Efficacy of a Guideline-Recommended Risk-Reduction Program to Improve Cardiovascular and Limb Outcomes in Patients With Peripheral Arterial Disease | Acute Coronary Syndromes | JAMA Surgery | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Study Flow Diagram
Study Flow Diagram

PAD indicates peripheral arterial disease; SAVR, Systematic Assessment of Vascular Risk.

Figure 2.  Observed Incidence of Acute Myocardial Infarction, Ischemic Stroke, and Death
Observed Incidence of Acute Myocardial Infarction, Ischemic Stroke, and Death

The error bars represent 95% CIs. SAVR indicates Systematic Assessment of Vascular Risk.

Figure 3.  Observed Incidence of Secondary Outcomes
Observed Incidence of Secondary Outcomes

The error bars represent 95% CIs. SAVR indicates Systematic Assessment of Vascular Risk.

Table 1.  Baseline Demographic and Clinical Characteristics of the Unmatched and Matched Cohorts
Baseline Demographic and Clinical Characteristics of the Unmatched and Matched Cohorts
Table 2.  Outcomes of Patients Enrolled in the SAVR Program and Control Patients
Outcomes of Patients Enrolled in the SAVR Program and Control Patients
1.
Joosten  MM, Pai  JK, Bertoia  ML,  et al.  Associations between conventional cardiovascular risk factors and risk of peripheral artery disease in men.  JAMA. 2012;308(16):1660-1667.PubMedGoogle ScholarCrossref
2.
Selvin  E, Erlinger  TP.  Prevalence of and risk factors for peripheral arterial disease in the United States: results from the National Health and Nutrition Examination Survey, 1999-2000.  Circulation. 2004;110(6):738-743.PubMedGoogle ScholarCrossref
3.
Fowkes  FG, Murray  GD, Butcher  I,  et al; Ankle Brachial Index Collaboration.  Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: a meta-analysis.  JAMA. 2008;300(2):197-208.PubMedGoogle ScholarCrossref
4.
Fowkes  FGR, Rudan  D, Rudan  I,  et al.  Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: a systematic review and analysis.  Lancet. 2013;382(9901):1329-1340.PubMedGoogle ScholarCrossref
5.
Anderson  JL, Halperin  JL, Albert  NM,  et al.  Management of patients with peripheral artery disease (compilation of 2005 and 2011 ACCF/AHA guideline recommendations): a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.  Circulation. 2013;127(13):1425-1443.PubMedGoogle ScholarCrossref
6.
Makowsky  MJ, McAlister  FA, Galbraith  PD,  et al; Alberta Provincial Program for Outcome Assessment in Coronary Heart Disease (APPROACH) Investigators.  Lower extremity peripheral arterial disease in individuals with coronary artery disease: prognostic importance, care gaps, and impact of therapy.  Am Heart J. 2008;155(2):348-355.PubMedGoogle ScholarCrossref
7.
Bhatt  DL, Steg  PG, Ohman  EM,  et al; REACH Registry Investigators.  International prevalence, recognition, and treatment of cardiovascular risk factors in outpatients with atherothrombosis.  JAMA. 2006;295(2):180-189.PubMedGoogle ScholarCrossref
8.
Subherwal  S, Patel  MR, Kober  L,  et al.  Missed opportunities: despite improvement in use of cardioprotective medications among patients with lower-extremity peripheral artery disease, underuse remains.  Circulation. 2012;126(11):1345-1354.PubMedGoogle ScholarCrossref
9.
Hasimu  B, Li  J, Yu  J,  et al.  Evaluation of medical treatment for peripheral arterial disease in Chinese high-risk patients.  Circ J. 2007;71(1):95-99.PubMedGoogle ScholarCrossref
10.
Flu  HC, Tamsma  JT, Lindeman  JHN, Hamming  JF, Lardenoye  JHP.  A systematic review of implementation of established recommended secondary prevention measures in patients with PAOD.  Eur J Vasc Endovasc Surg. 2010;39(1):70-86.PubMedGoogle ScholarCrossref
11.
CAPRIE Steering Committee.  A randomised, blinded, trial of Clopidogrel Versus Aspirin in Patients at Risk of Ischaemic Events (CAPRIE).  Lancet. 1996;348(9038):1329-1339.PubMedGoogle ScholarCrossref
12.
Yusuf  S, Sleight  P, Pogue  J, Bosch  J, Davies  R, Dagenais  G; The Heart Outcomes Prevention Evaluation Study Investigators.  Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients.  N Engl J Med. 2000;342(3):145-153.PubMedGoogle ScholarCrossref
13.
Heart Protection Study Collaborative Group.  MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20 536 high-risk individuals: a randomised placebo-controlled trial.  Lancet. 2002;360(9326):7-22.PubMedGoogle ScholarCrossref
14.
Al-Omran  M, Verma  S, Lindsay  TF.  Suboptimal use of risk reduction therapy in peripheral arterial disease patients at a major teaching hospital.  Ann Saudi Med. 2011;31(4):371-375.PubMedGoogle ScholarCrossref
15.
Carter  AA, Gomes  T, Camacho  X, Juurlink  DN, Shah  BR, Mamdani  MM.  Risk of incident diabetes among patients treated with statins: population based study.  BMJ. 2013;346:f2610.PubMedGoogle ScholarCrossref
16.
