Impact of Endovascular Treatment on Leg Muscle Mass in Peripheral Artery Disease: a Preliminary Report

Background: Peripheral artery disease (PAD), intermittent claudication, and impaired mobility lower patients’ quality of life, contributing to the loss of skeletal muscle. This study sought to investigate the impact of endovascular treatment (EVT) by measuring the mid-thigh level of muscle volume in above the knee PAD patients before and after EVT. Methods: In this prospective, observational study, symptomatic PAD patients with above the knee lesions who had intermittent claudication and were undergoing optimal medical treatment were enrolled. The mid-thigh level of muscle area was measured by computed tomography initially, and then 3 and 6 months after EVT. Patients were categorized into ipsilateral or bilateral based on clinical symptoms and initial ABI value. The muscle area in ischemic and non-ischemic legs were compared in ipsilateral PAD patients. The correlations between alterations in the total thigh muscle area and clinical characteristics were analyzed in univariable and multivariable analysis to investigate the factor contributing skeletal muscle loss. Results: A total of 22 patients were analyzed. The muscle area of the thighs increased after EVT. Fourteen patients had ipsilateral lesions and 8 had bilateral stenosis. In patients with ipsilateral lesions, the mid-thigh muscle area of ischemic lower limbs was signi�cantly lesser than that of non-ischemic lower limbs (118.2±16.5 cm 2 vs 124.0±17.3 cm 2 , p=0.0002). The thigh muscle area of ischemic lower limbs increased after EVT (before: 118.2±16.5 cm 2 vs 3 months: 124.0±18.7 cm 2 , p=0.0166; before vs 6 months: 123.0±17.7 cm 2 , p= 0.0566), but this was not the case for non-ischemic lower limbs. Multivariate regression analysis revealed that baseline glycated hemoglobin was the only factor that negatively correlated with the change in the muscle area after 3 (β= -3.74, 95% con�dence interval [CI] = -7.3 to -0.2, p=0.0417) and 6 months (β= -5.24, 95%CI = -10.1 to -0


Background
Skeletal muscle loss is a growing eld that is gaining attention among medical researchers [1].Patients with diabetes tend to have excessive loss of skeletal muscle in comparison with non-diabetic patients [2].
Atherosclerosis in the lower-limb peripheral arteries in patients with diabetes, dyslipidemia, and hypertension lead to peripheral arterial disease (PAD) [3,4] where reduced blood ow to the lower-limbs can induce metabolic myopathy of muscle in the diseased lower limbs [5] contributing to further skeletal muscle loss.The regional thigh muscle measurement is correlated with whole body skeletal muscle mass, implicating the role of thigh muscle loss as a surrogate of clinical fragility [6].Additionally, the thigh muscle area, as measured by computed tomography (CT), can be a predictor of mortality [7].A recent study showed that the proximal lesion location in PAD was associated with a poorer prognosis than that in the distal lesion location, presenting a negative association between lesion location and the volumetric area of muscle blood supply [8].As per the current clinical guidelines, endovascular treatment (EVT) is considered if the symptoms of patients with PAD persist after undergoing guideline-based medical treatment [9].
Assessment of the disease status utilizing only the cross-sectional single-slice CT measurements was associated with muscle volume for the identi cation of atrophy and focal sarcopenia [10].Therefore, in this preliminary study, we measured the muscle volume at the mid-thigh level in PAD patients with above the knee lesions before and after EVT to investigate the impact of EVT on patients with PAD.

Study Design
This was a single-center prospective observational study.Consecutive PAD patients with intermittent claudication under guideline derived pharmaco-physiological therapy who were scheduled to undergo elective EVT were enrolled.Patients who do not obey smoking prohibition, requiring maintenance hemodialysis, with critical ischemia or infection were excluded.The Institutional Review Board of the hospital reviewed and approved this study.This study had been performed in accordance with the Declaration of Helsinki obtaining informed consent from all patients.

