Introduction

Perioperative myocardial infarction (PMI) is a leading cause of postoperative morbidity and mortality in patients undergoing noncardiac surgery. Although it appears that its incidence and associated mortality rate have declined substantially over the past 10 to 15 years, likely due to improvements in preoperative risk stratification, perioperative management, and prophylaxis (e.g., beta-blockers and other sympatholytic strategies), in aggregate it remains a costly and largely preventable complication. Prior reviews have estimated associated costs in the billions of dollars from resources consumed and adverse outcomes. However, these estimates are poorly supported by hard data, and as of yet, no definitive large-scale prospective economic analyses have been reported.

Options/Therapies

A variety of diagnostic approaches are available for detecting myocardial infarction (MI). The strengths and limitations of the most commonly used modalities are presented in Table 57-1 . Older enzymes previously used to detect MI, including total creatine kinase (CK) (without MB fractionation), lactate dehydrogenase isoenzymes, and glutamic-oxaloacetic transaminase, are no longer recommended for clinical use because of their poor specificity.

Evidence

Studies dating back to the 1950s have reported that PMIs tended to occur with a peak incidence several days after surgery (postoperative days 2 and 3). Half were of the Q wave variety, and the remainder were non-Q wave; they rarely caused classic chest pain (although other associated cardiac signs, such as pulmonary edema, reduction in cardiac output, new ventricular dysrhythmias were common); and the associated mortality rate was high, averaging 50%. Patients undergoing vascular surgery or those with prior MI were at highest risk with incidences exceeding 5% and in some subgroups (e.g., high-risk vascular surgery) up to 20%. Patients sustaining PMI have been shown to have a substantially elevated long-term cardiovascular mortality rate over the first 1 to 2 years after surgery. More recent reports, in general, have reported lower rates of PMI, and a temporal shift in the peak incidence has been seen earlier, closer to the first postoperative day, or, in some studies, on the night of surgery. A distinct predominance of non-Q-wave MIs are reported, and the associated short-term mortality rate is appreciably lower, although long-term mortality and morbidity rates remain higher than in the non-MI population.

In the mid to late 1990s, a major shift occurred in the classic paradigms for diagnosing infarction. The rise to prominence of the troponins (cardiac structural protein markers with high sensitivity and of particular interest perioperatively) with nearly 100% specificity has radically changed cardiology practices and the epidemiologic implications of this diagnosis. Much of this is based primarily on clinical studies in patients with acute coronary syndromes (ACSs), where the need for rapid decision making regarding thrombolysis and revascularization strategies is critical. Several large studies of ACS patients support the clinical efficacy of troponin over the previous gold standard, the less specific cytoplasmic enzyme CK (and its MB fraction). Older studies used CK-MB elevations (determined by a mass assay that supplanted older activity-based assays) usually exceeding 5% of the total as diagnostic of MI when accompanied by at least one of the two following signs or symptoms: associated chest pain or electrocardiogram (ECG) changes (Q-wave or ST-T changes) as defined by the World Health Organization (WHO). These WHO criteria have been used in epidemiologic studies evaluating temporal patterns in coronary artery disease (CAD), and, as such, altering them has substantial implications. Perioperatively, it has long been appreciated that the low specificity of total CK mass (due to muscle injury) and even the CK-MB fraction (due to gene expression in injured muscle), a lack of classic chest pain (attributed in part to analgesic use, although not completely explained), and problems with ECG diagnosis (including sensitivity/specificity issues due to high resting sympathetic tone, changes in electrolyte and acid–base status, and patients with abnormal resting baseline ECGs) greatly complicated coding of MI using standard criteria. Despite these difficulties, it is important to understand that nearly all the well-accepted studies of clinical risk stratification are based, at least in part, on the diagnosis of PMI using adaptations of WHO critieria.

