Abstracts from the 2013 SVM Annual Scientific Sessions

In vivo electroporation of constitutively expressed HIF-1 plasmid DNA enhances neovascularization in a mouse model of hindlimb ischemia

Geoffrey O Ouma¹, Eduardo Rodriguez¹, Karuppiah Muthumani², David B Weiner², Robert L Wilensky¹, Emile R Mohler III¹

¹Hospital of the University of Pennsylvania, Philadelphia, PA, United States; ²Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States

Background: Hypoxia-inducible factor-1alpha (HIF-1α) is a transcription factor that stimulates angiogenesis during tissue ischemia. Electroporation (EP) enhances tissue DNA transfer. We evaluated the neovascularization efficacy of in vivo EP of a constitutively expressed HIF-1α DNA compared to intramuscular (IM) injection in a mouse model of limb ischemia.

Methods: Mice with ligated left femoral artery were assigned to one of the three groups: (1) HIF-EP (n = 13, in vivo EP of 20 µl HIF-1α plasmid DNA); (2) HIF-IM (n = 14, IM injection of 20 µl HIF-1α plasmid DNA); (3) pVAX-EP (n = 12, in vivo EP of 20 µl empty plasmid DNA). Limb perfusion recovery by Laser Doppler Perfusion Imager, limb function and limb necrosis were measured. Muscle tissues were stained for necrosis (H&E); capillary density (anti-CD31); and collateral vessels and size (anti-α-SMA).

Results: In vivo EP of HIF-1α DNA significantly boosted limb perfusion (HIF-EP: 41.03 ± 0.15 vs. HIF-IM: 0.78 ± 0.064; p < 0.05, vs. pVAX-EP: 0.41 ± 0.019; p < 0.001), limb function recovery (HIF-EP: 3.5 ± 0.58 vs. HIF-IM: 2.4 ± 1.14; p < 0.05, vs. pVAX-EP: 2.4 ± 1.14; p < 0.001), and reduced limb auto-amputation (HIF-EP: 77% ± 12% vs. HIF-IM: 43% ± 14%; p < 0.05 vs. pVAX-EP: 17% ± 11%; p < 0.01). Muscle necrosis declined (HIF-EP: 20.7% ± 1.75% vs. HIF-IM: 44% ± 3.73; p < 0.001, vs. pVAX-EP: 60.05% ± 2.17%; p < 0.0001), capillary growth improved (HIF-EP: 96.83 ± 5.72 vessels/hpf vs. HIF-IM: 62.87 ± 2.0 vessels/hpf; p < 0.001, vs. pVAX-EP: 39.37 ± 2.76 vessels/hpf; p < 0.0001), collateral vessels increased (HI-EP: 76.33 ± 1.94 vessels/hpf vs. HIF-IM: 37.5 ± 1.56 vessels/hpf; p < 0.0001, vs. pVAX-EP: 18.5 ± 1.34 vessels/hpf; p < 0.00001) and the collaterals were larger (HIF-EP: 15,521.67 ± 1298.16 µm² vs. HIF-IM: 7788.87 ± 392.04 µm²; p < 0.001 vs. pVAX-EP: 4640.25 ± 614.01 µm²; p < 0.0001) (Figure).

Conclusions: In vivo EP-mediated delivery of HIF-1α DNA is more effective in enhancing neovascularization than IM injection in a mouse model of limb ischemia. This modality warrants further studies in the treatment of critical limb ischemia.

OPEN IN VIEWER Figure

Leveraging the innate immunity pathway for transdifferentiation of fibroblasts to endothelial cells

Nazish Sayed, Wing Tak Wong, John P Cooke

Stanford University, Stanford, CA, United States

Background: Cell-based approaches to regenerate the endothelium holds promise, with one such source being induced pluripotent stem cells (iPSCs). We have investigated the potential of ECs derived from iPSCs to promote the perfusion of ischemic tissue in a murine model of PAD. However, it may be more efficient to transdifferentiate fibroblasts to ECs directly. Reports have suggested that direct reprogramming to ECs is feasible, however still requires the use of viral vectors encoding transcription factors, thus making them clinically unsafe. We recently discovered that retroviral vectors encoding the reprogramming factors, by activating the Toll-like receptor 3 (TLR3) pathway, make nuclear reprogramming possible by increasing epigenetic plasticity and favoring an open chromatin state (Lee and Sayed et al., Cell). Based on this recognition that innate immunity favors an open chromatin state, we hypothesized that activation of TLR3, together with external microenvironmental cues that drive EC specification, might induce transdifferentiation of fibroblasts into ECs (iECs).

Results: Intriguingly, our preliminary data showed that TLR3 agonist Poly IC, combined with exogenous EC growth factors, was sufficient to transdifferentiate fibroblasts into ECs (in the absence of viral vectors or transcription factors). These iECs exhibited all the characteristics of EC phenotype comparable to HMVEC including able to form capillary-like structures and incorporating acetylated-LDL. Furthermore, iECs significantly improved limb perfusion and neovascularization in the ischemic hindlimb compared to control groups.

Conclusion: This study is a first step toward development of a regenerative strategy for PAD on the use of ECs derived from small molecules without use of viral vectors encoding transcription factors. Moreover, it has allowed us to generate a small molecule strategy for therapeutic transdifferentiation that might be applied for direct reprogramming in vivo.

Diagnostic ability of duplex ultrasound to detect changes associated with carotid and renal fibromuscular dysplasia

Deborah A Hornacek, Heather L Gornik, Ruchi Sanghani, Esther SH Kim

Cleveland Clinic, Cleveland, OH, United States

Background: Fibromuscular dysplasia (FMD) is an uncommonly recognized, non-atherosclerotic arteriopathy of medium-sized arteries. Little data is available on the ability of duplex ultrasound (DUS) to detect extracranial and renal FMD compared to angiography.

Methods: Cases were identified through a retrospective review of FMD patients treated at a single medical center. Patients who had DUS performed and had correlative CTA, MRA, or catheter angiography within three months before or after duplex were eligible for inclusion. Cases were excluded for history of angioplasty or bypass. FMD involvement was determined by review of images by FMD clinic staff. DUS and angiography reports were reviewed for the presence of beading, aneurysms, pseudoaneurysms, and dissections.

Results: Of 262 FMD patients, 93 patients had correlative carotid studies and 62 had correlative renal studies. Test characteristics were determined using arteries as the unit of analysis (Table).

Conclusion: DUS has good sensitivity for detection of carotid FMD but is poor for detection of renal FMD. Characteristic abnormal findings on DUS have high PPV for detection of carotid and renal FMD on angiography. DUS is not a sensitive tool for the detection of pseudoaneurysm, aneurysm, or dissection. Angiography remains the gold standard for the diagnosis of FMD.

OPEN IN VIEWER

Table.

Analysis results for carotid and renal DUS as compared to angiography.

	Sensitivity of DUS	Specificity of DUS	Positive predictive value	Negative predictive value	Detection rate of DUS for aneurysms/pseudoaneurysms /dissections
Carotid arteries (n = 184)	85.1%	67.3%	87.7%	62.2%	30.4% (7/23)
Renal arteries (n = 121)	65.3%	93.5%	94.2%	62.3%	9.1% (1/11)

Exercise improves vascular endothelial function in patients with early diabetic kidney disease

Ulf G Bronas, Marc L Weber, Diane Treat-Jacobson, Daniel Duprez, Mark Rosenberg

University of Minnesota, Minneapolis, MN, United States

Background: Observational studies have shown an inverse association between physical function and CVD mortality in patients with chronic kidney disease (CKD) through unknown mechanisms. We have previously reported that patients with diabetic CKD (DKD) have a 64% reduction in brachial artery flow mediated vasodilation (FMD) compared to historical controls with type 2 diabetes. Endothelial dysfunction is known to play an important role in the pathophysiology of atherosclerotic CVD. The purpose of this study was therefore to be the first study to test the hypothesis that 12 weeks of walking exercise training would improve endothelial function in patients with stage 2–4 diabetic kidney disease.

Methods: We randomly assigned 85 participants (62 male, age 64.1 ± 9.2) with stage 2–4 DKD to either 12-weeks of moderate-intensity walking exercise training, 4×/wk for 45 minutes (n = 42) or a usual medical care control group (n = 43). The primary endpoint was change in endothelial function measured by brachial artery FMD using ultrasonography at 12 weeks.

Results: There were no differences between groups in baseline demographic, medical, or pharmacological variables. At 12-week follow-up, group comparisons were analyzed by ANCOVA using baseline as a covariate. FMD was significantly improved from 2.66% to 4.96% (p < 0.001) in the exercise-intervention group, with a mean difference in FMD of 3.2% between groups (p = < 0.001). VO2 peak improved by 3.11 ml/kg/min (p = < 0.001) in the exercise group, with a mean difference of 3.58 ml/kg/min (p = < 0.001) between groups. Change in FMD was moderately correlated with change in aerobic capacity (r = 0.35, p = 0.001). Secondary variables are presented in Table.

Conclusions: This study is the largest randomized, controlled study to date demonstrating that walking exercise training improves endothelial function in patients with DKD, potentially providing mechanistic insight into the cardioprotective effect of exercise in this population.

Clinical yield of carotid and vertebral artery duplex ultrasonography in patients evaluated for syncope

Ian Del Conde, Alexander Papolos, Daniella Kadian-Dodov, Jeffrey W Olin

Mount Sinai School of Medicine, New York, NY, United States

Syncope is a common condition characterized by a transient loss of consciousness and postural tone. The mechanism of syncope involves decreased perfusion to the brainstem regions responsible for consciousness. There is the common misbelief that carotid or vertebral artery disease can be the cause of syncope. The clinical yield of carotid artery duplex ultrasound (DUS), however, is largely unknown. The aim of this study was (1) to determine the diagnostic yield of DUS in patients (pts) with syncope, and (2) determine how often the study results would change the patient’s management in terms of secondary prevention of cardiovascular disease (CVD). We retrospectively identified 273 patients who had DUS for a primary indication of syncope at the Mount Sinai Medical Center between 20102012. The mean age was 70 years, and 48% were male. There was a history of CVD in 57% of patients. Only 3 of 273 (1%) pts had severe bilateral vertebral artery disease, and 2 (0.7%) had evidence of subclavian steal as a potential mechanism for syncope. Eleven (9%) patients who did not have a history of cardiovascular disease were found to have atherosclerotic plaque in at least one of the imaged vessels, qualifying them for therapies targeted at the secondary prevention of CVD. In conclusion, DUS had a very low diagnostic yield in the identification of mechanisms underlying syncope, however the identification of subclinical atherosclerosis in 9% of patients without known CVD justifies the performance of carotid and vertebral ultrasound in patients with syncope.

The prevalence of abdominal aortic aneurysm, carotid stenosis, peripheral arterial disease and atrial fibrillation among 280,000 screened British and Irish Adults

Richard Bulbulia¹, Mohssen Chabok², Mohammad Aslam³, Sara Lewington⁴, Paul Sherliker⁴, Andrew Manganaro⁵, Alison Halliday⁶

¹Vascular Surgical Unit, Cheltenham General Hospital/Clinical Trial Service Unit, University of Oxford, Oxford, United Kingdom; ²Life Line Screening, London, United Kingdom; ³Vascular Department, Hammersmith Hospital, Imperial College, London, United Kingdom; ⁴Clinical Trial Service Unit, University of Oxford, Oxford, United Kingdom; ⁵Life Line Screening, Cleveland, OH, United States; ⁶Nuffield Department of Surgery, John Radcliffe Hospital, Oxford, United Kingdom

Objective: The identification of asymptomatic vascular disease by screening may help reduce major vascular morbidity and mortality. However, reliable contemporary prevalence estimates for AAA and other serious vascular pathologies in men and women are lacking.

