Abstract

Editor’s Perspective
What We Already Know about This Topic
  • Perioperative hypotension is associated with cardiovascular events in patients having noncardiac surgery, but it is unknown if the severity of preexisting coronary artery disease determines susceptibility to the cardiovascular risks of perioperative hypotension

What This Article Tells Us That Is New
  • In patients who had risk factors for, or known, coronary artery disease and were having noncardiac surgery, perioperative hypotension and degree of coronary artery disease on preoperative coronary computed tomographic angiography were independently associated with perioperative cardiovascular death and myocardial infarction

  • Perioperative hypotension was associated with cardiovascular events regardless of the degree of coronary artery disease on preoperative coronary computed tomographic angiography

Background

Perioperative hypotension is associated with cardiovascular events in patients having noncardiac surgery. It is unknown if the severity of preexisting coronary artery disease determines susceptibility to the cardiovascular risks of perioperative hypotension.

Methods

In this retrospective exploratory analysis of a substudy of an international prospective blinded cohort study, 955 patients 45 yr of age or older with history or risk factors for coronary artery disease underwent coronary computed tomographic angiography before elective inpatient noncardiac surgery. The authors evaluated the potential interaction between angiographic findings and perioperative hypotension (defined as systolic blood pressure less than 90 mmHg for a total of 10 min or more during surgery or for any duration after surgery and for which intervention was initiated) on the composite outcome of time to myocardial infarction or cardiovascular death up to 30 days after surgery. Angiography assessors were blinded to study outcomes; patients, treating clinicians, and outcome assessors were blinded to angiography findings.

Results

Cardiovascular events (myocardial infarction or cardiovascular death within 30 days after surgery) occurred in 7.7% of patients (74/955), including in 2.7% (8/293) without obstructive coronary disease or hypotension compared to 6.7% (21/314) with obstructive coronary disease but no hypotension (hazard ratio, 2.51; 95% CI, 1.11 to 5.66; P = 0.027), 8.8% (14/159) in patients with hypotension but without obstructive coronary disease (hazard ratio, 3.85; 95% CI, 1.62 to 9.19; P = 0.002), and 16.4% (31/189) with obstructive coronary disease and hypotension (hazard ratio, 7.34; 95% CI, 3.37 to 15.96; P < 0.001). Hypotension was independently associated with cardiovascular events (hazard ratio, 3.17; 95% CI, 1.99 to 5.06; P < 0.001). This association remained in patients without obstructive disease and did not differ significantly across degrees of coronary disease (P value for interaction, 0.599).

Conclusions

In patients having noncardiac surgery, perioperative hypotension was associated with cardiovascular events regardless of the degree of coronary artery disease on preoperative coronary computed tomographic angiography.

More than 300 million people have major noncardiac surgery worldwide each year,1  and more than 3 million people suffer a perioperative cardiovascular death or a nonfatal myocardial infarction.2,3  While the pathophysiology of these events is likely multifactorial, they have been consistently associated with intraoperative4–6  and postoperative2,6–8  hypotension. Hypotension may lead to cardiac ischemia through supply–demand mismatch or promotion of coronary artery thrombosis.3  This may occur more commonly in patients with obstructive coronary artery disease than in patients without obstructive coronary artery disease, but this possibility has been difficult to address because many patients without a history of coronary artery disease may have significant anatomical disease without symptoms or history of a previous cardiovascular event.

Our group has previously reported results from 955 patients who completed the Coronary Computed Tomographic Angiography and Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (Coronary CTA VISION) substudy.9  In this study, the degree of preoperative coronary artery disease on coronary computed tomographic angiography was correlated with risk of perioperative myocardial infarction or cardiovascular death.

We undertook the present retrospective exploratory analysis of the Coronary CTA VISION substudy to better understand the relationship between perioperative hypotension and cardiovascular events in patients having noncardiac surgery and inform the pathophysiology of perioperative myocardial infarction. Our hypothesis was that patients who suffer cardiovascular events in the setting of perioperative hypotension may be susceptible to cardiovascular events because they have significant coronary artery disease—even if their disease has not been clinically apparent—while patients without significant disease remain unaffected or are affected much less frequently.

Materials and Methods

The research ethics board at each site approved the protocol before patient recruitment. The study was overseen by an independent data monitoring committee. Written informed consent was given by all patients.

After the main study results were published9  but before the current analyses were performed, a plan for the primary analysis was discussed to arrive at statistical methods and exposure and outcome definitions. A second discussion was held to plan sensitivity analyses after the results of the primary analysis were known. The analysis plan was not defined before data collection, and the study’s sample size was not explicitly intended to provide sufficient power for conducting these analyses. Thus, all analyses reported here—prespecified and post hoc—should be regarded as exploratory. This substudy was approved by the publication committee.

Study Design and Eligibility Criteria

Details of this cohort study were previously described.9,10  Patients were recruited from 12 centers from July 2008 to October 2013 and were eligible to participate if they were 45 yr of age or older, were undergoing elective vascular, orthopedic, thoracic, or abdominal surgery that required at least overnight admission after surgery, and had a history or risk factors for atherosclerotic disease or a history of congestive heart failure. All types of anesthesia were permitted. We excluded patients who were unable to undergo coronary computed tomographic angiography before surgery. Complete eligibility criteria are reported in Supplemental Digital Content 1 (http://links.lww.com/ALN/B884).

