Determinants of hospital length of stay after thoracoabdominal aortic aneurysm repair

PURPOSE: Extended hospital length of stay (LOS) and consequent high costs are associated with thoracic and thoracoabdominal aortic aneurysm (TAAA) surgery. In this study, we examined factors that may influence LOS after TAAA repair. METHODS: Five hundred forty thoracic and TAAA repairs were performed by one surgeon between 1990 and 1999. The data were analyzed with multiple linear regression with appropriate logarithmic transformation. The predictor variables included patient demographics, disease extent, severity indicators, intraoperative factors, and postoperative complications. RESULTS: The median LOS was 15 days. Postoperative creatinine level of greater than 2.9 was the most important predictor of LOS, followed by spinal cord deficit, age, and pulmonary complication (all statistically significant with P <.05). A second model constrained to preoperative risk factors showed both age and complete diaphragmatic division to be associated with increased LOS. Preservation of the diaphragm led to reduced LOS by an average of 4 days. The adjunct cerebrospinal fluid drainage and distal aortic perfusion was associated with a decrease in LOS, although it did not reach statistical significance. CONCLUSION: Renal failure, spinal cord deficit, and pulmonary complication were the major determinants of LOS in patients for TAAA repair. This study shows that the preservation of diaphragmatic function and the use of the adjunct distal aortic perfusion and cerebrospinal fluid drainage may reduce hospital LOS.     

Mid-term results of patients undergoing endovascular aortic aneurysm repair

BACKGROUND: The utilization of endovascular aneurysm repair (EAR) is increasing significantly; however, few papers have outlined mid-term outcomes. METHODS: Patients undergoing EAR with an AneuRx endograft between September 1997 and May 2001 were evaluated. Mean follow-up was 20.7 +/- 11.9 (SD) months. RESULTS: In all, 101 EAR devices were successfully deployed in 105 attempts. Four open conversions (2 acute, 2 delayed) were performed for complications of EAR. Technical, clinical, and 1 to 3 year continuing success rates were 75%, 73%, and 78% to 83%. When divided by the median date, significantly fewer patients in the later group required secondary procedures compared with the early group. Vascular insufficiency occurred in 12 patients; 11 were treated with a secondary procedure. For 9 type I and 9 type II persistent endoleaks, secondary procedures were attempted and successful in 10 patients. Of successful EAR deployments, including secondary interventions, 85% demonstrated no persistent leak, rupture, increase in aneurysm size, or migration at most recent follow-up. CONCLUSIONS: EAR is successful in selected individuals; however, continuing follow-up is of paramount importance.     

Relationship of proximal fixation to renal dysfunction in patients undergoing endovascular aneurysm repair

