Diagnostic Impact of Immunonutritional Indices in Peripheral Arterial Disease Severity and Complexity

Purpose We aimed to evaluate the relationships between the severity of peripheral arterial disease (PAD) and immune-nutritional status, as quantied by the prognostic nutritional index (PNI), controlling nutritional status score (CONUT), and modied Glasgow prognostic score (mGPS). Methods and results One hundred forty-eight patients with severe lower extremity PAD diagnosed by angiography constituted the study population. All patients had at least one severe supercial femoral arterial disease (SFA). According to the presence of a chronic total occlusion in SFA (SFA CTO), the study population was divided into two groups. Patients with SFA CTO (n: 84) had a signicantly lower PNI(43.4 ± 4.8 vs. 49.1 ± 4.7, p < 0.001), worse mGPS and CONUT scores than those without SFA CTO(n:64). A comparison of the three groups, those classied according to PNI tertiles (PNI ≥ 48, n:49; 43.8 < PNI < 48, n:49; PNI < 43.8, n: 50), revealed that a lower PNI was associated with the presence of a TASC C-D lesion in the femoropopliteal segment, SFA CTO, long lesion, and long CTO. Furthermore, PNI was found to be an independent predictor of SFA CTO in the logistic regression analysis. Diagnostic performance of PNI was better than mGPS and CONUT scores in ROC curve comparison. rst study demonstrating the relationship between PNI and PAD severity. Our results and the prognostic value of PNI should be validated in a larger prospective cohort.


Introduction
Lower extremity peripheral artery disease (PAD) is a manifestation of atherosclerotic vascular disease. Its severity and complexity is signi cantly associated with poor prognosis (1). Despite the omission of The Trans-Atlantic Inter-Society Consensus-II (TASC-II) classi cation in recent guidelines regarding various PAD types, it was used to assess PAD according to anatomic distribution and to re ect the severity and complexity of lesions (stenosis and occlusions) for many years (2). Clinical studies have revealed that the TASC-II classi cation has a prognostic impact in PAD and gives useful information when choosing a revascularization strategy (endovascular or surgical) (3)(4). The use of easily available diagnostic tools that can re ect the severity of PAD in clinical practice can allow patients to be detected and treated early and, consequently, improve their prognosis.
The increased prevalence of PAD in recent years, in addition to classical risk factors, hypertension, dyslipidemia, smoking, is believed to be mainly attributable to aging populations and the growing prevalence of risk factors, especially diabetes mellitus (5). However, traditional risk factors do not entirely describe the excess risk of PAD. The link between atherosclerosis and malnutrition increases the risk of cardiovascular mortality in elderly patients with poor nutritional status, as stated in previous studies (6).
Although it has been estimated to range from 20-50% in hospitalized patients (7)(8), malnutrition is commonly overlooked in clinical practice. Various methods and indices for nutritional evaluation have been advocated and used to predict unfavorable prognoses in some clinical settings (9)(10)(11). The modi ed Glasgow prognostic score (mGPS), prognostic nutritional index (PNI), controlling nutritional status (CONUT) score, and geriatric nutritional risk index (GNRI) are not only objective markers of malnutrition but also of immunological status (12). Since PAD is both an in ammatory disease and postulated to be more common in malnourished subjects, we hypothesized that these markers could be associated with PAD's complexity and severity as assessed by peripheral angiogram.

Study population
A consecutive 190 patients with PAD who underwent peripheral angiography at Gebze Fatih State Hospital between June 2019 and March 2020 were retrospectively screened. Exclusion criteria were active infections, a recent major surgical procedure, trauma, neoplastic disease, and previously diagnosed with systemic in ammatory disease, vasculitis, and malignancy. We also excluded patients with digital ulcers and with acute thromboembolic occlusions causing acute limb ischemia. The remaining 148 patients with signi cant PAD diagnosed via an angiogram constituted the study population. Demographic, clinical, laboratory, and procedural variables were obtained from reviewing our institution's electronic medical records.

Laboratory analysis
Before the peripheral angiography, blood samples were collected after twelve hours of fasting and analyzed in our institution's laboratory. Laboratory variables, including hemoglobin level, lymphocyte count, blood glucose, creatinine, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, and albumin evaluated. White blood cell, hematocrit, and platelet counts were measured as part of the automated complete blood count using a Coulter LH 780 Hematology Analyzer (Beckman Coulter Ireland, Inc, Galway, Ireland). The albumin levels were measured using a Roche Diagnostics Cobas 8000 c502 Analyzer (Indianapolis, USA). De nition and calculation of PNI (13), CONUT score (14), and mGPS (15) are presented in Table 1.

