- Email: [email protected]
Comparison of 4 Cardiac Risk Calculators in Predicting Postoperative Cardiac Complications After Noncardiac Operations
Accepted Manuscript Title: Comparison of 4 Cardiac Risk Calculators in Predicting Postoperative Cardiac Complications after Noncardiac Operations Author: Steven L. Cohn, Nerea Fernandez Ros PII: DOI: Reference:
S0002-9149(17)31603-X https://doi.org/doi:10.1016/j.amjcard.2017.09.031 AJC 22939
To appear in:
The American Journal of Cardiology
Received date: Accepted date:
20-6-2017 19-9-2017
Please cite this article as: Steven L. Cohn, Nerea Fernandez Ros, Comparison of 4 Cardiac Risk Calculators in Predicting Postoperative Cardiac Complications after Noncardiac Operations, The American Journal of Cardiology (2017), https://doi.org/doi:10.1016/j.amjcard.2017.09.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AJC-D-17-01624
Comparison of 4 Cardiac Risk Calculators in Predicting Postoperative Cardiac Complications After Noncardiac Operations
Steven L. Cohn, MD
a*
Nerea Fernandez Ros, MD, PhD
a
b
University of Miami Miller School of Medicine
Professor Emeritus Department of Medicine Director – Medical Consultation Service – Jackson Memorial Hospital Miami, FL
b
*
Clinica Universidad de Navarra, Pamplona, Spain
Corresponding author:
1120 NW 14th St – CRB-1140 Miami, FL 33136 Phone: 305-243-1960 Fax: 305-243-1538 Email: [email protected]
1 Page 1 of 19
AJC-D-17-01624
Abstract The 2014 American College of Cardiology/American Heart Association (ACC/AHA) Perioperative Guidelines suggest using the Revised Cardiac Risk Index (RCRI), myocardial infarction or cardiac arrest (MICA), or American College of Surgeons (ACS)-NSQIP calculators for combined patient-surgical risk assessment. There is no published data comparing their performance. This study compared these risk calculators and a reconstructed RCRI in predicting postoperative cardiac complications, both during hospitalization and 30-days after operation, in a cohort patients undergoing select surgical procedures in various risk categories. Cardiac complications occurred in 14/663 (2.1%) of patients, of which 11 occurred during hospitalization. Only 3/663 (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 comparison 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 conclusion, all were useful in defining low risk patients in whom further cardiac testing was unnecessary, and the MICA may be the most reliable in selecting higher risk patients.
Key words: preoperative, cardiac risk indices (or calculators), noncardiac surgery, postoperative cardiac complications
(word count of main text: 1799 words)
2 Page 2 of 19
AJC-D-17-01624
Introduction Cardiovascular complications after noncardiac surgeries represent an important cause of postoperative morbidity and mortality(1, 2). Identifying high risk patients has been an evolving topic in perioperative medicine since 1977 when Goldman et al. (3) published the first cardiac risk index to predict postoperative cardiac events. In 1999, Lee et al. (4) published a revised cardiac risk index (RCRI) which outperformed previous risk indices becoming the “goldstandard” tool. In 2013, Davis et al. (5) further improved prediction using a 5factor reconstructed RCRI (R-RCRI). The 2014 ACC/AHA Guideline on Perioperative
Cardiovascular
Evaluation
and
Management
of
Patients
Undergoing Noncardiac Surgery (6) suggested stratifying patients according to a combination of surgical and clinical risk factors into low (< 1 %) or elevated risk (≥ 1 %) for having postoperative major adverse cardiac events. They recommend using the RCRI or two newer tools created from the National Surgical Quality Improvement Program (NSQIP) database – the myocardial infarction or cardiac arrest (MICA) calculator (7) or the American College of Surgeons surgical risk calculator (ACS-SRC) (8). We are unaware of any published data comparing the performance of all 4 scores in the same cohort.
Methods We conducted a retrospective chart review of patients undergoing prespecified operations who were seen by a hospitalist in the UHealth Preoperative Assessment Center between September 2014 and June 2015. For each patient, we collected demographic data, comorbidities, functional status, and procedure-
3 Page 3 of 19
AJC-D-17-01624
related data including type of surgery, procedure risk, and length of stay . The local institutional review board approved this retrospective analysis. We calculated the scores and predicted risk for postoperative cardiac complications according to 4 different tools: the RCRI, R-RCRI, MICA, and ACS-SRC. Table 1 summarizes the cardiac risk calculators used for this study. We searched for postoperative cardiovascular complications defined as myocardial infarction, cardiac arrest, complete cardiac block, and pulmonary edema. Because of the differences in the outcomes measured in the original publications, we calculated the composite cardiac complications (myocardial infarction, cardiac arrest, complete heart block and pulmonary edema) occurring in-hospital (CC-IH) and within 30 days after surgery (CC-30day) and then only major cardiac complications (myocardial infarction/cardiac arrest) within 30 days after surgery (MCC-30day). We further analyzed cardiac complications based on patient classification of low or elevated risk according to the 2014 ACC/AHA guideline cut-off of 1 % for each of the tools. We calculated the performance of these four cardiac risk tools using receiver operating characteristic (ROC) curves (C-statistics), and compared them using a nonparametric method according to the technique by Hanley and McNeil (9). We compared the performance of the 4 scores in patients undergoing 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 exact tests (qualitative variables) or Student´s t-test for independent samples (quantitative variables). SPSS version 22.0 was used for all statistical analysis.
