The Association between Major Adverse Cardiovascular Events and Peripheral Artery Disease Burden

Objective: The aim of the present study was to investigate the possible relationship between the segmental burden of lower limb atherosclerosis and Major Adverse Cardiovascular Events (MACEs). Methods: All the consecutive symptomatic peripheral artery disease (PAD) patients admitted for digital subtraction angiography (DSA) at Turku University Hospital department of Vascular Surgery between 1 January 2009 and 30 July 2011 were retrospectively analyzed. Angiography due to symptomatic PAD was used as the index date for the inclusion in the study. The segmental burden of atherosclerosis based on DSA was divided into three categories according to the highest disease burden of the defined artery segment: aorto-iliac, femoropopliteal, or tibial segments. The major association for the study was MACEs (defined as a cerebrovascular event, heart failure (HF) and myocardial infarction requiring hospital admission). Demographic data and MACEs were obtained from the hospital electronic medical records system. Results. The lower limb atherosclerosis burden of tibial vessels was related to an increased probability for HF (OR 3.9; 95%CI 2.4–6.5) and for MACEs overall (OR 2.3; 95%CI 1.4–3.6). The probability of both HF and MACEs overall rose with the increasing severity of the atherosclerosis burden. Moreover, the more severe the tibial vessel atherosclerosis, the higher the risk of HF and MACEs. The most extensive tibial atherosclerosis patients had an OR 4.5; 95%CI 2.6–8.0 for HF and an OR 3.1; and 95%CI 1.7–5.6 for MACEs overall. The femoropopliteal disease burden was also associated with an increased risk of HF (OR 2.3; 95%CI 1.6–3.2) and MACE (OR 1.9; 95%CI 1.3–2.7). However, the increasing extent of atherosclerosis of the femoropopliteal segment solely increased the risk of MACEs. Conclusions: PAD patients with severe tibial atherosclerosis are likely to present with MACEs. The risk is further enhanced as the extent of tibial vessel atherosclerosis is increased. An association between MACE and severe atherosclerosis on the aortoiliac segment was not detected. However, when the femoropopliteal segment was the most affected artery segment, the risk of MACEs was increased.


Introduction
Peripheral artery disease (PAD) is a major risk factor for a poor cardiovascular outcome.It affects more than 200 million people worldwide [1] and in addition to mortality, it has a significant impact on patients' daily functioning.There is a strong body of evidence in the present literature that supports the use of non-invasive pressure measurements for predicting the overall and cardiovascular mortality [2][3][4][5][6][7].Both the non-invasive measures of the ankle brachial index (ABI) and the less utilized toe brachial pressure index (TBI) indirectly reflect the severity of atherosclerotic lesions between the descending aorta and the ankle or toe arteries.The possible significance of the extent of atherosclerosis on the lower limb vessels and patient outcome is poorly understood, however.
Peripheral arteries can be divided into three distinct segments and in each segment, the development of atherosclerosis is associated with distinctive risk factors [8].For example, smoking, male sex and younger age are associated with aortoiliac lesions [9], whereas male sex and smoking are associated with both internal and external iliac calcification [10].Both smoking and COPD are risk factors [11] associated with atherosclerosis of the femoropopliteal segment.Diabetes, hypertension, heart failure (HF) and CKD are associated with the tibial and pedal vessel disease [12].
Multiple studies have investigated the clinical significance of atherosclerosis in one or more of these artery segments.Abdominal aortic calcification is associated with cardiovascular mortality, and coronary and cerebrovascular events.The risks for coronary and cerebrovascular events therefore increase along with the extent of abdominal aortic atherosclerosis [13,14].A retrospective study of infrainguinal atherosclerosis assessed Bollinger scores and demonstrated that the Bollinger score is an independent predictor of mortality.Furthermore, the authors concluded that both the femoropopliteal and tibial Bollinger scores predicted mortality [15].Recent studies compared the extent of tibial atherosclerosis (crural index (CXi)) and observed an association between the extent of tibial atherosclerosis and several poor outcomes such as cardiovascular mortality, overall mortality amputation free survival, ischemic degenerative brain changes and poor outcome after thrombolysis [16][17][18][19].However, the cardiovascular outcome in relation to the burden of atherosclerosis on each lower limb artery segment still requires further investigation.
Since the association between extensive atherosclerosis and cardiovascular morbidity is still controversial, the aim of the present study was to investigate whether the extent of aortoiliac, femoropopliteal or tibial disease has an impact on the major adverse cardiovascular events (MACEs) in PAD patients with major adverse limb events (MALEs).

Patient Characteristics
The present study was a retrospective study of 732 symptomatic peripheral artery disease (PAD) patients at Turku University Hospital's department of vascular surgery admitted for digital subtraction angiography either for diagnostic digital subtraction angiography (DSA) or for endovascular treatment of lower limb atherosclerosis between 1 January 2009 and 30 July 2011.The angiography date was used as the index date for MALEs and an index date for inclusion in the study.Data were collected from hospital electronic databases.The study was approved by the Hospital District of South-Western Finland Ethics Committee (decision ID TK-53-1266-15).Patient consent was not required due to the retrospective nature of the study.This study conformed to the ethics guidelines of the 1975 Declaration of Helsinki.

