Abstract
Introduction:
The American Heart Association guidelines instruct use of post-operative telemetry (POT) should be reserved for patients undergoing cardiac procedures and/or those with ischemic cardiac symptoms, but acknowledge that major vascular procedures deserve unique consideration. Telemetry remains a limited resource in many hospitals, however it has been poorly defined which vascular patients have greatest need for POT. The purpose of this study is to define the rates of post-operative arrhythmias (POA) after major vascular operations using the Society for Vascular Surgery Vascular Quality Initiative (VQI) registry, identify independent predictors of POA, and determine the effect of POA on mortality in order to guide the use of POT in vascular patients.
Methods:
A retrospective cohort study was performed using the following VQI modules: open abdominal aortic aneurysm repair (oAAA), complex EVAR (TEVAR/c-EVAR), endovascular aneurysm repair (EVAR), suprainguinal bypass (SIB) and infrainguinal bypass (IIB). POA was defined in the VQI as a new rhythm disturbance requiring treatment with medication or cardioversion. The incidence of POA, pre-operative risk factors and demographics were determined for each procedure.
Results:
A total of 121,652 procedures were identified with an overall POA event rate of 5.1% (n=6,265). Procedure specific event rates for POA among VQI registries are as follows: oAAA 14.4%, TEVAR/c-EVAR 8.5%, EVAR 2.7%, SIB 6.2%, and IIB 3.8%. Across all procedure types, POA was associated with emergent operations and increased procedure time. Procedure specific multivariable regression revealed additional independent preoperative intraoperative factors associated with POA that were unique with each procedure. Across all procedural groups, the presence of POA was associated with increased rates of clinical MI and decreased survival on Kaplan Meier analysis.
Conclusion:
Rates of post-operative arrhythmia in patients undergoing vascular procedures appears higher than previously reported, and POA is associated with decreased survival. Our study elucidated patient and procedure specific predictor factors associated with POA that can be used to inform the use of post-operative telemetry.
Table of Contents Summary
In a retrospective review of Vascular Quality Initiative (VQI) data, the rate of post-operative arrhythmia (POA) was 5.1% and was associated with decreased survival across all procedural cohorts. Predicting vascular patients who require post-operative telemetry monitoring is clinically important given the association of POA with poor outcomes.
Introduction
In patients undergoing vascular surgery procedures the rate and clinical impact of postoperative arrhythmias (POA) is unknown. POA are not uncommon among surgical patients1 and are associated with poor clinical outcomes2 but this relationship has not been studied specifically within the vascular surgery population. Vascular patients have increased cardiovascular comorbidities3 and high rates of cardiac ischemia4,5 compared to the general surgical population and therefore POA may play a more significant role in clinical outcomes than previously realized.
Prior series have attempted to define the rates of POA in patients undergoing primarily open abdominal aortic aneurysm repair (oAAA) and endovascular abdominal aortic aneurysm repairs (EVAR) but were limited due to small sample size. In 2001, Valentine et al., studied 211 consecutive ICU patients who underwent elective aortic operations where atrial fibrillation (AF) was found in 10% (N=21) of patients6. Surprisingly, AF was not associated with increased morbidity or mortality in that study. This is in contrast to other studies showing that AF in patients undergoing noncardiac general surgery procedures was associated with increased mortality7–10.
Here we sought to use the national Society for Vascular Surgery Vascular Quality Initiative (SVS VQI) registry to evaluate the incidence, predictors, and effect of POA on survival in procedural subgroups of vascular patients.
Methods
This study was approved by the SVS VQI Research Advisory Council, as well as the Institutional Review Board (IRB) at the University of Alabama at Birmingham. The need for informed consent was waived given that the study analyzed existing clinical registry data with no direct patient contact and/or potential for harm.
Vascular Quality Initiative Database
The SVS VQI is the largest vascular quality improvement database and is comprised of over 700 participating centers11. VQI data entry is provided by individual physicians and validated by the VQI against hospital insurance claims to evaluate for missing procedures to ensure complete reporting. Perioperative and one-year follow-up data are collected for each procedure, and the Social Security Death Index (SSDI) is matched to the registry to allow long-term assessment of mortality. Not all procedural registries have existed over the duration of the VQI, and therefore not all procedures have the same number of eras available for analysis. Specifically, TEVAR/complex EVAR and suprainguinal bypass registries were not developed until 2009; therefore, the time frames for analysis in these two procedural groups begin at 2009 rather than 2003.
Study Design
A retrospective cohort study was performed using the following independent procedural registries for which POA data were recorded: open aortic repair (oAAA), thoracic and complex EVAR (TEVAR/c-EVAR), endovascular aneurysm repair (EVAR), suprainguinal bypass (SIB), and infrainguinal bypass (IIB). Demographics, preoperative, and intraoperative risk factors were compared. The primary outcome assessed was POA event rates (defined in the VQI as a new rhythm disturbance requiring treatment with medication or cardioversion). Secondary outcomes included mortality and clinical post-operative myocardial infarction (POMI). POMI was defined as a clinical MI with troponin and EKG changes.
