The course of vascular risk factors and the occurrence of vascular events in patients with symptomatic peripheral arterial disease

Study population 

Most patients with typical symptoms of IC (cramping pain in the lower legs during exercise) or with rest pain, nonhealing ulcers, or gangrene were referred by general practitioners (GPs) of the province Utrecht to the outpatient clinic of the Department of Vascular Surgery at the University Medical Center Utrecht, The Netherlands. If the vascular surgeon diagnosed IC (typical symptoms and a resting ankle-brachial pressure index [ABI] ≤0.90), patients were asked to participate in the SMART study (Second Manifestations of ARTerial disease). The SMART study runs in parallel to the care given by the vascular surgeon for PAD; decisions on revascularization procedures were independently made by the vascular surgeon. The SMART study is an ongoing single-center prospective cohort study with the purpose of investigating the prevalence and incidence of additional vascular disease in patients who already have a manifestation of arterial disease or who are otherwise at high risk (diabetes mellitus, hypertension, or hyperlipidemia) of developing symptomatic arterial disease. The rationale and the study design have been described previously.16 Briefly, patients aged 18 to 79 years who were willing to participate and gave their written informed consent underwent a standardized vascular screening including a health questionnaire, laboratory assessment, and ultrasonography. The ethics committee of our institution approved the study. For this study, the data of 461 consecutive patients with symptomatic PAD who were included in the vascular screening program from September 1996 to December 2000 were used. Patients were invited to visit the hospital again for a follow-up measurement from September 2003 to March 2005. Patients received usual care from a GP or a vascular specialist in the period between baseline and follow-up measurement.

Vascular screening 

Participating patients visited the hospital after an overnight fast of at least 8 hours and underwent the same screening at baseline and at follow-up. Patients completed a health questionnaire on history and symptoms of CVD, current medication use, and atherosclerotic risk factors (diabetes mellitus, hypertension, hyperlipidemia, smoking, diet, physical activity, and familial vascular history). The items on CHD and PAD were based on the Rose Questionnaire.17 The severity of PAD was classified according to the Fontaine classification.18 Height and weight were measured without shoes and heavy clothing. Body mass index (BMI) was calculated as weight divided by height squared. Waist circumference was measured halfway between the lower limb and the iliac crest. Blood pressure was measured two times in the sitting position on the right and left upper arm with a nonrandom sphygmomanometer. Fasting blood was sampled to determine serum glucose, total cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, creatinine, and homocysteine concentrations. Low-density lipoprotein (LDL) cholesterol was calculated with Friedewald’s formula. An early-morning urine sample was collected to measure albumin and creatinine concentrations. Resting ABI was measured for each leg by taking the ratios of the highest systolic blood pressure measured at the ankle to the highest systolic blood pressure in both arms with the patient in the supine position. The systolic blood pressure in the posterior tibial and dorsal pedal arteries (left and right) was measured with an 8-MHz continuous-wave Doppler probe connected to an IMEXLAB 9000 Vascular Diagnostic System (Imex Medical Systems Inc, Golden, Colo), and that in both brachial arteries was measured with a semiautomatic oscillometric device (Omega 1400; Invivo Research Laboratories Inc, Broken Arrow, Okla). The results of the vascular screening were discussed at weekly meetings of a multidisciplinary team consisting of an internist, vascular surgeon, cardiologist, vascular nurse practitioner, and, on request, a neurologist. An internist evaluated the results of the follow-up measurement. If risk factors were not within the target range, treatment was given according to the European Guidelines of Cardiovascular Disease Prevention.15 The vascular screening results and treatment recommendations were reported in writing to the treating vascular specialist (most often the vascular surgeon) and the GP, and further action was left to their discretion. The patients were informed by telephone about the results of the follow-up measurement.

Difference in risk factors 

One of our aims was to quantify how risk factors changed over time. Therefore, we determined the prevalence of hypertension, hyperlipidemia, diabetes mellitus, obesity, and smoking at baseline and at follow-up. The cutoffs for the different risk factors were according to the European Guidelines15 and were as follows: hypertension, blood pressure greater than 140/90 mm Hg; hyperlipidemia, total cholesterol greater than 4.5 mmol/L or LDL cholesterol greater than 2.5 mmol/L; diabetes mellitus, patients taking glucose-lowering agents; obesity, BMI greater than 25 kg/m2; and current smoking. The change in the number of risk factors was calculated by subtracting the number of risk factors at follow-up from the number at baseline.