In late fall 2007, shortly after the official release of the updated 2007 American Heart Association/American College of Cardiology (AHA/ACC) perioperative guidelines, the Joint European Society of Cardiology/American College of Cardiology Foundation/American Heart Association/World Heart Federation (ESC/ACCF/AHA/WHF) Task Force for the Redefinition of Myocardial Infarction released its extensive document providing a long-awaited “universal definition of myocardial infarction.” This task force essentially updated a widely cited and influential prior report that evaluated the changing diagnosis of MI given the rapidly expanding use of troponins in the late 1990s. The earlier report outlined recommendations for two specific categories: (1) acute, evolving, or recent MI and (2) established MI ( Box 57-1 ), specifically incorporating use of either troponin I (TnI) or troponin T (TnT), criteria that have in some instances dramatically increased sensitivity in the diagnosis of MI in ACS patients while appearing to maintain specificity. The updated ACC/AHA Guidelines for the Management of Patients with Unstable Angina/Non-ST Elevation Myocardial Infarction used these criteria, defining necrosis as elevation of troponin above the 99th percentile of normal and infarction as the latter along with a clinical finding such as ischemic ST- and T-wave changes, new left bundle branch block, new Q waves, percutaneous coronary intervention (PCI)–related marker elevation, or imaging showing a new loss of myocardium. Although these guidelines state that CK-MB and myoglobin may be useful for diagnosis of early infarct extension or periprocedural MI, it is likely that introduction of more sensitive TnI assays now commercially available will eventually supplant this recommendation. A major change in the new ESC universal guidelines is adoption of a clinical classification system for different types of MI into five major types: type 1, spontaneous MI related to ischemia due to a primary coronary event; type 2, MI secondary to ischemia due to increased demand or decreased supply; type 3, sudden cardiac death; type 4a, MI associated with PCI; type 4b, MI associated with coronary stent thrombosis; and type 5, MI associated with coronary artery bypass graft (CABG). The new definitions for diagnosis are presented in Box 57-1 .

Despite the initial enthusiasm resulting from the widespread availability of TnI, it was rapidly appreciated by clinicians and laboratory managers alike that there was substantial variability in the levels of detection and variability of measurement (coefficient of variation) between different vendors. This has prompted substantial ongoing efforts toward standardization. In 2007, the Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction and the National Academy of Clinical Biochemistry recommended that a TnI elevation greater than the 99th percentile of a normal reference range in the first 24 hours after the clinical event define cardiac injury; this included a coefficient of variation (CV) of 10% or less for the individual assay used. In addition, a rise or fall of the TnI level confirms acute cardiac injury because multiple chronic conditions may have an elevated TnI. Newer “high-sensitivity” TnI assays have dramatically lowered the limits of detection and thus lowered diagnostic levels. The clinical implications of these high-sensitivity assays include earlier detection and intervention but also include the risk of lower specificity with a higher false-positive rate. More recently, in 2010 the Biochemistry Subcommittee of the Joint ESC/ACCF/AHA/WHF Task Force recommended that the 99th percentile of a normal reference range be adopted for assays regardless of whether the CV is less than 10%. This change dramatically lowered the threshold for TnI diagnostic levels, therefore potentially increasing the number of patients receiving a diagnosis of MI. It is not clear that the patients now included with this lower threshold will benefit from improved outcomes based on interventions or management changes because prospective studies are lacking. Because troponin T is only available from one vendor, variability is not an issue.

Contemporary studies evaluating the efficacy of the troponins and CK-MB in detection of PMI are presented in Table 57-2 . In general, they note a higher specificity of the troponins over CK-MB (although conclusive demonstration of significant differences in sensitivity for MI are limited) and an apparent correlation of troponin leakage with either short- or intermediate-term outcomes (although not conclusively in all studies). The low outcome rates of most of the single-center studies limit statistical power; thus the positive predictive values of most markers studied is very limited.

TABLE 57-2

Contemporary Studies Evaluating Biochemical Markers of Perioperative Myocardial Infarction