Covariates Examined
Preoperative demographics analyzed included: age, race, gender, primary insurer and weight (kg). Comorbidities assessed included: coronary artery disease (CAD), chronic obstructive pulmonary disease (COPD), cerebrovascular disease (CVD), congestive heart failure (CHF), hypertension (HTN), dialysis-dependent end-stage renal disease (ESRD), preoperative creatinine, preoperative ejection fraction, stress test results, diabetes mellitus (insulin or non-insulin requiring), smoking (current/former/never), prior percutaneous coronary intervention (PCI) and/or coronary artery bypass grafting (CABG), ankle brachial index (ABI), preoperative ambulatory status and preoperative lower extremity symptoms (i.e., indication for intervention: claudication, chronic limb threatening ischemia), where available. Preoperative medications that were analyzed included use of aspirin (ASA), P2Y12 inhibitors, dual antiplatelet therapy (DAPT), statins, angiotensinogen converting enzyme (ACE) inhibitors, and beta blockers.
Capture of intraoperative variables was registry dependent. Where available, intraoperative variables analyzed included urgency of procedure (i.e. elective, urgent, emergent), procedure time (in hours), American Society of Anesthesiologists (ASA) class, intraoperative heparin administration, proximal extent of aortic disease by zone (grouped into zones 0–2, 3–5, and 6–9), number of aortic devices used, intravascular ultrasound use (IVUS), location of proximal and distal clamp for open aortic procedures, intraoperative crystalloid volume infused (in liters), intra-operative transfusions, intraoperative estimated blood loss (EBL in liters), return to operating room, concomitant procedures, and intraoperative complications.
Statistical Analysis
Patient preoperative characteristics and intraoperative variables were compared by presence or absence of POA using χ2 for categorical variables, t-tests for continuous variables. Multivariable logistic regression models predicting POA were created for each individual procedure in a forward stepwise variable reduction algorithm with probability to enter the model set at p<0.10. Postoperative survival was evaluated with Kaplan-Meier analysis.
Results
Between 2003 and 2018 a total of 121,652 individual vascular patients meeting inclusion criteria were identified: 10,603 oAAA, 10,820 TEVAR/c-EVAR, 40,077 EVAR, 14,747 SIB, and 44,705 IIB. Demographics for all of the identified groups are presented in Table 1. The overall clinically significant POA rate was 5.1% among all registries. Specifically, the POA rates for each subgroup were 14.4% in oAAA, 8.5% in TEVAR/c-EVAR, 2.7% in EVAR, 6.2% in SIB, and 3.8% in IIB (Figure 1).
Table 1:
Patient Demographics
| oAAA (n=10,603) | TEVAR/c-EVAR (n=10,820) | EVAR (n=40,777) | SIB (n=14,474) | IIB (n=44,705) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|||||||||||||||
| Variable | No POA (n=9,081) | POA (n=1,522) | p | No POA (n=9,990) | POA (n=920) | p | No POA (n=39,583) | POA (n=1,194) | p | No POA (n=13,560) | POA (n=914) | p | No POA (n=42,990) | POA (n=1,715) | p |
|
|
|||||||||||||||
| Age (years) | 69.4 ± 8.9 | 72.4 ± 8.3 | 0.82 | 67 ± 14.4 | 71 ± 12.1 | 0.79 | 73.3 ± 8.9 | 76.1 ± 8.6 | 0.3 | 63.6 ± 10.8 | 66.9 ± 10.1 | 0.77 | 66.5 ± 11.2 | 70.8 ± 10.1 | 0.66 |
| Male Gender | 74.2% | 74.7% | 0.67 | 64.5% | 64.6% | 0.95 | 80.9% | 76.7% | 0.65 | 59.3% | 60.8% | 0.38 | 68.9% | 68.1% | 0.48 |
| Caucasian Race | 90.6% | 91.6% | 0.61 | 73.8% | 77.2% | 0.31 | 90.3% | 88.3% | 0.86 | 86.2% | 88.2% | 0.2 | 81.2% | 81.8% | 0.82 |
| CAD | 25.9% | 28.9% | 0.013 | 20.9% | 23.9% | 0.07 | 29.3% | 30.3% | 0.66 | 26.9% | 32.0% | <0.001 | 31.3% | 38.0% | 0.15 |