Technological advancements have lead to dramatic improvements in stentgraft device design resulting in more trackable delivery systems and transrenal uncovered stents and barbs for better fixation. Transrenal bare-stents may limit stentgraft migration, particularly in patients with short or flared proximal aortic necks. However, potential disadvantages might be in worsening renal function, particularly in patients with preexisting renal insufficiency. We retrospectively analyzed our recent 7 year experience of patients undergoing endovascular aneurysm repair (EVAR) using a variety of stentgrafts with and without transrenal bare-stent fixation. Patients were divided into 2 groups; infrarenal fixation (IRF) vs transrenal fixation (TRF), or patients with preoperative serum Cr values that were normal (= or <1.5 mg/dl) vs slightly elevated (1.6-2 mg/dl), vs markedly elevated (2.1- 3.5 mg/dl). The exclusion criteria included patients with chronic renal insufficiency (CRI) on hemodialysis, and preoperative high-grade renal artery stenoses requiring angioplasty and stenting. Of 705 patients that underwent EVAR, 496 (IRF: 385 [78%], and TRF: 111 [22%]) were available with routine evaluations of serum Cr and CT scans. Preexisting comorbidities, mean procedure contrast volume, and postprocedure follow-up were similar in both groups. In the immediate postoperative period, mean serum Cr did not change significantly in either the IRF group (1.3+/-0.7 mg/dl to 1.2+/-0.9 mg/dl) or the TRF group (1.3+/-0.5 mg/dl to 1.3+/-0.6 mg/dl). Mean serum Cr did, however, significantly increase over longer follow-up in both groups: 1.4+/-0.8 mg/dl for IRF (P<0.03), and 1.5 +/- 0.8 mg/dl for TRF (P<0.01). Cr clearance was similarly unchanged in the immediate postoperative period (58+/-23 to 61+/-25 ml/min/1.73 m2 for IRF group, 53+/-17 to 55+/-17 ml/min/1.73 m2 for TRF group), but was significantly decreased in longer follow-up (53+/-23 ml/min/1.73 m2 for IRF, p<0.02: and 48+/-16 ml/min/1.73 m2 for TRF, P<0.01). There were no significant differences in serum Cr increase (p=0.19) or Cr clearance decrease (p=0.68) between the IRF and TRF groups. Small renal infarcts were noted in 6 patients (1.6%) in the IRF group, and in 8 patients (7%) in the TRF group (p=0.37). Of patients with normal preoperative renal function, renal dysfunction developed in 7.7% of IRF group and 6.1% of TRF group (p=0.76). In patients with preexisting CRI, renal dysfunction developed in 18.2% of IRF group, and 17.1% of TRF group (p=0.95). Eight patients with postoperative renal dysfunction, 5 (1.3%) from IRF group and 3 (2.7%) from TRF group subsequently required hemodialysis, and this difference was not statistically significant (p=0.91). We also analyzed 200 consecutive patients undergoing EVAR with intra-arterial contrast agents with and without preexisting CRI not on dialysis. The groups were identified on the basis of preprocedure serum Cr: group 1 (n=108), Cr less than 1.5 mg/dL (normal range); group 2 (n=65), Cr 1.5 to 2.0 mg/dL; group 3 (n=27), Cr 2.1 to 3.5 mg/dL. Routine precautions in patients with CRI included preoperative intravenous hydration with 2 L of normal saline solution, discontinuation of all nephrotoxic drugs, intraoperative administration of mannitol (0.5 g/kg intravenously), and use of nonionic, low osmolar intra-arterial contrast agent (Omnipaque 350). One-hundred and eight patients had normal renal function (group 1), and 92 patients had preexisting CRI with baseline Cr 1.5 to 2.0 mg/dL (group 2, n=65) or 2.1 to 3.5 mg/dL (group 3, n=27). Comorbid conditions included coronary artery disease (group 1, 51%; group 2, 49%; group 3, 59%), hypertension (group 1, 39%; group 2, 46%; group 3, 52%), and diabetes mellitus (group 1, 25%; group 2, 35%; group 3, 48%). In groups 1, 2, and 3, the mean volume of low osmolar contrast agent used was 210 cc, 160 cc, 130 cc, respectively; hemodynamic instability developed in 3, 1, and 1 patient, respectively. The incidence of postoperative complications between the 3 study groups was not statistically different. In grications between the 3 study groups was not statistically different. In group 1 a transient increase in serum Cr (>30% over baseline and >1.4 mg/dL) was noted in 3 patients (2.7%), 2 of whom (1.9%) required temporary hemodialysis and 1 (0.9%) who died of renal failure. In group 2 a transient increase in serum Cr was noted in 2 patients (3.1%); both patients (3.1%) required temporary hemodialysis, and 1 patient (1.5%) died of renal failure. In group 3 a transient increase in serum Cr was noted in 2 patients (7.4%); 1 patient (3.7%) required temporary hemodialysis, and 1 patient (3.7%) died of renal failure. Perioperative hypotension significantly increased the risk for elevated serum Cr and death (p<0.05), and larger contrast volume was associated with an increase in serum Cr (p<0.05) during the postoperative period. Following EVAR renal function declines slightly with both IRF and TRF. Our data show no overall difference between patients with IRF and TRF with respect to infarcts, decline in renal function, or onset of dialysis. There were a slightly greater number of renal infarcts in the TRF group, but these infarcts were clinically inconsequential. In patients with CRI, EVAR with intra-arterial radiographic contrast agents is believed to impair renal function, and CRI is considered a relative contraindication to the procedure. Results of our investigation indicate that risk for worsening renal insufficiency, dialysis, and death is only slightly and not significantly greater in patients with CRI compared with patients with normal renal function. With appropriate precautions of avoiding perioperative hypotension and limiting the volume of nonionic contrast agents, CRI need not be a contraindication for EVAR with intra-arterial contrast agents.     