Results 4 Page 4 of 19
AJC-D-17-01624
There were 663 patients meeting inclusion criteria. Tables 2 and 3 summarize the patient and surgical characteristics. The number of patients classified as elevated risk based on 2014 ACC/AHA Guideline criteria differed from one tool to another. The calculators classified 93-98 % of patients as low risk with corresponding overall complication rates of 0.3-1.2 %, and classified 27% of patients as elevated risk with complication rates ranging from 2.1-23.1 %. Fewer patients were classified as elevated risk patients by the MICA calculator (13/663) than for the other tools (47/663 for RCRI, 37/663 for R-RCRI, and 45/663 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 patients 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 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 pulmonary edema) occurred in 14 of 663 (2.1 %) within 30 days after surgery (CC-30day) of which 11 occurred during surgical hospitalization (CC-IH). However, there were only 3 (0.45%) major cardiac complications (myocardial infarction and/or cardiac arrest) within 30 days after surgery (MCC-30day). Table 5 shows the C-statistics of the original publications 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% CI) for RCRI and RRCRI showed good discrimination for cardiac complications as defined in the 5 Page 5 of 19
AJC-D-17-01624
original publications with no statistical difference between ROC curves. However, neither of these scores were predictive of MCC-30d. On the other hand, the ROC curves predicting MCC-30d were discriminative for MICA but not for ACS-SRC, and there were statistically significant 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 four cardiac risk scores to those with an expected length of stay ≥ 2 days (group B, n=446). Group B patients had statistically significantly 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 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 treatment, heart failure, chronic obstructive pulmonary disease, or history of ischemic heart disease. The performance of the RCRI and R-RCRI for predicting composite cardiac events was not discriminative in group A (C-statistics 0.58 and 0.57, respectively) and overestimated the risk of postoperative cardiac events, whereas in group B the C-statistics were similar to the overall cohort for the same period of time (0.79 and 0.81, respectively). The performance of MICA and ACS-SRC for predicting major cardiac events within 30 days after surgery remained discriminative (moderate to excellent) regardless of the group.
Discussion
6 Page 6 of 19
AJC-D-17-01624
Because cardiovascular complications are among the most important causes of morbidity and mortality in patients undergoing noncardiac surgery, the development of tools to predict these events has been an active area of interest in perioperative medicine. To our knowledge, this is the first study comparing the simultaneous performance of the different calculators suggested by the 2014 ACC/AHA Guidelines (6). The original RCRI publication by Lee et al. (4) and the 5-factor R-RCRI published by Davis et al. (5) evaluated patients >50 years undergoing elective noncardiac surgery with an expected length of stay >2 days, and defined postoperative cardiac complications as myocardial infarction, cardiac arrest, pulmonary edema or complete heart block occurring during the hospital stay. The two newer tools developed from the NSQIP database, MICA (7) and ACS-SRC (8), predicted major cardiac events defined as myocardial infarction or cardiac arrest within 30 days after surgery. Our results for the performance of each calculator are similar to those in the original publications (Table 5). Although these tools performed well predicting the outcomes for which they were designed, they tended to underestimate cardiac events in patients at elevated risk as per ACC/AHA guidelines. Despite this, our data shows that most patients did well after surgery. In the original publications, RCRI and R-RCRI rates for composite cardiac events were 2.5 and 2.1 %, respectively. In the MICA and ACS-SRC studies, the rates for only MI and cardiac arrest were 0.65% and 0.8% respectively. Our results - 1.7 % for composite cardiac events in-hospital and 0.45% for major cardiac events within 30 days after surgery - are comparable to these results even though we included patients with an expected length of stay <2 days. 7 Page 7 of 19
AJC-D-17-01624
Because the definitions for outcomes and timeframes used for developing these 4 risk calculators are different, a valid direct comparison of outcomes among them is not possible. This study confirmed that the performance of RCRI and R-RCRI was not different when predicting composite cardiac events in-hospital as in their original design. Regarding major cardiac events, the performance for MICA was very similar to ACS-SRC in the original publications (7,8), but in our cohort, MICA performed somewhat better. When the RCRI is used in a manner different from the way it was derived (either to predict only major cardiac complications or for a 30-day postoperative observation period), it did not perform as well and was not useful for predicting only MI or cardiac arrest in our cohort (ROC 0.57). Other studies evaluating the RCRI in various populations also reported somewhat lower area under the curve’s ranging from 0.62-0.75. (10) The RCRI performed worse than MICA (0.75 vs 0.87) when applied to the original 2008 data set from which the latter calculator was derived, although the RCRI and MICA were comparable (cstatistics 0.9 and 0.85) in a study of patients undergoing elective hip and knee surgery. (11) Ausset et al. (12) applied the RCRI to their cohort of patients undergoing hip surgery to predict myocardial infarction, and its performance was suboptimal (ROC 0.59), which is similar to our results. Because the MICA and ACS-SRC perform well in patients undergoing low risk procedures or those with an expected length of stay <2 days whereas the RCRI overestimates risk in these patients, we suggest using these calculators for this group of patients, particularly when the RCRI estimate is >1% and MICA or ACS-SRC estimates are <1%.
8 Page 8 of 19
AJC-D-17-01624
According to our results, all calculators were similar in classifying patients as low risk in whom further cardiac testing is unnecessary; however, in their classification of patients at elevated risk, complication rates varied, ranging from 2.1% to 23.1 %. The MICA calculator classified fewer patients as elevated risk, but it may be the most reliable calculator in selecting higher risk patients. In other studies of specific surgical subgroups, the ACS-SRC did not perform as well. It was a poor predictor of perioperative complications following major head and neck operations (13,14,15), gynecologic oncology procedures (16,17,18), and breast reconstruction (19). The ACS-SRC is being recalibrated after finding a slight tendency for predicted risk to be overestimated for lower and highest risk patients and underestimated for moderate risk patients.(20) The main limitations of this study are that it: 1) was retrospective; 2) was performed at a single academic medical center; 3) had a relatively small sample size and smaller number of elevated risk patients; and 4) had a low incidence of complications which underestimated the true incidence as there was no routine postoperative surveillance with troponin or electrocardiography. In conclusion, our results show that all 4 risk calculators performed well at defining low and elevated risk groups but tended to slightly underestimate cardiac events, although most patients did well.
