External Validation Of The Surgical Mortality Probability Model (S-MPM) In Patients Undergoing Non-Cardiac Surgery

Abstract

Background

Preoperative risk assessment is a key issue in the process of patient preparation for surgery and the control of quality improvement in health care and certification programs. Hence, there is a need for a prognostic tool, whose usefulness can be assessed only after validation in the center other than the home one. The aim of the study was to validate the Surgical Mortality Probability Model (S-MPM) for detecting deaths and complications in patients undergoing non-cardiac surgery and to assess its suitability for various surgical disciplines.

Methods

This retrospective study involved 38,555 adult patients undergoing non-cardiac surgery in a single center in 2012–2015. The observation period concerned in-hospital mortality.

Results

In-hospital mortality for the total population was 0.89%. Mortality in the S-MPM I class amounted to 0.26%, S-MPM II 2.51%, and in the S-MPM III class 22.14%. This result was in line with those obtained by the authors. The discriminatory power for in-hospital mortality was good (area under curve (AUC) = 0.852, 95% CI: 0.834–0.869, p = 0.0000). The scale was the most accurate in general surgery (AUC = 0.89, 95% CI: 0.858–0.922) and trauma (AUC = 0.89; 95% CI: 0.87–0.915). In the logistic regression analysis, the scale showed a perfect fit/goodness of fit in the cross-validation method (v-fold cross-validation): Hosmer–Lemeshow (HL) = 7.945; p = 0.159. This result was confirmed by the traditional derivation and validation data set method (1:3; 9712 vs 22.748 cases): HL test = 3.073 (p = 0.546) in the teaching derivation data set and 10.77 (p = 0.029) in the test sample (validation data set).

Conclusion

The S-MPM scale by Glance et al has proven to be a useful tool to assess the risk of in-hospital death and can be taken into account when considering treatment indications, patient information, planning post-operative care, and quality control.

Keywords:

• S-MPM
• early mortality
• in-hospital mortality
• ASA
• RCRI

Abbreviations

S-MPM, Surgical Mortality Probability Model; AUC, area under curve; HL test., Hosmer–Lemeshow test; ASA, American Society of Anesthesiology; RCRI, Revised Cardiac Risk Index; ESC, European Society of Cardiology; ESA, European Society of Anesthesiology; ACS-NSQIP, American College of Surgeons-National Surgical Quality Improvement Program; NSQIP, National Surgical Quality Improvement Program; NSQIP SRC, National Surgical Quality Improvement Program Surgical Clinical Reviewer; HIS, Hospital Information System; PCCL, Patient Clinical Complexity Level; BUPA, British United Provident Association; OR, odds ratio; ASA, American Society of Anesthesiology; ASA PS, American Society of Anesthesiologists physical status; V-POSSUM, Vascular-Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity; Vasc-S-MPM, Vascular-Surgical Mortality Probability Model; Trauma-S-MPM, Trauma-Surgical Mortality Probability Model; SRC, Surgery Risk Calculator; DRG, Diagnosis-Related Group; IR, interval; CHF, congestive heart failure; MI, myocardial infarction; AF, atrial fibrillation; CPR, cardio-pulmonary resuscitation; PAD, peripheral artery disease; TIA, transient ischemic attack; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary Artery Bypass Graft; COPD, chronic obstructive pulmonary disease; DM, type 2 diabetes mellitus; IDDM, insulin-dependent diabetes mellitus; NIDDM, non-insulin-dependent diabetes; liver cirrhosis, cirrhosis of any Child-Pugh level; HS, length of hospitalization; HS before OP, length of stay before surgery; Total Time OP, total time of durations of operations; CI, confidence interval; ROC, receiver operation curve; MACE, major accident cardiac event; Crude OR, odds ratio in univariate regression analysis; adj. OR, odds ratio after adjustment relative to the rest of explanatory variables in the multivariate regression analysis.

Limitations

The weakness of our study is the fact that we did not have the data needed to determine the suitability of the S-MPM scale to predict death within 30 days of surgery. Also, the classification of the severity of surgical procedures was not consistent with that used in the original work. One should remember about an additional problem regarding every scale trying to predict mortality and morbidity based on pre-operative data. An escalation of the surgery that cannot be predicted raises the actual risk compared to the risk that was estimated taking into account the original plan/scope of the operation. The retrospective nature of our study from the routine data excludes the possibility of creating various types of adulteration (“bias”). This fact speaks in favor of the tested scale, which worked well in the conditions of everyday routine in the external center. An ambiguous way of qualifying surgery for three risk classes, another in Glance’s work and in our study, did not significantly reduce the suitability of this scale to assess the risk of in-hospital death.

Data Availability

The first author has the data supporting the results reported in the manuscript and will make them available at the editor’s or readers’ request.

Author Contributions

All the authors approve of the version of the manuscript to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of the work are appropriately investigated and resolved. The specific contributions (conditions 1 and 2) of the authors is as follows:

Sebastian Kazimierczak: the conception and design of the work; the acquisition, analysis, and interpretation of data for the work; drafting the work. Anita Rybicka: the acquisition, analysis, and interpretation of data for the work; drafting the work. Jochen Strauss: the design of the work; the interpretation of data for the work; revising the work critically for important intellectual content. Malgorzata Schram: the acquisition and analysis of data for the work; drafting the work. Arkadiusz Kazimierczak: the conception or design of the work; the analysis and interpretation of data for the work; drafting the work. Elżbieta Grochans: the analysis and interpretation of data for the work; revising the work critically for important intellectual content.

Disclosure

The authors declare no competing interests of any kind in this work.