DSA Analysis and Description of the Crural Index (CXi)
All analyses of segment-specific atherosclerosis were based on DSA images.The index classification was as described in TASC II for aortoiliac and femoropopliteal segments (TASC A-D converted to indices I-IV).All three tibial vessels were analyzed separately for the CXi.Each crural vessel was coded as follows: no detectable occlusive disease or minor stenosis: 0; total occlusion of less than 5 cm: 1; total occlusion of less than 10 cm: 2; total occlusion of less than 15 cm: 3; total occlusion of more than 15 cm: 4.Only total occlusions were measured; other atherosclerotic lesions were not considered.The CXi was created by a sum of the three values that had been obtained from each individual tibial vessel.If the sum was 0, the index was 0, if the sum was between 1 and 3 the index was I, if the sum was 4-6 the index was II, if the sum was 7-9 the index was III and if the sum was 10-12 the index was IV.

Base Line Characteristics and MACEs
The baseline characteristics were collected from the hospital's patient electronic medical records, retrospectively.Comorbidities were recorded according to ICD-10 codes: coronary artery disease (CAD) (I20.0-I25.9),hypertension (I10.0-I10.9),(I48.0-I48.9),diabetes mellitus (E10.0-E11.9),chronic obstructive pulmonary disease (COPD) (J44.8),hypercholesterolemia (E78.0) and chronic kidney disease (CKD) (N18.1-N18.9).In addition, a smoking history and Rutherford class were recorded at the index date.The last values of the ABI and TBI before the index date for the study were obtained, and the value of the leg with the lower index was used for analyses.All non-invasive pressures were analyzed in a certified Angio laboratory as described earlier by Wickström and colleagues [21].
A MACE was defined as stroke, HF or another acute coronary syndrome.MACE data for the study cohort of 732 patients with a MACE were obtained from hospital digital patient files before and after the study index date.

Statistical Analyses
Statistical analyses were performed using the IBM SPSS ® version 29 statistics program.The Shapiro-Wilk test was used to test the normality of the study data.Continuous variables were expressed as mean ± standard deviation (SD) and the Kruskal-Wallis test was used for comparisons.Categorical variables were expressed as frequency and percentage and comparisons were performed using the Chi-square test.Age-adjusted logistic regression analyses were performed on MACE overall data and all selected MACEs separately as the outcome.Significant burdens were selected for the multilogistic regression analyses based on age-adjusted regression analyses.The following confounding variables were added to the model: Age, CAD, Diabetes, CKD.All authors had full access to all the study data.The corresponding author takes the responsibility for the integrity of the data analyses.Data cannot be shared publicly because of patient identification.However, data are available from the corresponding author for researchers who meet the criteria for access to confidential data.

The Risk of a MACE and Atherosclerosis Burden
An age-adjusted logistic regression analysis was performed to analyze the possible association between the atherosclerosis burden group and risk of defined outcomes.The aortoiliac or femoropopliteal segment did not present an increased OR (odds ratio) for any MACEs (Tables 5 and 6).Extensive tibial artery atherosclerosis was associated with MACEs (Table 7).When analyzed against the segment-specific burden of atherosclerosis, the infrainguinal disease femoropopliteal and tibial burden was associated with an increased risk compared to severe aortoiliac segment disease (Table 8).Based on the age-adjusted regression analyses, the CXi and lower limb artery disease burden was further analyzed via multilogistic regression analyses.The model contained significant comorbidities in regard to CAD, CKD and diabetes; see Table 9.

