Brazilian Journal of Anesthesiology 2022;72(3): 313−315

EDITORIAL

How to identify a high-risk surgical patient?

The number of surgical procedures continues to grow glob- congestive heart failure, cerebrovascular disease, preopera-
ally and there is a clear need to increase the availability of tive treatment with insulin, and preoperative serum creati-
safe, timely, and affordable surgery, especially in con- nine above 2 mg.dL 1.6 Unfortunately, it performs poorly
strained resources scenarios.1 As overall life expectancy when predicting cardiac events following vascular surgeries
continues to increase, it is more common to offer complex or all-cause mortality following noncardiac surgeries.6 The
procedures to patients displaying advanced age and serious ACS-NSQIP surgical risk calculator consists of 20 patient-spe-
comorbidities, with inevitable increases in morbidity and cific variables including ASA-PS, patient-reported functional
mortality after surgery despite the most recent technologi- capacity, and the planned surgery with over 1500 current
cal advances in anesthetic care. Although both postopera- procedural terminology codes that allow procedure-specific
tive morbidity and mortality vary significantly among estimation of postoperative risk.7 However, the ACS-NSQIP
patients, it is particularly higher in a vulnerable group of also displays limitations, not capturing important cardiovas-
high-risk patients. Therefore, accurate preoperative identi- cular complications, and it has not been extensively vali-
fication of high-risk patients is strongly recommended. dated out of the United States, impairing its
Risk stratification and early recognition of high-risk generalizability. Finally, the SORT model has been validated
patients may improve outcomes since it facilitates surgical in a multicenter study in the United Kingdom that used a
decision-making, preoperative optimization, and tailored specific surgical severity classification, comprising six main
intraoperative and postoperative management.2 However, variables: ASA-PS physical status, urgency of surgery (expe-
in order to enhance patient care, risk stratification tools dited, urgent, immediate), high-risk surgical specialty (gas-
should be validated to the target populations it will be trointestinal, thoracic, vascular), surgical severity (from
applied, and should be easily applicable at the bedside. Ulti- minor to complex major), cancer, and aged 65 years or
mately, risk stratification needs to encompass the complex over.8 The SORT model allows rapid and easy evaluation of
interaction between surgical and anesthetic procedures and mortality risk for individuals undergoing non-cardiac surgery.
patient specific features to be able to assess postoperative However, some recent studies have indicated that SORT per-
morbidity and mortality in different moments, including the formed poorly in other populations as compared to the origi-
preoperative, intra-operative, and postoperative periods.3,4 nal work9 and may not be an accurate predictor of adverse
Plenty of surgical risk models have been developed and outcomes in higher risk patients.10
investigated. American Society of Anesthesiologists’ Physical Of note, prediction of postoperative complications is
Status (ASA-PS) classification, Revised Cardiac Risk Index quite difficult. Predictors of perioperative outcomes are usu-
(RCRI), American College of Surgeons’ National Surgical ally categorized into two groups: patient-related and sur-
Quality Improvement Program Risk Calculators (ACS-NSQIP), gery-related factors. Patient age, comorbidities such as
and the Surgical Outcome Risk Tool (SORT) model are some cardiovascular and pulmonary diseases, functional status,
of the most commonly used preoperative scoring systems. frailty, and perioperative biomarkers may predict outcomes.
ASA-PS classification is the simplest scoring system and has Emergency or urgent surgeries significantly increase the risk
been used since 1941.5 Risk assignment is independent of of postoperative complications. Other surgery-related pre-
the surgical procedure and is based solely on subjective dictors include length of surgery, blood loss, and major sur-
assessment of a patients’ overall health status, leading to gery.11 Nevertheless, so far, most complication prediction
significant inter-rater reliability. The RCRI was designed to scores display moderate accuracy in predicting postopera-
focus on major cardiovascular mortality following noncar- tive complications, especially in some surgical subpopula-
diac surgeries, consisting of six independent predictors: tions.12 It is important to emphasize that these tools predict
high-risk surgery, history of ischemic heart disease, outcomes in a “typical patient”, but are limited in

https://doi.org/10.1016/j.bjane.2022.04.002
0104-0014/© 2022 Sociedade Brasileira de Anestesiologia. Published by Elsevier Editora Ltda. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 A.P. Schmidt and L.C. Stefani