Perioperative blood transfusion in anemic patients undergoing elective endovascular abdominal aneurysm repair

ObjectiveAlthough blood transfusion can be lifesaving in active hemorrhage or severe anemia, it is also associated with increased morbidity and mortality. Several trials have established this risk and therefore defined a restrictive standard for transfusion, but this threshold and the risk of transfusions have not been specifically examined in vascular surgery patients. We therefore sought to assess transfusion practices and outcomes of anemic patients undergoing elective endovascular aneurysm repair (EVAR).MethodsThe Vascular Quality Initiative database was queried for patients undergoing EVAR between the years 2008 and 2017. Anemic patients were included in the study and were further stratified into mild anemia, defined by a hemoglobin level of 10 to 13 g/dL in men or 10 to 12 g/dL in women, and moderate to severe anemia, defined by a hemoglobin level <10 g/dL. The primary study outcomes were in-hospital mortality and complications.ResultsAmong 27,777 EVAR patients, one-third (n = 9232) were anemic and included in the study. One-fifth (n = 1866) of anemic patients received a perioperative transfusion. Transfused patients were more likely to have a history of cardiovascular disease. In-hospital mortality was significantly higher for anemic patients who received transfusions, both in mild anemia (mortality, 3.6% vs 0.4% in no transfusion; P < .001) and in moderate to severe anemia (4.5% vs 1.3%; P < .01). Morbidity was also significantly higher, with anemic patients who received a transfusion having higher rates of myocardial infarction, congestive heart failure, dysrhythmias, renal complications, leg ischemia, respiratory complications, and reoperation compared with anemic patients who did not receive any transfusion. The 30-day mortality was also higher in transfused patients (P < .001). After adjustment for patients' demographics, comorbidities, and operative factors, transfusion in anemic patients was associated with a nearly 4.4-fold increased odds of in-hospital mortality (odds ratio [OR], 4.38; 95% confidence interval [CI], 2.72-7.05; P < .001) and 4.3-fold higher odds of any in-hospital complication (OR, 4.31; 95% CI, 3.47-5.34; P < .001). This was more pronounced among patients with mild anemia, with 5.7 times (OR, 5.7; 95% CI, 1.78-18.0) and 4.3 times (OR, 4.3; 95% CI, 3.46-5.29) the odds of in-hospital mortality and complications, respectively.ConclusionsAmong anemic patients undergoing elective EVAR, transfusion is associated with an increased risk of death and in-hospital complications, even after controlling for patients' comorbidities and operative factors. These data suggest that the restrictive use of blood transfusions might be safer in vascular surgery EVAR patients. Medical management of anemia may be warranted in these patients to reduce morbidity and mortality; however, further studies are needed to evaluate effectiveness.     

Sarcopenia predicts poor long-term survival in patients undergoing endovascular aortic aneurysm repair

Objective

Sarcopenia measured by decreased psoas muscle size has been used as a surrogate for frailty and correlates with adverse outcomes in both the short and long term after many major operations. Our aim was to evaluate this measure as a predictor of outcomes in patients undergoing endovascular aortic aneurysm repair (EVAR).

Any nonadherence to instructions for use predicts graft-related adverse events in patients undergoing elective endovascular aneurysm repair

Background

A variety of devices exist for endovascular aneurysm repair (EVAR). Device-specific instructions for use (IFU) detail anatomic constraints to application and deployment of devices and are developed from rigorous bench testing. Nonadherence to IFU occurs frequently to avoid open surgery. The purpose of this study was to determine if IFU violations are associated with increased risk of graft-related adverse events (GRAEs) during follow-up.