9 Page 9 of 19
AJC-D-17-01624
1. Vascular Events In Noncardiac Surgery Patients Cohort Evaluation (VISION) Study Investigators, Devereaux PJ, Chan MT, Alonso-Coello P, Walsh M, Berwanger O, Villar JC, Wang CY, Garutti RI, Jacka MJ, Sigamani A, Srinathan S, Biccard BM, Chow CK, Abraham V, Tiboni M, Pettit S, Szczeklik W, Lurati Buse G, Botto F, Guyatt G, Heels-Ansdell D, Sessler DI, Thorlund K, Garg AX, Mrkobrada M, Thomas S, Rodseth RN, Pearse RM, Thabane L, McQueen MJ, VanHelder T, Bhandari M, Bosch J, Kurz A, Polanczyk C, Malaga G, Nagele P, Le Manach Y, Leuwer M, Yusuf S. Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery. JAMA 2012;307:2295-2304. 2. Weiser TG, Regenbogen SE, Thompson KD, Haynes AB, Lipsitz SR, Berry WR, Gawande AA. An estimation of the global volume of surgery: a modelling strategy based on available data. Lancet 2008;372:139-144. 3. Goldman L, Caldera DL, Nussbaum SR, Southwick FS, Krogstad D, Murray B, Burke DS, O'Malley TA, Goroll AH, Caplan CH, Nolan J, Carabello B, Slater EE. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845-850. 4. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-1049. 5. Davis C, Tait G, Carroll J, Wijeysundera DN, Beattie WS. The Revised Cardiac Risk Index in the new millennium: a single-centre prospective cohort re-
10 Page 10 of 19
AJC-D-17-01624
evaluation of the original variables in 9,519 consecutive elective surgical patients. Can J Anaesth 2013;60:855-863. 6. Fleisher LA, Fleischmann KE, Auerbach AD, Barnason SA, Beckman JA, Bozkurt B, Davila-Roman VG, Gerhard-Herman MD, Holly TA, Kane GC, Marine JE, Nelson MT, Spencer CC, Thompson A, Ting HH, Uretsky BF, Wijeysundera DN. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;130:2215-2245. 7. Gupta PK, Gupta H, Sundaram A, Kaushik M, Fang X, Miller WJ, 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 2011;124:381-387. 8. Bilimoria KY, Liu Y, Paruch JL, Zhou L, Kmiecik TE, Ko CY, Cohen ME. Development and evaluation of the universal ACS NSQIP surgical risk calculator: a decision aid and informed consent tool for patients and surgeons. J Am Coll Surg 2013;217:833-842 e831-e833. 9. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 1983;148:839-843. 10. Duceppe E, Parlow J, MacDonald P, Lyons K, McMullen M, Srinathan S, Graham M, Tandon V, Styles K, Bessissow A, Sessler DI, Bryson G, Devereaux PJ. Canadian Cardiovascular Society Guidelines on Perioperative Cardiac Risk Assessment and Management for Patients Who Undergo Noncardiac Surgery. Can J Cardiol 2017;33:17-32.
11 Page 11 of 19
AJC-D-17-01624
11. Peterson B, Ghahramani M, Harris S, Suchniak-Mussari K, Bedi G, Bulathsinghala C, Foy A. Usefulness of the Myocardial Infarction and Cardiac Arrest Calculator as a discriminator of adverse cardiac events after elective hip and knee surgery. Am J Cardiol 2016;117:1992-1995. 12. Ausset S, Auroy Y, Lambert E, Vest P, Plotton C, Rigal S, Lenoir B, Benhamou D. Cardiac troponin I release after hip surgery correlates with poor long-term cardiac outcome. Eur J Anaesthesiol 2008;25:158-164. 13. Prasad KG, Nelson BG, Deig CR, Schneider AL, Moore MG. ACS NSQIP Risk Calculator: an accurate predictor of complications in major head and neck surgery? Otolaryngol Head Neck Surg 2016;155:740-742. 14. Schneider AL, Deig CR, Prasad KG, Nelson BG, Mantravadi AV, Brigance JS, Langer MP, McDonald MW, Johnstone PA, Moore MG. Ability of the National Surgical Quality Improvement Program Risk Calculator to predict complications following total laryngectomy. JAMA Otolaryngol Head Neck Surg 2016;142:972-979. 15. Arce K, Moore EJ, Lohse CM, Reiland MD, Yetzer JG, Ettinger KS. The American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator does not accurately predict risk of 30-day complications among patients undergoing microvascular head and neck reconstruction. J Oral Maxillofac Surg 2016;74:1850-1858. 16. Teoh D, Halloway RN, Heim J, Vogel RI, Rivard C. Evaluation of the American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator in gynecologic oncology patients undergoing minimally invasive surgery. J Minim Invasive Gynecol 2017;24:48-54.
12 Page 12 of 19
AJC-D-17-01624
17. Rivard C, Nahum R, Slagle E, Duininck M, Isaksson Vogel R, Teoh D. Evaluation of the performance of the ACS NSQIP surgical risk calculator in gynecologic oncology patients undergoing laparotomy. Gynecol Oncol 2016;141:281-286. 18. Szender JB, Frederick PJ, Eng KH, Akers SN, Lele SB, Odunsi K. Evaluation 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 American College of Surgeons NSQIP surgical risk calculator identify patients at risk of complications following microsurgical breast reconstruction? 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.