Methods: Between August 2008 and June 2012, 283,924 adults underwent 4 cardio-vascular screening tests at 5800 sites across the UK and Ireland: Aortic ultrasound scan (AAA defined as > 30 mm diameter); Carotid duplex (moderate stenosis: PSV = 125–230 cm/s; high-grade stenosis: PSV > 230 cm/s); Ankle brachial pressure index (peripheral arterial disease [PAD] defined as ABPI < 0.90); and an electrocardiogram to identify atrial fibrillation (AF). Major cardio-vascular risk factors and current medications were also recorded via questionnaires.

Results: 39.7% of attendees were male. The overall prevalence of AAA was 1.0% (2503/255,304): 2.1% in men and 0.3% in women. Significant carotid artery stenosis (CAS) was seen in 2.8% (7813/275,088) of participants (1.7% moderate and 1.1% high-grade), and 3.4% of men and 2.4% of women screened had significant CAS. PAD was detected in 2.2% (5972/269,745) of participants (2.4% men and 2.1% women). AF was observed in 1.2% (2672/224,149) of those screened (2.0% men and 0.7% women).

Conclusion: The prevalence of AAA was low in this self-selected population, particularly in women, but significant CAS, AF and PAD were observed more frequently. Point prevalence estimates and associations with major cardiovascular risk factors may help inform the development of a new targeted and more comprehensive vascular screening program, which could prevent heart attacks, strokes and amputations as well as AAA deaths in men and women.

Evaluation of the Venowave (Venous-Return Assist Device) on walking distance in peripheral arterial disease

Natasha Aleksova, Jeremy Paikin, Barbara Nowacki, Binod Neupane, John Eikelboom, Jeffrey Ginsberg, Jack Hirsh, Sonia Anand

McMaster University, Hamilton, ON, Canada

Background: Current non-surgical treatment to improve intermittent claudication is limited to supervised exercise programs and short-term use of cilostazol, patients are frequently non-compliant with both therapies. The primary objective of this pilot study was to determine if the venous-return assist device Venowave improves intermittent claudication in patients with PAD, as measured by absolute claudication distance (ACD) on a treadmill.

Methods: The Venowave is a portable lower-limb device that we hypothesized would increase venous outflow of the limb, causing a reduction in venous pressure, enhanced arterial blood flow and improved exercise tolerance in intermittent claudicants. We conducted a single-centre, double-blind, randomized, cross over trial of 25 participants. Participants had to have resting ABIs < 0.4, ACD ≤ 200 m, or rest pain. Each participant wore the active and sham device for 30 minutes at rest prior to and during a supervised treadmill walk on two separate visits.

Results: When compared to baseline ACD, there was no statistically significant difference between the active and sham devices (Table). Participants walked 14 meters further with the active device than sham device, but walking distance with both devices was still less than at the baseline visit. Initial claudication distance improved with both active and sham devices compared to baseline, but this difference was not significant. A significant period effect was observed with all participants walk time increasing at the first study visit after the baseline evaluation. Participants did report less difficulty walking with the active device.

Conclusion: Our pilot study was unable to show that the Venowave significantly improves walking distance in patients with severe PAD, although it may improve walking difficulty. A larger trial with a more select group of patients who wear the device for a longer period of time may be warranted.

OPEN IN VIEWER

Table.

Results for primary and secondary outcomes.

	Baseline	Active Device	Active Difference	Sham Device	Shame Difference	p-value
Primary outcome
Absolute claudication	299.9 (132.8)	291.1 (118.1)	–8.8 (86.5)	276.9 (113.1)	–22.9 (119.6)	0.52
Distance, meters (SD)
Secondary outcomes
Initial claudication	97.2 (57.1)	112.9 (49.9)	15.7 (45.0)	118.1 (48.6)	18.0 (45.6)	0.77
Distance, meters (SD)
Walk time, minutes (SD)	5.8 (2.5)	5.6 (2.1)	–0.2	5.6 (2.0)	–0.2	0.99
Modified WIQ score (SD)	N/A	32.3 (6.0)	1.8	30.5 (6.2)	1.8	0.05

Risk factors, clinical features and endovascular management of iliac artery fibromuscular dysplasia

Siva S Ketha, Sanjay Misra

Mayo Clinic, Rochester, MN, United States

Fibromuscular dysplasia (FMD) is a nonatherosclerotic, noninflammatory vascular disease that most frequently affects renal and carotid arteries. FMD of the iliac arteries is very rare with literature limited to case reports. Iliac FMD patients often present with claudication which can be bilateral. The incidence of FMD involving the iliac artery is unknown and large studies with long term follow-up are not available. Our study objectives were to identify the spectrum of clinical presentation of iliac artery FMD, to characterize its prevalence in patients with renal artery FMD and to review the experience of a single center with regard to its treatment. All patients with incident renal FMD January 1980 through December 2010 seen at Mayo Clinic were identified. There were 415 patients with renal FMD of which 10 patients (2.4%; mean age 56 ± 13.8 years, 87% women) had iliac FMD. Of these four patients were incidentally found to have iliac FMD as part of a renal artery evaluation and treatment. The remaining six patients had life style limiting claudication involving one or both extremities for which endovascular treatment (n = 10) was performed, including angioplasty (n = 8) and self-expanding stent placement (n = 2). All six patients were reported as mild PAD based on ABI measurements. Associated risk factors included hypertension (30%), dyslipidemia (20%), smoking history (20%), coronary artery disease (0%), diabetes (0%), migraines (5.7%), current oral contraceptive use (20%), and obesity (50%). One-year survival rate after endovascular treatment was 100% and 12-month primary patency rates were 90%. In conclusion, iliac FMD is rare in patients with renal FMD and may be found incidentally or may present with disabling claudication that is amenable to endovascular treatment.

Factors associated with delay in diagnosis of patients with fibromuscular dysplasia: A report from the United States Registry for Fibromuscular Dysplasia

Jordan Gavin¹, Xiaokui Gu¹, Heather L Gornik², Jeffrey W Olin³, Esther SH Kim², Pamela D Mace², Bruce H Gray⁴, J Michael Bachrach⁵, Michael R Jaff⁶, Robert D McBane⁷, Alan H Matsumoto⁸, Eva Kline-Rogers¹, James B Froehlich¹

¹University of Michigan, Ann Arbor, MI, United States; ²Cleveland Clinic, Cleveland, OH, United States; ³Mount Sinai, New York, NY, United States; ⁴Greenville Hospital System, Greenville, SC, United States; ⁵North Central Heart, Sioux Falls, SD, United States; ⁶Massachusetts General Hospital, Boston, MA, United States; ⁷Mayo Clinic, Rochester, MN, United States; ⁸University of Virginia, Charlottesville, VA, United States

Background: Fibromuscular dysplasia (FMD) is an uncommon disease of medium-sized arteries that may result in stenosis, dissection or aneurysm. It is unclear what factors affect delay between first signs/symptoms and diagnosis.

Methods: Patients enrolled in the FMD registry from ten U.S. sites were stratified into three groups based upon the length of time between first sign/symptom and diagnosis.

Results: Of the 615 total patients enrolled in the FMD registry, 538 patients had sufficient data for analysis. The mean length of time from first reported clinical sign/symptom to diagnosis was 3.6 ± 7.4 years. As outlined in the Table below, FMD patients with greater delay in diagnosis were younger at first sign/symptom and older by the time of diagnosis. Patients with a greater delay in diagnosis were more likely to present with hypertension, had earlier onset and greater family history of hypertension. Furthermore, they had a greater mean number of blood pressure medications and were more likely to take an ARB, diuretic or alpha blocker when compared to patients with a smaller gap between first sign/symptom and diagnosis. Conversely, FMD patients with a shorter time to diagnosis were more likely to have presented with a carotid or renal artery dissection. See Table.

OPEN IN VIEWER

Table.

	Time between first sign/symptom and diagnosis			p-value
< 3 years N (%)	3–5 years N (%)	> 5 years N (%)
N	383/538 (71.2)	55/538 (10.2)	100/538 (18.6)
Age at first sign/symptom (mean ± SD)	50.4±13.3	45.7±12.4	39.1±15.8	< 0.0001
Age at diagnosis (mean ± SD)	50.6±13.3	49.5±12.4	55.3±13.6	0.0041
Family history of hypertension	241/333 (72.4)	41/53 (77.4)	82/94 (87.2)	0.0087
Age at onset of hypertension (mean ± SD)	44.8±13.9	39.6±13.6	38.9±16.0	0.0056
Presenting symptoms or signs
Headache	195/339 (57.5)	31/53 (58.5)	54/94 (57.4)	1.0
Hypertension	229/360 (63.6)	34/54 (63.0)	75/96 (78.1)	0.021
Pulsatile tinnitus	106/321 (33.0)	20/50 (40.0)	31/88 (35.2)	0.60
Hemispheric TIA	35/330 (10.6)	4/51 (7.8)	7/93 (7.5)	0.67
Stroke	30/336 (8.9)	0/52 (0)	10/94 (10.6)	0.028
Cervical bruit	86/311 (27.7)	12/50 (24.0)	18/90 (20.0)	0.34
Carotid artery dissection	64/326 (19.6)	2/50 (4.0)	7/91 (7.7)	0.0008
Renal artery dissection	17/327 (5.2)	1/52 (1.9)	0/92 (0)	0.0350

Conclusion: The delay from first sign/symptom to diagnosis in FMD patients is prolonged and associated with having hypertension. This study suggests that the role of FMD is under-appreciated in patients who present with early onset hypertension. Patients with an acute arterial dissection are more likely to have a timely diagnosis of FMD. Further effort may be needed to increase physician awareness of FMD to more effectively diagnosis this disease and expedite appropriate treatment.

Underutilization of statin therapy in patients with peripheral artery disease – A long way to the finish line

Prasanna Sengodan, Harpreet Grewal, Sanjay Gandhi

MetroHealth Medical Center at Case Western Reserve University, Cleveland, OH, United States

Background: Hyperlipidemia is an important modifiable risk factor for peripheral arterial disease (PAD). The current ACC/AHA guidelines recommend treatment with a HMG CoA reductase inhibitor (statins) to achieve a target low-density lipoprotein level < 100 mg/dl in PAD patients. The purpose of our study was to compare the rates of statin utilization in patients with PAD to those with CAD in contemporary practice.

Methods: The Explorys database, which is an open private cloud platform, was used to obtain data from 2007–2012. Explorys integrates electronic non-identified patient data across 13 major healthcare systems comprising 31 million patients across the United States. The categorical variables between patients with PAD alone were compared to those of patients with CAD alone or combined CAD and PAD using 2 sided Mantel-Haenszel chi square test.

Results: 502,520 patients with PAD alone, 546,140 patients with CAD alone and 146,140 patients with both PAD and CAD were identified using power search. Patients with PAD alone were significantly more likely to women (55%, 41%, 42%, p < 0.0000001) be under 65 years of age (44%, 36%, 27%, p < 0.0000001) and had lower prevalence of diabetes (35%, 40%, 56%, p < 0.0000001) and hypertension (70%, 81%, 93%, p < 0.0000001) compared to patients with CAD or combined PAD and CAD respectively (Table). Significantly fewer patients with PAD alone were on statins compared to patients with CAD alone or combined CAD and PAD (41 %, 56%, 66%, p < 0.0000001) respectively.

Conclusions: Statins are underutilized in patients with both CAD and PAD. However, patients with PAD alone are significantly less likely to receive secondary prevention with statin therapy compared to patients with established CAD.

OPEN IN VIEWER

Table.

Baseline demographics and results.

