Background

In previous analyses, myocardial injury after noncardiac surgery, major bleeding, and sepsis were independently associated with most deaths in the 30 days after noncardiac surgery, but most of these deaths occurred during the index hospitalization for surgery. The authors set out to describe outcomes after discharge from hospital up to 1 yr after inpatient noncardiac surgery and associations between predischarge complications and postdischarge death up to 1 yr after surgery.

Methods

This study was an analysis of patients discharged after inpatient noncardiac surgery in a large international prospective cohort study across 28 centers from 2007 to 2013 of patients aged 45 yr or older followed to 1 yr after surgery. The study estimated (1) the cumulative postdischarge incidence of death and other outcomes up to a year after surgery and (2) the adjusted time-varying associations between postdischarge death and predischarge complications including myocardial injury after noncardiac surgery, major bleeding, sepsis, infection without sepsis, stroke, congestive heart failure, clinically important atrial fibrillation or flutter, amputation, venous thromboembolism, and acute kidney injury managed with dialysis.

Results

Among 38,898 patients discharged after surgery, the cumulative 1-yr incidence was 5.8% (95% CI, 5.5 to 6.0%) for all-cause death and 24.7% (95% CI, 24.2 to 25.1%) for all-cause hospital readmission. Predischarge complications were associated with 33.7% (95% CI, 27.2 to 40.2%) of deaths up to 30 days after discharge and 15.0% (95% CI, 12.0 to 17.9%) up to 1 yr. Most of the association with death was due to myocardial injury after noncardiac surgery (15.6% [95% CI, 9.3 to 21.9%] of deaths within 30 days, 6.4% [95% CI, 4.1 to 8.7%] within 1 yr), major bleeding (15.0% [95% CI, 8.3 to 21.7%] within 30 days, 4.7% [95% CI, 2.2 to 7.2%] within 1 yr), and sepsis (5.4% [95% CI, 2.2 to 8.6%] within 30 days, 2.1% [95% CI, 1.0 to 3.1%] within 1 yr).

Conclusions

One in 18 patients 45 yr old or older discharged after inpatient noncardiac surgery died within 1 yr, and one quarter were readmitted to the hospital. The risk of death associated with predischarge perioperative complications persists for weeks to months after discharge.

Editor’s Perspective
What We Already Know about This Topic
  • Previous literature demonstrates that myocardial injury after noncardiac surgery, major bleeding, and sepsis are independently associated with most deaths in the 30 days after noncardiac surgery, but most of these deaths typically occur during the index hospitalization for surgery

  • It is unclear to what extent complications during the index hospitalization are associated with postdischarge death up to 1 yr after surgery

What This Article Tells Us That Is New
  • In a secondary analysis of a prospective cohort study of 38,898 patients across 28 hospitals aged 45 yr or older who had inpatient noncardiac surgery between 2007 and 2013 and survived to hospital discharge, 2,165 (5.6%; cumulative incidence, 5.8%) patients died within 1 yr after surgery

  • Complications during the index hospitalization were associated with 33.7% (95% CI, 27.2 to 40.2%) of deaths up to 30 days after discharge and 15.0% (95% CI, 12.0 to 17.9%) of deaths up to 1 yr after discharge

  • Myocardial injury (6.4%), major bleeding (4.7%), and sepsis (2.1%) were associated with the most deaths within 1 yr

Advances in surgical technique, intraoperative monitoring, anesthesia, and early postoperative care have improved surgical safety in the hospital,1  but most patients are not at their baseline health at the time of discharge to home. Discharge requires that clinicians consolidate the events of the hospital stay and plan outpatient care. This includes making decisions about the timing of follow-up visits, utility and urgency of additional investigations, and medication prescribing. Some patients may benefit from enrollment in transition care programs. These decisions can be informed by the likelihood of complications postdischarge.

Our group has found that myocardial injury after noncardiac surgery, major bleeding, and sepsis occurring in the first 30 days after major noncardiac surgery were independently associated with most deaths occurring up to 30 days after surgery.2  The majority of these complications and the majority of deaths occurred during the index hospital stay after surgery.2  Other studies have found that early perioperative complications are associated with mortality in the longer term,3–11  but they have been small, had important methodologic limitations, or have not focused explicitly on risk after discharge. Further, studies have not characterized the proportion of longer-term deaths that are associated with early postoperative complications and have not adequately examined changes in risk over time.

The objectives of this study were to describe outcomes after discharge from hospital up to 1 yr after inpatient noncardiac surgery and to study the associations over time between predischarge complications and death up to 1 yr after surgery.

We report this study according to the Strengthening the Reporting of Observational Studies in Epidemiology statement.12 

Study Procedures

We have previously reported the Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) study design and methods.13,14  VISION was a prospective international cohort study that enrolled patients 45 yr old or older who had inpatient noncardiac surgery between August 2007 and October 2013 at 28 centers in 14 countries (ClinicalTrials.gov NCT00512109). The ethics review board at each participating center approved the VISION study protocol.

Study personnel identified potential participants during the day and night and on weekdays and on weekends through daily screening of patient lists in preoperative assessment clinics, surgical lists from the same and the previous day, lists on surgical wards and in intensive care units, and in preoperative holding areas. At some centers, the ratio of recruitment staff to potential patients exceeded the ability to recruit consecutive patients. At those centers, personnel recruited patients based on a statistician-generated schedule of randomly selected weeks or randomly selected surgical services proportional to the prevalence of the types of surgery at each center. At the end of each week, research personnel reviewed the surgical logbook and reported the number of patients who were eligible but not enrolled.

Eligible consenting patients answered a series of questions about their past medical and social history. Research personnel reviewed medical charts for further background history, noted outcome events throughout the hospital stay, and conducted a follow-up telephone interview with the patient or their next of kin 30 days after surgery and again 1 yr after surgery. Research staff obtained further documentation—including dates of events—if the interview indicated the occurrence of an outcome event. At the time of consent, patients provided contact details for two next of kin or close friends who did not reside together or with the patient. If the patient, their next of kin, or close friends could not be reached, research staff contacted family physicians, searched medical records, and consulted obituaries to obtain study outcomes. Researchers in Peru and Brazil also used national registries to identify deaths.

Study personnel followed patients prospectively while in the hospital, collecting data from patients and their charts. After hospital discharge, study personnel contacted patients or their next of kin at 30 days and 1 yr after surgery. Outcome definitions and ascertainment criteria were specified a priori. Cardiac troponin T measurements were to be done in all patients 6 to 12 h postoperatively and on days 1, 2, and 3 after the day of surgery, or until discharge (whichever was sooner). Patients enrolled between 12 and 24 h after surgery had blood drawn for troponin T measurement immediately; subsequent measurements followed the usual schedule. During the later phase of the study, high-sensitivity troponin T measurement was also performed before surgery. Research staff checked daily that measurements were done as scheduled. Clinical teams were encouraged to obtain electrocardiograms for several days after troponin T concentrations were elevated above the manufacturer’s reference limit or if patients experienced an ischemic symptom. Troponin T thresholds for myocardial injury after noncardiac surgery were subsequently derived from their associations with 30-day postoperative death and adjudicated centrally by physicians to exclude events potentially explained by nonischemic causes (e.g., sepsis, pulmonary embolism, atrial fibrillation, chronic elevation) as described previously.13–15  Physician adjudicators also evaluated all reported cases of myocardial infarction, stroke, venous thromboembolism, and new clinically important atrial fibrillation to ensure that they met study criteria. Supplemental Digital Content 1 (https://links.lww.com/ALN/D307) provides definitions of study outcomes.