Endovascular Treatment
Patients who were intolerant to certain medications or contrast agents did not participate in the study.
Dual antiplatelet therapy was administered to all patients for at least 7 days before EVT.At the start of the procedure, weight-adjusted intravenous heparin was administered with a target activated clotting time of >300 s.After crossing the guidewire through the target lesion, dilatation of the lesion was performed using a suitable balloon size.Stent placement was performed if ow limitation or residual stenosis >30% of elastic recoil remained in the target vessel.Twenty-one patients received stent placements.The size of the balloons was decided by the operator and the stents exceeded the reference diameter by 1 mm.Postdilatation was performed by a balloon shorter than the stent.Although efforts were made to nalize the procedure with a single stent placement, multiple stents were placed to achieve <30% residual stenosis in the target lesion.The stents that were used were EPIC (Boston Scienti c, Massachusetts), and SMART (Cordis Co, Florida).

Measurement of Thigh Muscle Area
A CT scan of the thighs was performed with a 64-slice Toshiba Aquilion 64 system (Toshiba Medical Systems, Otawara, Japan) [11].The thigh muscle area was measured by a commercial workstation (SYNAPSE; Fuji lm, Tokyo, Japan); further, the thigh muscle area was measured for both legs at the cross section of the midpoint of the upper margin of the distance between the upper edge, the midpoint of the upper margin of the greater trochanter and the lower margin of the femoral condyles; and the lower edge, the midpoint of the upper margin of the patella and the lower margin of the patella (Figure 1) [7].The investigators (TM and HK) who conducted the measurements were blinded to patients' clinical and laboratory data.

Ankle-Brachial Index Measurement
The ankle-brachial index (ABI) was measured using a Colin Wave-form analyzer (BP-203RPE III; Omron Colin, Tokyo, Japan), according to the recommendations of the American Heart Association [12].The ABI was measured at the time of admission and at discharge, as well as after 3 months and 6 months.

Statistical Analysis
Categorical data are presented as the frequency (percentage), and continuous variables are expressed as means ± standard deviations (SDs), as well as medians with inter-quartile ranges.All the statistical analyses were performed with R© (Ver 3.6.2,R Foundation for Statistical Computing, Vienna Austria).The Shapiro-Wilk test was used for assessing whether variables were normally distributed, and the Chi-square test was used for dichotomous variables.The mid-thigh area pre, 3 months post, and 6 months post EVT were compared by paired t-tests.The univariable and multivariable regression analyses between the change in (delta) the area of the thighs after 3 or 6 months and the baseline patient characteristics including the lesion location (bilateral), female sex, HbA1c, age, body mass index (BMI), Rutherford classi cation, hypertension, dyslipidemia, chronic kidney disease (CKD), and smoking history were performed to identify the baseline characteristics associated with the delta muscle area by the use of the MASS package ('Modern Applied Statistics with S', 4th edition, 2002).The quality of linear prediction models were assessed by the gvlma (Global Validation of Linear Models Assumptions) package [13].Statistical power analyses were evaluated with pwr (Basic Functions for Power Analysis) package [14].
Sixteen (53 %) lesions were located in the iliac artery, 9 (30 %) in the super cial femoral artery, 1 (3 %) in the common femoral artery, 3 (13.6 %) in the popliteal artery, and 1 (3 %) in both the super cial and popliteal artery.All patients were successfully discharged the day after EVT without any complications.
In relation to the delta thigh muscle area, the univariate regression analysis on each variable included lesion location, sex, baseline HbA1c levels, BMI, presence of hypertension (HT), dyslipidemia (DL), chronic kidney diseae (CKD), and past smoking history.The variables entered in the multivariate regression model were those with a relation in the univariate analysis (p ≤ 0.10), revealing that the baseline HbA1c was the only factor correlated with the delta muscle area of the thighs after 3 months (-7.3 cm 2 to -0.2 cm 2 per HbA1c 1 increase, p = 0.04) and 6 months (-10.1 to -0.4 cm 2 per HbA1c 1 increase, p = 0.036) (Table 3).