accurately predicting risk for an individual patient as spe- Morbidity and mortality for high-risk surgical patients are
cific factors related to the patient and the surgery must be often high, especially in low-resource settings. The develop-
considered. Alternatively, subspecialty prediction models ment of several risk calculators has enhanced our ability to
may be more accurate for high-risk patients and new risk comprehensively quantify the risk of adverse postoperative
models addressing individual high-risk groups are in constant events, especially surgery-related mortality. The association
development and validation. At the end of the day, the ulti- of these tools with new perioperative biomarkers, frailty
mate goal of any prediction tool is providing adequate and scores, and a more comprehensive assessment of functional
clear information to patients and clinicians in order to pre- status may further refine our ability to detect high-risk surgi-
emptively discuss management options, rescue strategies, cal patients. As knowledge continues to grow in this area,
and, in a more extreme scenario, end-of-life decisions. new research should focus on the implementation of mitiga-
As demonstrated, several assessment tools have been tion strategies to reduce adverse events after surgery. For
implemented to identify high-risk surgical patients. How- those patients who are identified as having increased surgi-
ever, most of these models have been developed and vali- cal risk, perioperative strategies to mitigate risk may need a
dated in high income countries. The feasibility of reliable whole new surgical model associated with an enhanced peri-
risk assessment is particularly important when resources are operative care with the potential to reduce preventable
limited, especially in low- and middle-income countries deaths and the risk of postoperative major adverse cardiac
(LMICs) where primary care is insufficient and advanced con- events. These strategies might include timing of surgery
ditions of diseases compound the surgical scenario. There- after cardiac events or interventions, an improved perioper-
fore, prospective validation is warranted across different ative management of ischemic or valvular heart disease,
geopolitical sectors to test the external validity of those hypertension, arrhythmia, and heart failure.23 Additionally,
scores. Brazil, as well as most countries in Latin America, postoperative comprehensive monitoring should be per-
present huge disparities in terms of healthcare access and formed at least in the first 48 hours after surgery in order to
medical available resources to assist surgical patients, fac- detect adverse events and implement early rescue strate-
tors that may significantly impact in poorer outcomes for gies. This postoperative enhanced care pathway for high-
individual patients undergoing surgery.13 risk surgical patients, or in other words, a “high-risk surgical
In this context, Gutierrez et al.,14 using a large Brazilian bundle” with a patient-centered decision-making, may sig-
surgical cohort, have developed and investigated a multivar- nificantly improve the patient’s experience and outcomes
iable logistic regression model, which predicts in-hospital through the surgical process for patients at higher risk of
mortality (the Ex-Care risk model). In this risk model, adverse events.24 In this context, a recent before-and-after
patient and perioperative predictors were considered, and cohort study with a clinical pathway based on enhanced sur-
its performance was compared to well-known surgical risk veillance for high-risk surgical patients has demonstrated a
tools, namely the Charlson comorbidity index (CCI), the significant reduction of in-hospital mortality.24 Particularly
RCRI, and the SORT risk model. The Ex-Care risk model was in this study, the “high-risk surgical bundle” has included six
very efficient at identifying high-risk surgical patients, dis- main elements, such as risk identification and communica-
playing superior accuracy than the RCRI and similar perfor- tion, adoption of a high-risk post anesthesia care unit dis-
mance as compared to the CCI and SORT models. Although charge checklist, prompt nursing admission to ward,
these findings are promising, the new surgical risk model intensification of vital signs monitoring, perioperative tropo-
needs to be further investigated in multicenter studies, nin measurement, and prompt access to medical support if
requiring an assessment of its accuracy in other national and required.24 New research is still warranted to further evalu-
international institutions. ate which strategies designed to enhance perioperative care
For all the reasons above, in this issue of the Brazilian can actually reduce morbidity and mortality in high-risk sur-
Journal of Anesthesiology we invite readers to access several gical subpopulations.
interesting studies providing new insights into the stratifica- In summary, the high-risk surgical patient is a growing
tion and management of high-risk surgical patients. These challenge to the modern anesthetic care. Perioperative risk
studies have addressed a myriad of topics related to high-risk stratification is currently a fundamental principle of an ade-
surgery, from measures to detect patients at higher risk for quate care for surgical patients. The surgical risk should be
complications to strategies focused on providing enhanced predicted for every patient in the preoperative period, and
perioperative care in high-risk surgical patients.15-22 risk models are valuable clinical tools for shared decision-
Among these studies, it is tempting to highlight the multi- making and the development of individualized care plans.
center study protocol described by Passos et al.,15 proposing Methods for stratifying individual risk include assessment
a large national investigation of the Ex-Care model as a new tools, measures of functional capacity and biomarker assays.
and valid risk tool for the Brazilian surgical population. This They have the potential to contribute to the delivery of a
is a retrospective, multicenter, cohort study which aims to high quality and up-to-date anesthetic and perioperative
build a national preoperative risk model based on Ex-Care care. Notably, the development and application of robust
model of probability of death within 30 days after surgery. tools to properly identify high-risk patients is essential to
In-hospital mortality within 30 days after surgery will be the ground future intervention studies toward improved out-
primary outcome. Importantly, to date, there is no surgical comes for all surgical patients.
risk model developed for the Brazilian population. Therefore,
the Ex-Care model may be a helpful tool to accurately strat-
ify the risk of death after surgery in Brazil, supporting profes- Conflicts of interest
sionals involved in perioperative care to identify high-risk
surgical patients and to better plan therapeutic strategies. The authors declare no conflicts of interest.