| HTN | 82.5% | 84.8% | 0.2 | 84.4% | 87.6% | 0.01 | 83.0% | 86.1% | 0.15 | 84.2% | 86.3% | 0.72 | 87.7% | 90.6% | 0.05 |
| CHF | 7.6% | 10.0% | 0.22 | 11.6% | 14.9% | 0.32 | 12.1% | 16.4% | 0.53 | 11.1% | 17.0% | 0.32 | 16.4% | 24.7% | <0.001 |
| Diabetes | 15.9% | 16.9% | 0.74 | 11.6% | 14.9% | 0.25 | 20.1% | 20.9% | 0.89 | 26.9% | 32.0% | 0.13 | 48.6% | 52.0% | <0.001 |
| Dialysis | 0.6% | 0.5% | 0.64 | 2.5% | 2.5% | 0.82 | 1.1% | 1.2% | 0.97 | 1.5% | 2.1% | 0.33 | 5.5% | 10.0% | <0.001 |
| COPD | 32.5% | 38.4% | <0.001 | 28.0% | 31.6% | 0.04 | 32.7% | 38.8% | <0.001 | 34.8% | 38.7% | 0.02 | 26.4% | 31.1% | 0.03 |
| Smoking Status | |||||||||||||||
| Never | 10.5% | 10.5% | 25.3% | 28.8% | 14.0% | 17.1% | 5.5% | 6.2% | 15.8% | 18.5% | |||||
| Prior | 45.2% | 45.5% | 0.99 | 43.1% | 40.7% | 0.06 | 53.9% | 51.1% | 0.82 | 37.3% | 42.1% | 0.83 | 43.3% | 48.9% | 0.52 |
| Current | 44.2% | 44.1% | 31.6% | 30.5% | 32.1% | 31.8% | 57.1% | 51.9% | 40.9% | 32.6% | |||||
| Hx of CABG | 28.7% | 29.9% | 0.12 | 22.1% | 24.0% | 0.06 | 33.9% | 32.8% | 0.78 | 26.9% | 32.5% | 0.15 | 34.4% | 39.8% | 0.12 |
| Pre-Op Medication | |||||||||||||||
| Anti-Platelet | 60.8% | 61.6% | 0.43 | 54.7% | 53.5% | 0.89 | 64.0% | 60.3% | 0.03 | 69.4% | 70.5% | 0.74 | 71.4% | 73.3% | 0.56 |
| Beta-Blocker | 57.9% | 58.9% | 0.14 | 62.1% | 64.0% | 0.13 | 54.8% | 54.7% | 0.92 | 51.6% | 59.0% | <0.001 | 58.3% | 66.8% | <0.001 |
| Statin | 64.8% | 62.0% | 0.24 | 55.6% | 54.5% | 0.53 | 68.4% | 63.1% | 0.02 | 69.8% | 73.8% | 0.03 | 70.7% | 73.3% | 0.22 |
| ACE inhibitor | 42.6% | 43.2% | 0.14 | 40.6% | 39.9% | 0.51 | 44.6% | 40.4% | 0.06 | 45.9% | 51.6% | 0.02 | 50.7% | 49.7% | 0.06 |
oAAA, open aortic repair; TEVAR/c-EVAR, Thoracic Endovascular Aortic Repair and complex endovascular repair; EVAR, endovascular aortic repair; SIB, Suprainguinal bypass; IIB, infrainguinal bypass; SD, Standard Deviation; WTKG, weight in kilograms; CAD, coronary artery disease; HTN, hypertension; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CABG, coronary artery bypass graft; ACE, angiotensinogen converting enzyme
Figure 1:
Temporal Changes in Post-operative Arrhythmia Event Rates. Key: EVAR- endovascular aneurysm repair, oAAA- open abdominal aortic aneurysm, SIB- Suprainguinal bypass, IIB- Infrainguinal bypass, TEVAR/C-EVAR- thoracic endovascular aortic repair/Complex-endovascular aneurysm repair
POA and Associations with Open Abdominal Aortic Aneurysm Repair (oAAA)
The POA rate in oAAA was 14.4% (n=1,522/10,603), which was the highest among all the procedural subgroups. Across the studied timeframe the rate of POA significantly decreased in oAAA (15.9% to 14.2%, p<0.001) (Figure 1). POA was significantly associated with increased rates of clinical post-operative MI (POMI) (10.6% vs 1.7%, p<0.001). ICU stay was associated with a significant difference in POA rate (14.7% vs 4.5%, p<0.001), however temporal relationships between POA and post-operative ICU care cannot be determined in this dataset. Decreased survival in patients with POA was seen as early at 30 days; Kaplan-Meier analysis demonstrated significantly reduced 2 year survival in patients with POA (68% vs 88%, log rank<0.001).
Multivariable regression analysis identified patient characteristics and intra-operative factors that were associated with POA in oAAA (Table 2). Notable patient characteristics associated with POA include age (OR 1.04 [1.03–1.05]; p<0.001) and reduced ejection fraction (30–50%) (OR 1.35[1.04–1.42]; p=.046). Intraoperative factors associated with POA included emergent procedure status (OR 2.11 [1.68–2.60]; p<0.001) and supra-celiac clamp compared to infrarenal clamp (OR 1.34 [1.06–1.69]; p=.014).