13 Page 13 of 19
AJC-D-17-01624
Figure 1. Postoperative cardiac complications based on predicted risk 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
14 Page 14 of 19
AJC-D-17-01624
Table 1. Cardiac Risk Calculators Revised Cardiac Risk Index (4) MI or Cardiac Arrest Calculator (MICA) (7) (MI/Cardiac Arrest, complete heart block, pulmonary edema during admission)
ACS NSQIP Surgical Risk Calculator (ACS-SRC) (8)
(MI/Cardiac Arrest within 30 days (MI/Cardiac Arrest within 30 days after surgery) after surgery)
High-risk surgery (3 categories)
Type of surgery (21 categories)
Surgical procedure (CPT codes)
Ischemic heart disease
Age
Age group
Congestive heart failure
Functional status
Functional status
Cerebrovascular disease
ASA class
ASA class
Renal insufficiency (Cr>1.5mg/dl)
Acute renal failure
Renal insufficiency (Cr>2mg/dl) Diabetes treated with insulin
Diabetes on oral meds or insulin Dialysis
Reconstructed-RCRI (5) (MI/Cardiac Arrest, complete heart block, pulmonary edema during admission)
Congestive heart failure (<30 days)
High-risk surgery (3 categories)
Dyspnea
Ischemic heart disease
Smoker (within past year)
Congestive heart failure
Severe COPD
Cerebrovascular disease
Ventilator dependent
Renal insufficiency (GFR<30cc/min)
Sepsis (within 48 hours) Disseminated cancer Hypertension requiring meds Wound class
Sex Steroid use (chronic) Ascites (within 30 days) BMI class
ACS-SRC-American College of Surgeons surgical risk calculator, ASA-American Society of Anesthesiology, BMI-body mass index, COPD-chronic obstructive pulmonary disease, GFR-
15 Page 15 of 19
AJC-D-17-01624 glomerular filtration rate, MI-myocardial infarction, MICA-myocardial infarction or cardiac arrest, RCRI-Revised Cardiac Risk Index, R-RCRI-reconstructed revised cardiac risk index
Table 2. Patient characteristics (n=663) Age (years, mean ± sd)
60.8 ± 14
Men
326 (49.2 %)
Race
Hispanic
193 (29.1 %)
White
376 (56.7 %)
Black
84 (12.7 %)
Other Smoker
10 (1.5 %)
Never
345 (52 %)
Former
258 (38.9 %)
Current Body Mass Index, mean ± sd (kg/m2)
60 (9.1 %) 29.7 ± 7.3
Hypertension
370 (55.8 %)
Creatinine (mg/dl), mean ± sd
0.98 ± 7.3
Glomerular filtration rate (GFR) (ml/min/m2), mean ±sd Chronic kidney disease (GFR < 60 ml/min/m2)
68.13 ± 18
Creatinine >1.5 mg/dl)
77 (11.6 %) 41 (6.4%)
Diabetes mellitus
No
547 (82.5 %)
Oral treatment
85 (12.8 %)
Insulin
15 (2.3 %)
Combo Prior stroke/Transient ischemic attack
16 (2.4 %)
Prior ischemic heart disease
74 (11.2 %)
Heart failure
15 (2.3 %)
Chronic obstructive pulmonary inhalers Obstructive sleep apnea
disease
26 (3.9 %)
on
Active cancer
33 (5 %) 84 (12.6 %) 330 (49.8 %)
American Society of Anesthesiology classification
1
76 (11.5%)
2
400 (60.3%) 16 Page 16 of 19
AJC-D-17-01624 3/4
187 (28.2%)
Functional capacity >4 METABOLIC EQUIVALENTS
548 (82.7%)
Table 3. Type of surgery
Otolaryngology (ENT) o
Head and neck cancer
77 (11.6 %)
o
Thyroid
53 (8.1 %)
o
Sinus surgery
33 (5 %)
Orthopedic o
Total joint replacement
100 %)
o
Spine surgery
87 (13.1 %)
Thoracic
(15.1
73 (11 %)
Urologic o
Kidney
50 (7.5 %)
o
Radical prostatectomy
50 (7.5 %)
o
Radical cystectomy
31 (4.7 %)
Breast
51 (7.7 %)
Sleeve gastrectomy
58 (8.7 %)
Procedural risk*
High risk (<1%)
104 %)
(15.7
Intermediate risk (1- 417 %) 5%)
(62.9
Low risk (>5%)
142 %)
(21.4
*Note: 2007 guidelines listed 3 risk classes of procedures; 2014 guidelines classify the 17 Page 17 of 19
AJC-D-17-01624 combined patient and surgical risk as low (<1%) or elevated (>1%) risk
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%)
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%)
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%)
All cardiac events in hospital Major cardiac events 30-day
*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-myocardial infarction or cardiac arrest, RCRI-Revised Cardiac Risk Index, R-RCRI-Reconstructed revised cardiac risk index
Table 5. Receiver operating characteristic (ROC) curve values for each risk calculator for predicting in-hospital and 30-day postop cardiac complications.