Study Procedures

Study personnel obtained data on patients’ baseline characteristics. Patients had cardiac troponin T measurements at 6 to 12 h after surgery and on the first, second, and third days after surgery. An electrocardiogram was obtained immediately after detection of troponin elevation. Study personnel followed patients throughout their hospital stay, reviewed medical records to ensure the study protocol was followed, and noted any outcomes. They interviewed patients or their next of kin by telephone 30 days after surgery and, if the interview indicated that the patient experienced an outcome or had been admitted to a hospital, they obtained the appropriate documentation from the attending physicians.

Outcomes

Our primary outcome was time (in days) from surgery to occurrence of a composite of cardiovascular death or nonfatal myocardial infarction within 30 days after surgery. The diagnosis of myocardial infarction used the third universal definition that required a typical rise of troponin concentration associated with at least one of the following: ischemic signs or symptoms, ischemic changes on electrocardiography, or new imaging abnormalities suggestive of myocardial infarction.11  Full outcome definitions are included in Supplemental Digital Content 1 (http://links.lww.com/ALN/B884).

A panel of clinicians who were blinded to the results of coronary computed tomographic angiography adjudicated the outcomes of cardiovascular death and myocardial infarction. Cases were first adjudicated by an investigator at the local study site. A central adjudicator reviewed each case, and if their assessment disagreed with the site assessment, the case was forwarded to a second central adjudicator. If the second central adjudicator disagreed with the first central adjudicator, the two central adjudicators discussed the case to reach consensus. At any step in the process, additional documentation could be requested from the study site.

Exposures

Coronary Computed Tomographic Angiography.

Expert evaluators (a cardiologist or radiologist with level 3 training in coronary computed tomographic angiography interpretation) read each angiogram using a 17-segment model of the coronary arteries without knowledge of the clinical data.12  Each scan was scored as normal (i.e., no evidence of coronary atherosclerosis), nonobstructive disease (i.e., evidence of at least one coronary artery plaque with less than 50% stenosis), obstructive disease (i.e., at least one coronary artery plaque with a 50% or greater stenosis), or extensive obstructive disease (i.e., 50% or greater stenosis in two coronary arteries including the proximal left anterior descending artery, 50% or greater stenosis in three coronary arteries, or 50% or greater stenosis in the left main coronary artery).10  Patients with previous coronary artery bypass grafting surgery were assessed for the number of unprotected coronary territories (bypass graft with 50% or greater stenosis and native coronary artery with 50% or greater).13  Those with no or one unprotected coronary territory were classified as having obstructive coronary artery disease, and patients with two or three unprotected coronary territories were classified as having extensive obstructive disease. The severity classification was based on the a priori observation outside the perioperative literature that people with previous coronary artery bypass grafting who had no or one unprotected coronary territory were at appreciable and similar risk of cardiovascular events; those with two or three unprotected categories had significantly higher risk of cardiovascular events.13  Each angiography study was assessed by one evaluator. Evaluators rated their confidence in the angiographic assessment on a scale of 1 (least confident) to 7 (most confident).

Patients with a 50% or greater stenosis in the left main coronary artery had the results of coronary computed tomographic angiography reported immediately to their attending physicians and were included in the study if they had noncardiac surgery without preoperative coronary revascularization (six patients). Potentially important incidental noncardiac findings (e.g., pulmonary embolism, cancer) were disclosed immediately after the scan was interpreted. All other patients had their results withheld from the care team until 30 days after surgery.

Blinding of treating physicians was important for the objective of the parent study (i.e., to accurately assess the prognostic capabilities of preoperative coronary computed tomographic angiography). This was consistent with previous studies of noninvasive tests in vascular surgery.14,15  The Coronary Artery Revascularization Prophylaxis trial found no benefit for prophylactic coronary revascularization before noncardiac surgery.16  Thus, the research ethics boards agreed that coronary computed tomographic angiography results were not needed to guide treatment. Patients with hemodynamically significant left main coronary disease were excluded from the Coronary Artery Revascularization Prophylaxis trial; therefore, treating physicians were unblinded to this coronary computed tomographic angiography finding in our study.15 

Perioperative Hypotension.

We studied intraoperative and postoperative hypotension as dichotomous variables. Depending on the analysis, they were combined into a single undifferentiated variable (i.e., occurrence of either intra- or postoperative hypotension) or as two separate variables. Intraoperative hypotension was defined as systolic blood pressure less than 90 mmHg for at least 10 min in total during surgery and for which intervention was initiated (including initiation or intensification of intravenous crystalloids or colloids, use of vasopressors or inotropes, blood transfusion, or intraaortic balloon pump). Intraoperative blood pressures were captured either continuously or every few minutes. We required a cumulative hypotensive duration of 10 min or greater during surgery to avoid capturing durations that may not significantly influence cardiac ischemia. Our previous analysis of the VISION study demonstrated, in a sensitivity analysis, that intraoperative hypotension of duration 1 to 10 min was not associated with the composite of death or vascular events.8  Postoperative hypotension was defined as systolic blood pressure less than 90 mmHg of any duration for which intervention was initiated but must have occurred from immediately after surgery up until the end of postoperative day 3 and on or before the day on which the primary outcome occurred. We relied on routine measurements of postoperative blood pressure. This typically occurs every 4 to 12 h but may be more frequent in the early postoperative period before patients are discharged from the postanesthesia care unit to the ward. We did not include postoperative hypotension that occurred on a day after a nonfatal myocardial infarction because it could not contribute to the event. The exact day for hypotension events occurring on or after postoperative day 4 was not recorded; thus, these few events were not considered in the analyses. We made these decisions before performing these analyses.