Investigators reviewed and approved data at each site. Research personnel submitted case report forms and supporting documentation directly to the coordinating center. Data monitoring involved central data consistency checks, statistical monitoring, and on-site monitoring for all centers. For on-site monitoring, the central coordinator randomly selected participants with and without a perioperative complication for audit by independent monitors.

We limited our analyses to patients who survived to discharge from the hospital after surgery. Some patients were transferred from the hospital at which the index surgery was performed to another acute care hospital; for these analyses, we only included these patients if they were discharged alive from the acute care hospital to which they were transferred. We included patients discharged to a rehabilitation facility.

Outcomes

The primary outcome was death from any cause occurring between discharge from acute hospital care after surgery up to 1 yr after surgery. We also describe the cumulative incidence of death from vascular causes (i.e., believed to be due to myocardial infarction, asystole, ventricular fibrillation, pulseless electrical activity, other sudden or arrhythmic death, sustained ventricular tachycardia, cardiogenic shock, congestive heart failure, other cardiac causes, stroke, peripheral vascular disease, aortic dissection/rupture, pulmonary embolus, bleeding, other vascular cause, or unknown cause), death from nonvascular causes (i.e., believed to be due to sepsis, pneumonia or other infection, renal failure, cancer, or other nonvascular causes), rehospitalization, myocardial infarction, congestive heart failure, stroke, amputation, venous thromboembolism, nonfatal cardiac arrest, coronary revascularization, new diagnosis of dementia, new diagnosis of diabetes, new cancer, and recurrent cancer.

Exposures

The exposures of interest were perioperative complications that occurred before discharge from acute hospital care for the index surgery. Complications of primary interest were sepsis, infection without sepsis, myocardial injury after noncardiac surgery that met the criteria for myocardial infarction, myocardial injury after noncardiac surgery that represented an isolated ischemic myocardial injury (i.e., an elevated cardiac troponin T measurement of presumed ischemic origin that did not meet criteria for myocardial infarction), and major bleeding that met diagnostic criteria we previously developed for bleeding independently associated with mortality after noncardiac surgery (i.e., bleeding during and up to 30 days after noncardiac surgery that resulted in hemoglobin less than 70 g/l or any red cell transfusion, or that was imminently fatal [although, this analysis excluded patients with fatal predischarge bleeds]).16  Complications of secondary interest were stroke, new clinically important atrial fibrillation, venous thromboembolism, acute congestive heart failure, amputation, and acute kidney injury managed with dialysis. These complications affected a smaller number of patients; their associations with mortality were treated as exploratory.

Statistical Analyses

We developed a statistical analysis plan before conducting the analyses and briefly summarize the methods here. The analysis plan was written after data were accessed to determine feasibility (e.g., by assessing number of primary outcome events, number of patients with various predischarge complications). We used R version 4.1.0 (R Core Team, 2021; Austria) and Stata version 17.0 (StataCorp, 2021; USA) for the analyses.

Missing Data.

Data regarding one or more variables were missing in 12.9% of patients, mostly preoperative estimated glomerular filtration rate (6.7%), preoperative hemoglobin (6.5%), and postoperative troponin measurements (5.5%). We imputed these data using multiple imputation across 20 imputed datasets for the association analysis.17  We present adjusted hazard ratios and percentages of associated deaths based on a single imputed dataset with model coefficients that closely approximated those pooled across the 20 imputed datasets.

Cumulative Incidence of Complications after Discharge.

We calculated the cumulative incidence of events that occurred after discharge from the index surgical hospitalization up to 1 yr after surgery. For outcomes other than all-cause death, we accounted for competing risk of death, and we treated patients who did not experience the outcome as censored at the time of death or last known vital status. We used the cumulative incidence function instead of the complement of the Kaplan–Meier survival function to estimate the crude incidence of outcomes because the complement of the Kaplan–Meier function overestimates incidence in the presence of competing events.18  We performed all analyses based on the time to the first event after discharge.

Analyses of the Association between Predischarge Complications and Mortality after Discharge.

We fit a flexible parametric survival model with a smooth transformation of the log cumulative hazard function using restricted cubic splines.19  We included as independent variables those listed in table 1, including predischarge complications, preoperative patient characteristics, type of the index surgery (major vascular, major general, major thoracic, major urologic or gynecologic, major neurosurgery, major orthopedic), surgical approach (i.e., open versus endoscopic), and the urgency of surgery. We also adjusted for the study center. We did not include nonfatal cardiac arrest in the model because only two patients who had this complication before discharge died after discharge. We report detailed surgical procedure type (e.g., lobectomy versus pneumonectomy) for transparency, but modeling included the broader categories of index surgery mentioned above (e.g., major thoracic surgery). We modeled continuous variables using restricted cubic splines. We presented adjusted hazard ratios and corresponding 95% CIs describing the association between each in-hospital complication and time to death after discharge. We performed direct adjustment by regression standardization averaged over the distribution of covariates to estimate standardized survival functions. This method averages all patient-specific survival curves estimated at each time point across the year of follow-up from all patients in the study, with each survival curve adjusted for that patient’s characteristics under hypothetical conditions under which they did and did not have each predischarge complication of interest.20  We allowed the effects of predischarge complications, type and urgency of surgery, and surgical approach to vary with time because we expected a priori that their contribution to the hazard of mortality would decline over time. We treated the association of venous thromboembolism and acute kidney injury managed with dialysis as constant over time due to problems with model convergence at the time of analysis because few patients had these complications before discharge, and few of them died after discharge.

Table 1.

Description of Patients Discharged from Hospital after Surgery

Description of Patients Discharged from Hospital after Surgery
Description of Patients Discharged from Hospital after Surgery

We also compared the adjusted association with postdischarge mortality for myocardial injury after noncardiac surgery that met criteria for myocardial infarction to that of myocardial injury after noncardiac surgery that represented an isolated ischemic myocardial injury against the null hypothesis that their coefficients in the model were equal.

To help with interpretation of the time-dependent adjusted hazard ratios, we estimated average absolute adjusted differences in the risk of postdischarge mortality associated with each predischarge complication over time.

Hypothesis Testing.

The test for statistical significance of the association between each complication and time to postdischarge death within 1 yr after surgery was an F test adapted for multiple imputation.21  We calculated a single global P value for each the association between each predischarge complication and postdischarge death. All hypothesis tests were two-sided without multiplicity adjustments; P < 0.05 denoted statistical significance.

Percentage of Postdischarge Deaths Associated with Predischarge Complications.

We calculated the percentages of deaths that occurred between discharge and 1 yr after surgery that were potentially attributable to each predischarge complication and with all predischarge complications combined (i.e., attributable fractions).22  We also estimated attributable hazard fractions that represent the percentage of the instantaneous hazard of death associated with predischarge complications at each day throughout the year after discharge.22  Because three of the predischarge complications were associated with a disproportionately large percentage of the postdischarge deaths, we then grouped them together to communicate their prominence.

Prespecified Sensitivity Analyses.

In the first sensitivity analysis, we changed the censoring times when calculating cumulative incidence of times to event outcomes other than death from the time of death or last known vital status (the primary analysis) to the time of last study contact when the patient was known not to have the event. The method used in the primary analysis might underestimate the cumulative incidence, while this sensitivity analysis might overestimate the cumulative incidence.23  In the second sensitivity analysis, we repeated the association analyses in patients without missing data.

Post Hoc Analyses in Response to Reviewer Comments.

We examined with logistic regression the relationship between the occurrence of censoring on day 30 after surgery (the day on which 94% of censoring occurred) and the independent variables included in our model for postdischarge death (i.e., predischarge complications, preoperative patient characteristics, type and timing of surgery, surgical approach, and the study center). We provided an estimate of the cumulative incidence of postdischarge death due to cardiac causes. This post hoc definition in response to a reviewer comment included deaths due to myocardial infarction, due to cardiac arrest, or after coronary revascularization, but excluded death due to congestive heart failure.24  We also presented the associations between adjustment variables and postdischarge death. Finally, we provided estimates of the cumulative incidence of postdischarge outcomes by study center.