Al-Omran  M, Mamdani  MM, Lindsay  TF, Melo  M, Juurlink  DN, Verma  S; Systematic Assessment of Vascular Risk Investigators.  Suboptimal use of statin therapy in elderly patients with atherosclerosis: a population-based study.  J Vasc Surg. 2008;48(3):607-612.PubMedGoogle ScholarCrossref
17.
Park-Wyllie  LY, Juurlink  DN, Kopp  A,  et al.  Outpatient gatifloxacin therapy and dysglycemia in older adults.  N Engl J Med. 2006;354(13):1352-1361.PubMedGoogle ScholarCrossref
18.
Al-Omran  M, Tu  JV, Johnston  KW, Mamdani  MM, Kucey  DS.  Outcome of revascularization procedures for peripheral arterial occlusive disease in Ontario between 1991 and 1998: a population-based study.  J Vasc Surg. 2003;38(2):279-288.PubMedGoogle ScholarCrossref
19.
Fan  J, Arruda-Olson  AM, Leibson  CL,  et al.  Billing code algorithms to identify cases of peripheral artery disease from administrative data.  J Am Med Inform Assoc. 2013;20(e2):e349-e354.PubMedGoogle ScholarCrossref
20.
Austin  PC.  Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.  Stat Med. 2009;28(25):3083-3107.PubMedGoogle ScholarCrossref
21.
Austin  PC, Grootendorst  P, Anderson  GM.  A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study.  Stat Med. 2007;26(4):734-753.PubMedGoogle ScholarCrossref
22.
Cohen  J.  Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
23.
Armstrong  EJ, Chen  DC, Westin  GG,  et al.  Adherence to guideline-recommended therapy is associated with decreased major adverse cardiovascular events and major adverse limb events among patients with peripheral arterial disease.  J Am Heart Assoc. 2014;3(2):e000697.PubMedGoogle ScholarCrossref
24.
Ardati  AK, Kaufman  SR, Aronow  HD,  et al.  The quality and impact of risk factor control in patients with stable claudication presenting for peripheral vascular interventions.  Circ Cardiovasc Interv. 2012;5(6):850-855.PubMedGoogle ScholarCrossref
25.
Feringa  HH, van Waning  VH, Bax  JJ,  et al.  Cardioprotective medication is associated with improved survival in patients with peripheral arterial disease.  J Am Coll Cardiol. 2006;47(6):1182-1187.PubMedGoogle ScholarCrossref
26.
Ebrahim  S, Taylor  F, Ward  K, Beswick  A, Burke  M, Davey Smith  G.  Multiple risk factor interventions for primary prevention of coronary heart disease.  Cochrane Database Syst Rev. 2011;(1):CD001561.PubMedGoogle Scholar
27.
Lager  KE, Mistri  AK, Khunti  K, Haunton  VJ, Sett  AK, Wilson  AD.  Interventions for improving modifiable risk factor control in the secondary prevention of stroke.  Cochrane Database Syst Rev. 2014;5:CD009103.PubMedGoogle Scholar
28.
Hirsch  AT, Murphy  TP, Lovell  MB,  et al; Peripheral Arterial Disease Coalition.  Gaps in public knowledge of peripheral arterial disease: the first national PAD public awareness survey.  Circulation. 2007;116(18):2086-2094.PubMedGoogle ScholarCrossref
29.
Lovell  M, Harris  K, Forbes  T,  et al; Peripheral Arterial Disease Coalition.  Peripheral arterial disease: lack of awareness in Canada.  Can J Cardiol. 2009;25(1):39-45.PubMedGoogle ScholarCrossref
30.
Hirsch  AT, Criqui  MH, Treat-Jacobson  D,  et al.  Peripheral arterial disease detection, awareness, and treatment in primary care.  JAMA. 2001;286(11):1317-1324.PubMedGoogle ScholarCrossref
31.
Chan  AW, Henderson  MA.  Reorganization of vascular care in Canada.  Can J Cardiol. 2010;26(1):43-44.PubMedGoogle ScholarCrossref
32.
Holbrook  A, Pullenayegum  E, Thabane  L,  et al.  Shared electronic vascular risk decision support in primary care: Computerization of Medical Practices for the Enhancement of Therapeutic Effectiveness (COMPETE III) randomized trial.  Arch Intern Med. 2011;171(19):1736-1744.PubMedGoogle ScholarCrossref
33.
Vernooij  JW, Kaasjager  HA, van der Graaf  Y,  et al; SMARTStudy Group.  Internet based vascular risk factor management for patients with clinically manifest vascular disease: randomised controlled trial.  BMJ. 2012;344:e3750.PubMedGoogle ScholarCrossref
34.
Evans  CD, Eurich  DT, Taylor  JG, Blackburn  DF.  The Collaborative Cardiovascular Risk Reduction in Primary Care (CCARP) Study.  Pharmacotherapy. 2010;30(8):766-775.PubMedGoogle ScholarCrossref
35.
Brouwer  BG, Visseren  FLJ, Algra  A,  et al.  Effectiveness of a hospital-based vascular screening programme (SMART) for risk factor management in patients with established vascular disease or type 2 diabetes: a parallel-group comparative study.  J Intern Med. 2010;268(1):83-93.PubMedGoogle Scholar
36.
Juurlink  DN, Preyra  C, Coxford  R,  et al.  Canadian Institute for Health Information Discharge Abstract Database: a Validation Study. Toronto, ON, Canada: Institute for Clinical Evaluative Sciences; 2006.
37.
Kokotailo  RA, Hill  MD.  Coding of stroke and stroke risk factors using International Classification of Diseases, Revisions 9 and 10.  Stroke. 2005;36(8):1776-1781.PubMedGoogle ScholarCrossref
Original Investigation
August 2016

Efficacy of a Guideline-Recommended Risk-Reduction Program to Improve Cardiovascular and Limb Outcomes in Patients With Peripheral Arterial Disease