314
 Brazilian Journal of Anesthesiology 2022;72(3): 313−315

References
Perioperative man-

17. Segura-Salguero JC, Díaz-Bohada L, Ruiz AJ.
agement of patients undergoing tracheal resection and recon-
1. Meara JG, Leather AJ, Hagander L, et al. Global Surgery 2030: struction: a retrospective observational study. Braz J
evidence and solutions for achieving health, welfare, and eco- Anesthesiol. 2022;72:331−7.
nomic development. Lancet. 2015;386:569−624. 18. Borovac-Pinheiro A, Branda ~o MJN, Argenton JLP, et al. Anesthe-
2. Bose S, Talmor D. Who is a high-risk surgical patient? Curr Opin sia technique and postpartum hemorrhage: a prospective
Crit Care. 2018;24:547−53. cohort study. Braz J Anesthesiol. 2022;72:338−41.
3. Stefani LC, Gutierrez CS, Castro SMJ, et al. Derivation and vali- 19. Salgado-Filho MF, Sachetto R, Carmona MJC. Use of software to
dation of a preoperative risk model for postoperative mortality guide the management of intraoperative hemodynamic instabil-
(SAMPE model): an approach to care stratification. PLoS One. ity. Braz J Anesthesiol. 2022;72:418−9.
2017;12:e0187122. 20. Petran J, Ansems K, Rossaint R, et al. Effects of hypercapnia
4. Stefani LC, Gamermann PW, Backof A, et al. Perioperative mor- versus normocapnia during general anesthesia on outcomes: a
tality related to anesthesia within 48 h and up to 30 days follow- systematic review and meta analysis. Braz J Anesthesiol.
ing surgery: a retrospective cohort study of 11,562 anesthetic 2022;72:398−406.
procedures. J Clin Anesth. 2018;49:79−86. 21. Silva CRD, De Armond LDS, Gentil Filho AC. Assessment of peri-
5. Sankar A, Johnson SR, Beattie WS, et al. Reliability of the Amer- operative cognitive disorders in elderly patients undergoing
ican Society of Anesthesiologists physical status scale in clinical elective surgery. Braz J Anesthesiol. 2022;72:420−2.
practice. Br J Anaesth. 2014;113:424−32. 22. Tesoro R, Suppan M, Dupuis A, et al. Futility of end-of-life and
6. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and emergency surgery in extreme high-risk patients: anaesthetists’
prospective validation of a simple index for prediction of car- versus surgeons’ perspective. Braz J Anesthesiol. 2022;72:434−6.
diac risk of major noncardiac surgery. Circulation. 23. Selwood A, Senthuran S, Blakely B, et al. Improving outcomes
1999;100:1043−9. from high-risk surgery: a multimethod evaluation of a patient-
7. Bilimoria KY, Liu Y, Paruch JL, et al. Development and evalua- centred advanced care planning intervention. BMJ Open.
tion of the universal ACS NSQIP surgical risk calculator: a deci- 2017;7:e014906.
sion aid and informed consent tool for patients and surgeons. J 24. Stahlschmidt A, Passos SC, Cardoso GR, et al. Enhanced peri-
Am Coll Surg. 2013;217. 833-42.e1-3. operative care to improve outcomes for high-risk surgical
8. Protopapa KL, Simpson JC, Smith NC, et al. Development and patients in Brazil: a single-centre before-and-after cohort
