The Relationship between the SARC-F Score and the Controlling Nutritional Status Score in Gastrointestinal Diseases

We sought to examine the relationship between the SARC-F score and the Controlling Nutritional Status (CONUT) score in patients with gastrointestinal diseases (GDs, n = 735, median age = 71 years, and 188 advanced cancer cases). The SARC-F score ≥ 4 (highly suspicious of sarcopenia) was found in 93 cases (12.7%). Mild malnutritional condition was seen in 310 cases (42.2%), moderate in 127 (17.3%) and severe in 27 (3.7%). The median SARC-F scores in categories of normal, mild, moderate and severe malnutritional condition were 0, 0, 1 and 1 (overall p < 0.0001). The percentage of SARC-F score ≥ 4 in categories of normal, mild, moderate and severe malnutritional condition were 4.4%, 12.9%, 26.8% and 25.9% (overall p < 0.0001). The SARC-F score was an independent factor for both the CONUT score ≥ 2 (mild, moderate or severe malnutrition) and ≥5 (moderate or severe malnutrition). In the receiver operating characteristic (ROC) curve analysis for the CONUT score ≥ 2, C reactive protein (CRP) had the highest area under the ROC (AUC = 0.70), followed by the SARC-F score (AUC = 0.60). In the ROC analysis for the CONUT score ≥ 5, CRP had the highest AUC (AUC = 0.79), followed by the SARC-F score (AUC = 0.63). In conclusion, the SARC-F score in patients with GDs can reflect malnutritional status.


Introduction
Sarcopenia is characterized by generalized loss of muscle mass and muscle functional decline, resulting in physical frailty, cachexia and mortality [1,2]. Malnutrition, reticence, advanced malignancy-bearing status and persistent inflammatory status frequently observed in gastrointestinal diseases (GDs) are representative features associated with sarcopenia [1][2][3][4][5][6][7][8][9]. Reduced daily dietary intakes and deterioration of nutritional status can be also often seen in patients with GDs [8], and sarcopenia in patients with GDs is associated with poorer patient quality of life (QOL) and prognosis [8,10]. GD is a prime example of secondary sarcopenia due to the disease burden [10]. Nishigori, et al. reported that out of 199 patients with esophageal cancer, 149 patients (75%) had sarcopenia [11]. Huang, et al. reported that out of 173 patients with gastric cancer, 52 (30%) had sarcopenia [12]. In patients with colon cancer, 39-48% patients have been reported to involve sarcopenia [13,14]. In patients with pancreatic cancer, approximately 60% have been reported to have sarcopenia [15]. In patients with hepatocellular carcinoma, 11-66% have been reported to have sarcopenia [6]. 2 of 13 SARC-F (strength (S), assistance walking (A), rising from a chair (R), climbing stairs (C), and falls (F)) is a questionnaire with five questions for the purpose of screening for sarcopenia [16,17]. Subjects are asked to reply to the five questions on a scale of 0 to 2, and the sum is calculated [16]. Subjects with a SARC-F score ≥ 4 are considered to be highly suspicious cases for sarcopenia [16]. The SARC-F score can be linked to physical functional reserve and QOL [18,19]. SARC-F is recommended to be used as an initial screening method for sarcopenia in the current international guidelines [20,21], whereas its lower sensitivity for sarcopenia may be a problem [22][23][24]. On the other hand, the controlling nutritional status (CONUT) score is a nutritional assessment tool, reported in 2002, in which serum albumin value, peripheral blood lymphocyte count, and total cholesterol value are scored and a total score is calculated [25,26]. The level of malnutrition is evaluated according to four levels: normal, mild, moderate, and severe. The higher the score, the more severe the malnutrition [25,26]. The CONUT score has the advantage of being simple and quick to calculate. The CONUT is a well validated screening tool for malnutrition and it correlates well with patient-generated subjective global assessment [26].
To our knowledge, however, there have been no reports regarding the relationship between the SARC-F score and the CONUT score in patients with GDs. The SARC-F does not include questions about nutrition. These issues deserve to be addressed, which urged us to conduct the current study.

