CCS Perioperative Guidelines When to order a BNP and What to do with - - PowerPoint PPT Presentation

CCS Perioperative Guidelines When to order a BNP and What to do with a Positive Troponin

• Dr. Vikas Tandon

Associate Professor, Cardiology McMaster University November 1, 2017

Canadian Society of Internal Medicine

Annual Meeting 2017

Toronto, ON

CSIM Annual Meeting 2017

Conflict Disclosures

I have the following conflicts to declare:

Company/Organization Details Advisory Board or equivalent

X

X

Speakers bureau member

X

X

Payment from a commercial organization. (including gifts or other consideration or ‘in kind’ compensation)

X

X

Grant(s) or an honorarium

X

X

Patent for a product referred to or marketed by a commercial organization.

X

X

Investments in a pharmaceutical

• rganization, medical devices company or

communications firm.

X

X

Participating or participated in a clinical trial

McMaster University

Participated in periop research studies including VISION, POISE-2, MANAGE

CSIM Annual Meeting 2017

The following presentation represents the views of the speaker at the time of the presentation. This information is meant for educational purposes, and should not replace other sources

• f information or your medical judgment.

Learning Objectives:

• Understand the importance of perioperative risk assessment
• Review the utility of current risk stratification tools
• Examine the utility of BNP/nt-pro-BNP in the preoperative setting
• Understand the significance of the postoperative troponin elevation and develop

an approach to management

Perioperative Care Congress: Science, Evidence and Practice

Save the date: Perioperative Care Congress 2018 May 11-13, 2018 Toronto, Ontario CANADA Visit our website http://periopcongress.org/

• r follow us on twitter @periopcongress

More information to follow!

• 72 y/o F with significant OA
• Referred for upcoming total knee arthroplasty
• Cardiac risk factors
• DM, HTN, Chol, previous NSTEMI 2003
• Otherwise asymptomatic, N vitals, N labs
• Meds: ASA, Atorvastatin, Coversyl, Bisoprolol
• OR date – July 19, 2016

Case – Mrs. B.W.

• What should be done next?
• 1. Send for cath
• 2. Take pt straight to the OR, no other consult

required

• 3. Cancel surgery – too high risk
• 4. Consider for a perioperative consult by

medicine and/or cardiology teams

Case – Mrs. B.W.

Is the preoperative consult useful?

• Worldwide >200,000,000 major noncardiac

surgical procedures annually

• 1:20 suffer myocardial injury/infarction or cardiac

arrest/death within 30 days

• Perioperative cardiac complications account for

≥1/3 of perioperative deaths

Scope of problem

Is the preoperative consult useful?

Yes!

• 1. Patients:

– ethical obligation to patients to give accurate risk assessment for informed decision making

• 2. Physicians:

– Gauge CV risk to guide management – Further testing if needed – Instructions re: medications – Postop monitoring – Shared care model

Is the preoperative consult useful?

• 1. Clear estimation of risk
• 2. Clear recommendation re: further testing
• 3. Clear recommendations for medications
• 4. Clear direction as to degree of post op monitoring

– i.e. ward bed w tele vs CCU/ICU/Step down bed, trops

• 5. Clear communication of who will do what

Good Pre-op Consults Specify:

• RCRI – most validated; simplest to use

– CAD, stroke, CHF, DM, high risk surgery, Creatinine – Does not take into account emergency surgeries – underestimates cardiac risk by 50%

• NSQIP – likely superior to RCRI

– Requires an online calculator – Underestimates risk as routine troponin screening not done

• All risk scores – will underestimate in >40% pts

– Limited mobility so pts won’t manifest symptoms

Risk Scores

CCS Recommendation

When evaluating cardiac risk, we suggest clinicians use RCRI over other available clinical risk prediction scores Conditional recommendation low-quality evidence