Author Affiliations
  • 1Division of Vascular Surgery, St Michael’s Hospital, Toronto, Ontario, Canada
  • 2Department of Surgery, University of Toronto, Toronto, Ontario, Canada
  • 3Li Ka Shing Knowledge Institute, St Michael’s Hospital, Toronto, Ontario, Canada
  • 4King Saud University–Li Ka Shing Collaborative Research Program
  • 5Department of Surgery, King Saud University, Riyadh, Saudi Arabia
  • 6Institute of Health Policy, Management, and Evaluation, University of Toronto, Toronto, Ontario, Canada
  • 7Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, Ontario, Canada
  • 8Applied Health Research Centre, Toronto, Ontario, Canada
  • 9Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
  • 10Division of Vascular Surgery, Peter Munk Cardiac Centre, University Health Network, Toronto, Ontario, Canada
  • 11Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
  • 12Division of Cardiac Surgery, St Michael’s Hospital, Toronto, Ontario, Canada
JAMA Surg. 2016;151(8):742-750. doi:10.1001/jamasurg.2016.0415
Abstract

Importance  Patients with peripheral arterial disease (PAD) are at a high risk for cardiovascular events, yet, to our knowledge, no studies have examined the effect of a comprehensive risk-reduction program on long-term outcomes for patients with PAD.

Objective  To investigative whether a program that focuses on 8 major guideline-recommended risk-management therapies reduces cardiovascular and limb events in patients with PAD.

Design, Setting and Participants  An observational cohort study with up to 7 years of follow-up was conducted using data from administrative databases from Ontario, Canada, between July 1, 2004, and March 31, 2013. Patients with symptomatic PAD who were enrolled in the Systematic Assessment of Vascular Risk (SAVR) program at a single tertiary vascular center in Ontario between July 2004 and April 2007 were matched with up to 2 (control) patients with PAD from other Ontario tertiary vascular centers not enrolled in the program using propensity score methods. Cox proportional hazards regression analysis was used to compare outcomes.

Exposures  Program that promoted antiplatelet agents, statins, angiotensin-converting enzyme inhibitors, blood pressure control, lipid control, diabetic glycemic control, smoking cessation, and target body mass index by engaging vascular surgeons, family physicians, and patients with PAD.

Main Outcomes and Measures  The primary outcome was a composite risk ratio of death, acute myocardial infarction, or ischemic stroke. Secondary outcomes included rates of lower limb amputations, bypass surgical procedures, and peripheral angioplasties with and without a stent.

Results  A total of 791 patients were studied after propensity score matching; the mean (SD) age of patients in the SAVR group (n = 290) was 67.9 (10.4) years and 68.2 (11.2) years in the control group (n = 501). During follow-up, the SAVR group experienced the primary outcome at a significantly lower rate than the control group (adjusted hazard ratio [HR], 0.63; 95% CI, 0.52-0.77). Patients in the SAVR group were also less likely to have major amputation (adjusted HR, 0.47; 95% CI, 0.29-0.77), minor amputation (adjusted HR, 0.26; 95% CI, 0.13-0.54), bypass surgery (adjusted HR, 0.47; 95% CI, 0.30-0.73), or hospitalization due to heart failure (adjusted HR, 0.73; 95% CI, 0.53-1.00). The rate of peripheral angioplasty with or without a stent was higher among the SAVR group (adjusted HR, 2.97; 95% CI, 2.15-4.10).

Conclusions and Relevance  A guideline-recommended risk-reduction program targeted at patients with PAD was associated with fewer cardiovascular and limb events over the long-term. This finding emphasizes the need for well-designed prospective studies to develop and examine the effect of such programs on reducing PAD-related morbidity, mortality, and health care costs.

Introduction

Severe systemic atherosclerosis in combination with multiple cardiovascular risk factors puts patients with peripheral arterial disease (PAD) at high risk for mortality and major coronary events.1-3 Furthermore, there has been a rapid increase in the number of people affected by PAD over the last decade, with 202 million people globally living with the disease in 2010.4 Current risk-management guidelines (eTable 1 in the Supplement) recommend that patients with PAD receive 8 main therapies: (1) antiplatelet agents, (2) statins, (3) angiotensin-converting enzyme (ACE) inhibitors, (4) blood pressure control, (5) lipid control, (6) diabetic glycemic control, (7) smoking cessation, and (8) target body mass index (BMI, calculated as weight in kilograms divided by height in meters squared).5 Despite clear guidelines, several studies have shown that patients with PAD are routinely undertreated for these risk factors,6-10 which may contribute to high rates of morbidity and mortality.

Benefits of individual pharmacotherapies in reducing the risk of cardiovascular events in patients with PAD have been reported in previous studies11-13; however, to our knowledge, no study has examined the effect of a program that focuses on all 8 guideline-recommended risk-reduction therapies. In 2004, we established the Systematic Assessment of Vascular Risk (SAVR) program to provide comprehensive risk management to patients with PAD because this was not the standard of care at the time. Our initial results showed that patients in the SAVR program were undertreated for all 8 risk factors at baseline prior to receiving targeted interventions for risk reduction.14

In this follow-up study, we sought to investigate whether patients with PAD enrolled in the SAVR program experienced lower rates of cardiovascular and limb events than patients with PAD not enrolled in this program.

Box Section Ref ID

Key Points

  • Question Does a risk-reduction program that promotes all 8 major guideline-recommended therapies reduce cardiovascular events in patients with peripheral arterial disease?

  • Findings In this cohort study, the composite risk of death, acute myocardial infarction, and ischemic stroke was 37% lower for those enrolled in a risk-reduction program compared with those not enrolled in a program, a significant difference.

  • Meaning Greater focus is needed on developing structured programs for improving control over risk factors for patients with peripheral arterial disease, as this may reduce cardiovascular morbidity and mortality over the long-term.

