Comparison of 4 Cardiac Risk Calculators in

Predicting Postoperative Cardiac Complications After

Noncardiac Operations
Steven L. Cohn, MDa,b,*, and Nerea Fernandez Ros, MD, PhDc

The 2014 American College of Cardiology/American Heart Association Perioperative Guide-

lines suggest using the Revised Cardiac Risk Index, myocardial infarction or cardiac arrest,
or American College of Surgeons—National Surgical Quality Improvement Program cal-
culators for combined patient-surgical risk assessment. There are no published data comparing
their performance. This study compared these risk calculators and a reconstructed Revised
Cardiac Risk Index in predicting postoperative cardiac complications, both during hospi-
talization and 30 days after operation, in a patient cohort who underwent select surgical
procedures in various risk categories. Cardiac complications occurred in 14 of 663 pa-
tients (2.1%), of which 11 occurred during hospitalization. Only 3 of 663 patients (0.45%)
had a myocardial infarction or cardiac arrest. Because these calculators used different risk
factors, different outcomes, and different durations of observation, a true direct compari-
son is not possible. We found that all 4 risk calculators performed well in the setting they
were originally studied but were less accurate when applied in a different manner. In con-
clusion, all calculators were useful in defining low-risk patients in whom further cardiac
testing was unnecessary, and the myocardial infarction or cardiac arrest may be the most
reliable in selecting higher risk patients. © 2017 Elsevier Inc. All rights reserved. (Am J
Cardiol 2018;121:125–130)

Cardiovascular complications after noncardiac surgeries unaware of any published data comparing the performance
represent an important cause of postoperative morbidity and of all 4 scores in the same cohort.
mortality.1,2 Identifying high-risk patients has been an evolv-
ing topic in perioperative medicine since 1977 when Goldman Methods
et al3 published the first cardiac risk index to predict post-
We conducted a retrospective chart review of patients who
operative cardiac events. In 1999, Lee et al4 published a
underwent prespecified operations who were seen by a
Revised Cardiac Risk Index (RCRI), which outperformed pre-
hospitalist in the UHealth Preoperative Assessment Center
vious risk indices becoming the “gold standard” tool. In 2013,
between September 2014 and June 2015. For each patient,
Davis et al5 further improved prediction using a 5-factor re-
we collected demographic data, co-morbidities, functional
constructed RCRI (R-RCRI). The 2014 American College of
status, and procedure-related data including type of surgery,
Cardiology/American Heart Association (ACC/AHA) Guide-
procedure risk, and length of stay. The local institutional review
line on Perioperative Cardiovascular Evaluation and
board approved this retrospective analysis.
Management of Patients Undergoing Noncardiac Surgery6 sug-
We calculated the scores and predicted risk for postop-
gested stratifying patients according to a combination of
erative cardiac complications according to 4 different tools:
surgical and clinical risk factors into low (<1%) or elevated
the RCRI, R-RCRI, MICA, and ACS-SRC. Table 1 summa-
risk (≥1%) for having postoperative major adverse cardiac
rizes the cardiac risk calculators used for this study.
events. They recommend using the RCRI or 2 newer tools
We searched for postoperative cardiovascular complica-
created from the National Surgical Quality Improvement
tions defined as myocardial infarction, cardiac arrest, complete
Program (NSQIP) database—the myocardial infarction or
cardiac block, and pulmonary edema. Because of the differ-
cardiac arrest (MICA) calculator7 or the American College
ences in the outcomes measured in the original publications,
of Surgeons surgical risk calculator (ACS-SRC).8 We are
we calculated the composite cardiac complications (myocar-
dial infarction, cardiac arrest, complete heart block, and
pulmonary edema) occurring in-hospital (CC-IH) and within
a
30 days after surgery (CC-30day) and then only major cardiac
Division of Hospital Medicine, Department of Medicine, University of complications (myocardial infarction/cardiac arrest) within
Miami Miller School of Medicine, Miami, Florida; bMedical Consultation
30 days after surgery (MCC-30day). We further analyzed
Service, Jackson Memorial Hospital, Miami, Florida; and cHospital Medi-
cine Division, Department of Internal Medicine, Clinica Universidad de
cardiac complications based on patient classification of low
Navarra, Pamplona, Spain. Manuscript received June 20, 2017; revised manu- or elevated risk according to the 2014 ACC/AHA guideline
script received and accepted September 19, 2017. cutoff of 1% for each of the tools.
See page 129 for disclosure information. We calculated the performance of these 4 cardiac risk tools
*Corresponding author: Tel: 305 243 1960; fax: 305 243 1538. using receiver operating characteristic (ROC) curves (C-
E-mail address: scohn@med.miami.edu (S.L. Cohn). statistics), and compared them using a nonparametric method