RCRI
R-RCRI
MICA
ACS SRC
18 Page 18 of 19
AJC-D-17-01624
Original Publication Results All Cardiac Complications* in-hospital (CC-IH) Major Cardiac Complications** 30-day (MCC-30d)
.76-.81
.79
-
-
-
-
.87-.88
>.80
.85
.87
.76
.92
.78
.80
.78
.89
.55
.56
.94
.77
Current Study Results All Cardiac Complications in-hospital (CC-IH) All Cardiac Complications 30-day (CC-30day) Major Cardiac Complications 30-day (MCC-30day)
*Myocardial infarction, cardiac arrest, pulmonary edema, or complete heart block. **Myocardial infarction or cardiac arrest. 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
19 Page 19 of 19
S0002-9149(17)31603-X https://doi.org/doi:10.1016/j.amjcard.2017.09.031 AJC 22939
To appear in:
The American Journal of Cardiology
Received date: Accepted date:
20-6-2017 19-9-2017
Please cite this article as: Steven L. Cohn, Nerea Fernandez Ros, Comparison of 4 Cardiac Risk Calculators in Predicting Postoperative Cardiac Complications after Noncardiac Operations, The American Journal of Cardiology (2017), https://doi.org/doi:10.1016/j.amjcard.2017.09.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AJC-D-17-01624
Comparison of 4 Cardiac Risk Calculators in Predicting Postoperative Cardiac Complications After Noncardiac Operations
Steven L. Cohn, MD
a*
Nerea Fernandez Ros, MD, PhD
a
b
University of Miami Miller School of Medicine
Professor Emeritus Department of Medicine Director – Medical Consultation Service – Jackson Memorial Hospital Miami, FL
b
*
Clinica Universidad de Navarra, Pamplona, Spain
Corresponding author:
1120 NW 14th St – CRB-1140 Miami, FL 33136 Phone: 305-243-1960 Fax: 305-243-1538 Email: [email protected]
1 Page 1 of 19
AJC-D-17-01624
Abstract The 2014 American College of Cardiology/American Heart Association (ACC/AHA) Perioperative Guidelines suggest using the Revised Cardiac Risk Index (RCRI), myocardial infarction or cardiac arrest (MICA), or American College of Surgeons (ACS)-NSQIP calculators for combined patient-surgical risk assessment. There is no published data comparing their performance. This study compared these risk calculators and a reconstructed RCRI in predicting postoperative cardiac complications, both during hospitalization and 30-days after operation, in a cohort patients undergoing select surgical procedures in various risk categories. Cardiac complications occurred in 14/663 (2.1%) of patients, of which 11 occurred during hospitalization. Only 3/663 (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 comparison 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 conclusion, all were useful in defining low risk patients in whom further cardiac testing was unnecessary, and the MICA may be the most reliable in selecting higher risk patients.
Key words: preoperative, cardiac risk indices (or calculators), noncardiac surgery, postoperative cardiac complications
(word count of main text: 1799 words)
2 Page 2 of 19
AJC-D-17-01624
Introduction Cardiovascular complications after noncardiac surgeries represent an important cause of postoperative morbidity and mortality(1, 2). Identifying high risk patients has been an evolving topic in perioperative medicine since 1977 when Goldman et al. (3) published the first cardiac risk index to predict postoperative cardiac events. In 1999, Lee et al. (4) published a revised cardiac risk index (RCRI) which outperformed previous risk indices becoming the “goldstandard” tool. In 2013, Davis et al. (5) further improved prediction using a 5factor reconstructed RCRI (R-RCRI). The 2014 ACC/AHA Guideline on Perioperative
Cardiovascular
Evaluation
and
Management
of
Patients
Undergoing Noncardiac Surgery (6) suggested stratifying patients according to a combination of surgical and clinical risk factors into low (< 1 %) or elevated risk (≥ 1 %) for having postoperative major adverse cardiac events. They recommend using the RCRI or two newer tools created from the National Surgical Quality Improvement Program (NSQIP) database – the myocardial infarction or cardiac arrest (MICA) calculator (7) or the American College of Surgeons surgical risk calculator (ACS-SRC) (8). We are unaware of any published data comparing the performance of all 4 scores in the same cohort.
Methods We conducted a retrospective chart review of patients undergoing prespecified operations who were seen by a hospitalist in the UHealth Preoperative Assessment Center between September 2014 and June 2015. For each patient, we collected demographic data, comorbidities, functional status, and procedure-
3 Page 3 of 19
AJC-D-17-01624
related data including type of surgery, procedure risk, and length of stay . The local institutional review board approved this retrospective analysis. We calculated the scores and predicted risk for postoperative cardiac complications according to 4 different tools: the RCRI, R-RCRI, MICA, and ACS-SRC. Table 1 summarizes the cardiac risk calculators used for this study. We searched for postoperative cardiovascular complications defined as myocardial infarction, cardiac arrest, complete cardiac block, and pulmonary edema. Because of the differences in the outcomes measured in the original publications, we calculated the composite cardiac complications (myocardial infarction, cardiac arrest, complete heart block and pulmonary edema) occurring in-hospital (CC-IH) and within 30 days after surgery (CC-30day) and then only major cardiac complications (myocardial infarction/cardiac arrest) within 30 days after surgery (MCC-30day). We further analyzed cardiac complications based on patient classification of low or elevated risk according to the 2014 ACC/AHA guideline cut-off of 1 % for each of the tools. We calculated the performance of these four cardiac risk tools using receiver operating characteristic (ROC) curves (C-statistics), and compared them using a nonparametric method according to the technique by Hanley and McNeil (9). We compared the performance of the 4 scores in patients undergoing 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 exact tests (qualitative variables) or Student´s t-test for independent samples (quantitative variables). SPSS version 22.0 was used for all statistical analysis.