	PAD (n = 502,520)	CAD (n = 546,140)	PAD & CAD (n = 146,140)	p-value
Age (%) > 65 years 18–64 years	56 44	64 36	73 27	< 0.000001 < 0.000001
Men (%)	45	59	58	< 0.0000001
HTN (%)	70	81	93	< 0.0000001
Diabetes (%)	35	40	56	< 0.0000001
Smoking (%)	21	21*	28**	*= 0.9997 **< 0.0000001
Statins use (%)	41	56	66	< 0.0000001

Etiology of acute limb ischemia in cancer patients

Elie Mouhayar, Reem Aoun, Peter Kim, Cezar Iliescu

UT/MD Anderson Cancer Center, Houston, TX, United States

Background: The etiology of acute limb ischemia (ALI) in patients with underlying malignancy is ill defined. Tumor embolization has been suggested as a potential cause based on few case reports. Our aim was to investigate the mechanism of ALI in our cancer population.

Methods: Single center, retrospective chart and pathology review of patients with known active malignancy and undergoing surgical revascularization for acute limb ischemia.

Results: A total of 26 patients met inclusion criteria. Pathology results were available for 24 who were included in the final analysis. Majority (19, 79%) were men and 17 (70%) were > 65 years old. Genitourinary (8, 33%) and lung (6, 25%) cancers were the most commonly indentified malignancies. Metastatic disease was present in 13 (54%) patients. Four (16%) patients had previously documented peripheral arterial disease and none had atrial fibrillation. Lower extremities were the site of ALI in 21 (87%) of the cases. Embolectomy was the most commonly performed procedure (22, 91%).

Thrombus was identified in 16 (67%) patients. Severe atherosclerotic changes with associated thrombus clot was noted in another 4 (16%) patients. Tumor extrinsic invasion of the arterial wall was documented in 2 (8%) cases. One patient with leukemia had a pathologic specimen showing clot made of leukemic cells.

Conclusion: ALI in patients with underlying malignancy is more likely to be due to thrombosis or local tumor invasion of the arterial wall than tumor embolization.

Gender and ethnicity in patients with premature peripheral artery disease (PAD)

Pavel J Levy, Jeanette S Andrews, Billy Chacko, George W Plonk, Matthew Edwards

Wake Forest School of Medicine, Winston-Salem, NC, United States

Background: PAD in younger adults with premature lower extremity atherosclerosis (PLEA) has been diagnosed with increasing frequency. Previous studies provided little information on gender and ethnic aspects of PLEA.

Methods: We studied consecutive patients (patients) (mean age 49.4 ± 6.4 yrs) with severe PAD diagnosed at ≤ 55 yrs of age (mean 45.5 ± 6.9 yrs) who were treated between 1998–2010. Data was collected prospectively at initial evaluation for PAD care.

Results: Among 565 patients, 46% were female and 20% black. Female patients comprised > 50% of patients ≤ 40 yrs of age. Female patients when compared to males had greater frequency of isolated aorto-iliac disease and visceral (renal, mesenteric) artery disease (VisAD) (p < 0.05 for each parameter); and had comparable frequencies of cardiovascular risks and polyvascular disease (PVD). In multivariable logistic regression analysis (MLRA) (Table) only depression showed a significant association. Black patients when compared to whites had higher frequency of hypertension, systemic disease and prior amputation, but lower frequency of isolated aorto-iliac disease, VisAD and PVD (p < 0.05 for all parameters). By MLRA, hypertension and systemic disease were significantly positively associated with race, while VisAD and PVD showed significant negative associations.

Conclusions: More than 1/2 of PLEA patients ≤ 40 were female. Importantly, female gender was highly associated with depression (OR > 3). Black patients had different anatomic pattern of PLEA and lower prevalence of other CVD.

OPEN IN VIEWER

Table.

Variable	Female %	Male %	OR (95% CI)	p- value*	Black %	White %	OR (95% CI)	p- value*
Hypertension	67.3	60.9	1.32 (0.87–2.03)	0.19	74.8	61.2	2.43 (1.41–4.19)	< 0.01
Hyperlipidemia	59.2	54.8	1.46 (0.97–2.22)	0.07	46.9	59.3	0.65 (0.39–1.08)	0.10
Diabetes	26.2	24.8	0.89 (0.57–1.38)	0.59	30.9	24.1	1.48 (0.87–2.52)	0.15
Polyvascular CVD	44.6	44.9	0.75 (0.49–1.14)	0.18	31.5	48.0	0.48 (0.29–0.82)	< 0.01
Visceral AD	20.2	12.3	1.47 (0.86–2.52)	0.16	6.3	18.3	0.36 (0.15–0.84)	0.02
Systemic Disease	12.5	10.9	1.06 (0.60–1.88)	0.84	17.1	10.3	2.24 (1.19–4.20)	0.01
Depression	51.8	25.6	3.37 (2.28–4.97)	< 0.01	30.0	39.5	0.62 (0.37–1.05)	0.07

*

p-value from multivariable model which adjusts for all variables in this table plus age at PAD dx, family history of any early CVD, and race (gender model) or gender (race model), none of which were statistically significant.

Lipid lowering and claudication: A meta-analysis

Rajmony Pannu¹, Stalin Subramaniam², Ruchi Rachmale³, Thom Rooke⁴

¹El Camino Hospital, Mountain View, CA, United States; ²Brookdale University Hospital and Medical Center, Brooklyn, NY, United States; ³Wayne State University, Detroit, MI, United States; ⁴Mayo Clinic, Rochester, MN, United States

Objective: To evaluate effects of lipid lowering therapy (LLT) on improving pain free walking distance in subjects with peripheral arterial disease (PAD).

Methods: Randomized controlled trials (RCT) of lipid lowering medication versus placebo on claudication in subjects with PAD were considered for review. Pain free walking distance (PFWD), maximum walking distance (MWD) & ankle brachial index (ABI) were assessed as outcomes. We performed a computerized search, without language restriction, to identify relevant articles from 1960 to December 2012 in the MEDLINE, Embase, ISI Web of Science, Cochrane Central Register of Controlled Trials and SCOPUS. The literature search was performed by a Mayo Clinic librarian and confirmed by two study authors who also performed data extraction. Jadad score was used to assess quality of trials and allocation concealment was assessed separately. Statistical analysis was performed using Revman 5.0. Predetermined subgroup analysis was performed for duration of treatment.

Results: Eight studies (seven in English and one in Czech) with 1373 participants met criteria (Table). There was marked heterogeneity between the studies. Compared with placebo, LLT increased PFWD by an average of 41 meters (m) (p = 0.0001) from baseline and MWD by 71 m (p = 0.00001). LLT had small increase in ABI 0.06 (p = 0.05). Treatment for three months did not increase PFWD but increase in PFWD was noted at six months and thereafter.

Conclusions: LLT, when administered for greater than six months, may improve pain free walking distance and maximum walking distance in people with PAD. There is limited direct evidence supporting use of specific LLT agent and increase in PWFT. A comprehensive evaluation of LLT in increasing PWFT among PAD population should be high research priority.

The influence of chronic kidney disease severity on ABI and walking distance in patients with peripheral arterial disease

Ana I Casanegra, David M Thompson, Polly S Montgomery, Andrew W Gardner University of Oklahoma, Oklahoma City, OK, United States

Background: Patients with chronic kidney disease (CKD) have a high prevalence of peripheral arterial disease (PAD). End stage CKD patients in dialysis often have calcified vessels, rendering the ankle brachial index (ABI) less reliable. Our aim is to determine the relationship between ABI and walking distance in patients with PAD across the spectrum of renal function.

Methods: Patients with PAD had their creatinine level and exercise performance assessed. We calculated the estimated glomerular filtration rate (GFR) with the four variable Modification of Diet in Renal Disease (MDRD) equation. We divided the patients into a high GFR tertile (101.4 ± 16 ml/min/1.73 m²) middle GFR tertile (76.5 ± 6.1 ml/min/1.73 m²), and low GFR tertile (52.6 ± 9.4 ml/min/1.73 m²). Patients with end stage CKD were excluded. Exercise performance was determined by measuring the absolute claudication distance (ACD) during a treadmill test.

Results: We enrolled 199 patients, 52% males, 58% Caucasian, ABI of 0.71 ± 0.23 (mean ± SD), and GFR of 77.1 ± 0.32 ml/min/1.73 m². In multivariate regression, ABI (p < 0.0001), age (p = 0.003), and GFR tertile (p = 0.018) were associated with ACD. In the equation predicting ACD, the regression coefficient for ABI was 295 m (95% CI = 103 m to 487 m; r² = 0.2, p = 0.0011), in the low tertile, which was significantly lower (p = 0.004) than the regression coefficient of 614 m (95% CI = 375 m to 854 m; r² = 0.32, p < 0.0001) in the high tertile.

Conclusions: After adjusting for age, the association between ABI and ACD is weakest in patients in the lowest tertile of GFR. Therefore, we conclude that the association between ABI and ACD in patients with PAD is impaired in those with progressively severe CKD.

Outcomes of ultrasound-guided thrombin injection for the treatment of brachial artery pseudoaneurysms in the vascular laboratory

Lee Joseph¹, Meisam Moghbelli², Aditya Sharma³, Marcelo Gomes¹, Douglas Joseph¹, Esther SH Kim¹, Neil Poria¹, Susan Whitelaw¹, Heather L Gornik¹

¹Cleveland Clinic, Cleveland, OH, United States; ²University of Mississippi Medical Center, Jackson, MO, United States; ³University of Virginia, Charlottesville, VA, United States

Background: Arterial pseudoaneurysm (PSA) is an established complication of arteriography. Case series of lower extremity PSA report a success rate of ultrasound-guided thrombin injection (UGTI) up to 94%. There is little published data regarding treatment of brachial arterial PSA.

Methods: Database query for upper extremity PSA identified within a single vascular laboratory between 1/1/2000 – 8/15/2012. Demographic and clinical data were obtained by medical record review. Ultrasound images were reviewed for PSA characteristics and outcomes.

Results: Of the 150 upper extremity PSA identified in 148 patients, 138 involved the brachial artery (92%). Ninety eight PSA (71%) occurred as a complication of coronary or peripheral angiography. Among the brachial PSA, initial treatment strategy was: UGTI 70/138 (50.7%), ultrasound-guided compression repair 31/138 (22.5%), surgery 16/138 (11.6%), observation 21/138 (15.2%). Of the 70 PSA treated initially with UGTI, mean chamber size, tract length and width were 2.9 cm, 0.6 cm, and 0.4 cm respectively, and 22.8% were multi-chambered. The primary success rate of UGTI (resolution without recurrence) was 73.5% (50/68). Secondary success of UGTI (including 10 patients who underwent 2nd UGTI) was 82.4% (56/68). Anticoagulation or antiplatelet therapies were not risk factors for UGTI failure. UGTI failure requiring surgical repair was needed in 9/70 (12.9%). Eight patients (8/70, 11.4%) experienced a complication related to UGTI, most of which were asymptomatic ultrasound findings: 4/70 (5.7%) brachial artery thrombus without hand ischemia treated conservatively, 1/70 (1.4%) brachial artery stenosis without thrombus, 1/70 (1.4%) brachial DVT, 1/70 (1.4%) reduced radial pulse. One patient (1.4%) required surgery for evacuation of large hematoma with skin compromise.

Conclusion: UGTI is an effective treatment for brachial PSA in a real world case series, though primary (73.5%) and secondary (82.4%) success rates are lower than those reported for lower extremity PSA.

CHA₂DS₂-VASc score and risk for reobstruction after endovascular treatment of the superficial femoral artery: Differences between balloon angioplasty and stenting

Thomas Gary, Klara Belaj, Philipp Eller, Gerald Hackl, Franz Hafner, Harald Froehlich, Ernst Pilger, Marianne Brodmann

Medical University Graz, Graz, Austria

Background: The CHA₂DS₂-VASc (congestive heart failure, hypertension, age > 75 years (doubled), type 2 diabetes, previous stroke, transient ischemic attack, or thromboembolism (doubled), vascular disease, age 65–75 years, and sex category) score was published as a predictive scoring model for stroke in atrial fibrillation patients. As multiple vascular risk factors are included in this score we evaluated the occurrence of reobstruction after endovascular treatment (percutaneous transluminal angioplasty (PTA) and stent) of the superficial femoral artery (SFA) in peripheral arterial occlusive disease (PAOD) patients according to their CHA₂DS₂-VASc score independent of a coexisting atrial fibrillation.