0002-9149/$ - see front matter © 2017 Elsevier Inc. All rights reserved. www.ajconline.org
https://doi.org/10.1016/j.amjcard.2017.09.031
126 The American Journal of Cardiology (www.ajconline.org)
Table 1
Cardiac risk calculators
Revised Cardiac Risk Index4 MI or Cardiac Arrest Calculator (MICA)7
(MI/Cardiac Arrest, complete heart block, (MI/Cardiac Arrest within 30 days after surgery)
pulmonary edema during admission)
High-risk surgery (3 categories) Type of surgery (21 categories)
Ischemic heart disease Age
Congestive heart failure Functional status
Cerebrovascular disease ASA class
Renal insufficiency (Cr > 2 mg/dl) Renal insufficiency (Cr > 1.5 mg/dl)
Diabetes treated with insulin
Reconstructed-RCRI5
(MI/Cardiac Arrest, complete heart block,
pulmonary edema during admission)
High-risk surgery (3 categories)
Ischemic heart disease
Congestive heart failure
Cerebrovascular disease
Renal insufficiency (GFR < 30 cc/min)
ACS-SRC = American College of Surgeons surgical risk calculator; ASA = American Society of Anesthesiology; BMI = body mass index; COPD = chronic
obstructive pulmonary disease; GFR = glomerular filtration rate; MI = myocardial infarction; MICA = myocardial infarction or cardiac arrest; RCRI = Revised
Cardiac Risk Index; R-RCRI = Reconstructed Revised Cardiac Risk Index.
according to the technique by Hanley and McNeil.9 We com-
pared the performance of the 4 scores in patients who
underwent low-risk procedures with those with an expected
length of stay ≥2 days, and analyzed differences in baseline
characteristics and cardiac complications between these groups
using Fisher’s exact tests (qualitative variables) or Stu-
dent’s t test for independent samples (quantitative variables).
SPSS version 22.0 (IBM Corp., Armonk, NY) was used for
all statistical analyses.
Results
There were 663 patients meeting the inclusion criteria.
Tables 2 and 3 summarize the patient and surgical charac-
teristics. The number of patients classified as elevated risk
based on the 2014 ACC/AHA Guideline criteria differed from
one tool to another. The calculators classified 93% to 98%
of patients as low risk with corresponding overall complica-
tion rates of 0.3% to 1.2%, and classified 2% to 7% of patients
as elevated risk with complication rates ranging from 2.1%
to 23.1%. Fewer patients were classified as elevated risk pa-
tients by the MICA calculator (13 of 663) than for the other
tools (47 of 663 for RCRI, 37 of 663 for R-RCRI, and 45
of 663 for ACS-SRC), but the incidence of cardiac events in
these patients was higher for MICA (23.1% of all cardiac
events and 7.7% of major cardiac events). Regarding pa-
tients classified as low risk by each of the calculators, the rates
of major cardiac events 30 days after surgery were all below
1% (Table 4 and Figure 1).
The mean scores ± SD (predicted % chance of having a
cardiac event) for RCRI, R-RCRI, MICA, and ACS-SRC in
ACS NSQIP Surgical Risk Calculator (ACS-SRC)8
(MI/Cardiac Arrest within 30 days after surgery)
Surgical procedure (CPT codes)
Age group
Functional status
ASA class
Acute renal failure
Diabetes on oral meds or insulin
Dialysis
Congestive heart failure
(<30 days)
Dyspnea
Smoker (within past year)
Severe COPD
Ventilator dependent
Sepsis (within 48 hours)
Disseminated cancer
Hypertension requiring meds
Wound class
Sex
Steroid use (chronic)
Ascites (within 30 days)
BMI class
our cohort were 1.06 ± 1.6, 1.66 ± 2.7, 0.19 ± 0.26, and
0.33 ± 0.58, respectively. Composite cardiac complications
(myocardial infarction, cardiac arrest, cardiac block, or pul-
monary edema) occurred in 14 of 663 patients (2.1%) within
30 days after surgery (CC-30day), of which 11 occurred during
surgical hospitalization (CC-IH). However, there were only
3 major cardiac complications (0.45%) (myocardial infarc-
tion and/or cardiac arrest) within 30 days after surgery
(MCC-30day).
Table 5 shows the C-statistics of the original publica-
tions and of our study for the different outcomes and for
different periods of time (in-hospital or 30 days after surgery).
In our study, the C-statistics (95% confidence interval) for
RCRI and R-RCRI showed good discrimination for cardiac
complications as defined in the original publications with no
statistical difference between ROC curves. However, neither
of these scores were predictive of MCC-30d. In contrast, the
ROC curves predicting MCC-30d were discriminative for
MICA but not for ACS-SRC, and there were statistically sig-
nificant differences between these ROC curves.
We analyzed a subgroup of patients with an expected length
of stay <2 days (group A, n = 217) and compared the number
of cardiac events and the performance of the 4 cardiac risk
scores with those with an expected length of stay ≥2 days
(group B, n = 446). Group B patients had statistically sig-
nificantly more cardiac events during hospitalization (2.5%
vs 0%, p = 0.02) and 30 days after surgery (2.9% vs 0.46%,
p = 0.04). They were older (64.3 years vs 53.7 years), had a
higher prevalence of hypertension (59% vs 49%), and were
less likely to have adequate functional status (78.9% vs 90.3%).
There was no difference in previous stroke, diabetes on any
 Miscellaneous/Comparison of 4 Cardiac Risk Calculators 127