Results 4 Page 4 of 19
AJC-D-17-01624
There were 663 patients meeting inclusion criteria. Tables 2 and 3 summarize the patient and surgical characteristics. The number of patients classified as elevated risk based on 2014 ACC/AHA Guideline criteria differed from one tool to another. The calculators classified 93-98 % of patients as low risk with corresponding overall complication rates of 0.3-1.2 %, and classified 27% of patients as elevated risk with complication rates ranging from 2.1-23.1 %. Fewer patients were classified as elevated risk patients by the MICA calculator (13/663) than for the other tools (47/663 for RCRI, 37/663 for R-RCRI, and 45/663 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 patients 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 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 pulmonary edema) occurred in 14 of 663 (2.1 %) within 30 days after surgery (CC-30day) of which 11 occurred during surgical hospitalization (CC-IH). However, there were only 3 (0.45%) major cardiac complications (myocardial infarction and/or cardiac arrest) within 30 days after surgery (MCC-30day). Table 5 shows the C-statistics of the original publications 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% CI) for RCRI and RRCRI showed good discrimination for cardiac complications as defined in the 5 Page 5 of 19
AJC-D-17-01624
original publications with no statistical difference between ROC curves. However, neither of these scores were predictive of MCC-30d. On the other hand, the ROC curves predicting MCC-30d were discriminative for MICA but not for ACS-SRC, and there were statistically significant 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 four cardiac risk scores to those with an expected length of stay ≥ 2 days (group B, n=446). Group B patients had statistically significantly 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 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 treatment, heart failure, chronic obstructive pulmonary disease, or history of ischemic heart disease. The performance of the RCRI and R-RCRI for predicting composite cardiac events was not discriminative in group A (C-statistics 0.58 and 0.57, respectively) and overestimated the risk of postoperative cardiac events, whereas in group B the C-statistics were similar to the overall cohort for the same period of time (0.79 and 0.81, respectively). The performance of MICA and ACS-SRC for predicting major cardiac events within 30 days after surgery remained discriminative (moderate to excellent) regardless of the group.
Discussion
6 Page 6 of 19
AJC-D-17-01624
Because cardiovascular complications are among the most important causes of morbidity and mortality in patients undergoing noncardiac surgery, the development of tools to predict these events has been an active area of interest in perioperative medicine. To our knowledge, this is the first study comparing the simultaneous performance of the different calculators suggested by the 2014 ACC/AHA Guidelines (6). The original RCRI publication by Lee et al. (4) and the 5-factor R-RCRI published by Davis et al. (5) evaluated patients >50 years undergoing elective noncardiac surgery with an expected length of stay >2 days, and defined postoperative cardiac complications as myocardial infarction, cardiac arrest, pulmonary edema or complete heart block occurring during the hospital stay. The two newer tools developed from the NSQIP database, MICA (7) and ACS-SRC (8), predicted major cardiac events defined as myocardial infarction or cardiac arrest within 30 days after surgery. Our results for the performance of each calculator are similar to those in the original publications (Table 5). Although these tools performed well predicting the outcomes for which they were designed, they tended to underestimate cardiac events in patients at elevated risk as per ACC/AHA guidelines. Despite this, our data shows that most patients did well after surgery. In the original publications, RCRI and R-RCRI rates for composite cardiac events were 2.5 and 2.1 %, respectively. In the MICA and ACS-SRC studies, the rates for only MI and cardiac arrest were 0.65% and 0.8% respectively. Our results - 1.7 % for composite cardiac events in-hospital and 0.45% for major cardiac events within 30 days after surgery - are comparable to these results even though we included patients with an expected length of stay <2 days. 7 Page 7 of 19
AJC-D-17-01624
Because the definitions for outcomes and timeframes used for developing these 4 risk calculators are different, a valid direct comparison of outcomes among them is not possible. This study confirmed that the performance of RCRI and R-RCRI was not different when predicting composite cardiac events in-hospital as in their original design. Regarding major cardiac events, the performance for MICA was very similar to ACS-SRC in the original publications (7,8), but in our cohort, MICA performed somewhat better. When the RCRI is used in a manner different from the way it was derived (either to predict only major cardiac complications or for a 30-day postoperative observation period), it did not perform as well and was not useful for predicting only MI or cardiac arrest in our cohort (ROC 0.57). Other studies evaluating the RCRI in various populations also reported somewhat lower area under the curve’s ranging from 0.62-0.75. (10) The RCRI performed worse than MICA (0.75 vs 0.87) when applied to the original 2008 data set from which the latter calculator was derived, although the RCRI and MICA were comparable (cstatistics 0.9 and 0.85) in a study of patients undergoing elective hip and knee surgery. (11) Ausset et al. (12) applied the RCRI to their cohort of patients undergoing hip surgery to predict myocardial infarction, and its performance was suboptimal (ROC 0.59), which is similar to our results. Because the MICA and ACS-SRC perform well in patients undergoing low risk procedures or those with an expected length of stay <2 days whereas the RCRI overestimates risk in these patients, we suggest using these calculators for this group of patients, particularly when the RCRI estimate is >1% and MICA or ACS-SRC estimates are <1%.
8 Page 8 of 19
AJC-D-17-01624
According to our results, all calculators were similar in classifying patients as low risk in whom further cardiac testing is unnecessary; however, in their classification of patients at elevated risk, complication rates varied, ranging from 2.1% to 23.1 %. The MICA calculator classified fewer patients as elevated risk, but it may be the most reliable calculator in selecting higher risk patients. In other studies of specific surgical subgroups, the ACS-SRC did not perform as well. It was a poor predictor of perioperative complications following major head and neck operations (13,14,15), gynecologic oncology procedures (16,17,18), and breast reconstruction (19). The ACS-SRC is being recalibrated after finding a slight tendency for predicted risk to be overestimated for lower and highest risk patients and underestimated for moderate risk patients.(20) The main limitations of this study are that it: 1) was retrospective; 2) was performed at a single academic medical center; 3) had a relatively small sample size and smaller number of elevated risk patients; and 4) had a low incidence of complications which underestimated the true incidence as there was no routine postoperative surveillance with troponin or electrocardiography. In conclusion, our results show that all 4 risk calculators performed well at defining low and elevated risk groups but tended to slightly underestimate cardiac events, although most patients did well.