Sample Size.

The model to evaluate adjusted associations between in-hospital perioperative complications and mortality after discharge involved 101 parameters including the intercept, coefficients for each variable, and segments of restricted cubic splines and their interactions with variables that have time-dependent effects. Simulation studies suggest that, depending on the conditions, between 5 and 20 events per parameter are adequate to minimize bias in the estimates of association across a variety of scenarios.25  We exceeded these requirements.

Figure 1 summarizes the flow of participants through the study. VISION included 40,004 patients between 2007 and 2013. We excluded 1,106 patients (2.8% of the total 40,004) from these analyses, including 684 (1.7% of the total 40,004) because they died before hospital discharge. Table 1 describes the characteristics of the 38,898 patients included in these analyses. Their median age was 63 yr, 49.9% were women, 5.0% required help with activities of daily living, 12.9% had a history of coronary artery disease, 24.4% had active cancer, 10.0% had urgent or emergency surgery, 19.6% had major general surgery, 17.6% had major orthopedic surgery, and 5.7% had major vascular surgery. Patients were followed for a median of 362 days after hospital discharge. Of these patients, 96.5% (37,544) completed follow-up at 1 yr after surgery or died before this time. Of the remaining 1,354 patients (who were censored on the day that their vital status was last known), 1,275 (94.1%) were censored at day 30 after surgery, and 72 (5.3%) were censored between days 345 and 364 after surgery.

Fig. 1.

Participant eligibility flow diagram. VISION, Vascular Events in Noncardiac Surgery Patients Cohort Evaluation.

Fig. 1.

Participant eligibility flow diagram. VISION, Vascular Events in Noncardiac Surgery Patients Cohort Evaluation.

Close modal

Among the 2,165 patients (5.6%) who died between hospital discharge after surgery and 1 yr after surgery, the median (25th, 75th percentiles) time to death after discharge from acute hospital care after surgery was 141 days (60, 240) and after the index surgery was 154 days (75, 252).

Complications before Discharge

The median length of the index acute care hospital stay after surgery was 5 days (25th percentile, 2 days; 75th percentile, 9 days). Before hospital discharge, 14.2% of patients had major bleeding, 3.1% developed sepsis, 3.2% developed infection without sepsis, 3.1% had myocardial injury after noncardiac surgery that met criteria for myocardial infarction, and 10.4% had myocardial injury after noncardiac surgery that represented an isolated ischemic myocardial injury. Each of the other predischarge complications occurred in fewer than 1% of patients who were ultimately discharged from the hospital.

Outcomes after Discharge

The 1-yr cumulative incidence of postdischarge death was 5.8% (95% CI, 5.5 to 6.0%) from any cause, 2.0% (95% CI, 1.9 to 2.2%) from vascular causes, and 3.8% (95% CI, 3.6 to 4.0%) from nonvascular causes (fig. 2A). The cumulative incidence of first hospital readmission was 24.7% (95% CI, 24.2 to 25.1%); it increased most quickly in the first 45 days (fig. 2B). The cumulative incidences of new diabetes, dementia, new cancer, and recurrent cancer increased at a near constant rate over time (Figure S1, Supplemental Digital Content 1, https://links.lww.com/ALN/D307). The cumulative incidence of most vascular events increased faster in the first 60 to 120 days (Supplemental Figure S1, https://links.lww.com/ALN/D307) and was less than 1% for each. Table 2 summarizes the incidence of first events, and Supplemental Figure S2 (https://links.lww.com/ALN/D307) shows their distributions over time.

Table 2.

One-year Postdischarge Outcomes

One-year Postdischarge Outcomes
One-year Postdischarge Outcomes
Fig. 2.

Cumulative incidences of death and rehospitalization after hospital discharge from surgery. Competing risk of death was accounted for by the cumulative incidence function.

Fig. 2.

Cumulative incidences of death and rehospitalization after hospital discharge from surgery. Competing risk of death was accounted for by the cumulative incidence function.

Close modal

Association between Predischarge Complications and Mortality after Discharge.

Figure 3 shows the adjusted associations between predischarge complications and postdischarge mortality over time. Although statistically significant for the 1-yr period overall, the adjusted hazard ratios associating several predischarge complications with death after discharge declined with time from discharge: myocardial infarction (30 days, 1.70 [95% CI, 1.32 to 2.18]; 180 days, 1.15 [95% CI, 0.95 to 1.40]; 365 days, 1.00 [95% CI, 0.75 to 1.32]; overall P < 0.001), sepsis (30 days, 1.62 [95% CI, 1.28 to 2.05]; 180 days, 1.28 [95% CI, 1.06 to 1.54]; 365 days, 1.16 [95% CI, 0.89 to 1.53]; overall P < 0.001), and major bleeding (30 days, 1.53 [95% CI, 1.30 to 1.81]; 180 days, 1.10 [95% CI, 0.98 to 1.25]; 365 days, 0.97 [95% CI, 0.81 to 1.15]; overall P < 0.001). The adjusted hazard ratio for isolated ischemic myocardial injury showed a similar trend but decreased more slowly (30 days, 1.49 [95% CI, 1.25 to 1.78]; 180 days, 1.26 [95% CI, 1.11 to 1.42]; 365 days, 1.19 [95% CI, 0.99 to 1.42]; overall P < 0.001) compared to myocardial infarction. Although the comparison is imprecise, there was no significant difference in the risk of death after discharge when comparing myocardial injury after noncardiac surgery that met the criteria for myocardial infarction to that which represented an isolated ischemic myocardial injury among patients who survived to hospital discharge (30 days, 1.14 [95% CI, 0.87 to 1.48]; 180 days, 0.92 [95% CI, 0.75 to 1.13]; 365 days, 0.84 [95% CI, 0.62 to 1.13]; overall P = 0.81). The hazard ratio for infection without sepsis was negligible early after discharge but increased over time (30 days, 1.05 [95% CI, 0.80 to 1.37]; 180 days, 1.33 [95% CI, 1.12 to 1.59]; 365 days, 1.51 [95% CI, 1.17 to 1.94]; overall P = 0.01).

Fig. 3.

Associations between predischarge perioperative complications of primary interest and mortality up to 1 yr after discharge from noncardiac surgery. Adjusted hazard ratios are presented; parentheses contain their 95% CIs. P values are from global tests for association between each predischarge complication and postdischarge mortality over the entire follow-up period.

Fig. 3.

Associations between predischarge perioperative complications of primary interest and mortality up to 1 yr after discharge from noncardiac surgery. Adjusted hazard ratios are presented; parentheses contain their 95% CIs. P values are from global tests for association between each predischarge complication and postdischarge mortality over the entire follow-up period.

Close modal

To aid in the interpretation of these changing hazard ratios, Figure S3 in Supplemental Digital Content 1 (https://links.lww.com/ALN/D307) shows changes over time after discharge in the average differences in the absolute adjusted cumulative risk of mortality associated with each predischarge complication. As the hazard ratios decrease over time, differences in the cumulative risks of mortality increase more slowly over time. For example, the average adjusted absolute increase in the cumulative risk of mortality associated with myocardial infarction that occurred before discharge was 0.55% (95% CI, 0.21 to 0.88%) at 30 days after discharge and increased to 1.45% (95% CI, 0.68 to 2.23%) at 180 days after discharge, but only increased to 1.56% (95% CI, 0.51 to 2.6%) by 365 days after discharge. These absolute differences in the risk of mortality illustrate an average effect for patients in our study and may be higher or lower for any one patient depending on their baseline risk.