Methods
SAVR Program Overview

From July 1, 2004, to April 30, 2007, we enrolled 459 consecutive patients with symptomatic PAD into the SAVR program at a single tertiary vascular care center in Toronto, Ontario, Canada. Briefly, the program had 4 key components. First, a baseline risk assessment was done by a vascular surgeon and nurse at a vascular clinic to collect data on all 8 major risk factors. Our previous results showed that patients in the SAVR program had suboptimal rates of antiplatelet use (78%), statin use (61%), ACE inhibitor use (45%), blood pressure control (37%), lipid control (47%), diabetic glycemic control (49%), smoking cessation (67%), and adequate BMI level (28%) at baseline.14 Second, we formally educated patients in the SAVR program in clinic about PAD, their specific risk factors, and management strategies; this information was reinforced through written materials. Patients also received risk-factor optimization via pharmacotherapy initiation and referral to appropriate specialists and programs targeting all 8 risk factors. Third, we provided written material on clinical practice guidelines on PAD risk management to family physicians of patients in the SAVR program, and specific suggestions for risk management were made for their patients. Finally, ongoing follow-up risk assessments were organized with either the patients’ family physicians or vascular surgeons. In addition, a formal in-clinic follow-up was arranged for all patients in the SAVR program prior to the end of the program between July 1, 2013, and November 30, 2014, for a complete reassessment of all 8 risk factors. eMethods 1 in the Supplement provides a more detailed overview of the program.

Study Design

To determine adverse cardiovascular and limb event rates among patients in the SAVR program compared with those with PAD not enrolled in the program, we conducted a retrospective population-based cohort study using linked Ontario health administrative databases. The 13.5 million residents of Ontario have access to universal health care services funded by a single-payer system, which allows for all patient interactions with health care to be captured using linked administrative data. The University Health Network Research Ethics Board in Toronto approved this study. Patients in the SAVR program provided written informed consent for long-term clinic follow-up; informed consent was waived for the control group because their data were obtained from deidentified databases.

Data Sources

We accessed a prospectively maintained database of consecutive patients with PAD enrolled in the SAVR program between July 1, 2004, and April 30, 2007, and linked it to Ontario administrative data on the basis of patient date of birth, sex, and Ontario Health Insurance Plan number. We then used several linked administrative data sets to conduct this study (eMethods 2 in the Supplement). These databases are linked using an encrypted identifier unique to each patient, and they are routinely used for population-based research.15-18

Patient Cohort and Propensity-Based Matching

The SAVR cohort consisted of 459 patients with symptomatic PAD, defined as one of the following: (1) a clinical history of intermittent claudication, ischemic rest pain, or tissue loss in conjunction with an ankle brachial index of 0.9 or less; (2) previous peripheral arterial bypass surgery; (3) previous percutaneous transluminal angioplasty (PTA) with or without a stent; or (4) a lower limb amputation for ischemia.14 Vascular surgeons made the diagnoses and established patient eligibility for enrollment into the program.

We constructed a control cohort of patients not enrolled in the SAVR program for comparison by selecting all individuals who were seen by academic vascular surgeons in Ontario during the same period and carried a diagnosis of PAD within 3 years prior to that visit. Academic vascular surgeon was defined as any vascular surgeon with a clinical practice address associated with one of the tertiary vascular care centers in Toronto, Ottawa, Hamilton, London, or Kingston, Ontario. Patients seen by vascular surgeons at our institution-affiliated hospitals were excluded to minimize data contamination. We used a validated model-based billing code algorithm to identify patients with PAD in administrative data sets (eTable 2 in the Supplement).19 We defined the index date as the date on which the patient was enrolled in the program for the SAVR cohort and as the first encounter between the patient with PAD and a vascular surgeon between July 1, 2004, and April 30, 2007, for the control cohort. To restrict the analysis to patients with atherosclerotic disease, we excluded those who were younger than 40 years.

We used propensity score matching to match each SAVR patient with up to 2 control patients with PAD from other Ontario tertiary vascular centers not enrolled in the program. This approach balanced measured confounders between the intervention and control groups.20,21 We constructed logistic regression models using all 47 variables listed in Table 1 to calculate propensity scores and matched patients in each cohort on the basis of this score (within 0.2 SDs) and cohort entry date. If only 1 suitable match was found for a SAVR patient, we retained the pair for analysis; we did not include patients in the SAVR program for whom no control match was found. Further details on the baseline covariates measured are provided in eMethods 3 of the Supplement.

Outcomes

The primary outcome was a composite risk ratio of death or hospitalization for acute myocardial infarction (MI) or ischemic stroke because the primary goal of cardiovascular risk reduction is to reduce these 3 major ischemic events. Each individual component of the primary outcome was examined as a secondary outcome. Additional secondary outcomes included major amputation (above knee, below knee, or ankle level), minor amputation (through foot or toe), peripheral arterial bypass surgery, and PTA with or without a stent. Tertiary outcomes included hospitalization for heart failure, carotid endarterectomy, abdominal aortic aneurysm repair, coronary artery bypass graft, percutaneous coronary intervention, and initiation of dialysis. We also examined a tracer outcome (hospitalization for trauma to the head or upper extremities) that should not be affected by whether patients with PAD were enrolled in a risk-reduction program. The codes used to define outcomes and comorbid conditions are presented in eTable 3 in the Supplement. Patients were followed up in the databases until March 31, 2013, or until they experienced the primary end point or reached the maximum follow-up period of 7 years.

Clinical Follow-up of Patients in the SAVR Program

Because of limitations of the administrative data sets, we were not able to determine the rates of adherence to risk-reduction therapies in our cohort using administrative data. Therefore, we used the prospectively maintained SAVR database to (1) compare the baseline and follow-up risk control of patients in the SAVR program and (2) determine the causes of death for those who had died.