Risk factors of early and late complications in patients undergoing endovascular aneurysm repair.

OBJECTIVES: to identify pre-operative factors that could predict complications following from transluminal repair of abdominal aortic aneurysms (AAA). METHODS: during a 5-year period, 96 consecutive patients underwent elective endovascular treatment of a AAA. In all patients, helical CT and/or Magnetic Resonance Imaging (MRI), and plain abdominal roentgenogram were performed at 1, 3, 6, 12, 18, and 24 months and yearly thereafter. Angiography was performed systematically 1 year after the stent-graft implantation, or earlier if helical CT or MRI diagnosed an increase in the maximal transverse diameter or a high flow endoleak. RESULTS: early (<30 days) morbidity (12%) was significantly increased by pre-operative renal insufficiency (p < 0.01). Early mortality (2%) correlated with ASA score (p = 0.01). Median follow-up was 27 months (range 3-66). Mortality (12%) during follow-up was correlated to the pre-operative coronary status (p = 0.01). A type I endoleak was diagnosed in 18 patients (19%). Common iliac artery diameter was correlated with the presence of type I endoleak (p < 0.001). A type II endoleak was diagnosed in 47 (49%) patients. The diagnostic of type II endoleak was significantly increased (p = 0.001) in patients with pre-operative patent IMA associated with more than four patent lumbar arteries. The anatomic characteristics of the aneurysm were correlated to the additional endovascular procedures during stentgraft implantation (p = 0.01), and to the implantation of a complementary iliac limb extension during follow-up (p = 0.01). CONCLUSIONS: the risk factors determined by this statistical analysis could help surgeons to select more accurately patients suitable for endovascular treatment.     

Financial implications of coding inaccuracies in patients undergoing elective endovascular abdominal aortic aneurysm repair

Objective

Previous cost analyses have found small to negative margins between hospitalization cost and reimbursement for endovascular aneurysm repair (EVAR). Hospitals obtain reimbursement on the basis of Medicare Severity Diagnosis Related Group (MS-DRG) coding to distinguish patient encounters with or without major comorbidity or complication (MCC). This study's objective was to evaluate coding accuracy and its effect on hospital cost for patients undergoing EVAR.

Extending endovascular aneurysm repair to more patients without better outcomes

ObjectiveThis study quantifies the survival and outcomes associated with endovascular aneurysm repair (EVAR) patients treated in two eras. We hypothesized that both end points will improve over time.MethodsPatients receiving EVAR between January 2003 and May 2017 contained within the Vascular Quality Initiative data set were examined. Those patients treated between January 2003 and December 2007 were included in the early era, and those treated between January 2012 and December 2015 were considered late era. Baseline demographics, technical variables, and outcomes were compared with appropriate statistical tests. Survival was estimated with Kaplan-Meier life-table analysis. Cox proportional hazards modeling analyzed the relationship between repair era and survival; the repair era's significance was further examined in matched cohorts generated by coarsened exact matching and propensity scoring.ResultsEarly era (n = 787) patients demonstrate decreased estimated survival in comparison to those treated in the late era (n = 20,066; log-rank, P < .001). Repair in the late era was not an independent predictor of survival in the Cox model (hazard ratio, 1.06; 95% confidence interval, 0.23-4.95; P = .94). Three different matching methods confirmed a nonsignificant contribution of treatment era to survival suggested by the initial Cox model (all P > .05). Total hospital stays were longer in the late era (3.1 vs 4.2 days; P < .001). Postoperative myocardial infarction and surgical site infections decreased in frequency in the late group, although postoperative vasopressor use became more frequent (all P < .05). Operations became 20% faster, required 27% less contrast material and 29% less crystalloid, and lost 30% less blood over time (all P < .05).ConclusionsAlthough EVAR volume increased significantly over time, post-EVAR outcomes of patients do not differ between the early era, 2003 to 2007, and the late era, 2012 to 2015. Aspects of surgical performance improved, but this did not translate into a measurable benefit to patients. Length of hospital stay unexpectedly increased over time.     

