Age-stratified outcomes in elderly patients undergoing open and endovascular procedures for aortoiliac occlusive disease

Novel bypass risk predictive tool is superior to the 5-Factor Modified Frailty Index in predicting postoperative outcomes

OBJECTIVE

This study aimed to develop risk predictive models of 30-day mortality, morbidity, and major adverse limb events (MALE) after bypass surgery for aortoiliac occlusive disease (AIOD) and to compare their performances with a 5-Factor Frailty Index.

METHODS

The American College of Surgeons National Surgical Quality Improvement Program 2012-2017 Procedure Targeted Aortoiliac (Open) Participant Use Data Files were queried to identify all patients who had elective bypass for AIOD: femorofemoral bypass, aortofemoral bypass, and axillofemoral bypass (AXB). Outcomes assessed included mortality, major morbidity, and MALE within 30 days postoperatively. Major morbidity was defined as pneumonia, unplanned intubation, ventilator support for >48 hours, progressive or acute renal failure, cerebrovascular accident, cardiac arrest, or myocardial infarction. Demographics, comorbidities, procedure type, and laboratory values were considered for inclusion in the risk predictive models. Logistic regression models for mortality, major morbidity and MALE were developed. The discriminative ability of these models (C-indices) were compared with that of the 5-Factor Modified Frailty Index (mFI-5): a general frailty tool determined from diabetes, functional status, history of chronic obstructive pulmonary disease, history of congestive heart failure, and hypertension. Calculators were derived using the most significant variables for each of the three risk predictive models.

RESULTS

A total of 2612 cases (mean age 65.0, 60% male) were identified, of which 1149 (44.0%) were femorofemoral bypass, 1138 (43.6%) were aortofemoral bypass, and 325 (12.4%) were axillofemoral bypass. Overall, the rates of mortality, major morbidity, and MALE were 2.0%, 8.5%, and 4.9%, respectively. Twenty preoperative risk factors were considered for incorporation in the risk tools. Apart from procedure type, age was the most statistically significant predictor of both mortality and morbidity. Preoperative anemia and critical limb ischemia were the most significant predictors of MALE. All three constructed models demonstrated significantly better discriminative ability (P < .001) on the outcomes of interest as compared with the mFI-5.

CONCLUSIONS

Our models outperformed the mFI-5 in predicting 30-day mortality, major morbidity, and adverse limb events in patients with AIOD undergoing elective bypass surgery. Calculators were created using the most statistically significant variables to help calculate individual patient's postoperative risks and allow for better informed consent and risk-adjusted comparison of provider outcomes.

Younger patients have worse outcomes after peripheral endovascular interventions for suprainguinal arterial occlusive disease

OBJECTIVE

The choice of intervention for treating suprainguinal arterial disease, open bypass vs endovascular intervention, is often tempered by patient age and comorbidities. In the present study, we compared the association of patient age with 1-year major adverse limb events (MALE)-free survival and reintervention-free survival (RFS) rates among patients undergoing intervention for suprainguinal arterial disease.

METHODS

The Vascular Quality Initiative datasets for bypass and peripheral endovascular intervention (PVI; aorta and iliac only) were queried from 2010 to 2017. The patients were divided into two age groups: <60 and ≥60 years at the procedure. Age-stratified propensity matching of patients in bypass and endovascular procedure groups by demographic characteristics, comorbidities, and disease severity was used to identify the analysis samples. The 1-year MALE-free survival and RFS rates were compared using the log-rank test and Kaplan-Meier plots. Proportional hazard Cox regression was used to perform propensity score-adjusted comparisons of MALE-free survival and RFS.

RESULTS

A total of 14,301 cases from the Vascular Quality Initiative datasets were included in the present study. Propensity matching led to 3062 cases in the ≥60-year group (1021 bypass; 2041 PVI) and 2548 cases in the <60-year group (1697 bypass; 851 PVI). In the crude comparison of the matched samples, the older patients undergoing bypass had had significantly greater in-hospital (4.6% vs 0.9%; P < .001) and 1-year (10.5% vs 7.5%; P = .005) mortality compared with those who had undergone endovascular intervention. The rates of MALE (7.5% vs 14.3%; P < .001) and reintervention (6.7% vs 12.7%; P < .001) or death were significantly higher for the younger group undergoing PVI than bypass at 1 year. However, the rates of MALE (12.9% vs 14.3%; P = .298) and reintervention (12.7% vs 12.9%; P = .881) or death for were similar both procedures for the older group. Both log-rank analyses and the adjusted propensity score analyses of MALE-free survival and RFS in the two age groups confirmed these findings. The adjusted comparison of outcomes using propensity score matching favored PVI at 1-year survival (hazard ratio, 1.4; 95% confidence interval, 1.1-1.9; P = .003) for the older group but was not different for the younger group (hazard ratio, 0.6; 95% confidence interval, 0.3-1.0; P = .054).