9 Page 9 of 19
AJC-D-17-01624
1. Vascular Events In Noncardiac Surgery Patients Cohort Evaluation (VISION) Study Investigators, Devereaux PJ, Chan MT, Alonso-Coello P, Walsh M, Berwanger O, Villar JC, Wang CY, Garutti RI, Jacka MJ, Sigamani A, Srinathan S, Biccard BM, Chow CK, Abraham V, Tiboni M, Pettit S, Szczeklik W, Lurati Buse G, Botto F, Guyatt G, Heels-Ansdell D, Sessler DI, Thorlund K, Garg AX, Mrkobrada M, Thomas S, Rodseth RN, Pearse RM, Thabane L, McQueen MJ, VanHelder T, Bhandari M, Bosch J, Kurz A, Polanczyk C, Malaga G, Nagele P, Le Manach Y, Leuwer M, Yusuf S. Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery. JAMA 2012;307:2295-2304. 2. Weiser TG, Regenbogen SE, Thompson KD, Haynes AB, Lipsitz SR, Berry WR, Gawande AA. An estimation of the global volume of surgery: a modelling strategy based on available data. Lancet 2008;372:139-144. 3. Goldman L, Caldera DL, Nussbaum SR, Southwick FS, Krogstad D, Murray B, Burke DS, O'Malley TA, Goroll AH, Caplan CH, Nolan J, Carabello B, Slater EE. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845-850. 4. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-1049. 5. Davis C, Tait G, Carroll J, Wijeysundera DN, Beattie WS. The Revised Cardiac Risk Index in the new millennium: a single-centre prospective cohort re-
10 Page 10 of 19
AJC-D-17-01624
evaluation of the original variables in 9,519 consecutive elective surgical patients. Can J Anaesth 2013;60:855-863. 6. Fleisher LA, Fleischmann KE, Auerbach AD, Barnason SA, Beckman JA, Bozkurt B, Davila-Roman VG, Gerhard-Herman MD, Holly TA, Kane GC, Marine JE, Nelson MT, Spencer CC, Thompson A, Ting HH, Uretsky BF, Wijeysundera DN. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;130:2215-2245. 7. Gupta PK, Gupta H, Sundaram A, Kaushik M, Fang X, Miller WJ, 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 2011;124:381-387. 8. Bilimoria KY, Liu Y, Paruch JL, Zhou L, Kmiecik TE, Ko CY, Cohen ME. Development and evaluation of the universal ACS NSQIP surgical risk calculator: a decision aid and informed consent tool for patients and surgeons. J Am Coll Surg 2013;217:833-842 e831-e833. 9. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 1983;148:839-843. 10. Duceppe E, Parlow J, MacDonald P, Lyons K, McMullen M, Srinathan S, Graham M, Tandon V, Styles K, Bessissow A, Sessler DI, Bryson G, Devereaux PJ. Canadian Cardiovascular Society Guidelines on Perioperative Cardiac Risk Assessment and Management for Patients Who Undergo Noncardiac Surgery. Can J Cardiol 2017;33:17-32.
11 Page 11 of 19
AJC-D-17-01624
11. Peterson B, Ghahramani M, Harris S, Suchniak-Mussari K, Bedi G, Bulathsinghala C, Foy A. Usefulness of the Myocardial Infarction and Cardiac Arrest Calculator as a discriminator of adverse cardiac events after elective hip and knee surgery. Am J Cardiol 2016;117:1992-1995. 12. Ausset S, Auroy Y, Lambert E, Vest P, Plotton C, Rigal S, Lenoir B, Benhamou D. Cardiac troponin I release after hip surgery correlates with poor long-term cardiac outcome. Eur J Anaesthesiol 2008;25:158-164. 13. Prasad KG, Nelson BG, Deig CR, Schneider AL, Moore MG. ACS NSQIP Risk Calculator: an accurate predictor of complications in major head and neck surgery? Otolaryngol Head Neck Surg 2016;155:740-742. 14. Schneider AL, Deig CR, Prasad KG, Nelson BG, Mantravadi AV, Brigance JS, Langer MP, McDonald MW, Johnstone PA, Moore MG. Ability of the National Surgical Quality Improvement Program Risk Calculator to predict complications following total laryngectomy. JAMA Otolaryngol Head Neck Surg 2016;142:972-979. 15. Arce K, Moore EJ, Lohse CM, Reiland MD, Yetzer JG, Ettinger KS. The American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator does not accurately predict risk of 30-day complications among patients undergoing microvascular head and neck reconstruction. J Oral Maxillofac Surg 2016;74:1850-1858. 16. Teoh D, Halloway RN, Heim J, Vogel RI, Rivard C. Evaluation of the American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator in gynecologic oncology patients undergoing minimally invasive surgery. J Minim Invasive Gynecol 2017;24:48-54.
12 Page 12 of 19
AJC-D-17-01624
17. Rivard C, Nahum R, Slagle E, Duininck M, Isaksson Vogel R, Teoh D. Evaluation of the performance of the ACS NSQIP surgical risk calculator in gynecologic oncology patients undergoing laparotomy. Gynecol Oncol 2016;141:281-286. 18. Szender JB, Frederick PJ, Eng KH, Akers SN, Lele SB, Odunsi K. Evaluation 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 American College of Surgeons NSQIP surgical risk calculator identify patients at risk of complications following microsurgical breast reconstruction? 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.