Table 2 A (C-statistics 0.58 and 0.57, respectively) and overesti-

Patient characteristics (n = 663) mated the risk of postoperative cardiac events, whereas in
Age (years, mean ± sd) 60.8 ± 14 group B, the C-statistics were similar to the overall cohort
Men 326 (49.2 %) for the same period of time (0.79 and 0.81, respectively). The
Race performance of MICA and ACS-SRC for predicting major
• Hispanic 193 (29.1 %) cardiac events within 30 days after surgery remained dis-
• White 376 (56.7 %) criminative (moderate to excellent) regardless of the group.
• Black 84 (12.7 %)
• Other 10 (1.5 %)
Smoker Discussion
• Never 345 (52 %)
• Former 258 (38.9 %) Because cardiovascular complications are among the most
• Current 60 (9.1 %) important causes of morbidity and mortality in patients who
Body Mass Index, mean ± sd (kg/m2) 29.7 ± 7.3 underwent noncardiac surgery, the development of tools to
Hypertension 370 (55.8 %) predict these events has been an active area of interest in
Creatinine (mg/dl), mean ± sd 0.98 ± 7.3 perioperative medicine. To our knowledge, this is the first study
Glomerular filtration rate (GFR) (ml/min/m2), mean ±sd 68.13 ± 18 comparing the simultaneous performance of the different cal-
Chronic kidney disease (GFR < 60 ml/min/m2) 77 (11.6 %) culators suggested by the 2014 ACC/AHA Guidelines.6 The
Creatinine >1.5 mg/dl) 41 (6.4%) original RCRI publication by Lee et al4 and the 5-factor
Diabetes mellitus
R-RCRI published by Davis et al5 evaluated patients aged ≥50
• No 547 (82.5 %)
• Oral treatment 85 (12.8 %)
years who underwent elective noncardiac surgery with an ex-
• Insulin 15 (2.3 %) pected length of stay ≥2 days, and defined postoperative cardiac
• Combo 16 (2.4 %) complications as myocardial infarction, cardiac arrest, pul-
Prior stroke/Transient ischemic attack 26 (3.9 %) monary edema, or complete heart block occurring during the
Prior ischemic heart disease 74 (11.2 %) hospital stay. The 2 newer tools developed from the NSQIP
Heart failure 15 (2.3 %) database, MICA7 and ACS-SRC,8 predicted major cardiac
Chronic obstructive pulmonary disease on inhalers 33 (5 %) events defined as myocardial infarction or cardiac arrest within
Obstructive sleep apnea 84 (12.6 %) 30 days after surgery. Our results for the performance of each
Active cancer 330 (49.8 %) calculator are similar to those in the original publications
American Society of Anesthesiology classification
(Table 5). Although these tools performed well in predict-
•1 76 (11.5%)
•2 400 (60.3%)
ing the outcomes for which they were designed, they tended
• 3/4 187 (28.2%) to underestimate cardiac events in patients at elevated risk