Variables Pts Hx of IHD 1 Hx of CHF 1 Hx of CVD 1 Insulin for diabetes 1 Crt >177 µmol/L 1 High-risk surgery 1 Total RCRI points Risk of MI, cardiac arrest, or death 30 days after surgery 95% CI 3.9% 2.8%-5.4% 1 6.0% 4.9%-7.4% 2 10.1% 8.1%-12.6% ≥3 15.0% 11.1%-20.0%

* based on high-quality external validation studies

Revised Cardiac Risk Index

• Current guidelines:

– Pts with low functional capacity – Pts with risk of MI/death ≥ 1% – When result will change management

• Stress Nuclear and Stress Echo most common

– 9% of adults ≥ age 40 with int/high risk tested

Is Non-Invasive Testing Useful?

• Several studies, mostly small sample size and small

number of events

• Low quality of evidence

– most retrospective, few reported risk adjusted associations

• No study adequately assessed incremental value of

stress tests over well-established perioperative cardiac risk factors (e.g., RCRI)

Pharmacological stress echocardiography and radionuclide imaging

CCS Recommendations

We recommend against performing preoperative exercise stress test, pharmacological stress echocardiography,

• r preoperative radionuclide imaging

to enhance perioperative cardiac risk estimation

Is Cardiac CT Angiography Useful?

• Prospective cohort study

– 12 centers in 8 countries

• Evaluated whether preop CCTA enhances

perioperative risk prediction in 955 at-risk patients

• Physicians were blinded unless LM detected
• Systematic Postop Trop monitoring
• Primary outcome - CV death and nonfatal MI

– 74 patients (7.7%) within 30 days of surgery

VISION CCTA

• Although CCTA findings improve risk

estimation

– for patients who will suffer periop CV death or MI

• CCTA findings are more than 5 X as likely to

lead to inappropriate overestimation of risk

– among patients who will not suffer these

• utcomes

Interpretation of VISION CCTA results

CCS Recommendation

We recommend against performing preoperative coronary CT angiography to enhance perioperative cardiac risk estimation Strong recommendation, moderate-quality evidence

Biomarkers – NT pro-BNP

Individual data M-A of 2179 patients

• 235 suffered death or MI within 30 days after

noncardiac surgery

• Preop NT-proBNP ≥300 ng/l or BNP ≥92 ng/l strongest

independent preop predictor of death/MI – OR, 3.40; 95% CI, 2.57-4.47

• Compared to preop clinical model preop natriuretic

peptide improved risk estimation among patients who did and did not suffer primary outcome

• In sample of 1000 patients overall absolute NRI is 155

patients

Test result Risk estimate 95% CI

NT-proBNP <300 ng/L or BNP <92 mg/L

4.9% 3.9% - 6.1%

NT-proBNP value ≥300 ng/L or BNP ≥92 mg/L

21.8% 19.0% - 24.8%

Risk of death or MI at 30 days after noncardiac surgery, based on patient’s preoperative NT-proBNP

• r BNP

– compared to RCRI, preop NT-proBNP/BNP results improved risk classification in 155 patients in 1000 patient sample – based on risk categories <5%, 5-10%, >10-15%, >15%

NT-proBNP/BNP

• Compared to imaging, NT pro-BNP

– More accurate – Less expensive – Convenient and faster due to availability of point of care NT pro-BNP assays – due to cost differential b/w NT pro-BNP and consult may have role in determining who needs preop consult

Biomarkers – NT pro-BNP

We recommend measuring NT-proBNP or BNP before noncardiac surgery to enhance perioperative cardiac risk estimation in patients ≥65 years of age, 45 to 64 years of age with significant cardiovascular disease, or who have RCRI score ≥1 Strong recommendation, moderate-quality evidence

CCS Recommendation

x

• 72 y/o F with significant OA
• Referred for upcoming total knee arthroplasty
• Cardiac risk factors
• DM, HTN, Chol, previous NSTEMI 2003
• Otherwise asymptomatic, N vitals, N labs
• Meds: ASA, Atorvastatin, Coversyl, Bisoprolol
• OR date – July 19, 2016