Methods: We evaluated 773 PAOD (529 PTA and 244 stent) patients treated at our institution from 2005 to 2010. CHA₂DS₂-VASc score was calculated and the occurrence of a symptomatic reobstruction during a median follow up of 58 months was investigated. Furthermore all constituents of the score were individually investigated concerning their association with reobstruction.

Results: In PTA patients reobstruction rate increased with increasing CHA₂DS₂-VASc Score (p = 0.009). Arterial hypertension was associated with an increased risk for reobstruction (OR 2.4; 95%CI 1.7–3.4), as was type 2 diabetes (OR 1.9; 95%CI 1.5–2.3). In stent patients reobstruction rate was high, but independent of the CHA₂DS₂-VASc score (p = 0.5). Its constituents were not associated with an increased risk for reobstruction.

Conclusion: A high CHA₂DS₂-VASc score was associated with a high risk for reobstruction after PTA of the SFA. On the other hand the reobstruction risk in stent patients was high but independent of the CHA₂DS₂-VASc score, indicating that the pathophysiology of instent restenoses is different from reobstruction after PTA.

Clinical utility of cerebral angiography in the preoperative assessment of endocarditis

Peter P Monteleone, Nabin Shrestha, Jessen Jacob, Steven Gordon, Thomas Fraser, Dalia El Bejjani, Susan Rehm, Christopher Bajzer, Samir Kapadia, Mehdi Shishehbor

Cleveland Clinic, Cleveland, OH, United States

Background: In presentation with infective endocarditis (IE), concern arises for infectious mycotic aneurysm (MA). Cerebral angiography (CA), the gold standard for MA evaluation, is an invasive procedure. The literature to date poorly demonstrates the clinical significance of MA in IE. No guidelines exist to direct CA utilization in IE, a disease process with 10–15,000 new cases diagnosed in the US each year.

Methods: We identified all patients meeting Duke’s criteria between 2007 and 2011 who underwent cardiac surgery at Cleveland Clinic with a diagnosis of IE, and all patients who were discharged on IV antibiotics treating IE between 2007 and 2009.

Results: Of the 553 patients (481 surgical and 72 nonsurgical), 144 underwent CA. Frequency of CA performance increased steadily over the assessed interval. Seven patients had MA identified (1.2% of all patients or 4.9% of all CA patients).

MA+ and MA– patients were clinically similar with the exception that 71.4% of MA+ and only 14.5% of MA– patients were identified to have Strep viridans as the causative agent of IE (Table).

Of the seven MA+ patients, six underwent cardiac surgery. One was refused surgery due to multiple comorbidities including severe pulmonary disease. Of the surgical patients, four underwent MA embolization prior to surgery, one had repeat CA after antibiotic administration to confirm decreased size or stability of the MA and one had no further workup prior to surgery. Six survived to discharge with one surgical patient dying post-operatively secondary primarily to bowel ischemia and sepsis.

Conclusion: Though rare, MA is an important consideration in the clinical management and preoperative evaluation of IE. Early analysis suggests that Strep viridans bacteremia may convey some additional risk for MA. Identification of MA is not prohibitive of performance of successful cardiac surgery.

OPEN IN VIEWER

Table.

Clinical and echocardiographic characteristics of patients with mycotic aneurysm and without mycotic aneurysm on cerebral angiography.

Characteristic	Patients with mycotic aneurysm on CA (n = 7)	Patients without evidence of mycotic aneurysm on CA (n = 144)
Age (yrs, mean ± std dev)	51.9 ± 12.4	55.4 ± 15.3
Male sex % (no.)	85.7% (6)	68.1% (98)
Diabetes % (no.)	14.3% (1)	28.47% (41)
Hypertension % (no.)	57.1% (4)	51.4% (74)
Coronary Artery Disease % (no.)	28.6% (2)	44.4% (64)
Immunocompromised state % (no.)*	14.3% (1)	7.6% (11)
Baseline CKD present % (no.)	28.6% (2)	40.3% (58)
Prosthetic Valve Endocarditis % (no.)	28.6% (2)	43.1% (62)
Causative microorganism % (no.)
Viridans group streptococci	71.4% (5)	14.5% (21)
Other streptococci	0.0% (0)	11.1% (16)
Staphylococcus aureus	0.0% (0)	29.2% (42)
Enterococcus	0.0% (0)	12.5% (18)
Fungus	14.3% (1)	2.1% (3)
HACEK	14.3% (1)	2.8% (4)
Other pathogen	0.0% (0)	20.1% (29)
Pathogen not identified	0.0% (0)	7.6% (11)

*

Includes HIV, cirrhosis, s/p solid organ transplant, autoimmune disease on immunomodulator therapy.

Vessel involvement in Marfan syndrome patients

Wouter van der Pluijm¹, Marco Di Eusanio², Rossella Fattori², Matthias Voehringer³, Truls Myrmel⁴, Mark D Peterson⁵, Patrick O’Gara⁶, Mark J Russo⁷, Arturo Evangelista⁸, Daniel G Montgomery¹, Kim A Eagle¹, Eric M Isselbacher⁹, Christoph A Nienaber¹⁰

¹University of Michigan, Ann Arbor, MI, United States; ²University Hospital S Orsola, Bologna, Italy; ³Robert-Bosch Krankenhaus, Stuttgart, Germany; ⁴Tromso University Hospital, Tromso, Norway; ⁵St Michael’s Hospital, Toronto, ON, Canada; ⁶Brigham & Women’s Hospital, Boston, MA, United States; ⁷University of Chicago, Chicago, IL, United States; ⁸Hospital General Universitari Vall d’Hebron, Barcelona, Spain; ⁹Massachusetts General Hospital, Boston, MA, United States; ¹⁰University of Rostock, Rostock, Germany

Objective: Although previous studies have shown that aortic branch vessel dilatation is not uncommon in patients with Marfan Syndrome (MFS), it is relatively unknown whether the altered genetic substrate impacts the incidence of branch vessel involvement and related complications in patients with acute aortic dissection.

Methods: Among 3690 patients enrolled in the International Registry of Acute Aortic Dissection between 1996 and 2012, 147 were identified with MFS. Of these, 100 had acute Type A dissection (TAAAD) and the remaining 47 had acute Type B dissection (TBAAD). Vessel involvement was defined as extension of the dissection flap into the vessel ostium.

Results: There was no difference in branch vessel involvement noted on diagnostic imaging between TAAAD patients with and without MFS. For TBAAD patients, renal artery involvement was significantly more common in MFS patients. As well, concomitant, arch vessel involvement was seen more often in type B patients with MFS versus patients without MFS. No differences between groups were seen in branch vessel-related complications, including malperfusion or myocardial infarction. Among TAAAD patients who developed mesenteric ischemia, those without MFS were significantly more likely to die in-hospital than patients with MFS (71.0% vs. 0.0%, p = 0.027), which is likely due in part to the low number of MFS patients who developed mesenteric ischemia. No other differences between patients with and without MFS were identified. (Table)

Conclusion: TBAAD patients with MFS have significantly more renal artery and arch involvement than TBAAD patients without MFS.MFS patients do not otherwise appear to develop branch vessel or arch complications more frequently than patients without MFS.

OPEN IN VIEWER

Table.

	Type A Dissection			Type B Dissection
	No Marfan’s	Marfan’s	P value	No Marfan’s	Marfan’s	P value
Abdominal Vessel Involvement	464 (20.7%)	19 (19.0%)	.685	373 (30.3%)	22 (46.8%)	.016
Right Renal	161 (7.2%)	10 (10.0%)	.288	153 (12.4%)	11 (23.4%)	.027
Left Renal	274 (12.2%)	11 (11.0 %)	.717	176 (14.3%)	13 (27.7%)	.011
Complete	151 (6.7%)	3 (3.0%)	.211	129 (10.5%)	4 (8.5%)	.811
Arch Vessel Involvement	623 (30%)	30 (41.7%)	.648	78 (7.7%)	7 (17.5%)	.026
Coronary Vessel Involvement	198 (12.4%)	13 (18.1%)	.156	2 (0.2%)	1 (2.9%)	.096
Pre-op Mesenteric Ischemia	82 (3.8%)	2 (2.1%)	.581	66 (5.6%)	4 (6.3%)	.350
Pre-op Limb Ischemia	233 (10.8%)	6 (6.4%)	.170	106 (9%)	1 (2.0%)	.118
Pre-op or Post-op Mesenteric Ischemia or Infarction	138 (6.4%)	3 (3.1%)	.279	75 (6.4%)	4 (8.3%)	.546
Pre-op or Post-op Mesenteric Limb Ischemia	275 (12.8%)	7 (7.4%)	.126	116 (9.9%)	1 (2.0%)	.080
Pre-op or Post-op Mesenteric Myocardial Infarction/ Ischemia	269 (13.7%)	17 (17.7%)	.273	62 (5.3%)	0 (0.0%)	.169

Sleep characteristics in persons with peripheral arterial disease

Atul Jain¹, Peter J de Chavez², Lihui Zhao², Mary M McDermott², Mercedes Carnethon²

¹The University of Chicago, Chicago, IL, United States; ²Northwestern University, Chicago, IL, United States

Short duration and poor quality sleep is increasingly recognized as an adverse health behavior based on its association with atherosclerotic coronary heart disease and stroke. Less is known about the association of PAD with sleep duration and quality. We conducted a pilot study to investigate the cross-sectional association of PAD with sleep duration and sleep percentage (a measure of sleep quality). 28 men and women with PAD were recruited from Biomarker Risk Assessment in Vulnerable Outpatients (BRAVO), a prospective, observational study at Northwestern University. 27 men and women free of PAD were recruited from the Northwestern Aging Research Registry. Study participants underwent wrist actigraphy over seven consecutive days and nights to measure mean sleep percentage (% of time during sleep interval spent sleeping) and mean sleep duration. Linear regression was used to compare sleep duration and sleep percentage by PAD status. Unadjusted mean sleep duration did not differ between PAD and non-PAD participants (434.6 ± 11.7 minutes vs. 443.9 ± 10.0 minutes, p = 0.55) (Table a). Unadjusted mean sleep percentage was lower among PAD participants than among non-PAD participants (88.8 ± 1.0% vs. 91.4 ± 0.6%, respectively, p = 0.02). After adjusting for covariates (age, gender, race, body mass index, hypertension, and coronary heart disease), the mean sleep percentage among PAD and non-PAD participants was 89.6 ± 1.3% and 90.0 ± 1.3%, respectively, p = 0.77 (Table b). In a small, cross-sectional study of 55 participants, mean sleep duration did not differ between adults with and without PAD. Sleep percentage was lower in adults with PAD, but these findings did not persist after adjusting for covariates. Further investigation with larger sample sizes is warranted to conclusively identify any potential association between short duration and poor quality sleep and PAD.

OPEN IN VIEWER

Table.

(a) Mean sleep duration (minutes) of participants with and without PAD.

	PAD (n = 28)	Non-PAD (n = 27)	p-value
Unadjusted	434.6 ± 11.7	443.9 ± 10.0	0.552
Model I	432.10 ± 17.8	455.66 ± 16.9	0.235
Model II	439.9 ± 17.1	457.0 ± 16.7	0.401
(b) Mean sleep percentage of participants with and without PAD
	PAD (n = 28)	Non-PAD (n = 27)	p-value
Unadjusted	88.8 ± 1.0	91.4 ± 0.6	0.023
Model I	89.4 ± 1.2	90.2 ± 1.2	0.576
Model II	89.6 ± 1.3	90.0 ± 1.3	0.772

Values shown are mean ± standard error. Model I adjusted for age, gender, and race. Model II additionally adjusted for body mass index, hypertension, and coronary heart disease.