Patients and Our Study
In the Second Department of Internal Medicine of the Osaka Medical and Pharmaceutical University (OMPU) Hospital, sarcopenia risk was assessed using the SARC-F. As a rule, all hospitalized patients were asked to answer the SARC-F questionnaire on admission. Between May 2020 and April 2021, 735 Japanese GD patients with both the SARC-F score and the CONUT score could be found in our database. As described above, the SARC-F score and the CONUT score for each patient was evaluated. First, the SARC-F score and the percentage of patients with a SARC-F score ≥ 4 were compared based on the nutritional condition as evaluated by the CONUT score. Next, univariate and multivariate analyses for the malnutrition were performed. Finally, receiver operating characteristic curve (ROC) analysis for the malnutrition, etc. was performed. Cancer of Stage III or worse was defined as advanced cancer. The ethics committee of OMPU hospital provided ethical approval (approval number, 2021-095).

Statistics
In the analysis of continuous variables, the appropriate choice of Student's t test, Mann-Whitney U or the and Pearson correlation coefficient (r) was made to compare the two groups, and the appropriate choice between the ANOVA and Kruskal-Wallis tests was made to compare multiple groups. Continuous variables were shown as median (interquartile range, IQR). In the analysis of categorical variables, the Pearson χ 2 test was adopted to evaluate between-group difference. Multivariate logistic regression analysis related to the CONUT score ≥ 2 or the CONUT score ≥ 5 was done to extract independent variables using significant factors in the univariate analysis. Multivariate analysis using a cumulative logistic model was also performed. ROC analysis was done for the area under the ROC (AUC) and sensitivity/specificity, and the cutoff was adopted as the point where the sum of sensitivity and specificity is maximized. p = 0.05 was set at the significant level by the JMP ver. 15 (SAS Institute Inc., Cary, NC, USA). For the comparison between each group of two in multiple-group comparisons, the Bonferroni correction was adopted for adjusting type I error.

The Percentage of Subjects with SARC-F Score ≥ 4 Based on the Nutritional Condition
The percentage of SARC-F score ≥ 4 in categories of normal, mild, moderate and severe malnutritional condition were: 4.4% (12/271) in the normal, 12.9% (40/310) in the mild, 26.8% (34/127) in the moderate and 25.9% (7/127) in the severe, respectively, (p = Figure 1. The SARC-F score based on the nutritional condition.

The Percentage of Subjects with SARC-F Score ≥ 4 Based on the Nutritional Condition in Subjects with and without Advanced Cancer
The percentage of SARC-F score ≥ 4 in categories of normal condition, mild malnutritional condition, and moderate to severe malnutritional condition in patients with advanced cancer were: 4.2% (2/48) in the normal, 21.1% (16/76) in the mild, and 31.3% (20/64) in the moderate to severe, respectively, (p = 0.0090 (normal versus mild), p = 0.1803 (mild versus moderate to severe), p = 0.0002 (normal versus moderate to severe), and overall p = 0.0019) ( Figure 4A).

The Percentage of Subjects with SARC-F Score ≥ 4 Based on the Nutritional Condition in Subjects with and without Advanced Cancer
The percentage of SARC-F score ≥ 4 in categories of normal condition, mild malnutritional condition, and moderate to severe malnutritional condition in patients with advanced cancer were: 4.2% (2/48) in the normal, 21.1% (16/76) in the mild, and 31.3% (20/64) in the moderate to severe, respectively, (p = 0.0090 (normal versus mild), p = 0.1803 (mild versus moderate to severe), p = 0.0002 (normal versus moderate to severe), and overall p = 0.0019) ( Figure 4A). = 0.0019) ( Figure 4A).

The Percentage of Subjects with SARC-F Score ≥ 4 Based on the Nutritional Condition Stratified by the Anatomical Categories of Disease
The percentage of SARC-F score ≥ 4 in categories of normal condition, mild m tritional condition, and moderate to severe malnutritional condition in patients with

Multivariate Analysis Using Cumulative Logistic Model
The study cohort can be classified into four groups based on the CONUT score (0-1, 2-4, 5-8, and 9-12). Thus, we further performed multivariate analysis for the CONUT score using the cumulative logistic model. As shown in Table 3, SARC-F score, the SARC-F score, BMI, eGFR and CRP were independent predictors for the CONUT score.