Among the predischarge complications of secondary interest, congestive heart failure and amputation were significantly associated with death after discharge (Supplemental Figure S4, https://links.lww.com/ALN/D307) but were uncommon and associated with a small percentage of deaths (Figure S5 in Supplemental Digital Content 1, https://links.lww.com/ALN/D307). Estimates of association were too imprecise to determine if new clinically important atrial fibrillation, venous thromboembolism, stroke, or acute kidney injury managed with dialysis that occurred before discharge were associated with death after discharge.

Considered together, the studied predischarge complications had an attributable fraction of 33.7% (95% CI, 27.2 to 40.2%) for deaths occurring between hospital discharge and 30 days after discharge, 20.7% (95% CI, 17.2 to 24.2%) for deaths between hospital discharge and 180 days after discharge, and 15.0% (95% CI, 12.0 to 17.9%) of all deaths up to 1 yr after discharge (fig. 4). Myocardial injury after noncardiac surgery (either myocardial infarction or isolated ischemic myocardial injury), sepsis, and major bleeding accounted for most postdischarge deaths associated with predischarge complications and for most of the instantaneous hazard of death associated with predischarge complications at any time after discharge (fig. 4). Isolated ischemic myocardial injury was more common than myocardial infarction and was independently associated with a greater percentage of deaths after discharge (Figure S5 in Supplemental Digital Content 1, https://links.lww.com/ALN/D307) and a greater percentage of the hazard of death at any time after discharge (Figure S6 in Supplemental Digital Content 1, https://links.lww.com/ALN/D307).

Fig. 4.

Percentage of postdischarge deaths and of the hazard of death independently associated with predischarge complications up to 1 yr after discharge from noncardiac surgery. In the top panels, calculations are for events that occurred from the day of hospital discharge up to the specified time after discharge. For example, the calculation at 365 days considered all deaths that occurred during the follow-up period, while the calculation at 30 days considered only deaths that occurred in the first 30 days after discharge. In the bottom panels, calculations are for the instantaneous hazard of death on specific days; they pertain to deaths predicted to occur on those days. Predischarge complications included myocardial injury after noncardiac surgery, major bleeding, sepsis, infection without sepsis, acute congestive heart failure, amputation, new clinically important atrial fibrillation, acute kidney injury managed with dialysis, stroke, and venous thromboembolism. Parentheses contain 95% CIs.

Fig. 4.

Percentage of postdischarge deaths and of the hazard of death independently associated with predischarge complications up to 1 yr after discharge from noncardiac surgery. In the top panels, calculations are for events that occurred from the day of hospital discharge up to the specified time after discharge. For example, the calculation at 365 days considered all deaths that occurred during the follow-up period, while the calculation at 30 days considered only deaths that occurred in the first 30 days after discharge. In the bottom panels, calculations are for the instantaneous hazard of death on specific days; they pertain to deaths predicted to occur on those days. Predischarge complications included myocardial injury after noncardiac surgery, major bleeding, sepsis, infection without sepsis, acute congestive heart failure, amputation, new clinically important atrial fibrillation, acute kidney injury managed with dialysis, stroke, and venous thromboembolism. Parentheses contain 95% CIs.

Close modal

Results of Prespecified Sensitivity Analysis.

Table S1 in Supplemental Digital Content 1 (https://links.lww.com/ALN/D307) reports results of the first sensitivity analysis where cumulative incidences for events other than death were calculated such that patients who did not have an event were censored at the time of last contact instead of at the time of death or last known vital status. Cumulative incidences calculated this way were similar to the primary analysis. Figures S7 and S8 in Supplemental Digital Content 1 (https://links.lww.com/ALN/D307) show association results from the complete case analysis; these were similar to the primary analyses with imputed data.

Results of Post Hoc Analyses.

In the post hoc analysis to identify predictors of censoring, we found that study center was the strongest predictor (Supplemental Digital Content 1, Table S2, https://links.lww.com/ALN/D307). Additionally, the odds of censoring were modestly higher with older age, lower preoperative hemoglobin, and open surgery; the odds were lower with orthopedic surgery. All of these variables were already included in the model estimating the association between predischarge complications with postdischarge mortality. None of the postoperative predischarge complications studied were associated with censoring.

The cumulative incidence of postdischarge death from cardiac causes was 0.34% (95% CI, 0.28 to 0.40%) per year.

Table S3 in Supplemental Digital Content 1 (https://links.lww.com/ALN/D307) presents the associations between adjustment variables and postdischarge death. Some associations are in general as would be expected: older patients and those with various chronic diseases are more likely to die after discharge. Patients with active cancer before surgery had the highest hazard ratio with mortality. However, there was no association between postdischarge mortality and preoperative history of coronary disease, stroke or transient ischemic attack, or congestive heart failure. The risk of death was lower for patients with a history of hypertension than in those without. Men were more likely than women to die in the year after discharge. We caution that these associations were not the focus of our study and are fraught with risk of bias.26,27 

Table S4 in Supplemental Digital Content 1 (https://links.lww.com/ALN/D307) presents cumulative incidences of postdischarge outcomes at each study center.

In this large, international, prospective cohort study, approximately 1 in 18 patients aged 45 yr or older who were discharged from the hospital after inpatient noncardiac surgery died at 1 yr follow-up compared to 1 in 58 who died during the index surgical hospitalization. One in three deaths up to 30 days after discharge and one in five deaths up to 6 months after discharge were independently associated with perioperative complications that occurred before discharge. Predischarge myocardial injury after noncardiac surgery, sepsis, and major bleeding were the complications associated with the largest proportion of postdischarge deaths.

Numerous studies broadly corroborate our finding that early perioperative complications are associated with mortality in the longer term, but our study adds significantly more information about these associations and uniquely focuses on information available to healthcare providers at the transition from hospital to home. A study of patients who had predominantly orthopedic surgery at one United Kingdom center found increased long-term mortality associated with postoperative complications but lacked granularity with respect to the type of complications.3  A larger sample from the Veterans Administration Surgical Quality Improvement Program showed associations between the number of postoperative complications and long-term mortality but ignored deaths occurring in the first 90 days after surgery4 —a period when deaths were most strongly associated with predischarge complications in our study. In a prospective cohort study of 2018 patients having noncardiac surgery, myocardial injury after noncardiac surgery was associated with death occurring up to 30 days after surgery (adjusted hazard ratio, 2.28; 95% CI, 1.19 to 4.36) and up to 1 yr after surgery (adjusted hazard ratio, 1.48; 95% CI, 1.07 to 2.06).5  That study also found other early postoperative complications (i.e., sepsis, stroke, and pneumonia) to be associated with death up to 1 yr after surgery. An observational analysis of a large randomized trial found that postoperative anemia is associated with death or disability up to 90 days after major abdominal surgery.6  Recent analysis of 9,733 patients pooled from two cohort studies found an association between complications occurring in the first 30 days after surgery and mortality up to 1 yr after surgery.7  These included complications that occurred during the index postoperative hospitalization. Evolution of the association between complications and mortality was only crudely characterized as hazard ratios before and after postoperative day 20, and attributable fractions were not estimated.