PAD, peripheral arterial disease.

Low ABI predicts higher coronary syntax scores and myocardium at risk, but not incomplete coronary revascularization

Michael S McMurtry, Meghan Sebastianski, Seshasayee Narasimhan, Olga Toleva, Jay Shavadia, Seraj Abdualnaja, Ross Tsuyuki, Michelle M Graham

University of Alberta, Edmonton, AB, Canada

Background: Peripheral arterial disease is associated with coronary artery disease (CAD) and poor outcomes after coronary revascularization. We hypothesized that patients with low ankle brachial index (ABI ≤ 0.90) have more complex CAD and myocardium at risk than patients with normal ABI (1.00 ≤ ABI ≤ 1.40), and that their coronary revascularization is less complete.

Methods: 728 consecutive patients were drawn from a prospective cohort of adults referred for coronary angiography. ABI was measured bilaterally using Doppler ultrasound prior to angiography. Blinded reviewers measured Syntax score, Duke Jeopardy score at baseline and at three months. Data was analyzed using one-way ANOVA and multinomial logistic regression. Thresholds for high syntax score and high Duke Jeopardy score (pre and post revascularization) were calculated using the entire sample means plus one standard deviation, rounded to the nearest integer.

Results: Of 728 patients, 56 had ABI ≤ 0.90, 57 had 0.90 < ABI < 1.00, 563 had 1.00 ≤ ABI ≤ 1.40, and 49 had ABI > 1.40. After adjustment for age, sex, hypertension, dyslipidemia, diabetes and smoking status, subjects with ABI ≤ 0.90 had an odds ratio for high syntax score of 3.2 (95% CI 1.2–8.8) compared with the normal ABI group. Similarly, after adjustment the odds ratio for high baseline Duke Jeopardy score was 3.3 (95% CI 1.3–8.1) in the ABI ≤ 0.90 group. The odds ratio for high post-revascularization Duke Jeopardy score was 2.6 (95% CI 0.8–8.4; p = 0.104) in the ABI ≤ 0.90 group.

Conclusions: We found ABI ≤ 0.90 is associated with higher Syntax scores and more myocardium at risk in subjects referred for coronary angiography. We did not find that subjects with ABI ≤ 0.90 have less complete coronary revascularization.

Atherosclerotic and non-atherosclerotic spontaneous coronary artery dissection: Two distinct clinical entities

Asuka Ozaki, Sachin S Goel, Venu Menon, Esther SH Kim

Cleveland Clinic, Cleveland, OH, United States

Background: The differences between spontaneous, atherosclerotic (ACAD) and spontaneous, non-atherosclerotic (SCAD), coronary artery dissection are not well defined.

Methods: We queried the coronary angiography database at our institution for ‘coronary dissection.’ Medical records were reviewed for demographics, presenting symptoms, interventions, and follow-up. Angiograms (angios) were reviewed for presence of atherosclerosis, location of dissection, and interventions. Cases with prior intervention were excluded. Cases were deemed ACAD if atherosclerosis was present in any vessel.

Results: Of 110,223 angios over 20 years, 67 cases (0.06%) of coronary dissection were identified. Of these, 26 were SCAD (73% females) and 41 ACAD (32% females). Mean age was lower in SCAD compared to ACAD (43 vs. 59 years, p < 0.001). Acute MI on presentation was significantly greater in patients with SCAD compared to ACAD (85% vs. 22%, p < 0.001). 56% of ACAD presented with chest pain or were found incidentally. SCAD was associated with strenuous activity (n = 6), post-partum (n = 4), or arteriopathy (n = 5). Etiology of coronary dissection was undetermined in 12 cases of SCAD. SCAD most commonly occurred in the LAD (62%), whereas ACAD most commonly occurred in the RCA (39%), followed by the LAD (32%). On follow up angios in 17 SCAD and 10 ACAD patients, new coronary dissections were seen in two SCAD patients and none in the ACAD group.

Conclusions: In a single-center retrospective review of 110,223 angios, spontaneous coronary dissection remains rare. SCAD and ACAD appear to be two distinct clinical entities, and investigation for arteriopathy may be warranted in cases of SCAD.

Determinants of quality of life among women with fibromuscular dysplasia

Natalia Fendrikova Mahlay, Mingyuan Shao, Esther SH Kim, Ruchi Sanghani, Leopoldo Pozuelo, Heather L Gornik

Cleveland Clinic, Cleveland, OH, United States

Background: Fibromuscular dysplasia (FMD) is an uncommon arterial disease which primarily affects women in the prime of life. We describe health related quality of life (HRQOL) in a cohort of women with FMD.

Methods: Short Form-36 v1 (SF-36) questionnaire was provided to new patients in a single-center FMD program. Physical component summary (PCS) and mental component summary (MCS) scores were calculated. Medical record review was undertaken for symptoms, vascular bed involvement, clinical events, and comorbidities. Regression modeling was used to identify risk factors for HRQOL impairment.

Results: 137 FMD patients underwent SF-36 assessment, of which 129 were female (94%). Among women with FMD, mean age was 53.3 ± 12.7 years. Median number of vascular beds involved was two with 82.7% (105/127) and 63.9% (78/122) of patients imaged with carotid/vertebral and renal involvement, respectively. Most common clinical symptoms were headache (78.7%), hypertension (67.2%), pulsatile tinnitus (56.7%), neck pain (42.7%), and dizziness (33.3%). Arterial dissection was reported in 23.8% of patients, 11.1% of patients had an aneurysm, and 23.6% had a major neurological event. 26.8% had been treated for anxiety or depression. Mean SF-36 scores were: PCS: 42.5 ± 11.5 and MCS: 48.1 ± 11.0, both of which were reduced compared to norms for the healthy and general US population but similar to a cohort of women seen in preventive cardiology clinic (PCS 42.0, MCS 48.2). Significant risk factors for reduced PCS among female FMD patients were headaches, neck pain, abdominal pain, and diabetes and for MCS were aneurysm and history of anxiety or depression.

Conclusions: Female FMD patients presenting to a referral center had decreased HRQOL compared to the general and healthy US population, though scores were surprisingly preserved given the high prevalence of symptoms and vascular events. Chronic pain (headache, neck pain, abdominal pain) was associated with impaired HRQOL, suggesting that pain control is an important goal of care for FMD patients.

Incidence of severe renal dysfunction among individuals taking warfarin: Potential implications for new oral anticoagulants

Deirdre Mooney, Christina Cove, Andrew Cowan, Darae Ko, Lori Henault, Yorghos Tripodis, Elaine Hylek

Boston Medical Center, Boston, MA, United States

Background: The lack of knowledge about incidence, severity and risk factors for severe renal dysfunction in patients requiring oral anticoagulation impedes development of strategies to mitigate risks of hemorrhage associated with renally cleared novel oral anticoagulants.

Methods: Patients taking warfarin for AF or VTE were consecutively enrolled 1/2007 – 12/2010. Eligible patients had ≥ 2 creatinine values during this period, with baseline glomelular filtration rates > 30 ml/min (GFR) as estimated by Cockcroft-Gault calculation. Censoring occurred with GFR < 30.

Results: Of 696 patients identified, 21 were excluded for baseline GFR < 30. At baseline, 51% had chronic kidney disease (CKD), 75% hypertension, 33% heart failure, 32% diabetes and 23% coronary artery disease (CAD). Severe renal dysfunction (GFR < 30) occurred in 64 patients (9.5%), occurring within six months in 25%. It occurred in 35% of those with GFR 30–59 compared to only 3% of those with GFR 60–89 or 90+ (Table). In multivariable models, baseline GFR 30–59 conferred 21-fold increase in risk of severe renal dysfunction (OR 20.9, 95% CI 7.7 to 56.9) and CAD a two-fold increased risk (OR 2.0, 95% CI 1.1 to 3.7). Adjusting for baseline renal function and CAD, age, HTN, CHF and prior stroke were not independently associated with severe renal impairment.

Conclusions: Acute and chronic renal dysfunction is common among individuals on chronic warfarin therapy. Better understanding of the fluctuations in renal function would inform patient selection and monitoring strategies for optimal use of novel anticoagulants.

OPEN IN VIEWER

Table.

	Frequency of GFR < 30 (n = 64)	Frequency of no GFR < 30 (n = 611)	p-value
Renal function*			< 0.0001
GFR 30–59	47 (35%)	87 (65%)
GFR 60–89	7 (3%)	207 (97%)
GFR 90+	10 (3%)	317 (97%)
Race			0.3597
White	23 (8%)	253 (92%)
Black	32 (11%)	268 (89%)
Hispanic	8 (12%)	57 (88%)
Female	31 (10%)	269 (90%)	0.4992
Age
< 65	17 (5%)	309 (95%)
65–74	12 (7%)	171 (93%)
75+	35 (21%)	131 (79%)
Hypertension	56 (11%)	451 (89%)	< 0.0001
CAD	28 (18%)	130 (82%)	< 0.0001
Diabetes	22 (10%)	195 (90%)	0.6885
Heart failure	33 (15%)	193 (85%)	0.0013
Prior stroke	5 (5%)	88 (95%)	0.1456

*

Cockcroft-Gault calculation.

Influence of peripheral artery disease and statin therapy on apolipoprotein profiles

Omar L Esponda¹, Petar Alaupovic², Donald E Parker³, Polly S Montgomery⁴, Ana I Casanegra¹, Andrew W Gardner⁵

¹University of Oklahoma Health Sciences Center, Oklahoma City, OK, United States; ²Oklahoma Medical Research Foundation, Oklahoma City, OK, United States; ³College of Public Health, OUHSC, Oklahoma City, OK, United States; ⁴Reynolds Oklahoma Center on Aging, OUHSC, Oklahoma City, OK, United States; ⁵Reynolds Oklahoma Center on Aging, OUHSC & Veterans Affairs Medical Center, Oklahoma City, OK, United States

Background: Apolipoprotein B is a stronger predictor of myocardial infarction than LDL-Cholesterol, and it is inversely related to physical activity and modifiable with exercise training. As such, apolipoprotein measures may be of particular relevance for subjects with PAD and claudication. We compared plasma apolipoprotein profiles in 29 subjects with peripheral artery disease (PAD) and intermittent claudication and in 39 control subjects. Furthermore, we compared the plasma apolipoprotein profiles of subjects with PAD either treated (n = 17) or untreated (n = 12) with statin medications.

Methods: Subjects were assessed on plasma apolipoproteins, lipids, medical history, and a physical examination.

Results: After adjusting for age, subjects with PAD had higher triglycerides (p < 0.01), LDL-cholesterol (p < 0.01), total cholesterol (p < 0.05), and lower HDL-cholesterol (p < 0.05) than the controls. For the apolipoprotein sub-particle analyses, subjects with PAD had higher age-adjusted LpB:C (p < 0.05) and lower values of LpA1A2 (p < 0.05) than controls. The PAD group taking statins had lower age-adjusted values for apoB (p < 0.05), LpA2:B (p < 0.05), LpB:E:CE (p < 0.05), LpB:C (p < 0.05), and LpA1 (p < 0.05) than the untreated PAD group.

Conclusions: Subjects with PAD have worse lipid profiles and impaired apolipoprotein profiles than controls, characterized by LpB:C and LpA1A2. Furthermore, PAD subjects on statin medications have a more favorable risk profile, particularly noted in multiple apolipoprotein sub-particles. The efficacy of statin therapy to improve cardiovascular risk appears more evident in the apolipoprotein sub-particle profile than in the more traditional lipid profile.

Diurnal and day-to-day variability of flow-mediated dilation

Umberto Campia, Yihua Liao, Kiang Liu, Mary McDermott, Douglas Vaughan, Donald Lloyd-Jones

Northwestern University, Chicago, IL, United States

Background: The measurement of flow-mediated dilation (FMD) of the brachial artery is often used to assess the effects of interventions on endothelial function. Determination of its variability is crucial for the design and interpretation of the findings of those investigations.