Discussion
The present study suggests that extensive tibial vessel atherosclerosis is associated with an increased risk of HF and MACE overall.In addition, both infrainguinal lower limb atherosclerosis burdens, i.e., of the femoropopliteal segment and the tibial segment, are associated with an increased risk of either HF or MACEs in general.In contrast, the aortoiliac burden could not be demonstrated to be associated with any consistent risk of any of the selected outcomes.Our earlier studies on the overall and cardiovascular mortality based on this cohort are in line with the present observations that the highest cardiovascular disease burden is associated with extensive and severe tibial vessel disease [16,17].
The inclusion criteria for this cohort were MALEs.Therefore, every patient in this cohort had significant PAD.The present study did not demonstrate an association between aortoiliac segment disease and MACEs.However, compared to the normal population, the cardiovascular burden of these patients was increased [22,23], and this was further borne out by the logistic regression analyses.The classification utilized for aortoiliac and femoropopliteal segments is based on the TASC A-D criteria.These criteria might not best present the extent of atherosclerosis on the aortoiliac segment of patients for all categories, but rather present the complicity of revascularization in that TASC class.Therefore, AI III and IV might especially comprise patients that had lesions that were technically difficult to treat but who did not have extensive atherosclerosis at that particular segment [24].
Even one in seven CAD or PAD patients present with a new MACE or MALE within 2 years of follow-up [25].Therefore, these patients have a large economic burden, and targeted efficient secondary preventive actions for patients with the highest MACE risk is therefore essential [25].Some general health-related predictors for MACEs and MALEs among patients with CAD and/or MALEs have previously been identified [25].Based on the present observations, the analyses of the segment-specific burden of atherosclerosis among MALE patients enable the identification of those with the highest risk of HF and MACEs overall.In accordance with the present observations, HF has been shown to be associated with low ABI indices [26].In that study, which was an Atherosclerosis Risk in Communities (ARIC) study, Gupta et al., 2014 also demonstrated a threshold value of 1.0 [26] for the increased risk of MACEs.In addition, borderline ABI values of 0.9-1.0 were shown to be associated with an increased risk of HF.However, the present observations are, to our knowledge, the first observations that suggest that the extent and anatomic distribution of atherosclerosis of the lower limb arteries can have a significant effect on the burden of HF among patients with symptomatic PAD.The mean ABI for patients at the highest risk of HF for CXi category IV was 0.57 and that for the segment-specific tibial vessel atherosclerosis burden was 0.97, which suggests that the ABI alone cannot distinguish patients with the highest risk.
The utilized parameters for disease extent and severity are based on the clinician's evaluation of DSA images.The categories and grading of atherosclerosis severity based on the TASC II classification was originally created to serve as a guide to select the strategy for revascularization.The CXi is based on the length of the occlusions of three tibial arteries.For example, a patient with the CXi IV disease has extensive severe lesions in all tibial arteries, and mild lesions are not taken into consideration when the index is calculated.All these aspects should be considered when interpreting the present observations.However, the present observations strongly suggest that the localization, extent and severity of atherosclerosis of three lower limb artery segments has a significant input on the patient's outcome.Therefore, the development of algorithms that can automatically sequence the lower limb arteries and objectively analyze the disease burden [27] is needed.Such algorithms will provide a quick tool for angiography-based risk analyses.The algorithms could predict not only selected MACEs, but also predict cardiovascular death and even overall death.
The present study further emphasizes the need for new means for the risk analyses for peripheral artery disease.The most utilized non-invasive lower limb pressure measurement ABI has many advantages.However, the present results suggest that the ABI together with other parameters such as the extent or severity of the disease burden would further enhance the risk analyses.In the presence of tibial disease, the ABI is especially pseudohypertensive in many patients; thus, it may not distinguish all patients with a significant tibial atherosclerotic burden.Whether the ABI or TBI is more sensitive for risk analyses together with automated artificially intelligent produced data on disease burden will be an interesting topic for further studies as well as the possibility to utilize cytokines and chemokines as biomarkers for the progression and risk of the acute onset of vascular disease in various organ systems [28][29][30].

Conclusions
According to the present study data, severe tibial vessel PAD is associated with high overall MACE risk and especially the risk of HF.Further studies are essential for creating artificial intelligence-assisted analyses of lower limb atherosclerosis in order to provide a rapid and effective tool for both risk analyses for individual patients and possibly helping the planning of the optimal revascularization for the affected limb.

Table 1 .
The demography and diagnosed MACE conditions of the MALE patients for each aortoiliac (AI) burden group for categorical values n (%) and continuous variables mean (SD).

Table 2 .
The demography and diagnosed MACE conditions of the MALE patients for each femoropopliteal (FP) burden group for categorical values n (%) and continuous variables mean (SD).

Table 3 .
The demography and diagnosed MACE conditions of the MALE patients at each tibial crural index (CXi) burden group for categorical values n (%) and continuous variables mean (SD).

Table 4 .
The demography and diagnosed MACE conditions the MALE patients according to the lower limb segment specific atherosclerosis burden groups for categorical values n (%) and continuous variables mean (SD).

Table 5 .
The age-adjusted hazard for the overall MACE and MACEs (heart failure, myocardial infarct, cerebrovascular event) in five aortoiliac (AI) burden groups (an OR with 95% confidence intervals and p values).

Table 6 .
The age-adjusted hazards for the overall MACE and MACEs (heart failure, myocardial infarct, cerebrovascular event) in the defined femoropopliteal (FP) burden groups are presented with 95% confidence intervals and p values.

Table 7 .
The age-adjusted hazards for the overall MACE and MACEs (heart failure, acute cardiac syndrome, cerebrovascular event) for the defined tibial burden groups (CXi) are presented with 95% confidence intervals and p values.

Table 8 .
The age-adjusted hazards for the overall MACE and MACEs (heart failure, acute cardiac syndrome, cerebrovascular event) in the most severely affected lower limb segment burden groups are presented with 95% confidence intervals and p values.

Table 9 .
The multilogistic regression analyses.The most severely affected lower limb segment and CXi are presented with 95% confidence intervals and p values.Model with significant comorbidities in regard to CAD, CKD and diabetes.