ROC Analysis for the Malnutrition as Evaluated by the CONUT Score
ROC analysis for the malnutrition was performed. In the ROC analysis for the CONUT score ≥ 2, CRP had the highest AUC (AUC = 0.70), followed by the SARC-F score (AUC = 0.60). (Table 3) In the ROC analysis for the CONUT score ≥ 5, CRP had the highest AUC (AUC = 0.79), followed by the SARC-F score (AUC = 0.63). (Table 4 and Figure 7) The sensitivity, specificity and optimal cutoff point in each variable are summarized in Table 3. 68.2 0.32 CONUT; Controlling Nutritional Status, BMI; body mass index, eGFR; estimated glomerular filtration rate, CRP; C reactive protein, AUC; area under the receiver operating characteristics curve.

Figure 7.
Receiver operating characteristic curves for the CONUT score ≥ 2 (A-E) and for the CO-NUT score ≥ 5 (F-J). The vertical axis indicates sensitivity, and the horizontal axis indicates 1-specificity.

ROC Analysis of the CONUT Score for the SARC-F Score 4 or More
ROC analysis of the CONUT score for the SARC-F score 4 or more (highly possibility of sarcopenia) was also performed ( Figure 8A). The best cutoff point of the CONUT score was 4. The AUC and sensitivity/specificity was 0.70 and 58.1/74.5%. The prevalence of patients with SARC-F score ≥ 4 in patients with the CONUT score ≥ 4 was significantly higher than that in patients with the CONUT score <4 (24.8% (54/218) vs. 7.5% (39/517), p < 0.0001) ( Figure 8B). 3.11. ROC Analysis of the CONUT Score for the SARC-F Score 4 or More ROC analysis of the CONUT score for the SARC-F score 4 or more (highly possibility of sarcopenia) was also performed ( Figure 8A). The best cutoff point of the CONUT score was 4. The AUC and sensitivity/specificity was 0.70 and 58.1/74.5%. The prevalence of patients with SARC-F score ≥ 4 in patients with the CONUT score ≥ 4 was significantly higher than that in patients with the CONUT score <4 (24.8% (54/218) vs. 7.5% (39/517), p < 0.0001) ( Figure 8B).

Discussion
Sarcopenia research is rapidly advancing as the disease attracts more atte around the world. The development of new drugs such as anti-myostatin antibodi improve sarcopenia is also underway [27]. As described earlier, SARC-F is recomme to use as a first screening tool for sarcopenia in the current international guide [17,20,21]. The SARC-F score is reported to be associated with physical functional re and QOL [18,19]. The CONUT score is a well validated screening tool for malnutr [25,26,[28][29][30]. The usefulness of the CONUT score as a prognostic factor in various m nancies has been demonstrated [28,29,[31][32][33][34]. A nutritional assessment by the CO score is also recommended in patients with COVID-19 [35]. However, to our knowl no reports can be found comparing the SARC-F score and the CONUT score in pat with GDs. The current study seems to be the initial effort at clarifying these research q tions. A large cohort (n = 735) is one of the major strengths of the present study.
Overall, the SARC-F score and the percentage of patients with a SARC-F scor