Angiographic analysis
PAD was evaluated on the peripheral angiogram, contralateral to the site of retrograde access, or ipsilateral to the site of antegrade access, in the lower extremities. Stenosis ≥ 70 % in diameter was de ned as a severe disease. The disease level was classi ed as suprapopliteal (iliac, femoral, and popliteal arteries), infrapopliteal (anterior tibial, peroneal, and posterior tibial arteries) arteries. Lesion characteristics were de ned regarding the angiographic ndings according to the TASC-II classi cation (3). A super cial femoral artery (SFA) lesion greater than 25 cm was described as a long lesion (16). Chronic total occlusion (CTO) is a lesion that was at least three months old. If the participant had never received imaging, a clinical de nition is used, which included the presence of an occluded SFA without lling defects in the body of the occlusion to indicate a more acute or subacute occlusion as well as having no clinical history of acute onset symptoms or other signs of acute limb ischemia (17).
Statistical analysis SPSS version 17.0 (SPSS Inc, Chicago, USA) was used to perform the statistical analysis. To assess the normal distribution of continuous variables, a Kolmogorov-Smirnov test was performed. Continuous variables with and without a normal distribution were expressed as a mean ± standard deviation and median, interquartile range. Categorical variables were expressed as percentages, and the chi-square test was used to compare the variables. Continuous and non-continuous variables between the two groups were compared using the student t-test and Mann-Whitney U test. Multivariable logistic regression analysis was used to determine independent predictors of the super cial femoral artery (SFA CTO). The predictive performance of PNI for diagnosing SFA CTO lesions was assessed using receiver-operating characteristic (ROC) curve analyses. The predictive validities of PNI, albumin, and lymphocyte count were quanti ed as the area under the ROC curves (c statistics). Comparing the c statistics were performed by the DeLong's test. A two-sided p-value of < 0.05 was considered signi cant.

Results
The cohort composed 148 patients (69 ± 10 years, 4.1% female) with severe PAD. All patients had a severe SFA lesion and participants were separated into two groups: patients with chronic total occlusion of SFA (n: 84) and patients without it (n: 64), as determined by peripheral angiography. The baseline characteristics and laboratory ndings of the patients were given in Table 2. Although 95.9% (n: 142) of study participants were male, female sex was more prevalent in patients with SFA CTO (p=0.029). The frequency of hypertension, diabetes mellitus, hyperlipidemia, coronary artery disease and smoking was comparable among patients with and without SFA CTO lesions. Ankle brachial indices of patients with SFA CTO were lower than without SFA CTO (0.629 ± 0.081 vs. 0.685±0.076, p < 0.001). Walking distance was shorter in patients with SFA CTO compared to those without SFA CTO [150 (120-185) vs. 190 meters (150-220), p < 0.001]. Comparison of laboratory parameters revealed that, subjects with SFA CTO had lower serum albumin levels compared to those without SFA CTO. Despite patients with SFA CTO had a lower leucocyte count, especially the lymphocyte subgroup, they had a higher neutrophil count and neutrophil to lymphocyte ratio in comparison to those without SFA CTO.
Nutritional indices and angiographic characteristics of all patients and comparison of two groups are listed in Table 3. Patients allocated in SFA CTO group not only had severe disease according to TASC classi cation, but also they had severe bilateral femoropopliteal disease, longer lesion and more prevalent long CTO and infrapopliteal disease compared to patients without SFA CTO. Among nutritional indices, PNI were signi cantly lower in patients with SFA CTO compared to those without SFA CTO (43.4 ± 4.8 vs. 49.1 ± 4.7, p < 0.001). Similarly, patients with SFA CTO had an increased mGPS and CONUT scores (Table 2). Furthermore, study population divided into three tertiles according to increased PNI values (PNI ≥ 48, n:49; 43.8 < PNI < 48, n:49; PNI < 43.8, n: 50). Patients in the lower PNI tertile tended to have SFA CTO more frequently when compared to those in the higher PNI tertile (80 vs. 69.4 vs. 20.4 %, p < 0.001, respectively). Patients with PNI less than 43.8 had worse femoropopliteal TASC-II grades compared to those with higher PNI values. Also, subjects in the lower tertile had more bilateral femoropopliteal TASC-II C-D lesions in comparison to those with higher PNI (8.2 vs. 36.7 vs. 34%, p = 0.005). Lesions longer than 25 cm, long CTOs and aortailiac disease were seen more frequently in patients with lower PNI when compared to those with higher PNI (p < 0.001). However, the frequency of infrapopliteal diseases was comparable among PNI groups (Table 4).
Multivariable logistic regression analysis were performed to examine the effect of confounding factors which chosen among comorbidities and laboratory parameters on the basis of our results and their clinical importance, on the performance of PNI in the diagnosis of SFA CTO. After adjusting clinically and statistically relevant confounding factors(female gender, ABI, walking distance, albumin, white blood cell and lymphocyte count), 1 (one) point decrease in PNI value was found to be associated with 23.0% increase in the diagnosis of SFA CTO in our study population(OR: 0.77; 95% CI; 0.69-0.85; p < 0.001).
ROC curve analysis showed that PNI ≤ 45.9 predicted the presence of SFA CTO with a sensitivity and speci city of 81.2% and 76.2% (Area under the curve (AUC): 0.807; 95% CI: 0.734-0.867 (p < 0.001)). Furthermore, ROC curve comparison revealed that PNI had better diagnostic performance than lymphocyte count, NLR, CONUT and mGPS scores (p<0.001 for each comparison) but not albumin (p=0,504); in predicting presence of SFA CTO (Figure 1). Result of ROC curve analysis of abovementioned parameters were listed in Table 5.