Table 2:
Risk Factos Associated with POA in oAAA
| Significant prognostic factors | Odds Ratio | 95% C.I. | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Preoperative F actors: | ||||
| Age (per yr) | 1.037 | 1.027 | 1.046 | <0.001 |
| EF 30–50% | 1.348 | 1.005 | 1.809 | 0.046 |
| COPD | 1.217 | 1.043 | 1.421 | 0.013 |
| Dual Antiplatelet | 1.567 | 1.201 | 2.044 | 0.001 |
| Operative Factors: | ||||
| Supraceliac Clamp** | 1.338 | 1.061 | 1.687 | 0.014 |
| Emergent | 2.111 | 1.683 | 2.647 | <0.001 |
| Return to OR | 1.758 | 1.437 | 2.151 | <0.001 |
| Total Procedure | 1.002 | 1.001 | 1.048 | <0.001 |
referenced against infrarenal clamp
Rates and Factors Associated with POA in TEVAR/c-EVAR
The POA rate in patients undergoing TEVAR/complex EVAR was 8.5% (n=920/10,820). Across the studied timeframe the rate of POA did not significantly decrease in TEVAR/c-EVAR (9% to 8.2%, p>0.05) (Figure 1). POA was significantly associated with increased rates of clinical post-operative MI (POMI) (8.2% vs 0.8%, p<0.001). ICU stay was associated with a significant difference in POA rate (9.6% vs 3.7%, p<0.001), however temporal relationships between POA and post-operative ICU care cannot be determined in this dataset. Decreased survival in patients with POA was seen as early at 30 days; Kaplan-Meier analysis demonstrated significantly reduced 2 year survival in patients with POA (62% vs 88%, log rank<0.001).
Patient and procedural characteristics associated with POA within TEVAR/c-EVAR are shown in Table 3. Patients that were non-ambulatory were at significantly increased risk for POA (OR 4.8 [1.63–14.41]; p=0.005). Emergent procedure status was significantly associated with POA (OR 1.55 [1.09–2.22]; p=0.014). Key intraoperative factors associated with increased odds of POA included location of proximal device placement (zone 0–2 deployment, OR 1.94[1.43–2.61]; p<0.001), greater number of aortic devices (OR 1.73 [1.05–2.86]; p=0.031), and increased hourly procedural time (OR 1.11 [1.06–1.17], p<0.001).
Table 3:
Risk Factors Associated with POA in TEVAR/c-EVAR
| Significant prognostic factors | Odds Ratio | 95 % C.I. | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Preoperative Factors: | ||||
| Creatine (per mg/dL) | 1.147 | 1.028 | 1.28 | 0.014 |
| Prior CABG <5 years ago | 0.556 | 0.313 | 0.989 | 0.046 |
| Nonambulatory* | 4.847 | 1.63 | 14.41 | 0.005 |
| OSH Transfer | 1.706 | 1.375 | 2.117 | <0.001 |
| Emergent Operation | 1.559 | 1.095 | 2.219 | 0.014 |
| ASA Class 4 | 4.911 | 1.202 | 20.065 | 0.027 |
| ASA Class 5 | 8.421 | 1.956 | 36.266 | 0.004 |
| Operative Factors: | ||||
| Proximal Aortic Disease Zone 0–2** | 1.935 | 1.434 | 2.611 | <0.001 |
| Proximal Aortic Zone Disease 3–5** | 1.516 | 1.182 | 1.943 | <0.001 |
| Number Aortic Devices (3–4)*** | 1.293 | 1.039 | 1.609 | 0.021 |
| Number Aortic Devices (5–6)*** | 1.733 | 1.05 | 2.858 | 0.031 |
| Crystalloid (per L) | 1.095 | 1.02 | 1.175 | 0.012 |
| Procedure time | 1.114 | 1.058 | 1.173 | <0.001 |
referenced against independently ambulatory
referenced against proximal aortic zone disease 6–9
referenced against 1–2 aortic devices
POA and Associations with Endovascular Aortic Aneurysm Repair (EVAR)
The POA rate in EVAR was 2.7% (n=1,194/44,077), the lowest among all the subgroups analyzed. Across the studied timeframe the rate of POA significantly decreased in EVAR (4% to 2.2%, p<0.001) (Figure 1). POA was significantly associated with increased rates of clinical post-operative MI (POMI) (8.5% vs 0.4%, p<0.001). ICU stay was associated with a significant difference in POA rate (5.8% vs 1.3%, p<0.001), however temporal relationships between POA and post-operative ICU care cannot be determined in this dataset. Decreased survival in patients with POA was seen as early at 30 days; Kaplan-Meier analysis demonstrated significantly reduced 2 year survival in patients with POA (70% vs 91%, log rank<0.001).
Patient and procedural factors associated with POA in patients undergoing EVAR are summarized in Table 4. Pre-operatively, patients that presented with symptomatic (OR 1.63[1.14–2.32]; p=0.008) or ruptured aneurysms (OR 4.76[ 3.26–6.97]; p<0.001) were at significantly higher risk for POA. Procedural time (OR 1.004 [1.003–1.005]; p<0.001) and reoperation (OR 3.04[2.03–4.54]; p<.001) were both significantly associated with higher rates of POA.