Statistical Analysis

We used standardized differences to compare the baseline characteristics between the groups before and after matching. Standardized difference scores measure effect size between 2 groups independent of the sample size, making them ideal for use in population-based and propensity-matched studies. A standardized difference less than 0.2 generally indicates good balance between groups with minimal risk of confounding.22 For example, the standardized difference for age before matching was 0.22, indicating significant difference between the 2 groups. After matching, the standardized difference was reduced to 0.03, suggesting the groups were well-balanced for age.

We applied Cox proportional hazards to compare outcomes of patients in the SAVR group and control patients, with control as the reference group. We generated unadjusted hazard ratios (HRs) and HRs adjusted for any unbalanced baseline characteristics after matching by building multivariable Cox proportion hazards regression models. We tested the proportional hazards assumption by visually inspecting log-log survival curves and by examining the statistical significance of time-dependent covariates. We also generated Kaplan-Meier curves with 95% CIs to establish the cumulative incidence of the primary outcome and secondary limb outcomes.

Statistical significance was set at the .05 level, and all P values were 2-tailed. All statistical analyses were conducted using SAS version 9.4 (SAS Institute).

Results
Cohort Identification and Follow-up

Of the 459 patients enrolled in the SAVR program, 447 (97%) were successfully linked with administrative data. In addition, 2686 patients with PAD not enrolled in the SAVR program were identified as potential control individuals. After matching was applied, the final cohort contained 290 patients in the SAVR program (65%) and 501 control patients (Figure 1). Median follow-up for the cohort was 5.3 years (interquartile range, 1.8-6.9 years).

Baseline Characteristics

Before matching was applied, several baseline differences were evident between the 2 groups. After propensity score matching, all of these factors, except for the number of hospital admissions in the past 3 years, were equally distributed (Table 1).

Efficacy Outcomes

The primary outcome occurred in 289 matched control (58%) and 126 matched patients (43%) in the SAVR program (Table 2). After adjusting for mean number of hospitalizations in the past 3 years, patients in the SAVR program experienced the primary outcome at a significantly lower rate than control patients (adjusted HR, 0.63; 95% CI, 0.52-0.77; P < .001) (Figure 2).

With respect to secondary outcomes, patients in the SAVR program experienced lower mortality (adjusted HR, 0.60; 95% CI, 0.48-0.75; P < .001) than control patients, but no differences between the 2 groups existed in the risk of acute MI or ischemic stroke. Patients in the SAVR program were also less likely to undergo major amputation (adjusted HR, 0.47; 95% CI, 0.29-0.77; P = .002), minor amputation (adjusted HR, 0.26; 95% CI, 0.13-0.54; P < .001), or arterial bypass surgery (adjusted HR, 0.47; 95% CI, 0.30-0.73; P < .001) (Figure 3). The rate of PTA with or without a stent was higher among patients in the SAVR program (adjusted HR, 2.97; 95% CI, 2.15-4.10; P < .001). In addition, patients in the SAVR program were less likely to be hospitalized owing to heart failure (adjusted HR, 0.73; 95% CI, 0.53-1.00; P = .048), although no significant differences were found in any of the other tertiary outcomes.

Tracer Outcome

Both groups experienced the tracer outcome of head or upper body trauma at a similar rate (adjusted HR, 0.95; 95% CI, 0.47-1.90; P = .87).

Clinical Follow-up of Patients in the SAVR Program

Examination of the SAVR database revealed that 99 patients completed an end-of-program follow-up after a mean (SD) of 8.4 (0.8) years (eTable 4 in the Supplement). Others either died (n = 170) or did not complete an end-of-program assessment (n = 190). The most common causes of death were cardiovascular (n = 63 [37%]), neoplastic (n = 43 [25%]), and respiratory (n = 18 [11%]) (eTable 5 in the Supplement). Compared with baseline, patients in the SAVR program experienced significant improvements in adherence to 7 of the 8 therapies at follow-up: antiplatelet use (67% to 80%; P = .03), statin use (67% to 88%; P < .001), ACE inhibitor or angiotensin receptor blocker use (58% to 72%; P = .02), blood pressure control (35% to 52%; P = .02), low-density lipoprotein cholesterol control (63% to 86%; P = .002), smoking cessation (75% to 88%; P < .001), and target BMI (27% to 38%; P = .049). The proportion of patients with diabetes with adequate glycemic control also improved from 44% to 52%, but this did not reach statistical significance (P = .53) (eTable 6 and eFigure in the Supplement).

Discussion

The main finding from this study is that patients with PAD who were enrolled in a structured program promoting all 8 guideline-recommended risk-reduction therapies experienced a 37% (95% CI, 23%-48%) relative reduction in the rate of cardiovascular events (death, acute MI, or ischemic stroke) over 7 years. In addition, these patients experienced a 40% relative reduction in death (95% CI, 25%-52%), a 53% relative reduction in major amputations (95% CI, 23%-71%) and arterial bypass surgery (95% CI, 27%-70%), a 74% relative reduction in minor amputations (95% CI, 46%-87%), and a 27% relative reduction in hospitalization due to heart failure (95% CI, 0.2%-47%). However, the rate of angioplasty with or without a stent was 3-fold higher in the SAVR group; the reason for this observation is unclear but could be due to several factors. Physicians treating patients in the SAVR program at our institution may have had a lower threshold for offering endovascular interventions than physicians from other intuitions, or perhaps greater physician contact among the more intensively followed-up SAVR group led to earlier detection and treatment of disease in a more minimally invasive manner. In addition, we found that our program was effective in significantly improving the adherence of risk-reduction therapies in 7 of the 8 previously described areas, except for diabetic glycemic control.