validation of the Surgical Outcome Risk Tool (SORT). Br J Surg. study. Anaesthesia. 2022;77:416−27.
2014;101:1774−83.
9. Campbell D, Boyle L, Soakell-Ho M, et al. National risk predic- Andre
P. Schmidt a,b,c,d,e,f,
*, Luciana C. Stefani g
tion model for perioperative mortality in non-cardiac surgery. a
Hospital de Clínicas de Porto Alegre (HCPA), Serviço de
Br J Surg. 2019;106:1549−57.
ria, Porto Alegre, RS,
Anestesia e Medicina Perioperato
10. Oakland K, Cosentino D, Cross T, et al. External validation of the
Surgical Outcome Risk Tool (SORT) in 3305 abdominal surgery
Brazil
b
patients in the independent sector in the UK. Perioper Med Universidade Federal do Rio Grande do Sul (UFRGS), Insti-
(Lond). 2021;10:4. tuto de Cie^ncias Ba
sicas da Sau
de (ICBS), Departamento de
11. Adeleke I, Chae C, Okocha O, et al. Risk assessment and risk Bioquímica, Porto Alegre, RS, Brazil
stratification for perioperative complications and mitigation:
c
Universidade Federal de Cie ^ncias da Sau
de de Porto Alegre
Where should the focus be? How are we doing? Best Pract Res (UFCSPA), Santa Casa de Porto Alegre, Serviço de Anestesia,
Clin Anaesthesiol. 2021;35:517−29. Porto Alegre, RS, Brazil
12. Talmor D, Kelly B. How to better identify patients at high risk of d
Hospital Nossa Senhora da Conceiç a ~o, Serviço de Aneste-
postoperative complications? Curr Opin Crit Care. 2017;23:417 sia, Porto Alegre, RS, Brazil
−23. e
Universidade Federal do Rio Grande do Sul (UFRGS), Facul-
13. Stefani LC, Hajjar L, Biccard B, et al. The need for data describ-
dade de Medicina, Programa de Po
s-graduaç a ~o em Cie^ncias
ing the surgical population in Latin America. Br J Anaesth. 2022.
https://doi.org/10.1016/j.bja.2022.02.029. Online ahead of Pneumolo
gicas, Porto Alegre, RS, Brazil
print.
f
Faculdade de Medicina da Universidade de Sa ~o Paulo
14. Gutierrez CS, Passos SC, Castro SMJ, et al. Few and feasible pre- (FMUSP), Programa de Po
s-Graduaç a ~o em Anestesiologia,
operative variables can identify high-risk surgical patients: der- Cie^ncias Ciru
rgicas e Medicina Perioperato
ria, Sa
~o Paulo, SP,
ivation and validation of the Ex-Care risk model. Br J Anaesth. Brazil
2021;126:525−32. g
Universidade Federal do Rio Grande do Sul (UFRGS), Facul-
15. Passos SC, Stahlschmidt A, Blanco J, et al. Derivation and vali- dade de Medicina, Departamento de Cirurgia, Porto Alegre,
dation of a national multicenter mortality risk stratification RS, Brazil
model - the ExCare model: a study protocol. Braz J Anesthesiol.
2022;72:316−21. *
Corresponding author.
16. Yildiz GO, Hergunsel GO, Sertcakacilar G, et al. Perioperative
E-mail: aschmidt@ufrgs.br (A.P. Schmidt).
goal-directed fluid management using noninvasive hemody-
Received 6 April 2022; Accepted 18 April 2022
namic monitoring in gynecologic oncology. Braz J Anesthesiol.
2022;72:322−30. Available online 21 April 2022

315