Follow-up 

Patients were biannually asked to complete a questionnaire on hospitalizations and outpatient clinic visits. If a patient did not return the questionnaire, a research nurse phoned the patient, family, or GP to obtain the health status. The end point of interest for this study was a composite of first occurrence of vascular events: namely, mortality, ischemic stroke, and MI. Definitions of events are given in Table I. PAD-related amputation, surgical procedures (bypass), and percutaneous transluminal angioplasty were also recorded during follow-up. If a patient or family recorded such an event, we retrieved hospital discharge letters and the results of relevant laboratory and radiology examinations. Three members of the SMART Study Endpoint Committee independently audited all events on the basis of available information. This committee consisted of physicians from different departments. In case of disagreement, consensus was reached by consulting other members of the Endpoint Committee. The patients were followed up until they died or refused further participation.

Table I. Definitions of fatal and nonfatal events

	Event	Definition
	Vascular death	Sudden death: unexpected cardiac death occurring within 1 h after onset of symptoms or within 24 h given convincing circumstantial evidence
		Death from ischemic stroke
		Death from intracerebral hemorrhage (hemorrhage on CT scan)
		Death from congestive heart failure
		Death from myocardial infarction
		Death from rupture of abdominal aortic aneurysm
		Vascular death from other cause, such as sepsis after stent placement
	Ischemic stroke	Definite: relevant clinical features that have caused an increase in impairment of at least one grade on the modified Rankin scale, accompanied by a fresh ischemic infarction on a repeat brain scan
		Probable: clinical features that have caused an increase in impairment of at least one grade on the modified Rankin scale, without a fresh ischemic infarction on a repeat brain scan
	Myocardial infarction	Fatal or nonfatal myocardial infarction: at least two of the following criteria:
		1. Chest pain for at least 20 min, not disappearing after administration of nitrates
		2. ST elevation >1 mm in two following leads or a left bundle branch block on the ECG
		3. CK elevation of at least two times the normal value of CK and an MB fraction >5% of the total CK

CT, Computed tomography; ECG, electrocardiogram; CK, creatinine kinase; MB, myocardial band.

Data analysis 

Data are presented as means with standard deviations or as percentages. Cox proportional hazards analysis was used to assess the effect of risk factors on vascular events, and results are presented as hazard ratios and 95% confidence intervals (CIs). If a patient had multiple events, the first was used in the analysis. In model 1, the age-adjusted association between baseline risk factors and vascular events was examined. Additional adjustments for systolic and diastolic blood pressure, smoking, HDL cholesterol, and diabetes were made in model 2.

Differences between participants and those who refused to participate at follow-up measurement were tested with χ2 tests (categorical variables), unpaired t tests (continuous normally distributed variables), or Mann-Whitney U tests (continuous skewed variables). Changes in the proportion of patients in whom risk factor values were higher than the relevant cutoff level were determined and reported with the corresponding 95% CIs. Calculations were performed with SPSS version 12.0.1 (SPSS, Chicago, Ill).

Study population 

Patients were screened for the first time between September 1996 and December 2000. Baseline characteristics of the study population are given in Table II. The 461 patients with symptomatic PAD had a mean age of 60.2 years and were predominantly male (68%). Eighty-seven percent of the patients were classified as having Fontaine II and 13% as Fontaine III/IV, and the mean ABI value of the 461 patients was 0.82.

Table II. Baseline characteristics of the patients with symptomatic peripheral arterial disease (n = 461)

	Age (y)	60.2±10.9
	Male sex (%)	68
	Body mass index (kg/m2)	25.7±3.9
	Systolic blood pressure (mm Hg)	147±22
	Diastolic blood pressure (mm Hg)	80±11
	Total cholesterol (mmol/L)	5.9±1.2
	Glucose (mmol/L)	6.5±2.2
	Homocysteine (μmol/L)	15.0±9.0
	Creatinine clearance (Cockroft) (mL/min)	74±23
	Hypertension⁎ (%)	60
	Diabetes mellitus type 2† (%)	14
	Current smokers (%)	58
	Ankle-brachial pressure index	0.82±0.2
	Medication use
	Antiplatelet agents (%)	53
	Blood pressure–lowering agents (%)	46
	Lipid-lowering agents (%)	25
	ACE inhibitor and/or AIIA (%)	21
	Fontaine classification
	II (%)	87
	III + IV (%)	13
	Previous cardiovascular disease‡
	Abdominal aortic aneurysm (%)	7
	Cerebrovascular disease (%)	12
	Coronary heart disease (%)	27

ACE, angiotensin-converting enzyme; AIIA, angiotensin II-antagonist. Data are mean (SD) or %.

⁎ Systolic blood pressure >140 mm Hg or diastolic blood pressure >90 mm Hg. † Patients taking glucose-lowering agents. ‡ Not mutually exclusive.