CONCLUSIONS

Among the patients aged <60 years undergoing intervention for suprainguinal arterial disease, the choice of therapy should be open surgical intervention given the higher risk of reintervention and MALE with endovascular intervention. Endovascular intervention should be favored for patients aged ≥60 years because of reduced perioperative mortality.

The impact of combined iliac occlusive disease and aortic aneurysm on open surgical repair

OBJECTIVE

Severe aortoiliac occlusive disease is a relative contraindication for endovascular aneurysm repair, owing to an association with high stent graft-related complication and reintervention rates in this population. Open AAA repair requiring aortofemoral bypass (AFB), however, may represent a unique population with differing outcomes from standard open repair. We sought to compare the demographic and procedural characteristics, as well as outcomes of patients undergoing standard intra-abdominal repairs (STD) versus those requiring AFB.

METHODS

Using a prospectively maintained database, we retrospectively identified patients who underwent open AAA repair from 1994 to 2017. A total of 1087 consecutive cases were performed consisting of 981 STD (681 tube graft, 300 aortoiliac) and 106 AFB cases. Demographics, procedural data, postoperative complications, and long-term survival were analyzed.

RESULTS

The AFB cohort had more women (39.0 vs 22.8%; P = .001) and higher rates of hypertension (81.1 vs 69.8%; P = .015), chronic obstructive pulmonary disease (28.3 vs 17.4%; P = .006), and smoking (50.9 vs 36%; P = .002). The AFB group had smaller mean aortic (5.22 vs 5.77 cm; P = .001) and graft (17.08 vs 18.2 mm; P = .001) diameters. Proximal clamp position and blood loss were equivalent, although total anesthesia time was longer (295 vs 234 minutes; P = .001) in the AFB cohort. Overall 30-day postoperative morbidity (38.7 vs 24.8%; P = .002) was higher in the AFB group. Specifically, postoperative renal insufficiency (8.2 vs 3.4%; P = .032), wound infection (5.7 vs 1.2%; P = .005), and hematoma/seroma (5.7 vs 1.2%; P = .003) were more likely. Hospital length of stay was longer for AFB (11.9 vs 9.9 days; P = .007). The 30-day mortality (0.9% AFB vs 1.8% STD; P = .50) and major morbidity (17 vs 11.5%; P = .10) did not differ. Reintervention rate within 30 days of the initial surgery (12.3 vs 4.6; P = .001) and overall (33 vs 18.9%; P = .001) was higher in the AFB group. Long-term survival was lower in the AFB group (5-year survival: 63.1% AFB vs 71.9% STD; hazard ratio 0.76, log-rank P = .047). Multivariate regression analysis identified age, comorbid conditions, and aneurysm characteristics-rather than repair type-as independent predictors of 30-day reintervention and mortality at 5 years.

CONCLUSIONS

Patients requiring AFB for AAA owing to associated iliac occlusive disease have more preoperative comorbidities, postoperative complications, a longer length of stay, reintervention rates and shorter 5-year survival. Patient and aneurysm characteristics rather than surgical repair type appear to be responsible for these differences. Nevertheless, 30-day mortality and major morbidity were comparable, making AFB an attractive alternative to endovascular aneurysm repair in patients with advanced iliac occlusive disease.

A comparison between aortobifemoral bypass and aortoiliac kissing stents in patients with complex aortoiliac obstructive disease

OBJECTIVE

The aim of this study was to retrospectively compare early and late results of aortobifemoral bypass and endovascular recanalization with the kissing stent technique in the management of TransAtlantic Inter-Society Consensus II (TASC II) C and D lesions in the aortoiliac district in a multicenter study.