13 Page 13 of 19
AJC-D-17-01624
Figure 1. Postoperative cardiac complications based on predicted risk 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
14 Page 14 of 19
AJC-D-17-01624
Table 1. Cardiac Risk Calculators Revised Cardiac Risk Index (4) MI or Cardiac Arrest Calculator (MICA) (7) (MI/Cardiac Arrest, complete heart block, pulmonary edema during admission)
ACS NSQIP Surgical Risk Calculator (ACS-SRC) (8)
(MI/Cardiac Arrest within 30 days (MI/Cardiac Arrest within 30 days after surgery) after surgery)
High-risk surgery (3 categories)
Type of surgery (21 categories)
Surgical procedure (CPT codes)
Ischemic heart disease
Age
Age group
Congestive heart failure
Functional status
Functional status
Cerebrovascular disease
ASA class
ASA class
Renal insufficiency (Cr>1.5mg/dl)
Acute renal failure
Renal insufficiency (Cr>2mg/dl) Diabetes treated with insulin
Diabetes on oral meds or insulin Dialysis
Reconstructed-RCRI (5) (MI/Cardiac Arrest, complete heart block, pulmonary edema during admission)
Congestive heart failure (<30 days)
High-risk surgery (3 categories)
Dyspnea
Ischemic heart disease
Smoker (within past year)
Congestive heart failure
Severe COPD
Cerebrovascular disease
Ventilator dependent
Renal insufficiency (GFR<30cc/min)
Sepsis (within 48 hours) Disseminated cancer Hypertension requiring meds Wound class
Sex Steroid use (chronic) Ascites (within 30 days) BMI class
ACS-SRC-American College of Surgeons surgical risk calculator, ASA-American Society of Anesthesiology, BMI-body mass index, COPD-chronic obstructive pulmonary disease, GFR-
15 Page 15 of 19
AJC-D-17-01624 glomerular filtration rate, MI-myocardial infarction, MICA-myocardial infarction or cardiac arrest, RCRI-Revised Cardiac Risk Index, R-RCRI-reconstructed revised cardiac risk index
Table 2. Patient characteristics (n=663) Age (years, mean ± sd)
60.8 ± 14
Men
326 (49.2 %)
Race
Hispanic
193 (29.1 %)
White
376 (56.7 %)
Black
84 (12.7 %)
Other Smoker
10 (1.5 %)
Never
345 (52 %)
Former
258 (38.9 %)
Current Body Mass Index, mean ± sd (kg/m2)
60 (9.1 %) 29.7 ± 7.3
Hypertension
370 (55.8 %)
Creatinine (mg/dl), mean ± sd
0.98 ± 7.3
Glomerular filtration rate (GFR) (ml/min/m2), mean ±sd Chronic kidney disease (GFR < 60 ml/min/m2)
68.13 ± 18
Creatinine >1.5 mg/dl)
77 (11.6 %) 41 (6.4%)
Diabetes mellitus
No
547 (82.5 %)
Oral treatment
85 (12.8 %)
Insulin
15 (2.3 %)
Combo Prior stroke/Transient ischemic attack
16 (2.4 %)
Prior ischemic heart disease
74 (11.2 %)
Heart failure
15 (2.3 %)
Chronic obstructive pulmonary inhalers Obstructive sleep apnea
disease
26 (3.9 %)
on
Active cancer
33 (5 %) 84 (12.6 %) 330 (49.8 %)
American Society of Anesthesiology classification
1
76 (11.5%)
2
400 (60.3%) 16 Page 16 of 19
AJC-D-17-01624 3/4
187 (28.2%)
Functional capacity >4 METABOLIC EQUIVALENTS
548 (82.7%)
Table 3. Type of surgery
Otolaryngology (ENT) o
Head and neck cancer
77 (11.6 %)
o
Thyroid
53 (8.1 %)
o
Sinus surgery
33 (5 %)
Orthopedic o
Total joint replacement
100 %)
o
Spine surgery
87 (13.1 %)
Thoracic
(15.1
73 (11 %)
Urologic o
Kidney
50 (7.5 %)
o
Radical prostatectomy
50 (7.5 %)
o
Radical cystectomy
31 (4.7 %)
Breast
51 (7.7 %)
Sleeve gastrectomy
58 (8.7 %)
Procedural risk*
High risk (<1%)
104 %)
(15.7
Intermediate risk (1- 417 %) 5%)
(62.9
Low risk (>5%)
142 %)
(21.4
*Note: 2007 guidelines listed 3 risk classes of procedures; 2014 guidelines classify the 17 Page 17 of 19
AJC-D-17-01624 combined patient and surgical risk as low (<1%) or elevated (>1%) risk
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%)
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%)
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%)
All cardiac events in hospital Major cardiac events 30-day
*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-myocardial infarction or cardiac arrest, RCRI-Revised Cardiac Risk Index, R-RCRI-Reconstructed revised cardiac risk index
Table 5. Receiver operating characteristic (ROC) curve values for each risk calculator for predicting in-hospital and 30-day postop cardiac complications.
RCRI
R-RCRI
MICA
ACS SRC
18 Page 18 of 19
AJC-D-17-01624
Original Publication Results All Cardiac Complications* in-hospital (CC-IH) Major Cardiac Complications** 30-day (MCC-30d)
.76-.81
.79
-
-
-
-
.87-.88
>.80
.85
.87
.76
.92
.78
.80
.78
.89
.55
.56
.94
.77
Current Study Results All Cardiac Complications in-hospital (CC-IH) All Cardiac Complications 30-day (CC-30day) Major Cardiac Complications 30-day (MCC-30day)
*Myocardial infarction, cardiac arrest, pulmonary edema, or complete heart block. **Myocardial infarction or cardiac arrest. 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
19 Page 19 of 19