Methods: Thirty-one healthy adults participated in the study. FMD and NMD were assessed with high-resolution ultrasound. Occlusion cuff was placed on the upper arm and inflated for 5 minutes. Nitroglycerin 0.4 mg was given sublingually. FMD and nitroglycerin-mediated dilation were measured on three different time points: two in the same day between 7:00 and 8:30 and between 12:30 and 14:00; and one in another day between 7:00 and 8:30. T test was used to compare FMD and NMD between different time points. Variability of FMD and NMD was assessed by coefficient of variation (CV) between time points on the same day and on different days.

Results: Mean age of participants was 53±4.6. SBP and DBP were 128±13 and 80±7, respectively. BMI was 26.7±4.3. Total cholesterol was 189; LDL 113; HDL 53, and triglycerides 114. Mean FMD and NMD vales, t test comparisons and CVs are reported in the Table.

Conclusions: In healthy middle-aged adults, mean FMD and NMD do not differ between time points on the same day and on different days. However, FMD and, to a lesser extent, NMD appear to have moderate within-subject diurnal and day-to-day variability as assessed by CV. This variability should be considered in the design and interpretation of studies assessing the effects of interventions on endothelial function.

OPEN IN VIEWER

Table.

T test and coefficient of variation for Diurnal and day-to-day variation of FMD and NMD

	#	Day 1 Morning	Day 1 afternoon	Difference	T test P value	CV (%)
			% Increase
Same day FMD	30	9.14 ± 4.7	8.95 ± 4.14	0.19 ± 3.8	0.78	29.45
Same day NMD	28	21.37 ± 7.92	20.69 ± 6.54	0.69 ± 5.98	0.55	19.89
		Day 2 Morning	Day 2 afternoon	Difference	T test P value	CV (%)
			% Increase
Different day FMD	29	9.28 ± 4.70	9.37 ± 4.83	0.09 ± 4.19	0.91	31.46
Different day NMD	28	21.38 ± 7.92	21.78 ± 6.62	0.40 ± 6.3	0.74	20.27

A hybrid genomic and proteomic biomarker panel accurately predicts peripheral arterial disease status

Kevin T Nead¹, Kelly Putnam¹, John P Cooke¹, Jeffrey W Olin², Nicholas J Leeper¹

¹Stanford University, Stanford, CA, United States; ²Mount Sinai School of Medicine, New York, NY, United States

Background: Peripheral arterial disease (PAD) is a highly morbid condition affecting more than eight million Americans. Frequently, PAD patients are unrecognized and therefore do not receive appropriate therapies. Therefore, new methods to identify PAD have been pursued, but have thus far had only modest success. Here we describe a new approach combining genomic and proteomic information to enhance the diagnosis of PAD.

Methods and Results: We measured the genotype of the Chromosome 9p21 cardiovascular-risk polymorphism, rs10757269, as well as the proteomic biomarkers, C-reactive protein, Cystatin C and Beta-2-microglobulin in 389 patients undergoing coronary angiography. The rs10757269 allele was associated with PAD status (ankle-brachial index < 0.9) independent of proteomic biomarkers and traditional cardiovascular risk factors (odds ratio = 1.92; 95% confidence interval, 1.29–2.85). Importantly, compared to a model previously validated to predict PAD including smoking status, history of congestive heart failure, body mass index, age, gender, history of cerebrovascular disease, race, diabetes, history of coronary artery disease and hypertension state, the addition of proteomic biomarkers and rs10757269 significantly improved PAD risk prediction as assessed by the Net Reclassification Index (NRI, p = 0.001) and Integrated Discrimination Improvement (IDI, p = 0.017).

Conclusions: A model including a panel of biomarkers, which includes both genomic information (which is reflective of heritable risk) and proteomic information (which integrates environmental exposures), predicts the presence or absence of PAD better than established risk models, suggesting clinical utility for the diagnosis of PAD.

Cardiovascular risk assessment in operational firefighters

Elizabeth V Ratchford¹, Kathryn A Carson², M Dominique Ashen¹

¹Johns Hopkins University School of Medicine, Baltimore, MD, United States; ²Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States

Background: Cardiovascular disease (CVD) accounts for 45% of deaths among on-duty firefighters, in contrast to 15% of all deaths on conventional jobs. Early detection may prevent premature death and disability.

Methods: This prospective cohort study included 50 firefighters without known CVD or diabetes. Risk was assessed via coronary artery calcium (CAC) and carotid intima-medial thickness (cIMT) scans.

Results: Mean age was 47 (range 40 to 58), 90% were men, 92% were Caucasian, and 30% were former smokers. Only 4% were hypertensive; 48% were pre-hypertensive. Mean BMI was 30.4. Mean waist circumference (WC) was 39.2±4.9 inches. Only 14% had a normal BMI; 38% were overweight; 48% were obese; and 46% had an abnormal WC. Based on fasting glucose ≥ 100 mg/dL or hemoglobin A1C ≥ 5.6%, 52% had pre-diabetes (50%) or diabetes (2%). Mean total cholesterol was 200.3±44.13 mg/dL; mean hs-CRP was 1.66±2.10 mg/L. CAC was detected in 22%. Carotid plaque was detected in 36%. Using standard reference databases, 54% had cIMT > 75th percentile; 66% had carotid plaque and/or cIMT > 75th percentile as shown in the Figure. CAC was associated only with former smoking, which was marginally significant (p = 0.06, Fisher’s exact test). cIMT > 75th percentile was significantly associated with increased WC (p = 0.02), BMI (p = 0.02), total cholesterol (p = 0.048), and LDL (p = 0.01). More incidental findings were seen on CAC scans (58%) versus cIMT (26%). More participants will require follow-up for lung nodules (30%) versus thyroid abnormalities (10%).

Conclusion: Obesity and pre-diabetes are major issues facing firefighters, which could in part explain their elevated CVD risk. cIMT may be more informative than CAC in this young but high-risk population, with the added advantage of fewer incidental findings.

OPEN IN VIEWER Figure

Limb volume reduction utilizing advanced pneumatic compression treatment in the home

Satish Muluk, Elise Taffe

Allegheny General Hospital, Pittsburgh, PA, United States

Background: Pneumatic compression devices (PCDs) are widely used to treat chronic lymphedema. Postulated mechanisms include decreased capillary filtration, reduced venous reflux, and aiding the return of fluid to the circulatory system. However, this therapy has not been well studied. Our objectives were to examine the effectiveness of an advanced PCD (APCD) in reducing limb volume (LV) on patients of varying diagnoses, age and BMI. Clinician and patient reported outcomes were also evaluated.

Methods: A retrospective review of prospectively-collected data was conducted using the clinical records of 329 limbs among 197 lower extremity lymphedema patients who were prescribed an APCD (Table). Pre- and post-treatment limb measurements were recorded along with clinical outcomes, including skin changes, pain and function. Patient reported outcomes and satisfaction via a follow-up survey were also obtained.

Results: 90% of patients demonstrated LV reduction, and 38% had reduction >10%. Mean LV reduction across all patient groups was 1132.1 cc (median 789.9 cc), or 8.3% of pre-treatment LV (p < 0.0001). Larger BMI was a statistically significant predictor of LV reduction (p < 0.0001). Neither age nor etiology of lymphedema was a predictor of LV reduction. The majority of patients experienced a reduction in skin fibrosis and increase in function. Patient-reported outcomes showed a statistically significant increase in ability to control swelling at home and a decrease in the interference of pain. 66% of patients were ‘Very Satisfied’ with the APCD.

Conclusion: Results suggest that use of an APCD may be clinically effective when used in the home as measured by LV reduction, functional improvement, and patient reported treatment satisfaction.

Malignancy in patients with premature peripheral artery disease (PAD)

Tiffany Lin, Jeanette S Andrews, Matthew Edwards, Pavel J Levy

Wake Forest School of Medicine, Winston-Salem, NC, United States

Background: In younger individuals, PAD with premature lower extremity atherosclerosis (PLEA) is rare and is associated with poor prognosis. Atherosclerosis and cancer share multiple risk factors and mechanisms in their development and progression. We studied clinical characteristics in PLEA patients (pts) with malignancies (PLEA/+mal).

Methods: A retrospective, case-control study in PLEA pts ≤ 55 yrs of age (mean 49.4±6.9) with severe PAD treated between 1998–2010. PLEA/+mal pts had documented malignancy either at first evaluation, during follow-up, or as cause of death. Multivariable logistic regression modeling was used for associations with malignancy.

Results: Among 559 PLEA pts (47% female, 80% white), 85 (15%; 51% female) had malignancies diagnosed either before (‘early’; n = 42), or after (‘late’; n = 43) PAD. PLEA/+mal pts when compared to pts without malignancy, were relatively older at PAD diagnosis, had lower prevalence of hypertension (52% vs. 66%; p=0.011), diabetes (17% vs. 27; p = 0.038), lower amputation rate (4% vs. 11%; p < 0.045). No differences were found in family history of early cardiovascular disease, or malignancies, or anatomic distribution of PLEA. Hypertension was negative predictor of malignancy by multivariable analysis (OR = 0.53; p = 0.032). 5-year survival was significantly lower in PLEA/+mal pts (74% vs. 86%; p < 0.01). Among PLEA/+mal pts, those with ‘early’ compared with ‘late’ malignancies more frequently were female (67% vs. 35%; p < 0.003), had history of VTE (17% vs. 2%; p = 0.030) or aorto-iliac disease (p < 0.006).

Conclusions: Prevalence of malignancies in PLEA pts is higher than that of national statistics. PLEA/+mal pts had lower frequencies of traditional risks, suggesting that younger PLEA/+mal pts may have unique ‘proatherogenic’ mechanisms.

Renal artery stent fracture in patients with fibromuscular dysplasia: A cautionary tale

Manjunath Raju, Christopher Bajzer, Daniel Clair, Esther Kim, Heather Gornik

Cleveland Clinic Foundation, Cleveland, OH, United States

Background: Fibromuscular dysplasia commonly affects the renal arteries, and hypertension is one of the most common manifestations. We report a series of two patients with renal artery FMD who developed stent fracture following renal artery angioplasty and stenting.

Case 1: A 16-year-old girl was diagnosed with hypertension at age 13 and initially treated with medications. She underwent balloon angioplasty and ultimately placement of a drug-eluting stent in the right renal artery for uncontrolled hypertension. She was further referred to our FMD clinic. Renal arteriography revealed severe stenosis (estimated at 80%) in the right renal artery stent and intravascular ultrasound (IVUS) confirmed stent fracture. The patient underwent aorto-renal bypass grafting with venous conduit and did well postoperatively.

Case 2: A 41-year-old woman was seen for a second opinion regarding renal FMD and multi-drug resistant hypertension. She had previously donated her left kidney to her sister. Angioplasty and multiple stents were placed across a long segment of the right renal artery. She self-referred to our FMD clinic for worsening blood pressure control on six drugs. Angiography demonstrated a long area of stenting across the right renal artery with restenosis and stent fracture. Patient underwent aorto-renal bypass grafting.

Conclusion: In our two patients, stent fracture requiring surgical bypass likely resulted from the combination of a highly mobile right kidney and metal fatigue along a long area of stent deployment. Such long segment stenting should generally be avoided in FMD patients. Surgical revascularization remains a viable therapeutic option for patients with renal FMD with extensive disease not amenable to angioplasty.

Do behavior-change techniques increase walking in patients with intermittent claudication? A systematic review

Melissa N Galea, John A Weinman, Claire White, Lindsay M Bearne

King’s College London, London, United Kingdom

Background: Walking is a key component to treatment for people with intermittent claudication (IC). However, availability of supervised exercise programs is limited and most patients do not follow advice to walk. Therefore, people with IC do not maximize the benefits of walking. This systematic review identified and evaluated strategies to improve adherence to walking among people with IC.