METHODS

From January 2006 to December 2013, 210 open and endovascular interventions for TASC II class C and D aortoiliac obstructive lesions were performed at three Italian teaching hospitals. In all the cases, the intervention was performed for aortic and bilateral iliac involvement. An aortobifemoral bypass was performed in 82 patients (group 1); in the remaining 128, an endovascular recanalization with the kissing stent technique was carried out (group 2). Early results in the two groups were compared with the χ² test. Follow-up results were analyzed with Kaplan-Meier curves and compared with log-rank test. Univariate and multivariate (forward Cox regression) analysis to identify potentially significant predictors of primary patency in the whole study group was performed.

RESULTS

Patients in group 2 were more frequently female and more frequently had diabetes and arterial hypertension than patients in group 1. The indication for surgical intervention was the presence of critical limb ischemia in 29 cases in group 1 (35%) and in 31 cases in group 2 (24%; P = .07). Technical success in group 2 was 98.5%; two patients required immediate conversion to open surgery for iliac rupture. There was one perioperative death in group 1 (mortality rate, 1%; P = .2 in comparison with group 2). Four perioperative thromboses occurred, two in group 1 and two in group 2 (in one case requiring conversion to open surgical intervention), and no amputations at 30 days were recorded. Cumulative postoperative local and systemic complications occurred in 17 patients in group 1 (20.5%) and in 9 patients in group 2 (7%; P < .001). Mean duration of follow-up was 38 months (range, 1-96 months). Survival rates at 6 years were 65.5% (standard error [SE], 0.08) in group 1 and 83.5% (SE, 0.08) in group 2 (P = .08; log-rank, 2.2). At the same time interval, primary, assisted primary, and secondary patency rates were similar; reintervention rates were 6% in group 1 (SE, 0.05) and 11% in group 2 (SE, 0.04; P = .3; log-rank, 0.8). Univariate and multivariate analysis showed that only the presence of critical limb ischemia was independently associated with poorer primary patency during follow-up (hazard ratio, 2.4; 95% confidence interval, 0.9-6.4; P = .05).

CONCLUSIONS

In this multicenter experience, endovascular repair of aortoiliac complex lesions with the kissing stent technique provided similar satisfactory early and late results to those obtained with open surgery.

Impact of coexisting aneurysms on open revascularization for aortoiliac occlusive disease

OBJECTIVE

National data evaluating outcomes for occlusive abdominal aortic reconstructions are well described. The relative effect of operative indication as well as the presence of concomitant abdominal aortic aneurysm (AAA) on in-hospital mortality is not well defined.

METHODS

The Nationwide Inpatient Sample was queried to identify patients who underwent open aortic surgery (2003-2010). Indication for surgery was classified by International Classification of Diseases, Ninth Revision diagnostic codes to identify isolated occlusive indications as well as combined occlusive disease and AAA. Primary outcome was in-hospital mortality. Secondary outcomes were complications and discharge disposition.

RESULTS

Overall, 56,374 underwent aortic reconstruction, 48,591 for occlusive disease (86.2%) and 7783 for combined occlusive disease with AAA (13.8%). Intermittent claudication was the most common indication for intervention (60.9%), whereas 39.7% underwent intervention for critical limb ischemia (22.2% rest pain, 17.6% gangrene). Patients with intermittent claudication had more concomitant AAAs (17.3%) than did patients with critical limb ischemia (8.4%). The baseline characteristics for those with occlusive disease and combined occlusive with AAA disease were similar in terms of obesity (4.8% vs 4.2%; P = .27) and congestive heart failure (6.6% vs 6.3%; P = .65) but differed by age (62.2 years vs 68.4 years; P < .0001) and hypertension (65.4% vs 69.1%; P = .005). Patients with combined occlusive and AAA disease had higher mortality than those with occlusive disease alone (3.9% vs 2.7%; P = .01). On multivariable regression, factors associated with in-hospital mortality included gangrene with AAA compared with gangrene alone (odds ratio [OR], 2.8; 95% confidence interval [CI], 1.6-4.7; P < .0002), age >65 years age (OR, 3.1; 95% CI, 2.4-4.1; P < .0001), renal failure (OR, 2.7; 95% CI, 1.9-3.8; P < .0001), and concurrent lower extremity revascularization (OR, 1.3; 95% CI, 1.1-1.7; P < .02).

CONCLUSIONS

Intermittent claudication or critical limb ischemia with concomitant AAA carries a higher mortality than intermittent claudication or critical limb ischemia alone, especially in older patients with gangrene requiring revascularization and renal insufficiency. Preoperative risk stratification strategies should focus on the indication for surgery as well as the presence of concomitant AAA.