Table 4:
Risk Factors Associated with POA in EVAR
| Significant prognostic factors | Odds Ratio | 95% C.I. | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Preoperative Factors: | ||||
| Age (per yr) | 1.041 | 1.028 | 1.055 | <0.001 |
| OSH Transfer | 1.5 | 1.065 | 2.112 | 0.02 |
| Symptomatic Aneurysm | 1.625 | 1.136 | 2.324 | 0.008 |
| Ruptured Aneurysm | 4.763 | 3.257 | 6.965 | <0.001 |
| Unfit for oAAA Repair | 1.377 | 1.07 | 1.772 | 0.013 |
| Operative Factors: | ||||
| Major amputation due to embolization | 10.98 | 1.565 | 77.038 | 0.016 |
| Reoperation | 3.036 | 2.031 | 4.54 | <0.001 |
| Crystalloid (per L) | 1.097 | 1.009 | 1.191 | 0.029 |
| Total Procedure Time | 1.004 | 1.003 | 1.005 | <0.001 |
POA and Associations with Suprainguinal bypass (SIB)
The rate of POA in SIB was 6.2% (n=914/14,474). Across the studied timeframe the rate of POA significantly decreased in SIB (8.8% to 5.8%, p=0.003) (Figure 1). POA was significantly associated with increased rates of clinical post-operative MI (POMI) (9.8% vs 1.0%, p<0.001). ICU stay was associated with a significant difference in POA rate (8.6% vs 2.8%, p<0.001), however temporal relationships between POA and post-operative ICU care cannot be determined in this dataset. Decreased survival in patients with POA was seen as early at 30 days; Kaplan-Meier analysis demonstrated significantly reduced 2 year survival in patients with POA (70% vs 89%, log rank<0.001).
Patient characteristics and intraoperative factors associated with increased risk for POA in this patient population are shown in Table 5. Preoperatively, a non-ambulatory status (OR 1.66[1.19–2.33]; p=0.011) and a history of symptomatic congestive heart failure (OR 1.49; [95% CI], 1.09–2.02; p=.011) were significantly associated with POA. Intraoperative factors associated with POA included longer operative times (per hour increase: OR 1.14[1.09–1.18]; p<.001) and higher EBL (per liter, OR 1.13[1.13–1.22]; p=.002).
Table 5:
Risk Factors Associated with POA in SIB
| Significant prognostic factors | Odds Ratio | 95% C.I. | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Preoperative Factors: | ||||
| Age (per yr) | 1.028 | 1.019 | 1.037 | <0.001 |
| Preop Symptomatic CHF | 1.489 | 1.094 | 2.026 | 0.011 |
| Nonambulatory | 1.663 | 1.185 | 2.332 | 0.003 |
| ASA Class 5 | 3.028 | 1.352 | 6.783 | 0.007 |
| OSH Transfer | 1.463 | 1.153 | 1.855 | 0.002 |
| Operative Factors: | ||||
| Total Procedure Time | 1.135 | 1.091 | 1.182 | <0.001 |
| EBL (per L) | 1.131 | 1.048 | 1.221 | 0.002 |
POA and Associations with Infrainguinal Bypass (IIB)
IIB were found to have a POA rate of 3.8% (N=1,715/44,705). Across the studied timeframe the rate of POA did not significantly decreased in IIB (4% to 3.7%, p>0.05) (Figure 1). POA was significantly associated with increased rates of clinical post-operative MI (POMI) (13% vs 1.0%, p<0.001). ICU designation was not available for this registry. Decreased survival in patients with POA was seen as early at 30 days; Kaplan-Meier analysis demonstrated significantly reduced 2 year survival in patients with POA (66% vs 87%, log rank<0.001).
Patient characteristics and intraoperative factors associated with increased risk for POA in this patient population are shown in Table 6. Preoperative factors associated with POA in IIB included non-ambulatory status (1.13 [1.00–1.29]; p=0.016), dialysis dependence (OR 1.62[1.35–1.95]; p<0.001), and emergent operative status (OR 1.16[1.03–1.32]; p=0.018). Intraoperatively, total operative time (OR 1.05[1.03–1.08]; p<0.001) and reoperations (OR 2.00[1.76–2.45]; p<0.001) were associated with increased POA.
Table 6:
Risk Factors Associated with POA in IIB
| Significant prognostic factors | Odds Ratio | 95% C.I. | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Preoperative Factors: | ||||
| Age (per yr) | 1.037 | 1.032 | 1.042 | <0.001 |
| Hx COPD | 1.199 | 1.069 | 1.346 | 0.002 |
| Dialysis Dependent | 1.619 | 1.347 | 1.946 | <0.001 |
| Abnormal Stress Test | 1.574 | 1.342 | 1.846 | <0.001 |
| Nonambulatory | 1.129 | 1.006 | 1.291 | 0.04 |
| Emergent Operation | 1.162 | 1.026 | 1.315 | 0.018 |
| OSH Transfer | 1.279 | 1.067 | 1.533 | 0.008 |
| ASA Class 5 | 4.657 | 1.88 | 11.534 | 0.001 |
| Operative Factors: | ||||
| Total Procedure Time | 1.054 | 1.026 | 1.084 | <0.001 |
| EBL (per L) | 1.334 | 1.206 | 1.084 | <0.001 |
| Reoperation | 2.001 | 1.758 | 2.445 | <0.001 |
Discussion
Predicting vascular patients who require post-operative telemetry monitoring is clinically important given the association of post-operative arrhythmias with poor outcomes and the financial burden placed healthcare systems by improper telemetry deployment. This is the largest dedicated dataset of vascular surgery patients evaluated for post-operative arrhythmia, including a total of 121,652 procedures, with an overall POA event rate 5.2% (n=6,265), with POA rates paralleling the anticipated invasiveness of the procedure (from 14.4% in oAAA to 3.0% in EVAR). The overall event rate (5.2%) was lower than what has previously been shown in vascular literature (10–12%12,13), likely because other studies included all arrhythmias, while this study included only clinically-significant arrhythmias requiring intervention (per VQI definitions). Unsurprisingly, POA events were higher in individuals with ICU stays compared to non-ICU stay, though we are unable to determine the timing of the arrhythmia in relation to the ICU stay, and suspect that this relationship is due to both higher risk for POA being selected for ICU care and ICU care as a result of diagnosed clinically significant arrythmias. This work demonstrates that POA are quite prevalent across all vascular patients and procedures.