Functional capacity ≥4 METABOLIC EQUIVALENTS 548 (82.7%) as per the ACC/AHA guidelines. Despite this, our data show
that most patients did well after surgery.
In the original publications, RCRI and R-RCRI rates for
Table 3 composite cardiac events were 2.5% and 2.1%, respec-
Type of surgery
tively. In the MICA and ACS-SRC studies, the rates for only
Otolaryngology (ENT) MI and cardiac arrest were 0.65% and 0.8%, respectively. Our
o Head and neck cancer 77 (11.6 %) results—1.7% for composite cardiac events in-hospital and
o Thyroid 53 (8.1 %) 0.45% for major cardiac events within 30 days after surgery—
o Sinus surgery 33 (5 %) are comparable with these results even though we included
Orthopedic
patients with an expected length of stay <2 days.
o Total joint replacement 100 (15.1 %)
o Spine surgery 87 (13.1 %)
Because the definitions for outcomes and time frames used
Thoracic 73 (11 %) for developing these 4 risk calculators are different, a valid
Urologic direct comparison of outcomes among them is not possible.
o Kidney 50 (7.5 %) This study confirmed that the performance of RCRI and
o Radical prostatectomy 50 (7.5 %) R-RCRI was not different when predicting composite cardiac
o Radical cystectomy 31 (4.7 %) events in-hospital as in their original design. Regarding major
Breast 51 (7.7 %) cardiac events, the performance of MICA was very similar
Sleeve gastrectomy 58 (8.7 %) to ACS-SRC in the original publications,7,8 but in our cohort,
• Procedural risk* MICA performed somewhat better.
• High risk (>5%) 104 (15.7 %)
When the RCRI is used in a manner different from the way
• Intermediate risk (1–5%) 417 (62.9 %)
• Low risk (<1%) 142 (21.4 %)
it was derived (either to predict only major cardiac compli-
cations or for a 30-day postoperative observation period), it
* Note: 2007 guidelines listed 3 risk classes of procedures; 2014 guide- did not perform well and was not useful for predicting only
lines classified the combined patient and surgical risk as low (<1%) or elevated MI or cardiac arrest in our cohort (ROC 0.57). Other studies
(≥1%) risk. evaluating the RCRI in various populations also reported some-
what lower area under the curve ranging from 0.62 to 0.75.10
treatment, heart failure, chronic obstructive pulmonary disease, The RCRI performed worse than MICA (0.75 vs 0.87) when
or history of ischemic heart disease. applied to the original 2008 dataset from which the latter cal-
The performance of the RCRI and R-RCRI for predict- culator was derived, although the RCRI and MICA were
ing composite cardiac events was not discriminative in group comparable (C-statistics 0.9 and 0.85) in a study of patients
128 The American Journal of Cardiology (www.ajconline.org)