Analysis

We described the data using means and SDs for continuous variables and proportions for categorical variables. We presented data within subgroups of hypotension and coronary disease and assessed variables for imbalance among the subgroups using ANOVA for continuous variables and Pearson’s chi-square statistic for categorical variables. We imputed missing data for body mass index, preoperative hemoglobin, postoperative epidural analgesia, and duration of surgery for 68, 12, 1, and 2 patients, respectively, by single stochastic imputation with predictive mean matching17  from a model that included all other variables listed in table 1.

Table 1.

Description of Study Participants Overall and by Subgroups of CCTA Findings and Perioperative Hypotension

Description of Study Participants Overall and by Subgroups of CCTA Findings and Perioperative Hypotension
Description of Study Participants Overall and by Subgroups of CCTA Findings and Perioperative Hypotension

For the primary analysis, we constructed four multivariable Cox proportional hazards models in which the dependent variable was time to a composite of cardiovascular death or nonfatal myocardial infarction; death from other causes was treated as a competing event.18  The independent variables included the classification of coronary computed tomographic angiography findings of coronary artery disease and any perioperative hypotension meeting the above definitions. We collapsed the four categories of coronary computed tomographic angiography findings of coronary artery disease into two categories: no disease or nonobstructive disease versus obstructive disease and extensive obstructive disease. The independent variables in each of the four models from which the primary results were derived are as follows:

  • (1) Any hypotension + coronary computed tomographic angiography findings

  • (2) Any hypotension + coronary computed tomographic angiography findings + any hypotension × coronary computed tomographic angiography findings

  • (3) Intraoperative hypotension + postoperative hypotension + coronary computed tomographic angiography findings

  • (4) Intraoperative hypotension + postoperative hypotension + coronary computed tomographic angiography findings + intraoperative hypotension × coronary computed tomographic angiography findings + postoperative hypotension × coronary computed tomographic angiography findings

Using (1), we estimated the association (not specific to a coronary computed tomographic angiography subgroup) of hypotension with the primary outcome (adjusted for coronary computed tomographic angiography findings by including indicator variables for hypotension and coronary computed tomographic angiography findings in the model simultaneously). In (2), we added the interaction between hypotension and the coronary computed tomographic angiography findings to the independent variables in (1) to estimate subgroup-specific associations for hypotension. In (3), we differentiated between intraoperative and postoperative hypotension as two separate independent variables to estimate average association with the primary outcome (not subgroup-specific). In (4) we estimated coronary computed tomographic angiography–specific associations with the primary outcome by interacting intraoperative and postoperative hypotension with coronary computed tomographic angiography findings. We treated postoperative hypotension as a time-varying covariate.

We conducted several sensitivity analyses using the same approach. In the first sensitivity analysis, we changed the coronary computed tomographic angiography categorization from “no disease or nonobstructive disease versus obstructive disease or extensive obstructive disease” to “no disease, nonobstructive disease, or obstructive disease versus extensive obstructive disease.” In the second sensitivity analysis, we changed the definition of postoperative hypotension to hypotension that occurred on days before (but not including) the day of the outcome. In the third sensitivity analysis, we limited the outcome to nonfatal myocardial infarction and treated any death as a competing event. In the fourth sensitivity analysis, we categorized patients as having obstructive or extensive obstructive coronary artery disease if they had 70% or greater stenosis in place of the 50% or greater stenosis threshold used for the primary analysis. In the fifth sensitivity analysis, we categorized patients as having a complete occlusion (100% stenosis) in at least one of the following: the right coronary artery, the left anterior descending artery, or the left circumflex artery.

Sensitivity Analyses Undertaken in Response to Peer Review

We conducted three sensitivity analyses in response to comments from peer reviewers. In the first of these analyses, we adjusted the models from the primary analysis additionally for variables that were unbalanced at the P < 0.10 significance level, including age, sex, body mass index, history of coronary artery disease, history of peripheral vascular disease, history of hypertension, history of diabetes, requirement of assistance with activities of daily living, preoperative use of statins, ß-blockers, aspirin, and angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, duration of surgery, major vascular and major orthopedic surgery, postoperative use of epidural analgesia, and major bleeding (defined as bleeding associated with receipt of blood products or reoperation for reasons of bleeding). Major bleeding was treated as a time-varying covariate. We chose the P < 0.10 threshold for imbalance to limit the number of uninfluential covariates given the relatively small number of outcome events in our study.

In the second analysis, we repeated the primary analysis but recategorized patients who had previously undergone coronary artery bypass grafting surgery who had no unprotected territories on coronary computed tomographic angiography as having nonobstructive coronary artery disease in place of the categorization in the primary analyses in which they were categorized as having obstructive coronary artery disease. In the third analysis, we repeated the primary analysis but excluded patients with 50% or greater stenosis of the left main coronary artery for whom the treating physicians were unblinded in the study.

We used Stata MP version 14.2 (StataCorp, USA); two-sided P values less than 0.05 denoted statistical significance. We ensured that the proportionality assumption of the Cox proportional hazards model, whenever used, was met using log–log plots and global tests for proportionality (where P < 0.05 would indicate violation of the assumption).

Results

Of the 1093 enrolled patients scheduled to undergo preoperative coronary computed tomographic angiography, 955 were included in the study (fig. 1). Confidence ratings were available for the angiographic assessments of 940 patients. The median rating was 6 (interquartile range, 5 to 7), indicating a high degree of confidence in the assessments. Six patients with suspected left main coronary artery disease had noncardiac surgery without preoperative revascularization and were included in this substudy. Among the included patients, 99.5% completed the 30-day follow-up with the remaining five patients censored at discharge.

Fig. 1.

Flow of patients throughout study. CCTA, coronary computed tomographic angiography.

Fig. 1.