Among patients who survived to discharge, the association between myocardial injury after noncardiac surgery and mortality was similar when the event met the diagnostic criteria for myocardial infarction and when it did not (i.e., isolated ischemic myocardial injury). It is likely that isolated ischemic myocardial injury did not prompt the same secondary prevention measures as myocardial infarction. This highlights the importance of systematic postoperative troponin measurement because most patients with myocardial injury after noncardiac surgery have no other evidence of ischemia,13,14  and because observational and randomized trial evidence suggests that treating patients with myocardial injury after noncardiac surgery can reduce long-term vascular risk.28,29  Persistent inflammation may partially explain the time-dependent association between predischarge sepsis and the risk of death after discharge. Increased risk of death associated with perioperative infectious complications has been found to persist for months after cardiac surgery.30  Major bleeding may confer risk beyond hospitalization through dysregulation of coagulation and thrombosis, through withholding of antithrombotic and antiplatelet therapy, or because persistent anemia after discharge potentiates cardiovascular risk.

Prevention of predischarge complications and increased awareness that they confer persistent mortality risk in the weeks to months after discharge are an opportunity to help prevent postdischarge deaths that may have previously appeared unrelated to the early postoperative course. However, because most deaths after discharge were not associated with in-hospital complications, research focused on postdischarge care is warranted to identify opportunities to prevent more deaths.

This was a large, representative, multicenter, multicountry, prospective cohort study with careful characterization and adjudication of perioperative complications with little loss to follow-up at 1 yr after surgery. We adjusted for preoperative characteristics, the type and urgency of surgery, and several perioperative complications. Unlike most survival analyses, we did not assume proportional hazards but instead allowed the association between covariates and postdischarge mortality to vary over time.

The study had potential limitations. Residual confounding remains a concern despite adjustment. Some of the time-dependent change in hazard ratios may be due to differential depletion of susceptible patients, but this bias is negligible under our study conditions when the outcome occurs in a low proportion of patients.31  Our study may have underestimated the incidence of nonfatal postdischarge outcomes due to limitations of recall between 30 days and 1 yr after surgery. A sensitivity analysis that prevented attributing event-free time after the day of last study contact provided estimates similar to the primary analysis. Regional variation in healthcare contact after discharge may have also led us to underestimate the incidence of nonfatal postdischarge outcomes at some centers. This may explain some of the differences we observed between centers (Table S4 in Supplemental Digital Content 1, https://links.lww.com/ALN/D307), but differences in case mix and chance operating on low event rates can also exaggerate apparent differences in incidence.

We did not adjudicate some complications (i.e., major bleeding, sepsis, infection without sepsis, congestive heart failure, and acute kidney injury managed with dialysis), and this may have influenced the association between some of these complications and mortality.

Generalizability of absolute estimates of risk to any specific patient, surgical approach, or type of surgery is limited by the broad eligibility criteria in our study. Nevertheless, absolute risk estimates in broad populations remain useful for policy makers, administrators, and researchers to understand what outcomes have the most potential for impact if focused upon. The magnitudes of association between predischarge complications and postdischarge mortality are less likely to differ meaningfully across patient characteristics, surgical approaches, and types of surgery, and identifying differences in magnitude with certainty beyond chance would require a much larger and more costly study with a much greater burden of data collection. It is likely, however, that sepsis, myocardial injury, and major bleeding remain the main prognostically important complications with temporal risk patterns maintained across most patient groups.

We completed follow-up on vital status for 96.5% of participants up to 1 yr after surgery. Nearly all patients lost to follow-up were censored at 30 days after surgery. Medical records, obituaries, and—in one country—a national registry provided data for the occurrence and date of death, but we relied on next of kin for some participants. It is unlikely that reliance on next of kin had any meaningful effect on the estimate of mortality or in the estimate of its association with predischarge complications, and if there was an effect, it would most likely be to modestly underestimate both. Propensity for loss to follow-up was strongly dependent on study center and to a small extent on some preoperative characteristics. Our analyses adjusted for these variables and for study center, thereby minimizing risk of bias from the small proportion of patients lost to follow-up.

Although encouraged to do so, clinical teams did not obtain electrocardiograms in some patients with elevated troponin. As a result, some myocardial infarctions may have been misclassified as isolated ischemic myocardial injury. This would falsely increase the similarity in prognosis between the two myocardial injury categories. However, only a small proportion of patients who had an electrocardiogram after an elevated troponin measurement had ischemic changes. We would expect ischemic changes to be even less common among patients who did not have an electrocardiogram. Such misclassification is likely to be more common in practice, and this further highlights the need to take notice of myocardial injury after noncardiac surgery that is isolated ischemic myocardial injury. Similarly, we may have missed atrial fibrillation as a cause of myocardial injury in patients without electrocardiograms, but myocardial injury due to atrial fibrillation tends to occur with rapid ventricular response that typically prompts investigation. Not all patients with troponin elevation had imaging to rule out pulmonary embolism, but we expect most pulmonary emboli that cause myocardial necrosis to have other features that would prompt investigation.

We imputed a small amount of missing preoperative estimated glomerular filtration rate, preoperative hemoglobin, and myocardial injury status due to missing postoperative troponin for our primary association analyses to avoid loss of sample size from casewise deletion and to avoid bias. Sensitivity analyses without imputation found very similar results.

Conclusions

Approximately 1 in 18 patients 45 yr and older who are discharged from hospital after inpatient noncardiac surgery died within 1 yr after surgery. One quarter of patients were readmitted to the hospital. One third of the deaths that occurred in the first 30 days after discharge and one fifth that occurred in the first 6 months were independently associated with myocardial injury after noncardiac surgery, sepsis, and major bleeding that occurred before hospital discharge.

Acknowledgments

This study was coordinated by the Clinical Advances through Research and Information Translation project office in the Department of Health Research Methods, Evidence, and Impact (formerly the Department of Clinical Epidemiology and Biostatistics) at McMaster University and the Population Health Research Institute at Hamilton Health Sciences, and McMaster University, Hamilton, Canada.

Research Support

The VISION funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript. Roche Diagnostics provided the troponin T assays as well as financial support for the VISION Study. Funding for this study came from more than 70 grants for VISION and its substudies, as follows: Canada: Canadian Institutes of Health Research (seven grants); Heart and Stroke Foundation of Ontario (two grants); Academic Health Science Centers Alternative Funding Plan Innovation Fund Ontario; Population Health Research Institute; Clinical Advances through Research and Information Translation Research Group; McMaster University Department of Surgery Surgical Associates; Hamilton Health Science New Investigator Fund; Hamilton Health Sciences; Ontario Ministry of Resource and Innovation; Stryker Canada; McMaster University, Department of Anesthesiology (two grants); St. Joseph’s Healthcare, Department of Medicine (two grants); Father Sean O’Sullivan Research Center (two grants); McMaster University Department of Medicine (two grants); Roche Diagnostics Global Office (five grants); Hamilton Health Sciences Summer Studentships (six grants); McMaster University Department of Clinical Epidemiology and Biostatistics; McMaster University, Division of Cardiology; Canadian Network and Center for Trials Internationally; Winnipeg Health Sciences Foundation; Department of Surgery, University of Manitoba (two grants); Diagnostic Services of Manitoba Research; Manitoba Medical Services Foundation; Manitoba Health Research Council; University of Manitoba Faculty of Dentistry Operational Fund; University of Manitoba Department of Anesthesia; University Medical Group, Department of Surgery, University of Manitoba, Start-up Fund. Australia: National Health and Medical Research Council Program. Brazil: Projeto Hospitais de Excelência a Serviço do SUS grant from the Brazilian Ministry of Health in partnership with Hcor (Cardiac Hospital Sao Paulo–SP); National Council for Scientific and Technological Development (CNPq) grant from the Brazilian Ministry of Science and Technology. China: Public Policy Research Fund (grant CUHK-4002-PPR-3), Research Grant Council, Hong Kong Special Administrative Region; General Research Fund (grant 461412), Research Grant Council, Hong Kong Special Administrative Region; Australian and New Zealand College of Anaesthetists (grant 13/008). Colombia: School of Nursing, Universidad Industrial de Santander; Grupo de Cardiología Preventiva, Universidad Autónoma de Bucaramanga; Fundación Cardioinfantil–Instituto de Cardiología; Alianza Diagnóstica SA. France: Université Pierre et Marie Curie, Département d’anesthésie Réanimation, Pitié-Salpêtrière, Assistance Publique–Hôpitaux de Paris. India: St. John’s Medical College and Research Institute; Division of Clinical Research and Training. Malaysia: University of Malaya (grant RG302-14AFR); University of Malaya, Penyelidikan Jangka Pendek. Poland: Polish Ministry of Science and Higher Education (grant NN402083939). South Africa: University of KwaZulu-Natal. Spain: Instituto de Salud Carlos III; Fundació La Marató de TV3. United States: American Heart Association; Covidien; American Society of Nephrology Student Scholar Grant. United Kingdom: National Institute for Health Research. Dr. Borges is a recipient of a Research Early Career Award from Hamilton Health Sciences. Dr. Garg holds the Adam Linton Chair in Kidney Health Analytics from the Schulich School of Medicine & Dentistry, Western University. Dr. Chow is supported by a National Health and Medical Research Council (Australia) Investigator fellowship. Dr. Popova is funded by a research contract (SLT017/20/000089) from the Health Department, Generalitat de Catalunya, Spain.