To our knowledge, few studies have examined the effect of adherence to multiple guideline-recommended therapies on the outcomes of PAD, and these studies have focused primarily on pharmacotherapy. Armstrong et al23 examined 739 patients with PAD who underwent lower extremity angiography. They reported a 36% reduction in the rate of MI, stroke, and death and a 45% reduction in major amputation, thrombolysis, and surgical bypass in patients who adhered to 4 therapies (smoking cessation and aspirin, statin, and ACE inhibitor use). However, the effects of blood pressure control, lipid control, weight reduction, and diabetic glycemic control were not measured. In a retrospective study of 1357 peripheral vascular interventions for stable claudication, Ardati et al24 reported a 55% reduction in adverse peripheral vascular events in patients receiving aspirin and statin therapy compared with those who received neither. However, this study was limited to 6-month follow-up, and data on nonpharmacological risk-reduction therapies were also lacking. Similarly, a prospective observational cohort study of 2420 patients with PAD by Feringa et al25 showed that statins, β-blockers, aspirins, and ACE inhibitors were independently associated with a 56%, 32%, 28%, and 20% reduction in mortality, respectively. No other outcomes were measured in this study, and the effects of nonpharmacological therapies were also not examined.

Our study had several strengths that provide unique insight into the effect of a multifactorial risk-reduction program. First, our study focused on examining the effect of adherence to 8 major guideline-recommended therapies on long-term outcomes, whereas prior studies have only focused on pharmacological therapies and have been limited by short-term follow-up. Second, in our propensity score analysis, we matched the SAVR cohort to a control group of patients with PAD based on several potential confounding factors that, to our knowledge, have not been measured in prior studies, including socioeconomic status, health care service use, ambulatory care grouping, medication prescriptions filled, and comorbidity burden. This allowed near-perfect equalization of baseline characteristics between the intervention and control groups. Furthermore, to assess for confounding, rates of a tracer outcome (head or upper body trauma) were examined and found to be similar between the 2 groups. Third, the SAVR program had several unique features that made it both practical and effective. The educational interventions were targeted specifically for patients with PAD, whereas prior research has focused on risk control for patients with coronary artery disease or stroke.26,27 The program also focused on patients’ family physicians. Public and physician awareness of PAD is poor,28-30 and targeting both patients and their family physicians is critical in enhancing PAD awareness and management. Furthermore, because there is no dedicated specialty responsible for medically managing patients with PAD in Canada,31 this responsibility often falls to family physicians. In our approach, vascular surgeons were responsible for bringing attention to the risk-reduction guidelines during the initial consultation, whereas family physicians were primarily responsible for longitudinal management.

Finally, by conducting follow-up with patients in the SAVR program in clinic, we showed that adherence to the 8 guideline-recommended therapies improved with our approach. Several other approaches to improving risk-reduction therapy have been studied in patients with other forms of cardiovascular disease, with modest results. Holbrook et al32 randomized 1102 patients with cardiovascular disease to a web-based vascular risk advice support system or to usual medical care. At 1-year follow-up, the intervention group showed improvement only in antiplatelet therapy use (73% vs 65%), whereas other factors remained relatively unchanged. Other Internet-based nurse-led or pharmacist-guided risk-management interventions have also failed to show significant improvements in vascular risk control.33,34 A vascular screening and prevention program studied by Brouwer et al35 for patients with diabetes and cardiovascular disease demonstrated slightly better risk management at 16-month follow-up. However, only 81%, 60%, 39%, and 24% met targets for nonsmoking status, low-density lipoprotein cholesterol level, diabetic glycemic control, and BMI, respectively, compared with 88%, 86%, 52%, and 38% in the SAVR cohort.

Our study had several limitations that merit discussion. First, residual confounding may have biased our results despite propensity-score matching. However, this seems unlikely because we matched the groups based on 47 baseline variables, adjusted our analyses for any unbalanced characteristics, and examined a tracer outcome that was equally distributed between the 2 groups. Second, only 290 of patients (65%) in the SAVR program were matched with control patients. This may be because the 2 groups were quite different at baseline before matching and because we used several covariates for matching. Despite a lower match rate, this approach allowed near-perfect balancing of baseline covariates after matching. Third, low coding accuracy is a possible source of bias. To minimize this risk, we used codes that have been validated in our data sets through reabstraction studies, including acute MI (83% sensitivity and 87% positive predictive value), ischemic stroke (85% accuracy), heart failure (79% sensitivity and 85% positive predictive value), and arterial bypass surgery (87% sensitivity and 88% positive predictive value).36,37 Furthermore, we used a validated model-based billing code algorithm that has 86% sensitivity, 83% specificity, 91% positive predictive value, and 74% negative predictive value in identifying patients with PAD to establish our control group.19 Fourth, although we were able to determine rates of bypass surgery and PTA with and without a stent in our databases, we were not able to accurately capture other PAD-related open surgical and endovascular procedures. Fifth, our databases did not allow us to stratify the results according to PAD severity. However, given both the SAVR and control cohorts had broad inclusion criteria and included patients with varying severity of PAD, we are confident that our results are generalizable. Finally, we were not able to determine the rates of guideline adherence among the control cohort at follow-up, and results of the clinic follow-up of patients in the SAVR program should be interpreted with caution because only 99 patients completed an end-of-program assessment.

Conclusions

This study demonstrated that patients with PAD who receive an educational intervention focused on all 8 major guideline-recommended risk reduction therapies experience lower rates of cardiovascular and limb events over 7 years. Despite the limitations of this observational study, our results indicate a greater focus is needed on developing structured programs for improving risk-factor control in the PAD population. The effect of these programs on reducing PAD-related morbidity and mortality and potentially reducing health care costs requires further research in well-designed prospective studies or randomized clinical trials.

Back to top
Article Information

Corresponding Author: Thomas F. Lindsay, MDCM, MSc, Division of Vascular Surgery, Peter Munk Cardiac Centre, University Health Network, Toronto General Hospital, 200 Elizabeth St, 6EN228, Toronto, ON M5G 2C4, Canada (thomas.lindsay@uhn.ca).