Flow of patients throughout study. CCTA, coronary computed tomographic angiography.

Patient Characteristics

Table 1 summarizes patient characteristics. Results of angiography showed that 81/955 patients (8.5%) had normal coronary arteries, 371/955 (38.8%) had nonobstructive coronary artery disease, 357/955 (37.4%) had obstructive coronary artery disease, and 146/955 (15.3%) had extensive obstructive coronary artery disease. The primary composite outcome of cardiovascular death or myocardial infarction occurred in 74/955 patients (7.7%), of whom 8 (0.8%) experienced cardiovascular death and 71 (7.4%) experienced a myocardial infarction (68 [7.1%] had a nonfatal myocardial infarction). Five patients (0.5%) died of a noncardiovascular cause. Major vascular and orthopedic surgeries were most common; they were performed in 289/955 (29.4%) and 538/955 patients (56.3%), respectively. Major vascular surgeries accounted for 30/74 cardiovascular events (40.5%), and orthopedic surgeries accounted for 36/74 (48.7%) events.

Relationship between Hypotension, Coronary Disease, and Cardiovascular Events

Figure 2 shows the temporal distribution of hypotension and cardiovascular events throughout the perioperative period. Clinically significant perioperative hypotension occurred in 348/955 (36.4%) patients either during the intraoperative period (161/955, 16.9%) or early postoperative period (i.e., up to the end of the third day after surgery) (236/955, 24.7%). Hypotension was consistently associated with the primary outcome (fig. 3), whether undifferentiated (i.e., either intra- or postoperative hypotension; adjusted hazard ratio, 3.17; 95% CI, 1.99 to 5.06; P < 0.001) or differentiated into intraoperative hypotension (adjusted hazard ratio, 2.43; 95% CI, 1.50 to 3.95; P < 0.001) and postoperative hypotension (adjusted hazard ratio, 2.17; 95% CI, 1.35 to 3.49; P = 0.001). Greater burden of coronary artery disease on coronary computed tomographic angiography was also associated with perioperative cardiovascular events (adjusted hazard ratio, 2.05; 95% CI, 1.24 to 3.37; P = 0.005).

Fig. 2.

Timing of perioperative hypotension and cardiovascular events. We considered postoperative day (POD) 1 to be the period after surgery on the day of surgery. One hundred sixty-one patients (16.9%) experienced hypotension during surgery and 252 (26.4%) at least once after surgery. Among the 74 patients who suffered a myocardial infarction (MI) or cardiovascular (CV) death, 68 (91.9%) experienced this before POD 4. The statistical analysis included only hypotension occurring on or before POD 3 because the day of hypotensive episodes occurring on or after POD 4 was not recorded. Only 12 (1.3%) patients had their first episode of postoperative hypotension on or after POD 4. Two hundred thirty-six patients (24.7%) experienced at least one episode of postoperative hypotension on or before the day of their myocardial infarction or cardiac death. Five patients died of noncardiac causes, one on each of POD 6, 9, 22, 25, and 27. OR, operating room.

Fig. 2.

Timing of perioperative hypotension and cardiovascular events. We considered postoperative day (POD) 1 to be the period after surgery on the day of surgery. One hundred sixty-one patients (16.9%) experienced hypotension during surgery and 252 (26.4%) at least once after surgery. Among the 74 patients who suffered a myocardial infarction (MI) or cardiovascular (CV) death, 68 (91.9%) experienced this before POD 4. The statistical analysis included only hypotension occurring on or before POD 3 because the day of hypotensive episodes occurring on or after POD 4 was not recorded. Only 12 (1.3%) patients had their first episode of postoperative hypotension on or after POD 4. Two hundred thirty-six patients (24.7%) experienced at least one episode of postoperative hypotension on or before the day of their myocardial infarction or cardiac death. Five patients died of noncardiac causes, one on each of POD 6, 9, 22, 25, and 27. OR, operating room.

Fig. 3.

Relationship between perioperative hypotension and cardiovascular (CV) events by coronary computed tomographic angiography findings. Interaction values reflect statistical comparisons of the association between perioperative hypotension and cardiovascular events observed in patients with no or nonobstructive disease compared to the relationship observed in patients with obstructive or extensive obstructive disease. The association between postoperative hypotension and cardiovascular events in the subgroup of patients with obstructive or extensive obstructive disease was not significant, but the test for interaction did not suggest that the association in this subgroup differed significantly from the association in the subgroup of patients with no or nonobstructive disease. We interpret this result as reflective of imprecision in a subgroup estimate and not as evidence that postoperative hypotension is not associated with cardiovascular events in people with obstructive coronary disease. aHR, adjusted hazard ratio.

Fig. 3.

Relationship between perioperative hypotension and cardiovascular (CV) events by coronary computed tomographic angiography findings. Interaction values reflect statistical comparisons of the association between perioperative hypotension and cardiovascular events observed in patients with no or nonobstructive disease compared to the relationship observed in patients with obstructive or extensive obstructive disease. The association between postoperative hypotension and cardiovascular events in the subgroup of patients with obstructive or extensive obstructive disease was not significant, but the test for interaction did not suggest that the association in this subgroup differed significantly from the association in the subgroup of patients with no or nonobstructive disease. We interpret this result as reflective of imprecision in a subgroup estimate and not as evidence that postoperative hypotension is not associated with cardiovascular events in people with obstructive coronary disease. aHR, adjusted hazard ratio.