Discussion
Sarcopenia research is rapidly advancing as the disease attracts more attention around the world. The development of new drugs such as anti-myostatin antibodies to improve sarcopenia is also underway [27]. As described earlier, SARC-F is recommended to use as a first screening tool for sarcopenia in the current international guidelines [17,20,21]. The SARC-F score is reported to be associated with physical functional reserve and QOL [18,19]. The CONUT score is a well validated screening tool for malnutrition [25,26,[28][29][30]. The usefulness of the CONUT score as a prognostic factor in various malignancies has been demonstrated [28,29,[31][32][33][34]. A nutritional assessment by the CONUT score is also recommended in patients with COVID-19 [35]. However, to our knowledge, no reports can be found comparing the SARC-F score and the CONUT score in patients with GDs. The current study seems to be the initial effort at clarifying these research questions. A large cohort (n = 735) is one of the major strengths of the present study.
Overall, the SARC-F score and the percentage of patients with a SARC-F score ≥ 4 were well stratified by the malnutritional status as assessed by the CONUT score in all analyses. Additionally, the SARC-F score was an independent factor associated with both the CONUT score ≥ 2 and the CONUT score ≥ 5, and in the multivariate analysis using the cumulative logistic model, the SARC-F score was consistently significant for the CONUT score. These observations imply that the SARC-F score accurately reflects the nutritional condition of patients with GDs and sheds some light on the understanding of the relationship between malnutritional status and sarcopenia in patients with GDs. The SARC-F score and the percentage of patients with a SARC-F score ≥ 4 between moderate and severe malnutritional condition were similar, so there is no need to separate the two conditions in light of the incidence of sarcopenia.
In our data, 63.1% (464/735) had mild or more advanced malnutritional status as assessed by the CONUT score at baseline. Considering our data, the evaluation for nutritional status in patients with GDs seems to be mandatory. The percentage of subjects with a SARC-F score ≥ 4 in advanced cancer patients with normal nutritional conditions was almost similar to that in non-advanced patients with normal nutritional conditions (4.2% versus 4.5%). In GD patients with normal nutritional condition, advanced cancer itself may not affect the incidence of sarcopenia. The optimal cutoff points of the SARC-F score for the CONUT score ≥ 2 and ≥5 were 1 and 2 in our ROC analysis. Patients with a SARC-F score of 1 or higher should be evaluated for nutritional status as if they are under malnutritional condition. In addition to blood tests, it is also important to ask the patient if he or she is eating well. The sensitivity and specificity of the SARC-F score for the CONUT score ≥ 2 were 42.0% and 74.5%, whereas those for the CONUT score ≥ 5 were 43.5% and 81.4%. As mentioned earlier, relatively lower sensitivity of the SARC-F score for sarcopenia may be a concern, which is similar to our results [22][23][24]. To the contrary, the optimal cutoff point of the CONUT score for the SARC-F ≥4 (highly possibility for sarcopenia) was 4 points. In patients with a CONUT score ≥ 4, evaluation of muscle strength and muscle mass should be considered.
In the ROC analyses both for mild or more severe malnutrition and for the moderate or more severe malnutrition, CRP had the highest AUC. Proinflammatory cytokines include IL-1β, IL-6, TNF-α, and IFN-γ, and they cause decreased appetite and increased energy expenditure, which can be related to our results [36]. Cachexia is a common form of disease-related malnutrition associated with inflammation and loss of skeletal muscle mass. Although cachexia is common in the daily clinical practice of patients with chronic wasting or inflammatory diseases, it is not fully recognized by medical professionals, and many cases of undiagnosed cachexia result in further worsening nutritional status and loss of skeletal muscle mass and physical function [37,38]. Clinicians should be fully aware of these. In our ROC analysis, the AUC of SARC-F for the mild or more severe malnutrition and for the moderate or more severe malnutrition was higher than that of BMI, which is a good indicator of nutritional status [39]. The SARC-F score rather than BMI can be a good indicator for malnutrition. We would like to emphasize this point. eGFR, on the other hand, was an independent variable associated with both the CONUT score ≥ 2 and ≥5 in our multivariate analysis. Intravascular dehydration due to malnutrition may be related to the present results [40].
Several limitations must be acknowledged in the current analysis. First, the current study was a cross-sectional study at a single hospital with a retrospective nature. Second, the exact number of patients with a definite diagnosis of sarcopenia was unclear in our data. Third, our study cohort was highly heterogeneous including broad spectrum of GDs. Nevertheless, our study results indicated that although CRP had the highest AUC for the malnutrition, the SARC-F score can correlate with the CONUT score in patients with GDs as well as CRP.
In conclusion, the SARC-F score in patients with GDs can reflect malnutritional status. The SARC-F score can be associated with various clinical parameters in patients with GDs.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jcm11030582/s1, Table S1: Details of diagnosis names (n = 735). Informed Consent Statement: Patient consent was waived due to the retrospective nature of this study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to personal information.