scoring system for rapidly evaluating shock in adults. Cureus. 2023;15(4):38224.

13. van Rossum MC, Bekhuis REM, Wang Y. Early warning scores to support continuous wireless vital sign monitoring for complication prediction in patients on surgical wards: retrospective observational study. JMIR Perioper Med. 2023;6:44483. doi:10.2196/44483

14. Subbe CP, Kruger M, Rutherford P, Gemmel L. Validation of a modified Early Warning Score in medical admissions. QJM. 2001;94(10):521–526. doi:10.1093/qjmed/94.10.521

15. Bibi M, Chaker K, Ouanes Y, et al. Comparison of prognosis of five scoring systems in emphysematous pyelonephritis patients requiring intensive care. Int Urol Nephrol. 2023;55(12):3045–3050. PMID: 37556105; PMCID: PMC10611593. doi:10.1007/s11255-023-03733-8

16. Wu PH, Hung SK, Ko CA, et al. Performance of six clinical physiological scoring systems in predicting in-hospital mortality in elderly and very elderly patients with acute upper gastrointestinal bleeding in emergency department. Medicina. 2023;59(3):556. PMID: 36984556; PMCID: PMC10057917. doi:10.3390/medicina59030556

17. Sharif AF, Shaheen RS, Alsubaie DS, et al. Performance of several clinical scoring systems as predictors of adverse outcomes in acute exposure to toxic alcohols. Toxicol Res. 2024;13(3):tfae069. PMID: 38716411; PMCID: PMC11070727. doi:10.1093/toxres/tfae069

18. Su X, Zhang H, Yuan W, et al. Comparison of four early warning scores in predicting the prognosis of critically ill patients in secondary hospitals. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2023;35(10):1093–1098. Chinese. PMID: 37873716. doi:10.3760/cma.j.cn121430-20230614-00441

19. Villamor MAC, Alonso-Sanz M, López-Izquierdo R. Comparison of eight prehospital early warning scores in life-threatening acute respiratory distress: a prospective, observational, multicentre, ambulance-based, external validation study. Lancet Digital Health. 2024;6(3):166–175. doi:10.1016/S2589-7500(23)00243-1

20. Huang JL, Tang FR, Xie YY, et al. Research on mews correction in emergency pre-examination and trial diagnosis with information technology. Acta Med Mediterr. 2021;37(1):663.

21. Kruisselbrink R, Kwizera A, Crowther M, et al. Modified Early Warning Score (MEWS) identifies critical illness among ward patients in a resource restricted setting in kampala, uganda: a prospective observational study. PLoS One. 2016;11(3):e0151408. doi:10.1371/journal.pone.0151408

22. Kyriacos U, Jelsma J, James M, et al. Monitoring vital signs: development of a modified early warning scoring (MEWS) system for general wards in a developing country. PLoS One. 2014;9(1):e87073. doi:10.1371/journal.pone.0087073

23. Smith ME, Chiovaro JC, O’Neil M, et al. Early warning system scores for clinical deterioration in hospitalized patients: a systematic review. Ann Am Thorac Soc. 2014;11(9):1454–1465. doi:10.1513/AnnalsATS.201403-102OC

24. Veldhuis LI, van der Weide L, Nanayakkara P, et al. Real-world performance of early warning scores at emergency department triage: a prospective multinational cohort study. Eur J Emerg Med. 2023;30(4):258–266. doi:10.1097/MEJ.0000000000001034