Relationship between Hypotension and Cardiovascular Events by Degree of Coronary Disease

Figure 4 shows Kaplan–Meier curves for event-free survival over 30 postoperative days. The primary outcome occurred in 2.7% (8/293) without obstructive coronary artery disease or hypotension compared to 6.7% (21/314) with obstructive coronary artery disease but no hypotension (hazard ratio, 2.51; 95% CI, 1.11 to 5.66; P = 0.027), 8.8% (14/159) in patients with hypotension but without obstructive coronary artery disease (hazard ratio, 3.85; 95% CI, 1.62 to 9.19; P = 0.002), and 16.4% (31/189) with obstructive coronary artery disease and hypotension (hazard ratio, 7.34; 95% CI, 3.37 to 15.96; P < 0.001).

Fig. 4.

Kaplan–Meier event-free survival estimates across four subgroups. CAD, coronary artery disease.

Fig. 4.

Kaplan–Meier event-free survival estimates across four subgroups. CAD, coronary artery disease.

Hypotension and the presence of obstructive coronary artery disease each made independent contributions to the risk of cardiovascular events. Patients who experienced hypotension and had obstructive coronary artery disease bear the sum of those contributions, but there did not appear to be a synergistic effect where hypotension would make an even stronger contribution in patients with obstructive coronary artery disease than it does in patients without obstructive coronary artery disease. This is confirmed by the subgroup analyses and tests of interaction in figure 3. None of these analyses suggested that the effect of hypotension on cardiovascular events was significantly weaker in patients with less coronary artery disease than in those with more coronary artery disease (P values for interaction greater than 0.05; fig. 3). These results were qualitatively consistent in all sensitivity analyses (Supplemental Digital Content 1, http://links.lww.com/ALN/B884, fig. S1 and tables S1 to S8).

Discussion

Principal Findings

Perioperative hypotension and burden of coronary artery disease on coronary computed tomographic angiography were both independently associated with perioperative cardiovascular events in this study of 955 patients who had a history or risk factors for coronary artery disease. These effects were additive: patients with more coronary artery disease who experience hypotension were more likely to experience cardiovascular events than patients with either risk factor alone, just as they would be at higher risk of cardiovascular events if they carry other known risk factors. The principal finding in our study was the absence of evidence for a large multiplicative effect (i.e., interaction) where patients with more coronary artery disease would be at a disproportionately higher risk of cardiovascular events if they experience hypotension compared to what would be expected by summing the independent effects of coronary artery disease and hypotension. In our study, the increase in risk was consistent with the sum of independent effects. Therefore, there was insufficient evidence that perioperative hypotension may have less deleterious cardiovascular effects in patients with a lesser burden of coronary artery disease than in patients with a greater burden of coronary artery disease. These data support efforts for the prevention, monitoring, and treatment of perioperative hypotension regardless of the presence or absence of significant coronary artery disease.

Our Findings in the Context of Previous Research

Multiple large studies have reported associations between various definitions of hypotension and cardiovascular events or mortality.4–8  Recent analysis of the Perioperative Ischemic Evaluation (POISE)-2 trial found intraoperative hypotension to be associated with a composite of perioperative myocardial infarction or mortality (adjusted odds ratio, 1.08; 98.3% CI, 1.03 to 1.12) and postoperative hypotension on the remainder of the day of surgery (adjusted odds ratio, 1.03; 98.3% CI, 1.01 to 1.05) for every 10 min of hypotension. Postoperative hypotension (defined similarly to our study) was also associated with these outcomes (adjusted odds ratio, 2.83; 98.3% CI, 1.26 to 6.35).

We did not determine whether the etiology of myocardial infarction events in this study was supply–demand mismatch or coronary thrombosis. Members of our group recently reported a study using optical coherence tomography intravascular imaging which showed coronary artery thrombus in only 4 of 30 patients (13.3%) with perioperative myocardial infarction compared to 20 of 30 patients (66.7%) with nonoperative myocardial infarction.19  Therefore, we expect the majority of events in our study to to have resulted from supply–demand mismatch.

Strengths

Strengths of this work included nearly complete follow-up in a prospective international study, direct assessment of the degree of coronary artery disease by coronary computed tomographic angiography, blinding of coronary computed tomographic angiography assessors to study outcomes, and blinding of treating clinicians and outcome assessors to coronary computed tomographic angiography findings. We systematically measured postoperative troponins and obtained electrocardiograms to detect myocardial infarctions that might otherwise go undetected in the absence of ischemic symptoms.

Limitations

This secondary analysis was not planned before data collection and should be regarded as exploratory. Data regarding postoperative hypotension were based on routine clinical assessment and documentation and were not sufficiently granular to assess the impact of different degrees or durations of postoperative hypotension. We did not record diastolic blood pressure, which may be more relevant to coronary perfusion than systolic blood pressure. However, we typically observe that diastolic pressure falls in concert with systolic pressure, and systolic hypotension was associated with cardiovascular events in this and other studies.7,8 

It is possible that, in many patients with coronary computed tomographic angiography findings of nonobstructive disease, this degree of disease is still sufficient to make them similarly susceptible to the risks associated with perioperative hypotension as patients with more disease. Only 8.5% of patients (81/955) had no coronary artery disease on coronary computed tomographic angiography, and only three of them experienced the primary outcome, precluding reliable analysis of this subgroup.

We did not evaluate the effects of hypotension on perioperative stroke. It is possible that patients with less coronary disease on coronary computed tomographic angiography have less cerebrovascular atherosclerotic plaque burden and are less likely to suffer a stroke in the perioperative period than patients with more coronary disease on coronary computed tomographic angiography. Only nine patients suffered a stroke within 30 days of surgery in the current study, precluding reliable analysis.