Risk factors for endoleak and the evidence for stent-graft oversizing in patients undergoing endovascular aneurysm repair.

OBJECTIVES: the aim of this study was to assess the relationship between patient factors, the anatomy of the proximal aneurysm neck; the type of endovascular graft; and the consequences of graft/neck size mismatch and the occurrence of proximal endoleak. Design multicentre clinical study. MATERIALS: of a total of 2194 patients, 2146 underwent successful endovascular repair of infra-renal abdominal aortic aneurysms (AAA). METHODS: endoleaks were identified by radiological imaging immediately after completion of the procedure as per study protocols. Clinical and anatomical features of AAA in patients with endoleak were compared to patients without endoleak and data were analysed using the Chi-square test. A multivariate logistic regression model was constructed by selecting variables found to be significantly associated with complications in a univariate analysis. RESULTS: intra-operative endoleak was observed in 16.7% overall, and 3.3% were noted to have proximal endoleak. Aneurysm size larger than 60 mm (p =0.004), ex-smokers ( p =0.005) and age over 75 years ( p =0.01) were independently associated with endoleak of all types. Univariate and multivariate analysis revealed correlation between proximal endoleak and (i) diameter of the aneurysm neck-proximal (D2a), middle (D2b), distal (D2c), at all levels (p <0.005); (ii) proximal aortic neck length ( p =0.0001); (iii) aortic device diameter ( p =0.0024). No correlation was identified for angulation and form of the aortic neck. A model of the frequency of proximal endoleak, in relation to the ratio of the aortic device diameter to the distal aortic neck diameter, revealed that endoleak decreased when the aortic device diameter became oversized by more than 10% and confidence intervals remained tight for up to and over 20% oversize.     

Current hospital costs and medicare reimbursement for endovascular abdominal aortic aneurysm repair

OBJECTIVE: The purpose of this study was to analyze the current inpatient hospital cost and Medicare reimbursement of endovascular abdominal aortic aneurysm repair (EVAR) at different hospitals. METHODS: The cost of EVAR from October 2000 to October 2001 with two commercially available endografts (Ancure, Guidant Endovascular Solutions, Menlo Park, Calif; and AneuRx, Medtronic AVE, Santa Rosa, Calif) was retrospectively analyzed at seven hospitals. Three university (n = 111) and four community hospitals (n = 110) from different regions of the country participated in the survey. Consecutive cases with complete financial records were included. Hospital finance departments provided their best estimates of hospital costs, including overhead for operating room, endograft, medical supply, bed, radiology, laboratory, and pharmacy services and reimbursement on the basis of hospital-specific Diagnostic Related Groups (DRG) 110 or 111. Detailed hospital charges and International Clinical Diagnosis codes also were reviewed from Universal Billing-92 forms submitted to Medicare. An additional cost analysis was performed by the authors to validate the estimates of the hospital financial departments. Outliers of more than three standard deviations from the mean were excluded. RESULTS: The mean total hospital cost was $22,999, and mean reimbursement, weighted by case mix, was $20,837, resulting in a net loss of $2162. The majority of EVAR cost was from the device (57%) and other medical supplies (16%). EVAR was reimbursed on the basis of DRG 110 in 78% of cases and of DRG 111 in 22%. Reimbursement varied widely by hospital and location (mean, $20,837; range, $14,818 to $35,343; standard deviation, $5450). With the exclusion of one hospital where reimbursement was not based on the DRG, cases reimbursed with DRG 110 resulted in an average loss of $2200, while the average loss was $9198 with DRG 111. The mean net loss for hospitals reimbursed with the DRG system was $3898. CONCLUSION: EVAR reimbursement is presently inadequate to cover hospital expenses. Substantial financial losses occurred at four of the participating centers. University hospitals fared surprisingly better because of higher reimbursement.