A significant amount of literature exists on the rates of post-operative arrhythmias in nonvascular patients. Cardiac procedures have the highest rate of POA, but studies have shown that they also occur in noncardiac surgical patients at rates ranging from 0.5% to 10%14,17–20. The rate of non-cardiac post-operative arrhythmias varies widely based on patient population from 0.37%21 in Minimally Invasive Surgery patients to roughly 10% in colorectal patients22,23. In 2006, Gaunt et al performed a meta-analysis of eight studies including over 8,000 patients undergoing non-cardiac surgery and identified an overall post-operative arrhythmia rate of 7.8%24. While a large cohort was evaluated, its usefulness to vascular surgeons is limited because it includes vascular procedures among all general procedures and does not separate clinically significant arrhythmias from incidentally identified arrhythmias that do not require intervention. Current literature defining procedure-specific rates among vascular patients is sparse.
To date, the rate of POA in vascular patients has been poorly defined. Importantly, the American Heart Association has noted that defining post-operative arrhythmia event rates in vascular surgery patients should be of high priority, and have identified the vascular population as likely high risk for POA, but provide no specific recommendations regarding the use of telemetry to monitor these patients. In 2001, Valentine et al. found a 10% rate of post-operative atrial fibrillation in 211 patients undergoing elective open aortic operations6. Noorani et al. found a similar rate (10%) of atrial fibrillation in their 2009 study of 200 open aortic repairs12. These two studies focus exclusively on open aortic procedures and atrial fibrillation, but do not characterize POA in other vascular procedures. In 2010, Winkle et al. retrospectively reviewed 515 vascular patients after any vascular procedure and found an overall arrhythmia rate of 12%13. To date there has been no other large-scale study to identify rates of arrhythmias in the vascular population as a whole.
The existing literature suggests that most current clinical practices do not utilize telemetry in an impactful manner, likely due to an unknown rate of POA and benefit of POT, causing increased unnecessary financial burden on health care systems and offering no proven outcomes improvement14. Metanalysis have demonstrated that up to one-third of telemetry cases do not meet criteria for telemetry; with improper deployment of telemetry costing an average size hospital roughly $250,000 a year15, and that missed POA increase the cost per patient by $6,35614. In addition to the financial burden of haphazard telemetry use, multiple large-scale studies have demonstrated that POA are associated with longer hospitalizations2, increased risk for stroke16, and mortality. In this study, multivariate analysis was used to identify patient characteristics and intra-operative factors that were independently associated with increased risk of PAO. Age, non-ambulatory patient status, urgency of the procedure, and total intraoperative procedure time were found to be significant independent predictors of increased POA risk among most procedures. However, the impact of total procedure time, while significant, had a small effect of POA risk. Other preoperative and intraoperative factors were unique between procedures, highlighting that POA risk stratification will depend on both procedure type and require evaluation of procedure specific factors. Ultimately, these data can help inform the use of post-operative telemetry, which is not an unlimited resource in most health systems and does confer some additional cost to post-operative care.
Here we have demonstrated that POA is not uncommon in vascular surgery patients, and that high-risk patient populations can be identified using pre-operative risk factors and intraoperative variables unique to each procedure. We have additionally demonstrated that the presence of POA is associated with decreased survival, and serves as a useful marker for high-risk patients. Our intention is to ultimately develop a predictive tool that will guide physicians in their post-operative telemetry use. In the absence of a current predictive tool, factors most easily identified for quick assessment include advanced patient age, non-ambulatory status, and emergent nature of the procedure. Eventually a predictive tool could help appropriate distribute POT use and potentially mitigate unnecessary use of POT, decrease cost, and properly identify patients at risk for POA in order to improve clinical outcomes.
Limitations
While we are able to demonstrate associations, the settings of care throughout index hospitalization and temporal relationships of some of the perioperative factors with POA remains unclear within our data source. Despite being the largest repository of vascular procedural data, the VQI is not entirely comprehensive in capturing data prior to the national expansion in 201125. Missingness of data is therefore unavoidable when working with VQI data, but is anticipated to be of more limited scope based on the perioperative and mortality endpoints studied here and additionally appropriate statistical analysis tools can mitigate inaccuracies and ensure soundness of conclusions. The VQI addresses this limitation by comparing all data against hospital insurance claims.