Although this study included nearly 1,000 patients, the number of outcome events limited statistical power to evaluate interaction effects and the ability to control for a large number of potential confounders. In a sensitivity analysis, we adjusted for bleeding, type and duration of surgery, and unbalanced preoperative patient characteristics. We also controlled for preoperative use (within 7 days before surgery) and initiation or withdrawal (within 24 h before surgery) of medications previously associated with perioperative cardiovascular events2,8,20  or major bleeding21  in the POISE, POISE-2, and VISION studies (including angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, ß-blockers, aspirin, and statins). However, postoperative management of these and other medications was not documented with sufficient detail to determine when these agents were initiated, withdrawn, or reinstituted after surgery. Controlling for the variables we were able to control did not appreciably affect the results.

Our analyses cannot exclude subgroup differences in the magnitude of effect from hypotension but strongly suggest that it has an important effect even in people without obstructive coronary artery disease. The best estimate for the interaction hazard ratio for cardiovascular events attributable to any hypotension in patients with obstructive coronary artery disease compared to hypotension in patients without obstructive coronary artery disease was 0.76 and not statistically significantly different from 1 (no interaction). However, the 95% CI ranged from 0.27 to 2.13, suggesting that the hazard ratio relating hypotension to cardiovascular events may plausibly range from 73% smaller to 113% larger in people with obstructive coronary artery disease compared to people without obstructive coronary artery disease. Importantly, the best estimate of the hazard ratio for cardiovascular events attributable to hypotension in patients without obstructive coronary artery disease was 3.85, and the 95% CI ranged from 1.62 to 9.19, suggesting that—if the relationship is causal—hypotension increases the risk of cardiovascular events by at least 62% in relative terms, even in the absence of obstructive angiographic disease.

Conclusions

In an international blinded prospective cohort study of 955 patients who had risk factors for, or known, coronary artery disease and of whom 91.5% had coronary computed tomographic angiography evidence of nonobstructive, obstructive, or extensive obstructive coronary artery disease, perioperative hypotension and degree of coronary artery disease were independently associated with an increased risk of perioperative cardiovascular death and myocardial infarction. Although we undertook post hoc exploratory analyses and the estimates of association were imprecise, there was insufficient evidence that patients with less coronary artery disease on coronary computed tomographic angiography were less susceptible to the deleterious cardiovascular effects of hypotension as defined and captured by the VISION protocol. Perioperative hypotension was consistently associated with cardiovascular events in patients without obstructive coronary artery disease on coronary computed tomographic angiography. We acknowledge that inclusion of more complete physiologic data along with greater granularity of clinician-initiated interventions in response to them as well as specific details of postoperative medical and analgesic management that may impact on hemodynamics may lead to different conclusions and is necessary before widespread generalization of these results into possible clinical performance measures.

Acknowledgments

This study was coordinated by the Clinical Advances Through Research and Information Translation (CLARITY) project office and the Population Health Research Institute (PHRI) at Hamilton Health Sciences, McMaster University, Hamilton, Canada.

Research Support

The study was funded by the Canadian Institutes of Health Research (Ottawa, Canada), the Hamilton Health Sciences New Investigator Fund (Hamilton, Canada), the McMaster University Division of Cardiology (Hamilton, Canada), the University of Western Ontario Department of Radiology (London, Canada), the University of Western Ontario Division of General Internal Medicine (London, Canada), the Washington University Institute of Clinical and Translational Sciences (St. Louis, Missouri), the American Heart Association (Dallas, Texas; grant No. 09CRP2240001), and the Research Grant Council (Hong Kong Special Administrative Region; grant No. GRF461412). The study funders had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.

Competing Interests

Dr. Devereaux is part of a group that has a policy of not accepting honoraria or other payments from industry for their own personal financial gain. Members of this group accept honoraria or other payments from industry to support research endeavors and for reimbursement of costs to participate in meetings such as scientific or advisory committee meetings. Based on study questions Dr. Devereaux originated and grants he wrote, he has received grants from Abbott Diagnostics (Abbott Park, Illinois), AstraZeneca (Cambridge, United Kingdom), Bayer (Leverkusen, Germany), Boehringer Ingelheim (Ingelheim, Germany), Bristol-Myers Squibb (New York, New York), Covidien (Dublin, Ireland), Philips Healthcare (Andover, Massachusetts), Stryker (Waterdown, Ontario, Canada), and Roche Diagnostics (Mannheim, Germany). Dr. Devereaux has also participated in an advisory boarding meeting for GlaxoSmithKline (London, United Kingdom) and an expert panel meeting for AstraZeneca. Dr. Chow holds the Saul and Edna Goldfarb Chair in Cardiac Imaging Research (University of Ottawa, Ottawa, Canada). He receives research support from CV Diagnostix (Kanata, Canada) and educational support from TeraRecon Inc. (Foster City, California). The other authors declare no competing interests.