Accepted for Publication: February 8, 2016.

Published Online: April 6, 2016. doi:10.1001/jamasurg.2016.0415.

Author Contributions: Dr Lindsay and Ms Wang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Hussain and Al-Omran contributed equally.

Study concept and design: Hussain, Al-Omran, Mamdani, Premji, Lindsay.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Hussain, Wang.

Critical revision of the manuscript for important intellectual content: Hussain, Al-Omran, Mamdani, Eisenberg, Premji, Saldanha, Verma, Lindsay.

Statistical analysis: Hussain, Mamdani, Premji, Wang.

Obtained funding: Lindsay.

Administrative, technical, or material support: Al-Omran, Premji, Saldanha, Lindsay.

Study supervision: Al-Omran, Lindsay.

Conflict of Interest Disclosures: None reported.

Funding/Support: The Peter Munk Cardiac Centre of the University Health Network funded this study. In addition, this study was supported by the Institute for Clinical Evaluative Sciences, which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. Dr Hussain is also supported in part by a Canadian Institute for Health Research Canada Graduate Scholarship.

Role of Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The opinions, results, and conclusions reported in this article are those of the authors and are independent from the funding sources. No endorsement by the Institute for Clinical Evaluative Sciences or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred.

Previous Presentation: This study was presented as an oral abstract at the American Heart Association Scientific Sessions; November 8, 2015; Orlando, Florida.

Additional Contributions: We thank the Office of the Registrar General of Ontario for providing data from the Ontario Death Registry and Dr Zeyad Khoushhal, MBBS, MPH (Division of Vascular Surgery, St. Michael’s Hospital, Toronto, Ontario, Canada), for assistance with data analysis. Dr Khoushhal did not receive any compensation for his contribution.