Competing Interests

Roche Diagnostics provided troponin T assays and financial support for the study. Dr. Devereaux reports grants from Roche Diagnostics and Abbott Diagnostics during the conduct of the study, and grants from Siemens, Roche, Octopharma, Philips Healthcare, Stryker, Covidien, Boehringer Ingelheim, and devices from CloudDx, all outside the submitted work, as well as advisory board memberships with Bayer and Trimedic, and consultation to Renibus. Dr. Conen received consultation fees from Roche Diagnostics and Trimedics, as well as speaker fees from BMS/Pfizer and Servier, all outside of the submitted work. Dr. Xavier received grants from Cadila Pharmaceuticals, Boehringer Ingelheim, Astra Zeneca India, Sanofi Aventis, Pfizer, Bristol–Myers Squibb, Medical Research Council (United Kingdom), and Wellcome Trust outside the submitted work. Dr. Pearse has received research grants and/or honoraria from Edwards Lifesciences, Intersurgical, and GlaxoSmithkline. Dr. Polanczyk has received grants from Novartis and Roche, and honoraria for consulting and lecture fees from Bayer, Bristol, Pfizer and Novartis, and has participated in data safety and monitoring boards or advisory boards for Angem Pharmaceuticals and Novartis, all outside the submitted work. Dr. Popova has received speaker fees from Roche Diagnostics outside the submitted work. Dr. Spence has received an honorarium from PhaseBio for participating in a panel discussion. Dr. McIntyre reports speaking fees from Servier and Bayer outside of the submitted work. Dr. Borges reports an investigator-initiated grant from Siemens. Dr Berwanger reports fees from AstraZeneca, Pfizer, and Novartis, and research grants from Servier and Amgen. Dr. Chow reports speaker fees or compensation for travel from Novo Nordisk, Novartis, Eli Lilly, and Limbic, all of which are unrelated to this work. Dr. Whitlock reports grants from AtriCure, Abbott, and Artivian, and money paid to institution from Cytosorbents. Dr. Belley-Côte reports grants from Bayer, Roche Diagnostics, and BMS Pfizer, and honoraria from Trimedic Therapeutics Inc. The other authors declare no competing interests.