The VQI definition of POA leaves ambiguity for the possibility of POA occurring during a cardiac arrest, which could skew the relationship between POA and survival and introduce bias leading to an overestimation of the relationship. The VQI does not collect data on ACLS and cardiac arrest so there can be no direct comparison; however, it is the opinion of the authors that the contribution of ACLS patients to the POA patients is low.
Limitations outside of the VQI exist. Our study uses previously published work on non-cardiac surgery POA event rates to put our findings into context, but lacks the ability to directly compare these patients to patients in the VQI. Additionally, patients receiving care in both an ICU or non-ICU setting were included, which may confound issues of hospital policy with medical decision-making or patient monitoring needs.
Conclusion:
Rates of clinically significant post-operative arrhythmias in patients undergoing vascular procedures appears higher than previously reported in noncardiac series and the presence of a POA is associated with decreased survival. Whether the POA is causative of poor outcomes, or if the converse is true, early detection and treatment is certainly important, which may be impacted by the use of telemetry monitoring. Identifying patients at high risk for POA can inform the use of post-operative telemetry, which is a limited resource in many health systems. Future development of a procedure-specific risk calculator to define high risk subgroups will be developed to add to the current selection of VQI calculators available.
Figure 2:
Kaplan-Meier Survival Curve: Key: oAAA- open abdominal aortic aneurysm, TEVAR- thoracic endovascular aortic repair, SIB- Suprainguinal bypass, IIB- Infrainguinal bypass, EVAR- Endovascular aneurysm repair
ARTICLE HIGHLIGHTS.
Type of Research:
Retrospective review of prospectively collected Vascular Quality Initiative (VQI) data.
Key Findings:
The overall rate of post-operative arrhythmia (POA) was 5.1% (n=6,265/121,652).
Procedure specific POA rates range from 3.0% to 14.4%.
POA associated with significantly decreased survival in all cohorts (P<0.001).
Take Home Message:
Rates of post-operative arrhythmia are higher than previously reported.
The presence of POA is associated with decreased survival across all VQI procedural cohorts.
Footnotes
Author Disclosures: No relevant financial disclosures
Presented at the 32nd Annual Meeting for the Florida Vascular Society Hollywood, Florida, Thursday, April 25th, 2019
References:
- 1.Brathwaite D, Weissman C. The new onset of atrial arrhythmias following major noncardiothoracic surgery is associated with increased mortality. Chest. 1998; 114(2): 462–468. doi: 10.1378/chest.114.2.462 [DOI] [PubMed] [Google Scholar]
- 2.Bhave PD, Goldman LE, Vittinghoff E, Maselli J, Auerbach A. Incidence, predictors, and outcomes associated with postoperative atrial fibrillation after major noncardiac surgery. Am Heart J. 2012; 164(6): 918–924. doi: 10.1016/j.ahj.2012.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial Index of Cardiac Risk in Noncardiac Surgical Procedures. N Engl J Med. 1977; 297(6): 845–850. doi: 10.1056/NEJM197710202971601 [DOI] [PubMed] [Google Scholar]
- 4.Davenport DL, O’Keeffe SD, Minion DJ, Sorial EE, Endean ED, Xenos ES. Thirty-day NSQIP database outcomes of open versus endoluminal repair of ruptured abdominal aortic aneurysms. J Vasc Surg. 2010; 51(2):305–309. doi: 10.1016/j.jvs.2009.08.086 [DOI] [PubMed] [Google Scholar]
- 5.Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to revise the 2002 guideline. Circulation. 2007; 116(17):1667–1679. doi: 10.1161/CIRCULATIONAHA.107.185699 [DOI] [PubMed] [Google Scholar]
- 6.Valentine RJ, Rosen SF, Cigarroa JE, Jackson MR, Modrall JG, Clagett GP. The clinical course of new-onset atrial fibrillation after elective aortic operations. J Am Coll Surg. 2001; 193(5): 499–504. doi: 10.1016/S1072-7515(01)01028-6 [DOI] [PubMed] [Google Scholar]
- 7.Polanczyk CA, Goldman L, Marcantonio ER, Orav EJ, Lee TH. Supraventricular arrhythmia in patients having noncardiac surgery: Clinical correlates and effect on length of stay. Ann Intern Med. 1998; 129(4): 279–285. doi: 10.7326/0003-4819-129-4-199808150-00003 [DOI] [PubMed] [Google Scholar]
- 8.Seguin P, Signouret T, Laviolle B, Branger B, Mallédant Y. Incidence and risk factors of atrial fibrillation in a surgical intensive care unit. Crit Care Med. 2004; 32(2): 722–726. doi: 10.1097/01.CCM.0000114579.56430.E0 [DOI] [PubMed] [Google Scholar]
- 9.Sohn GH, Shin DH, Byun KM, et al. The incidence and predictors of postoperative atrial fibrillation after noncardiothoracic surgery. Korean Circ J. 2009; 39(3): 100–104. doi: 10.4070/kcj.2009.39.3.100 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kanji S, Williamson DR, Yaghchi BM, Albert M, McIntyre L. Epidemiology and management of atrial fibrillation in medical and noncardiac surgical adult intensive care unit patients. J Crit Care. 2012; 27(3): 326–328. doi: 10.1016/j.jcrc.2011.10.011 [DOI] [PubMed] [Google Scholar]
- 11.Society for Vascular Surgery Vascular Quality Initiative 2012 Validation Process for 100% Procedure Entry; 2012.