Table 4
Incidence of cardiac events in low or elevated risk groups as per 2014 ACC/AHA Guidelines
Overall cohort RCRI R-RCRI MICA ACS-SRC
Low risk Elevated risk* Low risk Elevated risk* Low risk Elevated risk Low risk Elevated risk
n 663 616 47 626 37 650 13 618 45
All cardiac events 30-day 14 7 7 7 7 11 3 6 8
(2.1%) (1.1%) (14.9%) (1.1%) (18.9%) (1.7%) (23.1%) (0.97%) (17.8%)
All cardiac events in hospital 11 5 6 5 6 8 3 4 7
(1.65%) (0.8%) (12.8%) (0.8%) (16.2%) (1.2%) (23.1%) (0.6%) (15.6%)
Major cardiac events 30-day 3 2 1 2 1 2 1 2 1
(0.45%) (0.3%) (2.1%) (0.3%) (2.7%) (0.3%) (7.7%) (0.3%) (2.2%)

* Considering class I as low risk.

ACC/AHA = American College of Cardiology/American Heart Association; ACS-SRC = American College of Surgeons surgical risk calculator; MICA = myo-
cardial infarction or cardiac arrest; RCRI = Revised Cardiac Risk Index; R-RCRI = Reconstructed Revised Cardiac Risk Index.

A) Number of patients (%) with major cardiac events 30-day after surgery in low or
elevated (high) risk group according to different cardiac scores.

B) Number of patients (%) with any cardiac events during admission in low or high risk
group according to different cardiac scores.

ACS-SRC-American College of Surgeons surgical risk calculator, MICA-myocardial infarction or cardiac

arrest, RCRI-Revised Cardiac Risk Index, R-RCRI-reconstructed revised cardiac risk index
Figure 1. Postoperative cardiac complications based on predicted risk.

who underwent elective hip and knee surgery.11 Ausset et al12
applied the RCRI to their patient cohort who underwent hip
surgery to predict myocardial infarction, and its perfor-
mance was suboptimal (ROC 0.59), which is similar to our
results. Because the MICA and ACS-SRC perform well in
patients who underwent low-risk procedures or those with an
expected length of stay <2 days whereas the RCRI overes-
timates risk in these patients, we suggest using these calculators
for this group of patients, particularly when the RCRI esti-
mate is ≥1% and MICA or ACS-SRC estimates are <1%.
According to our results, all calculators were similar in
classifying patients as low risk in whom further cardiac testing
 Miscellaneous/Comparison of 4 Cardiac Risk Calculators 129