References

References
1.
Weiser
TG
,
Haynes
AB
,
Molina
G
,
Lipsitz
SR
,
Esquivel
MM
,
Uribe-Leitz
T
,
Fu
R
,
Azad
T
,
Chao
TE
,
Berry
WR
,
Gawande
A
:
Estimate of the global volume of surgery in 2012: An assessment supporting improved health outcomes.
Lancet (London, England)
2015
;
385
(
suppl
):
S11
2.
Devereaux
PJ
,
Xavier
D
,
Pogue
J
,
Guyatt
G
,
Sigamani
A
,
Garutti
I
,
Leslie
K
,
Rao-Melacini
P
,
Chrolavicius
S
,
Yang
H
,
Macdonald
C
,
Avezum
A
,
Lanthier
L
,
Hu
W
,
Yusuf
S
;
POISE (PeriOperative ISchemic Evaluation) Investigators
:
Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: A cohort study.
Ann Intern Med
2011
;
154
:
523
8
3.
Devereaux
PJ
,
Sessler
DI
:
Cardiac complications in patients undergoing major noncardiac surgery.
N Engl J Med
2015
;
373
:
2258
69
4.
Walsh
M
,
Devereaux
PJ
,
Garg
AX
,
Kurz
A
,
Turan
A
,
Rodseth
RN
,
Cywinski
J
,
Thabane
L
,
Sessler
DI
:
Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: Toward an empirical definition of hypotension.
Anesthesiology
2013
;
119
:
507
15
5.
Monk
TG
,
Bronsert
MR
,
Henderson
WG
,
Mangione
MP
,
Sum-Ping
ST
,
Bentt
DR
,
Nguyen
JD
,
Richman
JS
,
Meguid
RA
,
Hammermeister
KE
:
Association between intraoperative hypotension and hypertension and 30-day postoperative mortality in noncardiac surgery.
Anesthesiology
2015
;
123
:
307
19
6.
Sessler
DI
,
Meyhoff
CS
,
Zimmerman
NM
,
Mao
G
,
Leslie
K
,
Vásquez
SM
,
Balaji
P
,
Alvarez-Garcia
J
,
Cavalcanti
AB
,
Parlow
JL
,
Rahate
PV
,
Seeberger
MD
,
Gossetti
B
,
Walker
SA
,
Premchand
RK
,
Dahl
RM
,
Duceppe
E
,
Rodseth
R
,
Botto
F
,
Devereaux
PJ
:
Period-dependent associations between hypotension during and for four days after noncardiac surgery and a composite of myocardial infarction and death: A substudy of the POISE-2 Trial.
Anesthesiology
2018
;
128
:
317
27
7.
Devereaux
PJ
,
Sessler
DI
,
Leslie
K
,
Kurz
A
,
Mrkobrada
M
,
Alonso-Coello
P
,
Villar
JC
,
Sigamani
A
,
Biccard
BM
,
Meyhoff
CS
,
Parlow
JL
,
Guyatt
G
,
Robinson
A
,
Garg
AX
,
Rodseth
RN
,
Botto
F
,
Lurati Buse
G
,
Xavier
D
,
Chan
MT
,
Tiboni
M
,
Cook
D
,
Kumar
PA
,
Forget
P
,
Malaga
G
,
Fleischmann
E
,
Amir
M
,
Eikelboom
J
,
Mizera
R
,
Torres
D
,
Wang
CY
,
Vanhelder
T
,
Paniagua
P
,
Berwanger
O
,
Srinathan
S
,
Graham
M
,
Pasin
L
,
Le Manach
Y
,
Gao
P
,
Pogue
J
,
Whitlock
R
,
Lamy
A
,
Kearon
C
,
Chow
C
,
Pettit
S
,
Chrolavicius
S
,
Yusuf
S
;
POISE-2 Investigators
:
Clonidine in patients undergoing noncardiac surgery.
N Engl J Med
2014
;
370
:
1504
13
8.
Roshanov
PS
,
Rochwerg
B
,
Patel
A
,
Salehian
O
,
Duceppe
E
,
Belley-Côté
EP
,
Guyatt
GH
,
Sessler
DI
,
Le Manach
Y
,
Borges
FK
,
Tandon
V
,
Worster
A
,
Thompson
A
,
Koshy
M
,
Devereaux
B
,
Spencer
FA
,
Sanders
RD
,
Sloan
EN
,
Morley
EE
,
Paul
J
,
Raymer
KE
,
Punthakee
Z
,
Devereaux
PJ
:
Withholding versus continuing angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers before noncardiac surgery: An analysis of the vascular events in noncardiac surgery patients cohort evaluation prospective cohort.
Anesthesiology
2017
;
126
:
16
27
9.
Sheth
T
,
Chan
M
,
Butler
C
,
Chow
B
,
Tandon
V
,
Nagele
P
,
Mitha
A
,
Mrkobrada
M
,
Szczeklik
W
,
Faridah
Y
,
Biccard
B
,
Stewart
LK
,
Heels-Ansdell
D
,
Devereaux
PJ
;
Coronary Computed Tomographic Angiography and Vascular Events in Noncardiac Surgery Patients Cohort Evaluation Study Investigators
:
Prognostic capabilities of coronary computed tomographic angiography before non-cardiac surgery: Prospective cohort study.
BMJ
2015
;
350
:
h1907
10.
Sheth
T
,
Butler
C
,
Chow
B
,
Chan
MT V.
,
Mitha
A
,
Nagele
P
,
Tandon
V
,
Stewart
L
,
Graham
M
,
Choi
GYS
,
Kisten
T
,
Woodard
PK
,
Crean
A
,
Abdul Aziz
YF
,
Karthikeyan
G
,
Chow
CK
,
Szczeklik
W
,
Markobrada
M
,
Mastracci
T
,
Devereaux
PJ
:
The coronary CT angiography vision protocol: a prospective observational imaging cohort study in patients undergoing non-cardiac surgery.