References
1.
Joosten  MM, Pai  JK, Bertoia  ML,  et al.  Associations between conventional cardiovascular risk factors and risk of peripheral artery disease in men.  JAMA. 2012;308(16):1660-1667.PubMedGoogle ScholarCrossref
2.
Selvin  E, Erlinger  TP.  Prevalence of and risk factors for peripheral arterial disease in the United States: results from the National Health and Nutrition Examination Survey, 1999-2000.  Circulation. 2004;110(6):738-743.PubMedGoogle ScholarCrossref
3.
Fowkes  FG, Murray  GD, Butcher  I,  et al; Ankle Brachial Index Collaboration.  Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: a meta-analysis.  JAMA. 2008;300(2):197-208.PubMedGoogle ScholarCrossref
4.
Fowkes  FGR, Rudan  D, Rudan  I,  et al.  Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: a systematic review and analysis.  Lancet. 2013;382(9901):1329-1340.PubMedGoogle ScholarCrossref
5.
Anderson  JL, Halperin  JL, Albert  NM,  et al.  Management of patients with peripheral artery disease (compilation of 2005 and 2011 ACCF/AHA guideline recommendations): a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.  Circulation. 2013;127(13):1425-1443.PubMedGoogle ScholarCrossref
6.
Makowsky  MJ, McAlister  FA, Galbraith  PD,  et al; Alberta Provincial Program for Outcome Assessment in Coronary Heart Disease (APPROACH) Investigators.  Lower extremity peripheral arterial disease in individuals with coronary artery disease: prognostic importance, care gaps, and impact of therapy.  Am Heart J. 2008;155(2):348-355.PubMedGoogle ScholarCrossref
7.
Bhatt  DL, Steg  PG, Ohman  EM,  et al; REACH Registry Investigators.  International prevalence, recognition, and treatment of cardiovascular risk factors in outpatients with atherothrombosis.  JAMA. 2006;295(2):180-189.PubMedGoogle ScholarCrossref
8.
Subherwal  S, Patel  MR, Kober  L,  et al.  Missed opportunities: despite improvement in use of cardioprotective medications among patients with lower-extremity peripheral artery disease, underuse remains.  Circulation. 2012;126(11):1345-1354.PubMedGoogle ScholarCrossref
9.
Hasimu  B, Li  J, Yu  J,  et al.  Evaluation of medical treatment for peripheral arterial disease in Chinese high-risk patients.  Circ J. 2007;71(1):95-99.PubMedGoogle ScholarCrossref
10.
Flu  HC, Tamsma  JT, Lindeman  JHN, Hamming  JF, Lardenoye  JHP.  A systematic review of implementation of established recommended secondary prevention measures in patients with PAOD.  Eur J Vasc Endovasc Surg. 2010;39(1):70-86.PubMedGoogle ScholarCrossref
11.
CAPRIE Steering Committee.  A randomised, blinded, trial of Clopidogrel Versus Aspirin in Patients at Risk of Ischaemic Events (CAPRIE).  Lancet. 1996;348(9038):1329-1339.PubMedGoogle ScholarCrossref
12.
Yusuf  S, Sleight  P, Pogue  J, Bosch  J, Davies  R, Dagenais  G; The Heart Outcomes Prevention Evaluation Study Investigators.  Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients.  N Engl J Med. 2000;342(3):145-153.PubMedGoogle ScholarCrossref
13.
Heart Protection Study Collaborative Group.  MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20 536 high-risk individuals: a randomised placebo-controlled trial.  Lancet. 2002;360(9326):7-22.PubMedGoogle ScholarCrossref
14.
Al-Omran  M, Verma  S, Lindsay  TF.  Suboptimal use of risk reduction therapy in peripheral arterial disease patients at a major teaching hospital.  Ann Saudi Med. 2011;31(4):371-375.PubMedGoogle ScholarCrossref
15.
Carter  AA, Gomes  T, Camacho  X, Juurlink  DN, Shah  BR, Mamdani  MM.  Risk of incident diabetes among patients treated with statins: population based study.  BMJ. 2013;346:f2610.PubMedGoogle ScholarCrossref
16.
Al-Omran  M, Mamdani  MM, Lindsay  TF, Melo  M, Juurlink  DN, Verma  S; Systematic Assessment of Vascular Risk Investigators.  Suboptimal use of statin therapy in elderly patients with atherosclerosis: a population-based study.  J Vasc Surg. 2008;48(3):607-612.PubMedGoogle ScholarCrossref
17.
Park-Wyllie  LY, Juurlink  DN, Kopp  A,  et al.  Outpatient gatifloxacin therapy and dysglycemia in older adults.  N Engl J Med. 2006;354(13):1352-1361.PubMedGoogle ScholarCrossref
18.
Al-Omran  M, Tu  JV, Johnston  KW, Mamdani  MM, Kucey  DS.  Outcome of revascularization procedures for peripheral arterial occlusive disease in Ontario between 1991 and 1998: a population-based study.  J Vasc Surg. 2003;38(2):279-288.PubMedGoogle ScholarCrossref
19.
Fan  J, Arruda-Olson  AM, Leibson  CL,  et al.  Billing code algorithms to identify cases of peripheral artery disease from administrative data.  J Am Med Inform Assoc. 2013;20(e2):e349-e354.PubMedGoogle ScholarCrossref
20.
Austin  PC.  Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.  Stat Med. 2009;28(25):3083-3107.PubMedGoogle ScholarCrossref
21.
Austin  PC, Grootendorst  P, Anderson  GM.  A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study.  Stat Med. 2007;26(4):734-753.PubMedGoogle ScholarCrossref
22.
Cohen  J.  Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
23.
Armstrong  EJ, Chen  DC, Westin  GG,  et al.  Adherence to guideline-recommended therapy is associated with decreased major adverse cardiovascular events and major adverse limb events among patients with peripheral arterial disease.  J Am Heart Assoc. 2014;3(2):e000697.PubMedGoogle ScholarCrossref
24.
Ardati  AK, Kaufman  SR, Aronow  HD,  et al.  The quality and impact of risk factor control in patients with stable claudication presenting for peripheral vascular interventions.  Circ Cardiovasc Interv. 2012;5(6):850-855.PubMedGoogle ScholarCrossref
25.
Feringa  HH, van Waning  VH, Bax  JJ,  et al.  Cardioprotective medication is associated with improved survival in patients with peripheral arterial disease.  J Am Coll Cardiol. 2006;47(6):1182-1187.PubMedGoogle ScholarCrossref
26.
Ebrahim  S, Taylor  F, Ward  K, Beswick  A, Burke  M, Davey Smith  G.  Multiple risk factor interventions for primary prevention of coronary heart disease.  Cochrane Database Syst Rev. 2011;(1):CD001561.PubMedGoogle Scholar
27.
Lager  KE, Mistri  AK, Khunti  K, Haunton  VJ, Sett  AK, Wilson  AD.  Interventions for improving modifiable risk factor control in the secondary prevention of stroke.  Cochrane Database Syst Rev. 2014;5:CD009103.PubMedGoogle Scholar
28.
Hirsch  AT, Murphy  TP, Lovell  MB,  et al; Peripheral Arterial Disease Coalition.  Gaps in public knowledge of peripheral arterial disease: the first national PAD public awareness survey.  Circulation. 2007;116(18):2086-2094.PubMedGoogle ScholarCrossref
29.
Lovell  M, Harris  K, Forbes  T,  et al; Peripheral Arterial Disease Coalition.  Peripheral arterial disease: lack of awareness in Canada.  Can J Cardiol. 2009;25(1):39-45.PubMedGoogle ScholarCrossref
30.
Hirsch  AT, Criqui  MH, Treat-Jacobson  D,  et al.  Peripheral arterial disease detection, awareness, and treatment in primary care.  JAMA. 2001;286(11):1317-1324.PubMedGoogle ScholarCrossref
31.
Chan  AW, Henderson  MA.  Reorganization of vascular care in Canada.  Can J Cardiol. 2010;26(1):43-44.PubMedGoogle ScholarCrossref
32.
Holbrook  A, Pullenayegum  E, Thabane  L,  et al.  Shared electronic vascular risk decision support in primary care: Computerization of Medical Practices for the Enhancement of Therapeutic Effectiveness (COMPETE III) randomized trial.  Arch Intern Med. 2011;171(19):1736-1744.PubMedGoogle ScholarCrossref
33.
Vernooij  JW, Kaasjager  HA, van der Graaf  Y,  et al; SMARTStudy Group.  Internet based vascular risk factor management for patients with clinically manifest vascular disease: randomised controlled trial.  BMJ. 2012;344:e3750.PubMedGoogle ScholarCrossref
34.
Evans  CD, Eurich  DT, Taylor  JG, Blackburn  DF.  The Collaborative Cardiovascular Risk Reduction in Primary Care (CCARP) Study.  Pharmacotherapy. 2010;30(8):766-775.PubMedGoogle ScholarCrossref
35.
Brouwer  BG, Visseren  FLJ, Algra  A,  et al.  Effectiveness of a hospital-based vascular screening programme (SMART) for risk factor management in patients with established vascular disease or type 2 diabetes: a parallel-group comparative study.  J Intern Med. 2010;268(1):83-93.PubMedGoogle Scholar
36.
Juurlink  DN, Preyra  C, Coxford  R,  et al.  Canadian Institute for Health Information Discharge Abstract Database: a Validation Study. Toronto, ON, Canada: Institute for Clinical Evaluative Sciences; 2006.
37.
Kokotailo  RA, Hill  MD.  Coding of stroke and stroke risk factors using International Classification of Diseases, Revisions 9 and 10.  Stroke. 2005;36(8):1776-1781.PubMedGoogle ScholarCrossref
×