Fatal and nonfatal events during follow-up 

During follow-up (until September 1, 2004), 108 (23%) of the 461 patients died (66 of a vascular event). The mean age of those who died was 66.9 years (SD, 9.7 years). Twenty-seven (6%) of the 461 patients experienced an ischemic stroke, and 57 (12%) of the 461 patients had an MI. The composite of ischemic stroke, MI, or vascular death occurred in 91 patients. There were 2739 person-years of follow-up. The cumulative incidence proportion was 29.1% (95% CI, 22.8%-35.4%) for the composite of subsequent vascular events, 22.1% (95% CI, 16.4%-27.8%) for vascular death, and 19.6% (95% CI, 13.9%-25.3%) for MI.

Amputation occurred in 38 legs of 35 patients during follow-up. Surgical procedures (bypass or endarterectomy) were performed the first time in 97 (21%) of the 461 patients involving the aorta/iliac, iliac, or femoral/popliteal arteries. Reoperations occurred most often (n = 30) in the femoral/popliteal segment and occurred in two patients at the iliac artery segment. Percutaneous transluminal angioplasty was performed the first time in 61 (13%) of the 461 patients, 32 times in the iliac segment, and 29 times in the femoral/popliteal segment. Endovascular reinterventions occurred five times in treated iliac arteries and three times in femoral/popliteal arteries.

Older age, increased homocysteine levels, impaired renal function, and a history of CHD (age adjusted) were associated with an increased risk of fatal or nonfatal vascular events (Table III). Male sex and decreased HDL levels were borderline significantly related to vascular events. The strengths of the relations for a first vascular event remained essentially the same after adjustment for classical risk factors (data not shown). By analyzing the data separately for each single event outcome, we found results in the same direction but based on a rather small number of events.

Table III. Hazard ratios and 95% CI of baseline characteristics for vascular events (n = 461)

	Variable	HR	95% CI	P value
	Continuous variables
	Age per 10 y	1.85	1.48-2.30	<.001
	Total cholesterol per 0.5 mmol/L	0.92	0.84-1.01	.07
	HDL cholesterol per 0.1 mmol/L	0.94	0.89-1.00	.06
	LDL cholesterol per 0.5 mmol/L	0.94	0.85-1.04	.9
	Triglycerides per 0.5 mmol/L	1.01	0.94-1.08	.8
	Systolic blood pressure per 10 mm Hg	1.01	0.91-1.11	.9
	Diastolic blood pressure per 10 mm Hg	1.00	0.81-1.22	.96
	Serum glucose per 1.0 mmol/L	1.06	0.98-1.15	.2
	Homocysteine per 1.0 μmol/L	1.03	1.02-1.05	<.001
	Body mass index per 1.0 kg/m2	1.01	0.96-1.07	.6
	Creatinine clearance (Cockroft) per 10 mL/min	0.79	0.69-0.91	.001
	Dichotomous variables⁎
	Sex (female)	0.64	0.40-1.03	.07
	Diabetes mellitus type 2 (present)	1.49	0.89-2.51	.1
	Smoking (current smoking)	1.13	0.98-1.31	.09
	History of cerebrovascular disease (present)⁎	1.46	0.85-2.52	.2
	History of coronary heart disease (present)⁎	2.30	1.51-3.51	<.001
	History of abdominal aortic aneurysm (present)⁎	1.52	0.76-3.03	.2

HR, Hazard ratio adjusted for age; CI, confidence interval; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

⁎ Reference category in parentheses.

Follow-up measurement 

At the follow-up visit, 108 (23%) of the 461 patients had died, and 20 (4%) patients were lost to follow-up. Of the remaining 333 (72%) patients, 221 (66%) came for follow-up measurement, and 112 (34%) patients did not participate because of comorbidity, lack of time, lack of motivation, and travel distance. In 8 of the 221 patients, a nonfatal vascular event occurred: 2 had an ischemic stroke, and 6 patients had an MI. A PAD-related amputation occurred in 12 (5%) of the 221 between baseline and follow-up measurement. Percutaneous transluminal angioplasty was performed the first time in 35 (16%) of the 221, and surgical procedures (bypass) involving the aorta/iliac, iliac, or femoral/popliteal arteries were performed for the first time in 42 (19%) of the 221 patients.

There were small differences between the baseline characteristics of the participants and those who refused to participate at follow-up measurement. Participants were younger (57.1 vs 60.5 years; P = .01), less often had a history of cerebrovascular disease (7% vs 16%; P = .01), had a lower mean glucose level (6.0 vs 7.0 mmol/L; P = .02), and had a lower mean systolic blood pressure (144 vs 150 mm Hg; P = .01). The mean follow-up of the participants was 5.5 years (SD, 1.3 years).