Case – Mrs. B.W.

x

Troponin monitoring

• POISE Trial (8351 patients)
• 65% of patients suffering perioperative MI do

not experience ischemic symptoms

• Presence or absence of signs/symptoms does

not change risk 30-day mortality

– symptomatic MI: aOR 4.76 (95% CI, 2.68-8.43) – asymptomatic MI: aOR 4.00 (95% CI, 2.65-6.06)

VISION Study (Botto 2014)

• Prospective international cohort study
• 15,065 in-hospital noncardiac surgery patients
• TnT measured postop days 1,2,3
• MINS Criteria TnT ≥ 0.03 ng/ml due to myocardial ischemia

– death at 30 days: MINS - 9.8%, No MINS - 1.1%

• 84% MINS asymptomatic

– undetected without troponin monitoring

• Asymptomatic perioperative TnT elevations adjudicated as myocardial

injuries due to ischemia – that did not fulfill Universal Definition of MI – were also associated with increased risk of 30-day mortality

– aHR, 3.30; 95% CI, 2.26–4.81

Recommendation

We recommend obtaining daily troponin measurements for 48 to 72 hours after noncardiac surgery in patients with baseline risk >5%* for cardiovascular death or nonfatal MI at 30 days after surgery Strong recommendation, moderate-quality evidence * Patients with an elevated NT-proBNP/BNP measurement before

surgery or, if there is no NT-proBNP/BNP measurement before surgery, in those who have an RCRI score ≥1, age 45 to 64 years with significant cardiovascular disease, or age ≥65 years

Approach to MINS

• Look for and correct physiological

abnormalities

– hypoxia, hypotension, tachycardia (if BP adequate), Hb if <70

• If no signs of bleeding initiate ASA 81 mg daily
• Initiate or intensify Statin therapy

Postoperative management of complications

• ASA and statin in patients suffering myocardial injury

after noncardiac surgery

• Prospective cohort study
• 415 noncardiac surgery patients who suffered postop MI
• ASA and statin at discharge reduced 30-day mortality

– ASA : aOR 0.54 (95% CI, 0.29-0.99) – Statin: aOR 0.26 (95% CI, 0.13-0.54)

Recommendations

We recommend initiation of long-term ASA and statin in patients who suffer myocardial injury or myocardial infarction after noncardiac surgery Strong recommendation, moderate-quality evidence

• 72 y/o F with significant OA
• Referred for upcoming total knee arthroplasty
• Cardiac risk factors
• DM, HTN, Chol, previous NSTEMI 2003
• Otherwise asymptomatic, N vitals, N labs
• Meds: ASA, Atorvastatin, Coversyl, Bisoprolol
• Follow up – 1, 6, 12, (18, 24) months

Case – Mrs. B.W.

Conclusions

1. Current clinical risk scores underestimate risk in substantial proportion of patients

– Revised risk estimations for RCRI in new CCS guidelines – Non-invasive testing probably adds little – CCTA has net overall effect of putting more patients in wrong risk category

2. NT pro-BNP is more accurate, convenient, faster, and less expensive than non-invasive testing 3. Troponin are strong independent predictor of 30-day mortality after noncardiac surgery

– 85% of MINS patients asymptomatic (4TH gen trop) – Up to 93% asymptomatic with hs-trops

4. ASA and Statins reduce 30 day mortality in patients with MINS

CSIM Annual Meeting 2017

Special thanks to Dr. PJ Devereaux

• Scientific Leader, Perioperative Research

Group, PHRI, McMaster University

• VISION, POISE 1, POISE 2
• MANAGE, HIP ATTACK, VISION 2, POISE 3
• Co-Chair, CCS Perioperative Guidelines