Methods: An electronic database search (Medline, PsychINFO, Embase, CINAHL, Web of Science) was conducted up to December 2012. Titles and abstracts were reviewed by two independent reviewers and randomized controlled trials comparing interventions incorporating behavior-change techniques with walking advice or exercise were included. Quality appraisal was conducted using the Cochrane Collaboration risk of bias tool. The primary outcome was maximal walking ability (MWA) ≥ 3 months after the start of an intervention. Secondary outcome variables included pain-free walking ability (PFWA), self-report walking ability (SRWA) and daily walking activity.

Results: A total of 3575 records were retrieved. Of these, six trials met the inclusion criteria. Due to substantial heterogeneity between studies, a meta-analysis was not conducted. Two high-quality trials found that behavior-change techniques were successful in increasing MWA and PFWA versus usual care. The evidence from four low-quality trials was conflicting and data on SRWA and daily walking activity was inconclusive. Eleven behavior-change techniques were identified: helping people identify and overcome barriers to walking, facilitating self-monitoring, and providing feedback on walking performance were most frequently reported.

Conclusions: There is a limited but growing body of evidence examining strategies that can be applied to support adherence to walking for people with IC. Rigorous, fully-powered trials are required to isolate the effect of specific behavior-change techniques on walking, beyond the effects of exercise therapy.

Patterns of medication use in 615 patients with fibromuscular dysplasia. A report of the United States Registry for Fibromuscular Dysplasia

Ido Weinberg¹, Xiaokui Gu², Jay Giri³, Jeffery W Olin⁴, James Froehlich⁵, Heather L Gornik⁶, Kevin R Rogers⁷, Michael R Jaff¹

¹The Institute for Heart, Vascular and Stroke Care, Massachusetts General Hospital, Boston, MA, United States; ²Cardiovascular Center, University of Michigan, Ann Arbor, MI, United States; ³Cardiology Division, University of Pennsylvania, Philadelphia, PA, United States; ⁴Zena and Michael A Wiener Cardiovascular Institute and Marie-Josée and Henry R Kravis Center for Cardiovascular Health Icahn School of Medicine at Mount Sinai New York, NY, United States; ⁵Cardiovascular Center, University of Michigan, Ann Arbor, MI, United States; ⁶Cleveland Clinic Heart and Vascular Institute and the Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH, United States; ⁷Cardiology Division, University of Colorado, Denver, CO, United States

Background: Fibromuscular dysplasia (FMD), a noninflammatory arterial disease, may lead to renovascular hypertension (HTN) and extracranial cerebrovascular (CV) disease. Little is known about medication use among patients with FMD.

Methods: Clinical features, presenting symptoms, vascular events and medication use were reviewed for patients in a national FMD registry (ten U.S. sites). Logistic regression analyses were performed to investigate covariate associations and medication use. Covariates included demographic characteristics, co-morbid conditions and vascular bed involvement.

Results: All 615 patients (91.5% female) in the database were included in the analysis. Mean age was 55.5±12.9 years, 440/597 (73.7%) had HTN and 40/521 (7.7%) had a history of coronary artery disease (CAD). Renal and CV artery involvement were 382/507 (75.3%) and 366/490 (74.7%), respectively. Antiplatelet (AP) agents were administered to 260/362 (71.8%) and 286/355 (80.6%) of patients with renal or CV involvement, respectively. The factors associated with greater AP agent use were older age (OR = 1.022 per year, p = 0.006) and CV artery FMD; in isolation (OR = 2.9, p = 0.005) or with renal artery involvement (OR = 2.2, p = 0.029). Anti-HTN medication use differed between vascular beds (Figure). Factors associated with greater anti-HTN use were older age (OR = 1.02 per year, p = 0.0022), history of HTN (OR = 15.1, p < 0.0001) and history of CAD (OR = 2.5, p = 0.014).

Conclusions: Most patients with FMD receive AP agents. The use of anti-HTN agents differs by vascular bed involvement, with very few patients with renal artery FMD requiring no anti-HTN agents, while these patients require > 3 medications twice as often as patients with CV FMD.

OPEN IN VIEWER Figure

An expanded cohort of individuals with critical limb ischemia: Patients not included in a prospective registry

Sue Duval¹, Wobo Bekwelem¹, Lindsay G Smith¹, Hong H Keo², Tamara J Winden³, Niki Oldenburg¹, Alan T Hirsch¹

¹University of Minnesota, Minneapolis, MN, United States; ²Kantonsspital, Aarau, Switzerland; ³Allina Health, Minneapolis, MN, United States

Background: Critical limb ischemia (CLI) contributes to > 150,000 ischemic amputations annually in the United States. Yet, methods to identify the CLI population who might benefit from revascularization, risk reduction therapies, or who might be enrolled in clinical trials has never been prospectively defined. The FRIENDS (‘FReedom from Ischemic Events – New Dimensions for Survival’) registry was designed to prospectively enroll sequential CLI patients in order to evaluate health system predictors of adverse outcomes. Whether enrolled registry patients were representative of the larger hospital catchment area was not known.

Methods: Vascular specialist identified CLI patients (n = 126) were enrolled into the registry over 35 months (February 2007 – December 2009). Using administrative claims data from the same institution, registry cases were age- and sex-matched 2:1 to controls (based on the electronic problem list). Billing codes for all patients were extracted. Six case-finding algorithms were defined on the basis of ICD-9 diagnosis codes for CLI, diabetes, and procedure codes for CLI. The algorithms were applied to the claims database during the same time period. Demographic, risk factors and sites of care were compared for registry-enrolled (RE) vs. non-enrolled (NE) subjects.

Results: The registry enrolled 4% of all subjects with presumed CLI according to the most inclusive algorithm (CLI diagnosis or CLI procedure) (Table). The registry CLI cohort was provided care at three key care sites, had more medical co-morbidities, and more health system care encounters.

Conclusions: Despite efforts to enroll all sequential CLI patients into an outcomes registry, there were 25-times more patients with CLI receiving care in this hospital. This larger CLI cohort represents a population that could potentially be identified in real time via administrative database methods, for provision of CLI quality improvement efforts or enrollment in CLI clinical trials.

OPEN IN VIEWER

Table.

Algorithm based on ICD-9 codes	Se	Sp	RE (n)	NE (n)	% capture
CLI diagnosis	0.75	1.0	64	1413	4.3
CLI diagnosis and CLI procedure	0.59	1.0	6	45	11.8
CLI diagnosis and diabetes diagnosis	0.33	1.0	10	31	24.4
CLI diagnosis and diabetes diagnosis and CLI procedure	0.29	1.0	14	32	30.4
CLI diagnosis or CLI procedure	0.92	0.99	105	2839	3.6
CLI diagnosis and (diabetes diagnosis or CLI procedure)	0.8	0.99	50	194	20.5

Management of intracranial bleeding associated with dabigatran use in a neuroscience hospital

Alejandro Perez, Lynda Thompson, Geno J Merli

Thomas Jefferson University Hospital, Philadelphia, PA, United States

Background: Dabigatran, an alternative to warfarin for prevention of stroke with non-valvular atrial fibrillation (AF), offers advantages of a fixed dosage, minimal laboratory monitoring and limited medication interactions. Dabigatran requires dosage adjustment in renal dysfunction and is contraindicated if severe dysfunction. No identified dabigatran reversal agent exists.

Methods: As part of an ongoing quality initiative, novel anticoagulant associated adverse events (AE) are monitored at a dedicated neuroscience hospital.

Results: 5 cases of intracranial bleeding associated with dabigatran occurred from 12/2011–4/2012. All patients were on anticoagulation for AF, the most common dose being 150 mg BID. Mean admission values were as follows: age 83.2 (range 79–90), serum creatinine 1.48 mg/dL (range 0.9–3.5), creatinine clearance 45.6 mL/min (18–59) and aPTT 49 seconds (range 32–60) (Table). Strategies for the management of bleeding included withholding dabigatran, supportive care, administration of blood products and hemodialysis, when required. Dialysis was initiated on 3 patients. One patient had 3 dialysis sessions in an effort to normalize coagulation assays and had transfusions with 10 units of platelets and 4 units of fresh frozen plasma in an effort to stabilize bleeding. One patient died. Mean time for aPTT to normalize when abnormal on admission was 30.8 hours (range 21–37).

Conclusions: Appropriate patient selection is required to prevent dabigatran associated AE, especially in the setting of advanced age and kidney dysfunction. aPTT values may remain prolonged for extended periods, despite efforts to normalize. Hospitals need a defined management plan for major bleeding associated with novel anticoagulants.

OPEN IN VIEWER

Table.

Characteristics of patients presenting to neuroscience hospital with intracranial bleeding associated with dabigatran

Patient	Age, Gender (M/F)	Dabigatran dose	Creatinine (mg/dL)/ Creatinine Clearance (mL/min)	Admission PT (seconds)/ INR	Admission aPTT (seconds)	Elevated aPTT duration (hours)	Admission Hemoglobin/ Hematocrit/ Platelets	Dialysis During admission (yes/ No)
1	79, M	75 mg BID	3.5/18	17.5/1.48	54	37	9.8/28.9/174	Yes
2	84, F	150 mg BID	0.7/49	18.1/1.54	60	37	11.5/34.5/253	No
3	80, F	150 mg BID	0.9/54	16.8/1.4	60	28	14.8/43.3/240	Yes
4	83, M (expired)	Unknown	1.4/48	15.9/1.3	32	0	12.8/39.5/192	No
5	90, F	150 mg BID	0.9/59	21.8/1.97	39	21	12.6/39.5/267	Yes

Factors predicting bleeding in patients undergoing perioperative bridging therapy in a tertiary care hospital in Singapore

Nihar Pandit, Veerendra Chadachan, Roland Boey, Theressa Choo, Jackie Tan, Jam Chin Tay

Tan Tock Seng Hospital, Singapore, Singapore

Background: Perioperative management of surgical patients who are on long-term anticoagulation require bridging therapy with heparin after temporary discontinuation of Warfarin. Bridging therapy is associated with significant increase in post-operative bleeding. To date there is paucity of evidence based recommendations and hence practice is based on expert guidelines. We performed a retrospective analysis to identify factors associated with increased risk of bleeding in the perioperative setting.

Methods: This retrospective study amongst patients who attended Anti Coagulation Clinic for perioperative bridging looked at the postoperative bleeding risk in patients given bridging therapy. Data were collected for reason for anticoagulation, nature of surgery, number of days off Warfarin and heparin administered, postoperative day on which anticoagulation was restarted and incidence and nature of bleeding in the immediate postoperative period.

Results: 54 patients (mean age 67 years), were included, of which 45 had Warfarin interrupted for procedures associated with a high bleeding risk. Major bleeding was seen in ten patients (eight males, two females), of which four each (all males) had abdominal and urological procedures. Most patients had bleeding in the operated cavity, wound hematomas and drop in Hemoglobin (> 2 gm%). Three patients required re-exploration of the operated site.

Conclusion: The risk of major postoperative bleeding associated with bridging therapy is significant and dependent on male sex and nature of surgery. Bleeding in the operated cavity, wound hematomas and hematuria (with urological procedures) are the most commonly encountered major bleeding problems and all were associated with a drop in hemoglobin of > 2 gm%.

Prevalence and risk factors for present on admission deep venous thrombosis among patients with cardiovascular disease transferred to a tertiary care center

Andrew T Strong, Jeevanantham Rajeswaran, Aditya Sharma, John R Bartholomew, Eugene Blackstone, Moses Anabila, Daniel G Clair, Umesh N Khot, Pamela Goepfarth, Bruce W Lytle, Steven E Nissen, Shannon Connor Phillips, Neil Poria, Joseph F Sabik, Lars G Svensson, Heather L Gornik

Cleveland Clinic, Cleveland, OH, United States

Background: Because venous thromboembolism (VTE) is a leading cause of preventable death among hospitalized patients, regulatory agencies recognize VTE as a key quality indicator. While incidence of deep venous thrombosis (DVT) among inpatients has been studied previously, the prevalence of and risk factors for present on admission (POA) DVT among patients transferred to tertiary care centers have not been established.