Reference	Cohort	Variables	Gold Standard	Perioperative Findings	Mortality/Long-Term	Comments
Adams (1994)	108 patients, vascular or spine surgery	ECG, total CK, CK-MB, cTnI	New akinesia or dyskinesia on postoperative TTE	Eight patients MI; sensitivity: cTnI 100% versus CK-MB 75%; specificity: cTnI 99% versus CK-MB 81%; CK-MB/total CK > 2.5: sensitivity 63%	Three deaths, all with elevated cTnI; perioperative FU only	First major study to evaluate perioperative use of cTnI
Lee (1996)	1175 NCS patients age >50	ECG; total CK, CK-MB, cTnT	CK, CK-MB, and ECG changes	17 patients MI; cTnT (>0.1 ng/mL); sensitivity: 87%; specificity: 84%; ROC analysis for MI: no difference CK-MB versus cTnT; ROC analysis for complications: cTnT superior	One sudden death with no elevation of either marker; perioperative FU only	cTnT very low PPV, 90% of patients with elevations without complications
Lopez-Jimenez (1997)	772 NCS patients, age >50	Same as Lee (1996)	Same as Lee (1996); cTnT >0.1 ng/mL postoperatively as risk factor for long-term outcome	12% of cohort had cTnT elevation postoperatively; higher rates of postoperative CHF and new arrhythmias	2.5% had cardiac outcomes by 6 mo; PPV, 9%; RR, 5.4; CK-MB not correlated with outcome	cTnT independent predictor of 6-mo cardiac outcomes
Metzler (1997)	67 patients, known CAD or risk factors, vascular and other NCS	ECG, cTnT, CK-MB, cTnI for patients with elevated cTnT	CK-MB >12 IU/L and Q waves	13 patients elevated cTnT and cTnI; earlier rise in cTnI; CTnT >0.6 ng/mL; PPV, 87%; NPV, 98%; CK-MB elevated in 14 patients (seven patients discordant)	No perioperative deaths; perioperative FU only	Favor cTnT with cutoff value of 0.6 ng/mL
Badner (1998)	323 NCS patients, age >50, known CAD	ECG, total CK, CK-MB, cTnT	Total CK >174 U/L and 2 of CK-MB >5%, new Q waves, cTnI >0.2 mcg/L, (+)pyrophosphate scan	18 patients with MI, 14 on POD 0-1, use of cTnT alone would double MIs	1-yr FU: two of 15 MI patients death or unstable angina	cTnT not used in first 92 patients, lower rate of long-term complications than other studies
Neill (2000)	80 vascular or orthopedic patients	Ambulatory ST monitoring, CK-MB, cTnI, cTnT	CK-MB >5 mcg/L and troponins >1 mcg/L, ECG changes	cTnT and I specificity for major complications 96%/97%, sensitivity 29%/43%	3-mo FU: cTnT best correlated with complications	No correlation of serum markers with ST-segment ischemia
Godet (2000)	329 vascular patients	cTnI	ST depression > 2 days or new Q wave or cTnI > 1.5 ng/mL	13 patients with cardiac complications; peak cTnI POD 1; 27 patients cTnI > 1.5 ng/mL; cTnI > 0.54 ng/mL; sensitivity, 75%; specificity, 89%	1-yr FU; nine patients (3%) with cardiac complications	1-yr FU: no correlation with cTnI
Haggart (2001)	59 vascular patients; 24 emergent	cTnI	WHO criteria	Elective: 10/35 cTnI detected, no CK-MB >5%; emergent: 14/24 cTnI detected, four CK-MB >5%	Perioperative FU only: no deaths elective group; eight deaths emergent group; three cTnI elevated	CK-MB low sensitivity
Jules-Elysee (2001)	85 patients CAD or risk factors, orthopedic surgery	CK-MB, cTnI	cTnI > 3.1 ng/mL and CK-MB index > 3.0	11 patients (+)CK-MB; five of 11 patients (+)cTnI; all others (−)cTnI; all (−)cTnI patients had uneventful course	No deaths; perioperative FU only	cTnI better specificity
Kim (2002)	229 vascular patients	cTnI	WHO criteria	Peak cTnI > 1.5 ng/mL: 12% postoperatively; two of nine ESRD patients (+)cTnI	OR, 5.9 cTnI > 1.5 ng/mL for 6-mo mortality; OR, 27.1 for MI; dose-response relation	Diabetes only preoperative predictor of cTnI elevation
Le Manach (2005)	1316 vascular patients	cTnI	Abnormal cTnI > 0.2-0.5 ng/mL; PMI cTnI > 1.5 ng/mL	Abnormal cTnI (14%), PMI (5%)	Inhospital mortality: early MI, 24%; delayed MI, 21%; abnormal, 7%; normal, 3%	Early MI: increase in cTnI less than 24 hr, delayed MI > 24-hr period of increased cTnI
Domanski (2011)	Meta-analysis including seven studies with 18,908 patients, CABG	cTnI, CK-MB	Enzyme elevation	Abnormal cTnI or CK-MB	Increased CK-MB or troponin ratio after CABG: increased intermediate- and long-term mortality	Enzyme ratio: peak/upper limit of normal
Levy (2011)	Meta-analysis including 14 studies with 3318 patients, NCS	cTnT, cTnI, CK-MB	Enzyme elevation	Abnormal cTnI or cTnT	459 deaths at 1-yr follow-up, increased troponin postoperatively is an independent predictor of mortality	Various troponin thresholds used in studies analyzed
VISION study investigators (2012)	15,133 NCS patients; age >45	cTnT	Peak cTNT ≥ 0.02 ng/mL	11.6% of patients had peak TnT ≥ 0.02 ng/mL	Peak postoperative TnT associated with 30-day mortality	Higher peak cTnT correlated with earlier mortality

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Is a Preoperative Screening Clinic Cost-Effective? Which Patient Should Have a Preoperative Cardiac Evaluation (Stress Test)? What Is the Optimal Airway Management in Patients Undergoing Gastrointestinal Endoscopy? What Works in a Patient with Acute Respiratory Distress Syndrome? Which Are the Best Techniques for Reducing the Incidence of Postoperative Deep Vein Thrombosis? Aspiration: Is There an Optimal Management Strategy?