Discussion
The key ndings in the present study are as follows: rst, PAD induces thigh muscle atrophy and EVT can restore it.The increment in thigh muscle area in the ischemic limbs after EVT, but not in the non-ischemic limbs in patients with ipsilateral PAD, supports the result.Second, the glycated hemoglobin level was negatively correlated with the change in thigh muscle area.Increases in the thigh muscle area after EVT are more pronounced in normoglycemic patients than in diabetic patients.
The loss of skeletal muscle is a growing eld of medical interest since the recognition is indispensable for evaluating the disease morbidity and mortality [1].In PAD patients, pathological analysis on the gastrocnemius skeletal muscle of patients revealed ischemic myopathy due to apoptosis [15].Two-thirds of PAD patients had multiple lesions [8], most frequently located in the super cial femoral arteries [16,17].
Because the super cial femoral arteries vascularize the calf muscles, the assumption of a direct negative effect of PAD on the calf muscle area is reasonable [18].More speci cally, patients with proximal lesions had a poorer prognosis than those with distal lesions [19].
In PAD patients, chronic repetitive ischemia induces metabolic changes in ischemic muscle that leads to alterations in muscle structure, and eventually atrophy.In the muscle tissue of patients with PAD, mitochondrial dysfunction is found, which comprises mitochondrial respiration, subsequent poor energy production by the mitochondria, increased oxidative stress, and a signi cant amount of generated reactive oxygen spices (ROS), limiting exercise performance in patients with PAD [5].Even after the successful revascularization operations and normalization of hemodynamic parameters in PAD patients, only mild improvement in a signi cant mitochondrial bioenergetic defect was observed [20].Although the phenomenon observed in the previous report may con ict with our results, we showed that the thigh muscle area signi cantly increased after EVT.Although we did not measure the burden of ROS, muscle volume could be restored by EVT in a selected population.Enrolled patients had relatively larger stenoses than those in previous studies, and its revascularization had a larger impact on the diseased limbs [20].The disuse of skeletal muscle, even in the short term, leads to muscle atrophy [21], and improvement of symptoms in almost all patients facilitated the use of the thigh muscles which contributed to the muscle restoration in the present study.If EVT has the capability of modifying increased ROS production and mitochondrial function defects, which attracts clinical attention since the main factor in the theory that ROS is associated with degenerative disease and aging.Furthermore, the selection of patients might have had an impact on the results.
In the present study, since the PAD patients with more proximal lesions of above the knee of iliac, femoral, and popliteal arteries were enrolled [15], the negative association between PAD and change in the thigh muscle area was elucidated.The ndings of the present study were in line with those of a previous reports wherein diabetic patients were at risk of losing their skeletal muscle mass [2,22].In another study, a large numbers of PAD patients suffered from peripheral neuropathy complicated with diabetes, which led to immobility losing their skeletal muscle [20].Immobilized PAD patients do not move and consult a doctor until the symptoms are aggravated to the level of ulceration with ischemic paint at rest [23].EVT is bene cial in a selected PAD population (such as in this study) without apparent immobility who are in the relatively early stage of PAD with proximal lesions.The prevalence of diabetes is higher in PAD patients [24], and diabetes itself is associated with excessive loss of skeletal muscle tissue [2].However, early intervention would either stop or at least delay the loss of skeletal muscles which may have bene cial effects in patients with PAD.This study will help to pave the way for other randomized clinical trials investigating the true role of EVT in patients with intermittent claudication resulting from proximal PAD lesions [25].
This study had several limitations.First, this study was preliminary and was conducted with small number of patients.Since this study was statistically underpowered due to slow recruitment and single-centered nature, a larger number of patients enrolled in multiple centers would unveil considerably more robust information concerning PAD patients.Secondly, the observational period was 6 months, which was relatively short.It is necessary to observe over a longer period to assess whether the effects of EVT continue.Thirdly, we excluded patients solely with below the knee lesions and patients who were receiving maintenance hemodialysis, in whom the impact of EVT on muscle volume is not clear.Fourthly, most of the patients were male.We recommend including a larger number of patients, as well as an equal proportion of both sexes in future studies.

Conclusions
PAD induces thigh muscle atrophy, which can be restored by EVT, speci cally in normoglycemic patients.This preliminary study would pave the way to establish the favorable role of EVT on the skeletal muscle indicated the need of large-scale trial.Figures Abbreviations ABI, ankle brachial index