Methods: Consecutive patients transferred to the Heart and Vascular Institute underwent clinical assessment of DVT risk and lower extremity venous duplex ultrasound (LE DUS) within 48 hours of transfer. Patients with increased clinical risk for upper extremity (UE) DVT underwent arm DUS. Data from 2011 cohort have been previously shown; we present data from a validation cohort (10/22–11/21/2012) and analysis of POA DVT risk factors using pooled data.

Results: During the study period, 750 patients were admitted, of which 625 (83%) were evaluated. Patient characteristics, VTE risk factors and DUS findings for the 2011, 2012, and pooled cohorts are shown in the Table. In the pooled dataset, 10% of patients were found to have POA DVT; prevalence of POA LE DVT was 8.4%, of which 62% were isolated calf vein thrombi, and UE DVT was present in 9.8% of patients imaged. Risk factors for POA LE DVT were age, history of LE DVT, length of outside hospital stay > 4 days, bedridden status > 3 days or recent surgery, and Well’s score > 3 (C-statistic = 0.81).

Conclusion: The prevalence of POA DVT among patients with cardiovascular disease transferred to a tertiary care center is higher than VTE-related quality indicator targets. Facilities whose venous duplex scanning protocols include calf evaluation may report higher DVT rates. In light of these findings, algorithms for VTE-related quality metrics should be re-designed to accurately reflect hospital-acquired DVT; furthermore clinical prediction tools, risk factors and surveillance standards for POA DVT need to be established.

OPEN IN VIEWER

Table.

Variable	2011	2012	Pooled Data
DEMOGRAPHICS	N = 330	N = 295	N = 625
Male	61% (201/330)	62% (182/295)	61% (383/625)
Age, yrs (mean ± sd)	64.8 ± 16.5	65 .4 ± 14	65.1 ± 15.1
Admitted to CV Medicine (vs. surgical service)†	88% (291/330)	93% (273/295)	91% (564/625)
ICU patient	40% (132/330)	38% (111/295)	39% (243/625)
Outside hospital length of stay, days (median (IQR))†	1 (3)	1 (2)	1 (3)
PRETEST RISK ASSESSMENT
History of DVT/PE†	15% (49/330)	8.5% (25/295)	12% (74/625)
Length of outside hospital stay > 4 days†	24% (77/330)	19% (54/295)	22% (131/625)
Calculated LE Modified Well’s score > 3	12% (36/330)	9.1% (24/295)	11% (60/625)
Calculated UE risk score > 1†	23% (68/330)	33% (93/295)	28% (161/625)
ULTRASOUND FINDINGS
Lower extremity duplex performed	100% (330/330)	100% (295/295)	100% (625/625)
POA LE DVT (proximal or calf)	10% (33/330)	6.8% (20/295)	8.4% (53/625)
Proximal leg POA DVT (± calf)	4.5% (15/330)	1.7% (5/295)	3.2% (20/625)
Calf only POA DVT	5.5% (18/330)	5.1% (15/295)	5.3% (33/625)
Upper extremity venous duplex ultrasound	22% assessed (71/330)	35% assessed (103/295)	28% assessed (174/625)
POA UE DVT†	17% (12/71)	4.9% (5/103)	9.8% (17/174)
DVT POA (UE or LE)	13% (42/330)	7.8% (23/295)	10% (65/625)

†

p < 0.05 for comparison between 2011 and 2012 cohort.

Endovascular therapy for lower extremity atherosclerosis

Ryan O Lakin¹, Lina Vargas², James F Bena², Daniel G Clair²

¹University Hospitals Case Medical Center, Cleveland, OH, United States; ²Cleveland Clinic, Cleveland, OH, United States

The purpose of this study was to review outcomes of percutaneous endovascular therapy (ET) for lower extremity, infrainguinal, occlusive disease (LEOD). Between 1/2005 and 12/2006, over 500 patients with a clinical diagnosis of LEOD underwent arteriography at the Cleveland Clinic in Cleveland, Ohio. Of these, 294 limbs in 245 patients (mean age 70 ± 12 years; 53% male) were treated for claudication (27%), rest pain (14%), and tissue loss (59%) with angioplasty and adjunctive stenting. Electronic medical records were reviewed for demographic data, clinical variables, and outcomes (Mean follow-up, 41.3 ± 19 months). Kaplan-Meier estimates and Cox proportional hazard univariate and multivariate models were calculated. Patients (n = 245) with significant co-morbidities (coronary disease 64%, diabetes 51%, smoking 68%) underwent ET for LEOD under local anesthesia/sedation (98%). Thirty-day morbidity (amputation 4%, MI/pulmonary failure/renal dysfunction 1.2%) and mortality (2%) rates were low. Symptoms improved in patients presenting with claudication (88%), rest pain (80%), and tissue loss (69%). Overall primary patency (via vascular lab testing, 78%) at three years was 37% (30, 44; 95% CI), secondary patency 51% (45, 59), and limb salvage 86% (80, 91). African-American patients had diminished patency (HR 1.8, p = 0.002) and limb salvage (HR 3.5, p < 0.001). Patients with tissue loss had lower limb salvage rates than patients with rest pain (78 vs. 94%, HR 16, p = 0.007). Despite patency, patients on dialysis (n = 36) had dismal limb salvage rates at three years (48%, HR 4.3, p < 0.001). Overall three year survival was 69% (64, 75) (Figure). Multivariable analyses indicated that patients presenting with tissue loss (HR 4; 2, 8.6, p < 0.001) and on dialysis (HR 4.1; 2.4, 7.1, p < 0.001) were associated with lower survival. Endovascular treatment of LEOD is safe. Despite poor durability, especially for African-American patients, ET improves symptoms and prevents limb loss for most patients. Renal failure negatively affects limb preservation and survival.

OPEN IN VIEWER Figure

Outcome of stenting in the subclavian vein for upper extremity edema

Mohsen Sharifi¹, Stacey Klyn², Curt Bay², Neema Mafi¹, Herby Ambroise¹

¹Arizona Cardiovascular Consultants, Mesa, AZ, United States; ²A.T. Still University, Mesa, AZ, United States

Background: Symptomatic unilateral upper extremity (UE) venous edema is frequently associated with veno-occlusive disease. There is usually some degree of deep venous thrombosis (DVT) which is often due to flow obstruction in the more cephalad segments. Venous stenosis in the subclavian vein due to repeated insults is the main etiology. Little is known on the outcome of stenting in the subclavian vein in this setting.

Methods: Fifty patients underwent stenting in their subclavian veins as part of percutaneous endovenous intervention for symptomatic unilateral UE edema. All had been diagnosed with ‘DVT’ on duplex imaging. A total of 62 stents were placed. Access was obtained through the brachial vein under ultrasound guidance. All patients were maintained on aspirin for a minimum of six months plus warfarin (20 patients) or dabigatran. Follow-up venous duplex was performed at one month and every six months. The mean duration of follow-up was 22 ± 8 months.

Results: Substantial improvement in UE symptoms was achieved with stenting of the SCV. However some degree of edema persisted in the affected UE in 32 patients even at long term follow-up. Patency of the stents was observed in all patients although stagnation of flow was noted at non stented sites due to venosclerosis. Almost always a diffuse pattern of venosclerosis was present. This occurrence did not lead to clinical deterioration. Venography at the time of stenting demonstrated some degree of thrombus but the main etiology was venous stenosis rather than acute DVT. This was in sharp contrast to the lower extremities in which in our experience, acute DVT has been far more common.

Conclusions: Unilateral UE edema is usually due to venous stenosis with some degree of DVT due to repeated insults to the central veins especially the SCV. Stenting in the SCV is an effective and safe approach which considerably improves symptoms. Due to presence of diffuse stenosis outside the stented segments, a component of persisting DVT exists which requires chronic anticoagulation.

Correlation between clinical severity and peak systolic velocity at the sapheno-femoral junction

Leslie R Asuncion, Jenny L Beltran

St Luke’s Medical Center, Quezon City, Philippines

Background: Primary venous insufficiency may lead to venous hypertension that causes debilitating symptoms. The goal of every clinician is to determine severity of the disease and how it is correlated with the results of the diagnostic examination requested. Hence, the aim of this study is to establish correlation between the severity of clinical category of CEAP (Clinical, Etiologic, Anatomical and Pathophysiological) classification and the sapheno-femoral junction (SFJ) reflux peak systolic velocity (PSV), SFJ reflux time, SFJ diameter and greater saphenous vein (GSV) diameter.

Methods: This is a cross-sectional prospective study involving 161 limbs in 101 patients. Limbs were categorized according to the clinical severity of the CEAP classification. Doppler ultrasound was done to assess venous reflux, PSV and reflux time of the SFJ. The GSV diameters were measured in standing position at the junction and at the proximal segments. Subjects with deep vein thrombosis (DVT) or history of DVT were excluded in the study.

Results: One hundred sixty-one (161) lower limbs with SFJ reflux in 101 patients were included in the study. The mean age of the study population was 51 ± 13.381. Majority of them were females (74.8%). Increasing age, weight and presence of hypertension were significantly associated with CEAP 4–6 with p-value of 0.000, 0.017 and 0.047 respectively. The presence of SFJ PSV of > 10 cm/sec, SFJ reflux time of ≥ 1.11 sec, SFJ diameter of ≥ 0.75 mm or GSV diameter of ≥ 0.65 mm showed statistically significant correlation with CEAP 4–6 (p-value 0.01, 0.008, 0.001 and 0.002, respectively). A scoring system was developed using the results of multivariate analysis. A cut off score of 3.5 and above was determined using the ROC curve and a score ≥ 3.5 has the odds of > 8 times of developing CEAP 4–6. The vein score was noted to have higher negative predictive value (92.9%) than positive predictive value (29.5%).

Conclusion: Increasing SFJ PSV, reflux time, diameter and GSV diameter demonstrated strong correlation with CEAP 4–6.

National Health Survey on Chronic Venous Disease among Filipinos

Lilibeth Maravilla, David Salvador, Maria Teresa Abola

Philippine Heart Center, Quezon City, Philippines

Background: Chronic venous disease (CVD) is an under recognized condition that results in significant morbidities with considerable costs. There is scarce data on the prevalence of chronic venous disease in the Asian population.

Objective: The objectives of this study are to determine the prevalence of CVD using the Southern Tagalog Venous Insufficiency Questionnaire (STVIQ), to determine the prevalence of CVD according to age, gender, urbanization and occupation, and to determine the prevalence of claudication among those with CVD.

Methods: A nationwide community-based cross-sectional nutritional and health survey in free-living populations, aged 20 years and older, in the Philippines was conducted in 2008–2009 using the STVIQ and the Edinburgh questionnaire for claudication.

Results: Among 7212 subjects randomly surveyed, the crude overall prevalence rate of chronic venous disease is 52.5%, reticular veins (36%), varicose veins (15%), edema (1.4%), stasis dermatitis (0.1%), and healed venous ulcers (0.05%). CVD was more common in females (~60%). Reticular veins are more prevalent in the youngest age group among females, while varicose veins are more prevalent with age. About 38.6% of the general population have typical symptoms of chronic venous insufficiency while 3.5% have atypical symptoms. Urbanization and type of occupation are not important determinants of prevalence. Among all patients with typical symptoms of CVD, 5.8% also had claudication.

Conclusion: The prevalence of chronic venous disease among Filipinos is similar to data from surveys done in Caucasian populations and contemporary Asian studies. Increasing the awareness among physicians and the public will contribute to significantly diminishing the burden of chronic venous disease.