Table 5 2. Weiser TG, Regenbogen SE, Thompson KD, Haynes AB, Lipsitz SR,
Receiver operating characteristic (ROC) curve values for each risk calculator Berry WR, Gawande AA. An estimation of the global volume of surgery:
for predicting in-hospital and 30-day postop cardiac complications a modelling strategy based on available data. Lancet 2008;372:139–
144.
RCRI R-RCRI MICA ACS SRC 3. Goldman L, Caldera DL, Nussbaum SR, Southwick FS, Krogstad D,
Murray B, Burke DS, O’Malley TA, Goroll AH, Caplan CH, Nolan J,
Original Publication Results Carabello B, Slater EE. Multifactorial index of cardiac risk in noncar-
All Cardiac Complications* .76-.81 .79 - - diac surgical procedures. N Engl J Med 1977;297:845–850.
in-hospital (CC-IH) 4. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA,
Major Cardiac Complications† - - .87-.88 >.80 Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE,
30-day (MCC-30d) Pedan A, Goldman L. Derivation and prospective validation of a simple
Current Study Results index for prediction of cardiac risk of major noncardiac surgery.
All Cardiac Complications .85 .87 .76 .92 Circulation 1999;100:1043–1049.
5. Davis C, Tait G, Carroll J, Wijeysundera DN, Beattie WS. The Revised
in-hospital (CC-IH)
Cardiac Risk Index in the new millennium: a single-centre prospective
All Cardiac Complications .78 .80 .78 .89 cohort re-evaluation of the original variables in 9,519 consecutive elec-
30-day (CC-30day) tive surgical patients. Can J Anaesth 2013;60:855–863.
Major Cardiac Complications .55 .56 .94 .77 6. Fleisher LA, Fleischmann KE, Auerbach AD, Barnason SA, Beckman
30-day (MCC-30day) JA, Bozkurt B, Davila-Roman VG, Gerhard-Herman MD, Holly TA,
Kane GC, Marine JE, Nelson MT, Spencer CC, Thompson A, Ting HH,
* Myocardial infarction, cardiac arrest, pulmonary edema, or complete heart Uretsky BF, Wijeysundera DN. 2014 ACC/AHA guideline on
block. perioperative cardiovascular evaluation and management of patients un-
†
Myocardial infarction or cardiac arrest. dergoing noncardiac surgery: a report of the American College of
ACS-SRC = American College of Surgeons surgical risk calculator; Cardiology/American Heart Association Task Force on Practice Guide-
MICA = myocardial infarction or cardiac arrest; RCRI = Revised Cardiac lines. Circulation 2014;130:2215–2245.
7. Gupta PK, Gupta H, Sundaram A, Kaushik M, Fang X, Miller WJ,
Risk Index; R-RCRI = Reconstructed Revised Cardiac Risk Index. Esterbrooks DJ, Hunter CB, Pipinos II, Johanning JM, Lynch TG, Forse
RA, Mohiuddin SM, Mooss AN. Development and validation of a risk
calculator for prediction of cardiac risk after surgery. Circulation
is unnecessary; however, in their classification of patients at 2011;124:381–387.
elevated risk, complication rates varied, ranging from 2.1% 8. Bilimoria KY, Liu Y, Paruch JL, Zhou L, Kmiecik TE, Ko CY, Cohen
to 23.1%. The MICA calculator classified fewer patients as ME. Development and evaluation of the universal ACS NSQIP surgi-
elevated risk, but it may be the most reliable calculator in se- cal risk calculator: a decision aid and informed consent tool for patients
and surgeons. J Am Coll Surg 2013;217:833–842, e831-e833.
lecting higher risk patients. In other studies of specific surgical 9. Hanley JA, McNeil BJ. A method of comparing the areas under re-
subgroups, the ACS-SRC did not perform well. It was a poor ceiver operating characteristic curves derived from the same cases.
predictor of perioperative complications after major head and Radiology 1983;148:839–843.
neck operations,13–15 gynecologic oncology procedures,16–18 and 10. Duceppe E, Parlow J, MacDonald P, Lyons K, McMullen M, Srinathan
breast reconstruction.19 The ACS-SRC is being recalibrated S, Graham M, Tandon V, Styles K, Bessissow A, Sessler DI, Bryson
G, Devereaux PJ. Canadian Cardiovascular Society guidelines on
after finding a slight tendency for predicted risk to be over- perioperative cardiac risk assessment and management for patients who
estimated for lower and highest risk patients and undergo noncardiac surgery. Can J Cardiol 2017;33:17–32.