Comments and Questions

M-A of dipyridamole stress perfusion prior to vascular surgery

M-A of dipyridamole stress perfusion prior to vascular surgery

Baseline risk = 7%

M-A of dipyridamole stress perfusion prior to vascular surgery

Baseline risk = 7%

M-A of dipyridamole stress perfusion prior to vascular surgery

Baseline risk = 7%

M-A of dipyridamole stress perfusion prior to vascular surgery

Baseline risk = 7%

M-A of dipyridamole stress perfusion prior to vascular surgery

Baseline risk = 7%

M-A of dipyridamole stress perfusion prior to vascular surgery

Baseline risk = 7%

• Limitations:

– Small studies, few events, clinicians not blinded – Almost half used a retrospective design – No evaluation independent prognostic value – Few systematically monitored for MI – None reporting net absolute reclassification

Is Non-Invasive Testing Useful?

Net Absolute Reclassification Index

• how well a new model reclassifies subjects -

either appropriately or inappropriately - as compared to an old model

• i.e. comparison of old model vs. old model +

1 new element

• RCRI alone vs. RCRI + non invasive test

Is Cardiac CT Angiography Useful?

VISION CCTA

• Prospective cohort study

– 12 centers in 8 countries

• Evaluated whether preop CCTA enhances

perioperative risk prediction in 955 at-risk patients

• Physicians were blinded unless LM detected
• Systematic Postop Trop monitoring
• Primary outcome - CV death and nonfatal MI

– 74 patients (7.7%) within 30 days of surgery

Model with CCTA and RCRI - C=0.66

# of Patients HR 95% CI P RCRI scores 1 2 ≥3 320 407 178 50 1.00 1.39 (0.74-2.61) 1.88 (0.94-3.79) 4.02 (1.80-8.98) 0.005

• 0.300

0.076 <0.001 CCTA findings Normal Non-obst Obstructive Extensive obst 81 371 357 146 1.00 1.51 (0.45-5.10) 2.05 (0.62-6.74) 3.76 (1.12-12.62) 0.014

• 0.509

0.238 0.032

Model with CCTA and RCRI - C=0.66

# of Patients HR 95% CI P RCRI scores 1 2 ≥3 320 407 178 50 1.00 1.39 (0.74-2.61) 1.88 (0.94-3.79) 4.02 (1.80-8.98) 0.005

• 0.300

0.076 <0.001 CCTA findings Normal Non-obst Obstructive Extensive obst 81 371 357 146 1.00 1.51 (0.45-5.10) 2.05 (0.62-6.74) 3.76 (1.12-12.62) 0.014

• 0.509

0.238 0.032

NRI for those who had event: 21.6% 95% CI 10.4-32.9) p<0.001 NRI for those who did not have event: -10.7% (-13.9- -7.5) p<0.001 Overall NRI: 11% (-0.73, 22.64), p=0.066

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29 17 pts appropriately reclassified 1 pt inappropriately reclassfied Net = 17-1 =16

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29 17 pts appropriately reclassified 1 pt inappropriately reclassfied Net = 17-1 =16

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29 17 pts appropriately reclassified 1 pt inappropriately reclassfied Net = 17-1 =16

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29 17 pts appropriately reclassified 1 pt inappropriately reclassfied Net = 17-1 =16

Net reclassification index

Models for 30 day probability of CV death and MI Model that included CCTA findings Patients who had an event Patients who did not have an event RCRI

• nly

<5% 5-15% >15% <5% 5-15% >15% <5% 5 10 191 114 5-15% 41 7 47 453 37 >15% 1 10 10 29 17 pts appropriately reclassified 57 pts appropriately reclassified 1 pt inappropriately reclassfied 151 pts inapprop. reclassified Net = 17-1 = 16 Net = 57-151 = -94

Interpretation of VISION CCTA results

• Although CCTA findings improve risk

estimation

– for patients who will suffer periop CV death or MI

• CCTA findings are more than 5 X as likely to

lead to inappropriate overestimation of risk

– among patients who will not suffer these

• utcomes

CCS Recommendation

We recommend against performing preoperative coronary CT angiography to enhance perioperative cardiac risk estimation Strong recommendation, moderate-quality evidence