Discussion
The present study revealed that PNI is signi cantly associated with PAD severity and is an independent predictor for severe and complex femoropopliteal PAD as determined by the TASC-II classi cation in subjects diagnosed with peripheral angiography. Furthermore, PNI predicted occluded SFA more accurately than lymphocyte count, NLR, CONUT and mGPS scores.
Since systemic in ammation and malnutrition are related to cardiovascular events, the appraisal of immune-nutritional status in patients with cardiovascular diseases has gained popularity in recent years. PNI is feasible to e ciently and economically assess the in ammation-based nutritional situation and subsequent outcomes of patients with PAD. It requires simple and widely present objective markers like serum albumin levels and total lymphocyte counts. Many studies are demonstrating the prognostic signi cance of PNI for various medical conditions (18). Also, PNI is associated with poor prognosis in stable coronary artery disease (CAD) and heart failure. To our knowledge, no study has investigated the possible relationship between PNI and PAD severity. Our study demonstrated that decreased PNI was signi cantly associated with severe and complex femoropopliteal arterial disease.
Our observation could be explained by the effects of malnutrition and in ammation on serum albumin and lymphocyte. Malnutrition reduces albumin concentration by decreasing its synthesis rate. It leads to a reduced response to acute disease, exacerbates chronic diseases, and decreases long-term survival.
Malnutrition is associated with atherosclerosis development and increased cardiovascular mortality in older patients (19,20). It appears to be an independent predictor of poorer outcomes in PAD patients (21-27). In a cohort of patients treated with endovascular therapy for PAD, preprocedural malnutrition was a signi cant predictor of adverse outcomes (28). In agreement with the studies mentioned above, we observed that a decreased albumin level is related to the severity and complexity of the femoropopliteal disease. Also, malnutrition leads to a de ciency of essential vitamins and amino acids. The lack of these vital nutrients can depress cell-mediated immunity or humoral immunity. The number and function of both B and T cells would be signi cantly diminished, leading to a lower lymphocyte count (29).
In ammation plays an important role both in the initiation and progression of PAD. The in ammatory mediators participating in this process are similar to those contributing to the development of CAD (30-34). Furthermore, in ammatory markers, particularly CRP, are useful in predicting future cardiovascular events for CAD patients. Although it did not have statistical signi cance, we observed that the SFA CTO group has numerically higher CRP values compared to patients without SFA CTO. The in ammatory activity can have a signi cant impact on the serum albumin levels. Some cytokines, such as IL-2 and IL-6, inhibit albumin's hepatic synthesis (35-37). Systemic in ammation makes a patient susceptible to malnutrition, and malnutrition exacerbates in ammation, which leads to a vicious cycle. Furthermore, the lymphocyte count is affected in in ammatory situations due to an increased breakdown and a common nding in response to increased corticosteroids in stress conditions. Hypo-responsiveness of albumin and lymphocytes to disease states indicates an impaired immunonutritional status.
Although many sophisticated imaging tools are used to determine PAD and ABI's diagnosis and prognosis and are easily obtainable and reproducible, ABI remains valid. The test's validity in clinical settings has been proven in previous studies (38-40). In agreement with the literature, we found ABI to be an independent predictor of femoropopliteal disease severity.
There was no prognostic data because of the observational nature of our study. Also, being a retrospective and single-center design makes the study prone to bias. Despite the statistical signi cance being high, our relatively small sample size is another drawback. PNIs were evaluated only once; their changes with time during the follow-up period were not assessed. Although diagnostic performance of PNI seemed to be independent from albumin in multivariable logistic regression analysis, there was no statistically signi cant difference between their AUC values in ROC curve comparison. Further studies with larger population is needed to validate if PNI has better diagnostic performance than albumin. Neither sophisticated tools more clearly revealing malnutrition nor a high sensitive CRP showing baseline chronic in ammation were used.

Conclusion
The PNI should be considered a clinical tool and marker of disease severity and complexity in PAD. It can be utilized to determine high-risk patients who may bene t from more aggressive therapy and close follow-up. As we explained in detail, a novel predictive parameter, PNI, was introduced to patients with PAD, which includes both albumin and lymphocyte counts. PNI brings these two essential measurements together and presents a unique prognostic value.  Tables   Table 1 De nition immunonutritional indices