underestimated for moderate-risk patients.20 11. Peterson B, Ghahramani M, Harris S, Suchniak-Mussari K, Bedi G,
The main limitations of this study are that it (1) was ret- Bulathsinghala C, Foy A. Usefulness of the myocardial infarction and
cardiac arrest calculator as a discriminator of adverse cardiac events after
rospective; (2) was performed at a single academic medical elective hip and knee surgery. Am J Cardiol 2016;117:1992–1995.
center; (3) had a relatively small sample size and smaller 12. Ausset S, Auroy Y, Lambert E, Vest P, Plotton C, Rigal S, Lenoir B,
number of elevated risk patients; and (4) had a low inci- Benhamou D. Cardiac troponin I release after hip surgery correlates with
dence of complications which underestimated the true poor long-term cardiac outcome. Eur J Anaesthesiol 2008;25:158–
incidence as there was no routine postoperative surveillance 164.
13. Prasad KG, Nelson BG, Deig CR, Schneider AL, Moore MG. ACS
with troponin or electrocardiography. NSQIP risk calculator: an accurate predictor of complications in major
In conclusion, our results show that all 4 risk calculators head and neck surgery? Otolaryngol Head Neck Surg 2016;155:740–
performed well at defining low and elevated risk groups but 742.
tended to slightly underestimate cardiac events, although most 14. Schneider AL, Deig CR, Prasad KG, Nelson BG, Mantravadi AV,
Brigance JS, Langer MP, McDonald MW, Johnstone PA, Moore MG.
patients did well. Ability of the National Surgical Quality Improvement Program Risk Cal-
culator to predict complications following total laryngectomy. JAMA
Disclosures Otolaryngol Head Neck Surg 2016;142:972–979.
15. Arce K, Moore EJ, Lohse CM, Reiland MD, Yetzer JG, Ettinger KS.
The authors have no conflicts of interest to disclose. The American College of Surgeons National Surgical Quality Improve-
ment Program Surgical Risk Calculator does not accurately predict
risk of 30-day complications among patients undergoing microvascu-
1. Vascular Events In Noncardiac Surgery Patients Cohort Evaluation lar head and neck reconstruction. J Oral Maxillofac Surg 2016;74:1850–
(VISION) Study Investigators, Devereaux PJ, Chan MT, Alonso- 1858.
Coello P, Walsh M, Berwanger O, Villar JC, Wang CY, Garutti RI, Jacka 16. Teoh D, Halloway RN, Heim J, Vogel RI, Rivard C. Evaluation of the
MJ, Sigamani A, Srinathan S, Biccard BM, Chow CK, Abraham V, American College of Surgeons National Surgical Quality Improve-
Tiboni M, Pettit S, Szczeklik W, Lurati Buse G, Botto F, Guyatt G, Heels- ment Program Surgical Risk Calculator in gynecologic oncology patients
Ansdell D, Sessler DI, Thorlund K, Garg AX, Mrkobrada M, Thomas undergoing minimally invasive surgery. J Minim Invasive Gynecol
S, Rodseth RN, Pearse RM, Thabane L, McQueen MJ, VanHelder T, 2017;24:48–54.
Bhandari M, Bosch J, Kurz A, Polanczyk C, Malaga G, Nagele P, Le 17. Rivard C, Nahum R, Slagle E, Duininck M, Isaksson Vogel R, Teoh D.
Manach Y, Leuwer M, Yusuf S. Association between postoperative tro- Evaluation of the performance of the ACS NSQIP surgical risk calcu-
ponin levels and 30-day mortality among patients undergoing noncardiac lator in gynecologic oncology patients undergoing laparotomy. Gynecol
surgery. JAMA 2012;307:2295–2304. Oncol 2016;141:281–286.
130 The American Journal of Cardiology (www.ajconline.org)

18. Szender JB, Frederick PJ, Eng KH, Akers SN, Lele SB, Odunsi K. Evalu-
ation of the National Surgical Quality Improvement Program Universal
Surgical Risk Calculator for a gynecologic oncology service. Int J Gynecol
Cancer 2015;25:512–520.
19. O’Neill AC, Bagher S, Barandun M, Hofer SO, Zhong T. Can the Ameri-
can College of Surgeons NSQIP surgical risk calculator identify patients
at risk of complications following microsurgical breast reconstruc-
tion? J Plast Reconstr Aesthet Surg 2016;69:1356–1362.
20. Liu Y, Cohen ME, Hall BL, Ko CY, Bilimoria KY. Evaluation
and enhancement of calibration in the American College of Surgeons
NSQIP Surgical Risk Calculator. J Am Coll Surg 2016;223:231–
239.