- 12.Noorani A, Walsh SR, Tang TY, et al. Atrial fibrillation following elective open abdominal aortic aneurysm repair. Int J Surg. 2009; 7(1): 24–27. doi: 10.1016/j.ijsu.2008.09.002 [DOI] [PubMed] [Google Scholar]
- 13.Winkel TA, Schouten O, Hoeks SE, et al. Risk factors and outcome of new-onset cardiac arrhythmias in vascular surgery patients. Am Heart J. 2010; 159(6): 1108–1115. doi: 10.1016/j.ahj.2010.03.035 [DOI] [PubMed] [Google Scholar]
- 14.Mitchell LB, Exner DV, Wyse DG, et al. Prophylactic oral amiodarone for the prevention of arrhythmias that begin early after revascularization, valve replacement, or repair. PAPABEAR: A randomized controlled trial. J Am Med Assoc. 2005; 294(24); 3093–3100. doi: 10.1001/jama.294.24.3093 [DOI] [PubMed] [Google Scholar]
- 15.Benjamin EM, Klugman RA, Luckmann R, Fairchild DG, Abookire SA. Impact of cardiac telemetry on patient safety and cost. Am J Manag Care. 2013; 19(6): 225–230. [PubMed] [Google Scholar]
- 16.Bateman BT, Schumacher HC, Wang S, Shaefi S, Berman MF. Perioperative acute ischemic stroke in noncardiac and nonvascular surgery: Incidence, risk factors, and outcomes. Anesthesiology. 2009; 110(2): 231–238. doi: 10.1097/ALN.0b013e318194b5ff [DOI] [PubMed] [Google Scholar]
- 17.Danelich IM, Lose JM, Wright SS, et al. Practical management of postoperative atrial fibrillation after noncardiac surgery. J Am Coll Surg. 2014; 43(3):235–244. doi: 10.1016/j.jamcollsurg.2014.02.038 [DOI] [PubMed] [Google Scholar]
- 18.Smith H, Yeung C, Gowing S, et al. A review and analysis of strategies for prediction, prevention and management of post-operative atrial fibrillation after non-cardiac thoracic surgery. J Thorac Dis. 2018; 10(23): 3799–3808. doi: 10.21037/jtd.2018.09.144 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gialdini G, Nearing K, Bhave PD, et al. Perioperative atrial fibrillation and the long-term risk of ischemic stroke. JAMA - J Am Med Assoc. 2014; 312(6): 616–622. doi: 10.1001/jama.2014.9143 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sandau KE, Funk M, Auerbach A, et al. Update to Practice Standards for Electrocardiographic Monitoring in Hospital Settings: A Scientific Statement From the American Heart Association. Circulation. 2017; 136(17): 273–344. doi: 10.1161/CIR.0000000000000527 [DOI] [PubMed] [Google Scholar]
- 21.Christians KK, Wu B, Quebbeman EJ, Brasel KJ. Postoperative atrial fibrillation in noncardiothoracic surgical patients. American Journal of Surgery. 2001; 182(6): 713–715. doi: 10.1016/S0002-9610(01)00799-1 [DOI] [PubMed] [Google Scholar]
- 22.Walsh SR, Oates JE, Anderson JA, Blair SD, Makin CA, Walsh CJ. Postoperative arrhythmias in colorectal surgical patients: Incidence and clinical correlates. Colorectal Dis. 2006; 8(3): 212–216. doi: 10.1111/j.1463-1318.2005.00881.x [DOI] [PubMed] [Google Scholar]
- 23.Batra GS, Molyneux J, Scott NA. Colorectal patients and cardiac arrhythmias detected on the surgical high dependency unit. Ann R Coll Surg Engl. 2001; 83(3): 174–176. [PMC free article] [PubMed] [Google Scholar]
- 24.Walsh SR, Tang T, Wijewardena C, Yarham SI, Boyle JR, Gaunt ME. Postoperative arrhythmias in general surgical patients. Ann R Coll Surg Engl. 2007; 89(2): 91–95. doi: 10.1308/003588407X168253 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Dansey K, de Guerre L, Li C, et al. Not All Databases Are Created Equal, a Comparison of Administrative Data and Quality Improvement Registries for Abdominal Aortic Aneurysm Repair. J Vasc Surg. 2020; 20(9): 315–338. doi: 10.1016/j.jvs.2020.04.338 [DOI] [PubMed] [Google Scholar]