BMJ Open
2012
;
2
:
e001474
11.
Thygesen
K
,
Alpert
JS
,
Jaffe
AS
,
Simoons
ML
,
Chaitman
BR
,
White
HD
,
Katus
HA
,
Lindahl
B
,
Morrow
DA
,
Clemmensen
PM
,
Johanson
P
,
Hod
H
,
Underwood
R
,
Bax
JJ
,
Bonow
RO
,
Pinto
F
,
Gibbons
RJ
,
Fox
KA
,
Atar
D
,
Newby
LK
,
Galvani
M
,
Hamm
CW
,
Uretsky
BF
,
Steg
PG
,
Wijns
W
,
Bassand
JP
,
Menasché
P
,
Ravkilde
J
,
Ohman
EM
,
Antman
EM
,
Wallentin
LC
,
Armstrong
PW
,
Simoons
ML
,
Januzzi
JL
,
Nieminen
MS
,
Gheorghiade
M
,
Filippatos
G
,
Luepker
RV
,
Fortmann
SP
,
Rosamond
WD
,
Levy
D
,
Wood
D
,
Smith
SC
,
Hu
D
,
Lopez-Sendon
JL
,
Robertson
RM
,
Weaver
D
,
Tendera
M
,
Bove
AA
,
Parkhomenko
AN
,
Vasilieva
EJ
,
Mendis
S
;
Joint ESC/ACCF/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction
:
Third universal definition of myocardial infarction.
Circulation
2012
;
126
:
2020
35
12.
Austen
WG
,
Edwards
JE
,
Frye
RL
,
Gensini
GG
,
Gott
VL
,
Griffith
LS
,
McGoon
DC
,
Murphy
ML
,
Roe
BB
:
A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association.
Circulation
1975
;
51
(
4 suppl
):
5
40
13.
Chow
BJ
,
Ahmed
O
,
Small
G
,
Alghamdi
AA
,
Yam
Y
,
Chen
L
,
Wells
GA
:
Prognostic value of CT angiography in coronary bypass patients.
JACC Cardiovasc Imaging
2011
;
4
:
496
502
14.
Virgilio
C de
,
Toosie
K
,
Elbassir
M
,
Donayre
C
,
Baker
JD
,
Narahara
K
,
Mishkin
F
,
Lewis
RJ
,
Chang
C
,
White
R
,
Mody
FV
:
Dipyridamole-thallium/sestamibi before vascular surgery: A prospective blinded study in moderate-risk patients.
J Vasc Surg
2000
;
32
:
77
89
15.
Vanzetto
G
,
Machecourt
J
,
Blendea
D
,
Fagret
D
,
Borrel
E
,
Magne
JL
,
Gattaz
F
,
Guidicelli
H
:
Additive value of thallium single-photon emission computed tomography myocardial imaging for prediction of perioperative events in clinically selected high cardiac risk patients having abdominal aortic surgery.
Am J Cardiol
1996
;
77
:
143
8
16.
McFalls
EO
,
Ward
HB
,
Moritz
TE
,
Goldman
S
,
Krupski
WC
,
Littooy
F
,
Pierpont
G
,
Santilli
S
,
Rapp
J
,
Hattler
B
,
Shunk
K
,
Jaenicke
C
,
Thottapurathu
L
,
Ellis
N
,
Reda
DJ
,
Henderson
WG
:
Coronary-artery revascularization before elective major vascular surgery.
N Engl J Med
2004
;
351
:
2795
804
17.
Landerman
LR
,
Land
KC
,
Pieper
CF
:
An empirical evaluation of the predictive mean matching method for imputing missing values.
Sociol Methods Res
1997
;
26
:
3
33
18.
Austin
PC
,
Lee
DS
,
Fine
JP
:
Introduction to the analysis of survival data in the presence of competing risks.
Circulation
2016
;
133
:
601
9
19.
Sheth
T
,
Natarajan
MK
,
Hsieh
V
,
Valettas
N
,
Rokoss
M
,
Mehta
S
,
Jolly
S
,
Tandon
V
,
Bezerra
H
,
Devereaux
PJ
:
Incidence of thrombosis in perioperative and non-operative myocardial infarction.
Br J Anaesth
2018
;
120
:
725
33
20.
Berwanger
O
,
Le Manach
Y
,
Suzumura
EA
,
Biccard
B
,
Srinathan
SK
,
Szczeklik
W
,
Santo
JA
,
Santucci
E
,
Cavalcanti
AB
,
Archbold
RA
,
Devereaux
PJ
;
VISION Investigators
:
Association between pre-operative statin use and major cardiovascular complications among patients undergoing non-cardiac surgery: the VISION study.
Eur Heart J
2016
;
37
:
177
85
21.
Devereaux
PJ
,
Mrkobrada
M
,
Sessler
DI
,
Leslie
K
,
Alonso-Coello
P
,
Kurz
A
,
Villar
JC
,
Sigamani
A
,
Biccard
BM
,
Meyhoff
CS
,
Parlow
JL
,
Guyatt
G
,
Robinson
A
,
Garg
AX
,
Rodseth
RN
,
Botto
F
,
Lurati Buse
G
,
Xavier
D
,
Chan
MT
,
Tiboni
M
,
Cook
D
,
Kumar
PA
,
Forget
P
,
Malaga
G
,
Fleischmann
E
,
Amir
M
,
Eikelboom
J
,
Mizera
R
,
Torres
D
,
Wang
CY
,
VanHelder
T
,
Paniagua
P
,
Berwanger
O
,
Srinathan
S
,
Graham
M
,
Pasin
L
,
Le Manach
Y
,
Gao
P
,
Pogue
J
,
Whitlock
R
,
Lamy
A
,
Kearon
C
,
Baigent
C
,
Chow
C
,
Pettit
S
,
Chrolavicius
S
,
Yusuf
S
;
POISE-2 Investigators
:
Aspirin in patients undergoing noncardiac surgery.
N Engl J Med
2014
;
370
:
1494
503