Supplemental Appendix, https://links.lww.com/ALN/D307

1.
Devereaux
PJ
,
Sessler
DI
:
Cardiac complications in patients undergoing major noncardiac surgery.
N Engl J Med
2015
;
373
:
2258
69
2.
Spence
J
,
LeManach
Y
,
Chan
MTV
,
Wang
CY
,
Sigamani
A
,
Xavier
D
,
Pearse
R
,
Alonso-Coello
P
,
Garutti
I
,
Srinathan
SK
,
Duceppe
E
,
Walsh
M
,
Borges
FK
,
Malaga
G
,
Abraham
V
,
Faruqui
A
,
Berwanger
O
,
Biccard
BM
,
Villar
JC
,
Sessler
DI
,
Kurz
A
,
Chow
CK
,
Polanczyk
CA
,
Szczeklik
W
,
Ackland
G
,
X
GA
,
Jacka
M
,
Guyatt
GH
,
Sapsford
RJ
,
Williams
C
,
Cortes
OL
,
Coriat
P
,
Patel
A
,
Tiboni
M
,
Belley-Côté
EP
,
Yang
S
,
Heels-Ansdell
D
,
McGillion
M
,
Parlow
S
,
Patel
M
,
Pettit
S
,
Yusuf
S
,
Devereaux
PJ
;
Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Study Investigators
:
Association between complications and death within 30 days after noncardiac surgery.
CMAJ
2019
;
191
:
E830
7
3.
Moonesinghe
SR
,
Harris
S
,
Mythen
MG
,
Rowan
KM
,
Haddad
FS
,
Emberton
M
,
Grocott
MP
:
Survival after postoperative morbidity: A longitudinal observational cohort study.
Br J Anaesth
2014
;
113
:
977
84
4.
Portuondo
JI
,
Itani
KMF
,
Massarweh
NN
:
Association between postoperative complications and long-term survival after non-cardiac surgery among veterans.
Ann Surg
2021
;
277
:
e24
32
5.
Puelacher
C
,
Lurati Buse
G
,
Seeberger
D
,
Sazgary
L
,
Marbot
S
,
Lampart
A
,
Espinola
J
,
Kindler
C
,
Hammerer
A
,
Seeberger
E
,
Strebel
I
,
Wildi
K
,
Twerenbold
R
,
du Fay de Lavallaz
J
,
Steiner
L
,
Gurke
L
,
Breidthardt
T
,
Rentsch
K
,
Buser
A
,
Gualandro
DM
,
Osswald
S
,
Mueller
C
;
BASEL-PMI Investigators
:
Perioperative myocardial injury after noncardiac surgery: Incidence, mortality, and characterization.
Circulation
2018
;
137
:
1221
32
6.
Myles
PS
,
Richards
T
,
Klein
A
,
Wood
EM
,
Wallace
S
,
Shulman
MA
,
Martin
C
,
Bellomo
R
,
Corcoran
TB
,
Peyton
PJ
,
Story
DA
,
Leslie
K
,
Forbes
A
;
RELIEF Trial Investigators
:
Postoperative anaemia and patient-centred outcomes after major abdominal surgery: A retrospective cohort study.
Br J Anaesth
2022
;
129
:
346
54
7.
Fowler
AJ
,
Wan
YI
,
Prowle
JR
,
Chew
M
,
Campbell
D
,
Cuthbertson
B
,
Wijeysundera
DN
,
Pearse
R
,
Abbott
T
:
Long-term mortality following complications after elective surgery: A secondary analysis of pooled data from two prospective cohort studies.
Br J Anaesth
2022
;
129
:
588
97
8.
Faizer
R
,
Dombrovskiy
VY
,
Vogel
TR
:
Impact of hospital-acquired infection on long-term outcomes after endovascular and open abdominal aortic aneurysm repair.
Ann Vasc Surg
2014
;
28
:
823
30
9.
Pang
HY
,
Zhao
LY
,
Wang
H
,
Chen
XL
,
Liu
K
,
Zhang
WH
,
Yang
K
,
Chen
XZ
,
Hu
JK
:
Impact of type of postoperative complications on long-term survival of gastric cancer patients: Results from a high-volume institution in China.
Front Oncol
2021
;
11
:
587309
10.
Vogel
TR
,
Dombrovskiy
VY
,
Lowry
SF
:
Impact of infectious complications after elective surgery on hospital readmission and late deaths in the U.S. Medicare population.
Surg Infect (Larchmt)
2012
;
13
:
307
11
11.
Yamamichi
T
,
Ichinose
J
,
Omura
K
,
Hashimoto
K
,
Matsuura
Y
,
Nakao
M
,
Okumura
S
,
Ikeda
N
,
Mun
M
:
Impact of postoperative complications on the long-term outcome in lung cancer surgery.
Surg Today
2022
;
52
:
1254
61
12.
Vandenbroucke
JP
,
von Elm
E
,
Altman
DG
,
Gøtzsche
PC
,
Mulrow
CD
,
Pocock
SJ
,
Poole
C
,
Schlesselman
JJ
,
Egger
M
;
STROBE Initiative
:
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and elaboration.
PLoS Med
2007
;
4
:
e297
13.
Botto
F
,
Alonso-Coello
P
,
Chan
MT
,
Villar
JC
,
Xavier
D
,
Srinathan
S
,
Guyatt
G
,
Cruz
P
,
Graham
M
,
Wang
CY
,
Berwanger
O
,
Pearse
RM
,
Biccard
BM
,
Abraham
V
,
Malaga
G
,
Hillis
GS
,
Rodseth
RN
,
Cook
D
,
Polanczyk
CA
,
Szczeklik
W
,
Sessler
DI
,
Sheth
T
,
Ackland
GL
,
Leuwer
M
,
Garg
AX
,
Lemanach
Y
,
Pettit
S
,
Heels-Ansdell
D
,
Luratibuse
G
,
Walsh
M
,
Sapsford
R
,
Schünemann
HJ
,
Kurz
A
,
Thomas
S
,
Mrkobrada
M
,
Thabane
L
,
Gerstein
H
,
Paniagua
P
,
Nagele
P
,
Raina
P
,
Yusuf
S
,
Devereaux
PJ
,
Devereaux
PJ
,
Sessler
DI
,
Walsh
M
,
Guyatt
G
,
McQueen
MJ
,
Bhandari
M
,
Cook
D
,
Bosch
J
,
Buckley
N
,
Yusuf
S
,
Chow
CK
,
Hillis
GS
,
Halliwell
R
,
Li
S
,
Lee
VW
,
Mooney
J
,
Polanczyk
CA
,
Furtado
MV
,
Berwanger
O
,
Suzumura
E
,
Santucci
E
,
Leite
K
,
Santo
JA
,
Jardim
CA
,
Cavalcanti
AB
,
Guimaraes
HP
,
Jacka
MJ
,
Graham
M
,
McAlister
F
,
McMurtry
S
,
Townsend
D
,
Pannu
N
,
Bagshaw
S
,
Bessissow
A
,
Bhandari
M
,
Duceppe
E
,
Eikelboom
J
,
Ganame
J
,
Hankinson
J
,
Hill
S
,
Jolly
S
,
Lamy
A
,
Ling
E
,
Magloire
P
,
Pare
G
,
Reddy
D
,
Szalay
D
,
Tittley
J
,
Weitz
J
,
Whitlock
R
,
Darvish-Kazim
S
,
Debeer
J
,
Kavsak
P
,
Kearon
C
,
Mizera
R
,
O’Donnell
M
,
McQueen
M
,
Pinthus
J
,
Ribas
S
,
Simunovic
M
,
Tandon
V
,
Vanhelder
T
,
Winemaker
M
,
Gerstein
H
,
McDonald
S
,
O’Bryne
P
,
Patel
A
,
Paul
J
,
Punthakee
Z
,
Raymer
K
,
Salehian
O
,
Spencer
F
,
Walter
S
,
Worster
A
,
Adili
A
,
Clase
C
,
Cook
D
,
Crowther
M
,
Douketis
J
,
Gangji
A
,
Jackson
P
,
Lim
W
,
Lovrics
P
,
Mazzadi
S
,
Orovan
W
,
Rudkowski
J
,
Soth
M
,
Tiboni
M
,
Acedillo
R
,
Garg
A
,
Hildebrand
A
,
Lam
N
,
Macneil
D
,
Mrkobrada
M
,
Roshanov
PS
,
Srinathan
SK
,
Ramsey
C
,
John
PS
,
Thorlacius
L
,
Siddiqui
FS
,
Grocott
HP
,
McKay
A
,
Lee
TW
,
Amadeo
R
,
Funk
D
,
McDonald
H
,
Zacharias
J
,
Villar
JC
,
Cortés
OL
,
Chaparro
MS
,
Vásquez
S
,
Castañeda
A
,
Ferreira
S
,
Coriat
P
,
Monneret
D
,
Goarin
JP
,
Esteve
CI
,
Royer
C
,
Daas
G
,
Chan
MT
,
Choi
GY
,
Gin
T
,
Lit
LC
,
Xavier
D
,
Sigamani
A
,
Faruqui
A
,
Dhanpal
R
,
Almeida
S
,
Cherian
J
,
Furruqh
S
,
Abraham
V
,
Afzal
L
,
George
P
,
Mala
S
,
Schünemann
H
,
Muti
P
,
Vizza
E
,
Wang
CY
,
Ong
GS
,
Mansor
M
,
Tan
AS
,
Shariffuddin
II
,
Vasanthan
V
,
Hashim
NH
,
Undok
AW
,
Ki
U
,
Lai
HY
,
Ahmad
WA
,
Razack
AH
,
Malaga
G
,
Valderrama-Victoria
V
,
Loza-Herrera
JD
,
De Los Angeles Lazo
M
,
Rotta-Rotta
A
,
Szczeklik
W
,
Sokolowska
B
,
Musial
J
,
Gorka
J
,
Iwaszczuk
P
,
Kozka
M
,
Chwala
M
,
Raczek
M
,
Mrowiecki
T
,
Kaczmarek
B
,
Biccard
B
,
Cassimjee
H
,
Gopalan
D
,
Kisten
T
,
Mugabi
A
,
Naidoo
P
,
Naidoo
R
,
Rodseth
R
,
Skinner
D
,
Torborg
A
,
Paniagua
P
,
Urrutia
G
,
Maestre
ML