Risk factors at baseline and follow-up 

The characteristics of the 221 patients who completed the baseline and follow-up measurements are given in Table IV. The mean ABI value was 0.85 at baseline and 0.89 at follow-up measurement. The mean serum glucose level (6.0 mmol/L) was comparable at both measurements, but more patients quantified for the diagnosis diabetes mellitus type 2 at follow-up (8% vs 16%). The mean levels of total cholesterol (6.0 vs 5.2 mmol/L), LDL cholesterol (4.0 vs 2.9 mmol/L), HDL cholesterol (1.1 vs 1.4 mmol/L), and triglycerides (2.1 vs 1.8 mmol/L) were more favorable at follow-up. The percentage of current smokers had decreased from 59% to 37%. The mean systolic and diastolic blood pressure, homocysteine level, BMI, and waist circumference had increased after a mean follow-up of 5.5 years (Table IV).

Table IV. Characteristics of patients with symptomatic peripheral arterial disease at baseline and follow-up

	Characteristic	Baseline (n = 221)	Follow-up (n = 221)
	Male sex (%)	67	67
	Age (y)	57.1±9.9	62.6±10.0
	Total cholesterol (mmol/L)	6.0±1.1	5.2±1.2
	LDL cholesterol (mmol/L)	4.0±1.0	2.9±1.0
	HDL cholesterol (mmol/L)	1.1±0.3	1.4±0.4
	Triglycerides (mmol/L)	2.1±1.4	1.8±1.0
	Systolic blood pressure (mm Hg)	144±21	152±23
	Diastolic blood pressure (mm Hg)	80±10	83±11
	Fasting serum glucose (mmol/L)	6.0±1.5	6.0±1.7
	Diabetes mellitus type 2⁎ (%)	8	16
	Homocysteine (μmol/L)	13.9±5.2	14.5±10.5
	Body mass index (kg/m2)	25.6±3.8	26.6±4.2
	Waist circumference (cm)	95±11	98±12
	Current smoking (%)	59	37
	Creatinine clearance (Cockcroft; mL/min)	78±20	82±29
	Albuminuria ≥3.0 mg/mmol (%)	14	25
	Ankle brachial pressure index	0.85±0.2	0.89±0.2
	Medication use
	Antiplatelet agents (%)	47	80
	Blood pressure–lowering agents (%)	31	44
	Lipid-lowering agents (%)	22	66
	ACE inhibitor and/or AIIA (%)	20	35
	Folic acid (%)	2	11

HDL, high-density lipoprotein; LDL, low-density lipoprotein; ACE, angiotensin-converting enzyme; AIIA, angiotensin II-antagonist. Data are mean (SD) or %.

⁎ Patients taking glucose-lowering agents.

During this period, the number of risk factors in individual patients remained the same in 93 (42%) patients, decreased in 71 (32%) patients, and increased in 57 (26%) patients. The decrease in risk factors was mainly caused by patients who quit smoking (from 59% to 37%; 95% CI, −13% to −31%) and by fewer patients with increased lipid levels (from 96% to 73%; 95% CI, −29% to −16%; Table V). The increase in risk factors was caused by more hypertension (19%; 95% CI, 10%-28%), more obesity (12%; 95% CI, 3%-21%), and more diabetes mellitus (8%; 95% CI, 2%-14%). Medication use in all categories increased with time (Table V).

Table V. Difference in risk factors and use of drugs between follow-up and baseline (n = 221)

	Variable	% (n) at baseline	% (n) at follow-up	% Increase (follow-up − baseline)	95% CI	P value
	Hypertension⁎	51(113)	70(155)	19	10to28	<.001
	Hyperlipidemia†	96(211)	73(161)	−23	−29to−16	<.001
	Diabetes mellitus‡	8(17)	16(35)	8	2to14	<.001
	Obesity§	54(120)	67(147)	12	3to21	<.001
	Current smoking	59(130)	37(82)	−22	−13to−31	<.001
	Antiplatelet agents	47(99)	79(167)	32	24to41	<.001
	Blood pressure–lowering drugs	31(67)	44(95)	13	4to22	<.001
	Lipid-lowering drugs	22(47)	66(139)	44	35to52	<.001
	ACE inhibitor and/or AIIA	20(43)	35(75)	15	7to23	<.001

CI, Confidence interval; ACE, angiotensin-converting enzyme; AIIA, angiotensin II-antagonist.

⁎ Systolic blood pressure >140 mm Hg or diastolic blood pressure >90 mm Hg. † Total cholesterol >4.5 mmol/L or LDL cholesterol >2.5 mmol/L. ‡ Patients taking glucose-lowering agents. § Body mass index >25 kg/m2.