,
Santaló
M
,
Gonzalez
R
,
Font
A
,
Martínez
C
,
Pelaez
X
,
De Antonio
M
,
Villamor
JM
,
García
JA
,
Ferré
MJ
,
Popova
E
,
Alonso-Coello
P
,
Garutti
I
,
Cruz
P
,
Fernández
C
,
Palencia
M
,
Díaz
S
,
Del Castillo
T
,
Varela
A
,
de Miguel
A
,
Muñoz
M
,
Piñeiro
P
,
Cusati
G
,
Del Barrio
M
,
Membrillo
MJ
,
Orozco
D
,
Reyes
F
,
Sapsford
RJ
,
Barth
J
,
Scott
J
,
Hall
A
,
Howell
S
,
Lobley
M
,
Woods
J
,
Howard
S
,
Fletcher
J
,
Dewhirst
N
,
Williams
C
,
Rushton
A
,
Welters
I
,
Leuwer
M
,
Pearse
R
,
Ackland
G
,
Khan
A
,
Niebrzegowska
E
,
Benton
S
,
Wragg
A
,
Archbold
A
,
Smith
A
,
McAlees
E
,
Ramballi
C
,
Macdonald
N
,
Januszewska
M
,
Stephens
R
,
Reyes
A
,
Paredes
LG
,
Sultan
P
,
Cain
D
,
Whittle
J
,
Del Arroyo
AG
,
Sessler
DI
,
Kurz
A
,
Sun
Z
,
Finnegan
PS
,
Egan
C
,
Honar
H
,
Shahinyan
A
,
Panjasawatwong
K
,
Fu
AY
,
Wang
S
,
Reineks
E
,
Nagele
P
,
Blood
J
,
Kalin
M
,
Gibson
D
,
Wildes
T
;
Vascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) Writing Group, on behalf of The Vascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) Investigators
:
Myocardial injury after noncardiac surgery: A large, international, prospective cohort study establishing diagnostic criteria, characteristics, predictors, and 30-day outcomes.
Anesthesiology
2014
;
120
:
564
78
14.
Devereaux
PJ
,
Biccard
BM
,
Sigamani
A
,
Xavier
D
,
Chan
MTV
,
Srinathan
SK
,
Walsh
M
,
Abraham
V
,
Pearse
R
,
Wang
CY
,
Sessler
DI
,
Kurz
A
,
Szczeklik
W
,
Berwanger
O
,
Villar
JC
,
Malaga
G
,
Garg
AX
,
Chow
CK
,
Ackland
G
,
Patel
A
,
Borges
FK
,
Belley-Cote
EP
,
Duceppe
E
,
Spence
J
,
Tandon
V
,
Williams
C
,
Sapsford
RJ
,
Polanczyk
CA
,
Tiboni
M
,
Alonso-Coello
P
,
Faruqui
A
,
Heels-Ansdell
D
,
Lamy
A
,
Whitlock
R
,
LeManach
Y
,
Roshanov
PS
,
McGillion
M
,
Kavsak
P
,
McQueen
MJ
,
Thabane
L
,
Rodseth
RN
,
Buse
GAL
,
Bhandari
M
,
Garutti
I
,
Jacka
MJ
,
Schünemann
HJ
,
Cortes
OL
,
Coriat
P
,
Dvirnik
N
,
Botto
F
,
Pettit
S
,
Jaffe
AS
,
Guyatt
GH
;
Writing Committee for the VISION Study Investigators
:
Association of postoperative high-sensitivity troponin levels with myocardial injury and 30-day mortality among patients undergoing noncardiac surgery.
JAMA
2017
;
317
:
1642
51
15.
Devereaux
PJ
,
Chan
MTV
,
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
;
Vascular Events In Noncardiac Surgery Patients Cohort Evaluation (VISION) Study Investigators
:
Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery.
JAMA
2012
;
307
:
2295
304
16.
Roshanov
PS
,
Eikelboom
JW
,
Sessler
DI
,
Kearon
C
,
Guyatt
GH
,
Crowther
M
,
Tandon
V
,
Borges
FK
,
Lamy
A
,
Whitlock
R
,
Biccard
BM
,
Szczeklik
W
,
Panju
M
,
Spence
J
,
Garg
AX
,
McGillion
M
,
VanHelder
T
,
Kavsak
PA
,
de Beer
J
,
Winemaker
M
,
Le Manach
Y
,
Sheth
T
,
Pinthus
JH
,
Siegal
D
,
Thabane
L
,
Simunovic
MRI
,
Mizera
R
,
Ribas
S
,
Devereaux
PJ
:
Bleeding Independently associated with Mortality after noncardiac Surgery (BIMS): An international prospective cohort study establishing diagnostic criteria and prognostic importance.
Br J Anaesth
2021
;
126
:
163
71
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.
Royston
P
,
Parmar
MKB
:
Flexible parametric proportional-hazards and proportional-odds models for censored survival data, with application to prognostic modelling and estimation of treatment effects.
Stat Med
2002
;
21
:
2175
97
20.
Hernán
MA
:
The hazards of hazard ratios.
Epidemiology
2010
;
21
:
13
5
21.
Li
K-H
,
Meng
X-L
,
Raghunathan
TE
,
Rubin
DB
:
Significance levels from repeated P-values with multiply-imputed data.
Stat Sin
1991
;
1
:
65
92
22.
Samuelsen
SO
,
Eide
GE
:
Attributable fractions with survival data.
Stat Med
2008
;
27
:
1447
67
23.
Lesko
CR
,
Edwards
JK
,
Moore
RD
,
Lau
B
:
Censoring for loss to follow-up in time-to-event analyses of composite outcomes or in the presence of competing risks.
Epidemiology
2019
;
30
:
817
24
24.
Beattie
WS
,
Lalu
M
,
Bocock
M
,
Feng
S
,
Wijeysundera
DN
,
Nagele
P
,
Fleisher
LA
,
Kurz
A
,
Biccard
B
,
Leslie
K
,
Howell
S
,
Landoni
G
,
Grocott
H
,
Lamy
A
,
Richards
T
,
Myles
P
;
StEP COMPAC Group
:
Systematic review and consensus definitions for the Standardized Endpoints in Perioperative Medicine (StEP) initiative: Cardiovascular outcomes.
Br J Anaesth
2021
;
126
:
56
66
25.
Peduzzi
P
,
Concato
J
,
Feinstein
AR
,
Holford
TR
:
Importance of events per independent variable in proportional hazards regression analysis. II. Accuracy and precision of regression estimates.
J Clin Epidemiol
1995
;
48
:
1503
10
26.
Westreich
D
,
Greenland
S
:
The table 2 fallacy: Presenting and interpreting confounder and modifier coefficients.
Am J Epidemiol
2013
;
177
:
292
8
27.
Gaskell
AL
,
Sleigh
JW
:
An introduction to causal diagrams for anesthesiology research.
Anesthesiology
2020
;
132
:
951
67
28.
Devereaux
PJ
,
Duceppe
E
,
Guyatt
G
,
Tandon
V
,
Rodseth
R
,
Biccard
BM
,
Xavier
D
,
Szczeklik
W
,
Meyhoff
CS
,
Vincent
J
,
Franzosi
MG
,
Srinathan
SK
,
Erb
J
,
Magloire
P
,
Neary
J
,
Rao
M
,
Rahate
PV
,
Chaudhry
NK
,
Mayosi
B
,
de Nadal
M
,
Iglesias
PP
,
Berwanger
O
,
Villar
JC
,
Botto
F
,
Eikelboom
JW
,
Sessler
DI
,
Kearon
C
,
Pettit
S
,
Sharma
M
,
Connolly
SJ
,
Bangdiwala
SI
,
Rao-Melacini
P
,
Hoeft
A
,
Yusuf
S
;
MANAGE Investigators
:
Dabigatran in patients with myocardial injury after non-cardiac surgery (MANAGE): An international, randomised, placebo-controlled trial.
Lancet
2018
;
391
:
2325
34
29.
Foucrier
A
,
Rodseth
R
,
Aissaoui
M
,
Ibanes
C
,
Goarin
JP
,
Landais
P
,
Coriat
P
,
Le Manach
Y
:
The long-term impact of early cardiovascular therapy intensification for postoperative troponin elevation after major vascular surgery.
Anesth Analg
2014
;
119
:
1053
63
30.
Robich
MP
,
Sabik
JF
,
Houghtaling
PL
,
Kelava
M
,
Gordon
S
,
Blackstone
EH
,
Koch
CG
:
Prolonged effect of postoperative infectious complications on survival after cardiac surgery.
Ann Thorac Surg
2015
;
99
:
1591
9
31.
Wyss
R
,
Gagne
JJ
,
Zhao
Y
,
Zhou
EH
,
Major
JM
,
Wang
SV
,
Desai
RJ
,
Franklin
JM
,
Schneeweiss
S
,
Toh
S
,
Johnson
M
,
Fireman
B
:
Use of time-dependent propensity scores to adjust hazard ratio estimates in cohort studies with differential depletion of susceptibles.
Epidemiology
2020
;
31
:
82
9