Nutritional status and depressive symptoms in patients with hepatocellular carcinoma undergoing transcatheter arterial chemoembolization: association with inflammation

Abstract

Purpose: To evaluate whether there are relationships among nutritional status, depressive symptoms and inflammation in patients with hepatocellular carcinoma (HCC) undergoing transcatheter arterial chemoembolization (TACE).

Methods: One hundred thirty patients with HCC undergoing TACE were recruited for this study. Nutritional status and depressive symptoms were assessed after TACE by using questionnaires derived from the Patient-Generated Subjective Global Assessment (PG-SGA) and Hospital Anxiety Depression Scale depression subscale (HADS-D), respectively. Systemic inflammation was evaluated using the systemic immune-inflammation index (SII), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR).

Results: A total of 89.2% of the patients were malnourished, and 58.5% of the patients presented with depressive symptoms. Poor nutritional status was significantly correlated with depressive symptoms (r=0.651, p<0.001), whereas no significant correlations were found between nutritional status and NLR (r=0.085, p=0.066), PLR (r=0.162, p=0.066), or SII (r=0.130, p=0.140). Depression was significantly correlated with PLR (r=0.294, p=0.001) and SII (r=0.197, p=0.024), but there was no correlation between depressive symptoms and NLR (p >0.05). A multiple linear regression model showed that the PG-SGA score (β=0.469 (95% CI: 0.351-0.586), p<0.01) was independently associated with depression symptoms, but there was no significant association between PLR and depression (β=0.003 (95% CI: -0.002-0.007)), p=0.230). There was no significant interaction effect of PG-SGA × PLR on depression (β=0.000 (95% CI: -0.001-0.001), p=0.596). A similar multiple linear regression model indicated no independent effect of a SII or PG-SGA × SII interaction on depression (β=0.000 (95% CI: -0.001-0.001), p=0.926, β = 0.000 (95% CI: 0.000-0.000), p=0.513), but PG-SGA was associated with depression (β=0.472 (95% CI: 0.353-0.591), p＜0.001).

Conclusion: Depression was significantly correlated with nutrition, PLR and SII, although the potential mechanism underlying the correlation between nutritional status and depression is not clearly related to inflammation.

Introduction

Hepatocellular carcinoma (HCC) is the most frequent type of primary liver cancer and the third most common cause of cancer-associated death worldwide [1]. Currently, liver resection is a potential radical treatment for HCC [2]. However, most patients cannot undergo HCC resection due to hepatic dysfunction, advanced clinical stage, or poor performance status [3]. Transcatheter arterial chemoembolization (TACE) is recommended as a palliative therapy for these patients to relieve symptoms and extend survival [3, 4]. Nevertheless, TACE treatment may cause malnutrition due to postembolization syndrome (e.g., transient fever, abdominal pain, nausea, and vomiting), leading to a prolonged length of stay [5], poor health-related quality of life [6], increased rate of readmission [7], and worse prognosis.

In addition, patients undergoing TACE often experience anxiety disorders and depression, which likely contributes to their poor nutritional status [8–10].

The inflammatory response plays an important role in the pathogenesis of nutrition and depression. The neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR), as indicators of the systemic inflammatory state [11], have been proven to be connected with depression symptoms and nutritional status [12]. Recently, the systemic immune-inflammation index (SII), a new biological parameter, was also proposed for the evaluation of inflammatory status [13, 14].

Previous studies reported that nutritional status was statistically associated with inflammatory markers in patients with advanced cancer [15] and patients with chronic radiation enteritis [16]. Recent studies suggest that the NLR is related to nutritional status in patients with cancer [17, 18]. The NLR value can decline due to improved nutritional status [19].

High NLR and PLR values were found to be associated with suicidal behavior in depressed and anxious children and adolescents [20]. Significant correlations were found among the PLR, SII and depression [21, 22]. However, few studies have focused on the NLR, PLR and SII as inflammatory markers in patients with hepatocellular carcinoma undergoing TACE.

Hence, this study investigated the association between nutritional status and depressive symptoms in patients with HCC who had undergone TACE treatment and examined whether the potential mechanism underlying this relationship is associated with inflammation.

Materials And Methods

Study population

A prospective cohort study was conducted in our hospital from November 2021 to April 2022, and 177 patients with HCC undergoing TACE were recruited. The inclusion criteria were as follows: 1) HCC diagnosed by a gastroenterologist, 2) age ≥ 20 years, 3) liver function categorized as Child–Pugh class A or B, 4) the administration of TACE treatment, and 5) agreement to participate in this study. Patients with liver function categorized as Child–Pugh class C (n = 14), extrahepatic metastases (n = 31)and mental disorders (n = 2༉were excluded. Finally, 130 patients were included in this study (Fig. 1).

Data collection

Standardized protocols were used to collect sociodemographic characteristics and clinical data (age, sex, hypertension, diabetes mellitus, Child–Pugh classification (Child–Pugh class A and B), type of TACE, number of TACE sessions and laboratory tests). The type of TACE included conventional TACE (c-TACE) and drug-eluting bead TACE (DEB-TACE). The final stage of the data collection involved the collection of nutritional status and depressive symptom information with the Patient-Generated Subjective Global Assessment (PG-SGA) [23] and HADS-D subscale [24], respectively.

Assessment of nutritional status

We evaluated the nutritional status of cancer patients in clinical practice using the PG-SGA. The PG-SGA contains two parts that are used to evaluate the nutritional status of cancer patients. The first part (including anthropometric data, changes in diet, symptoms, activity and function (Boxes A-D)) was filled out by the patients to assess their nutritional status. The second part was completed by medical staff to evaluate the disease status, metabolic stress, and physical examination results [25]. The PG-SGA categorizes nutritional status into three classes: well-nourished (SGA-A, score 0–1), suspected or moderately malnourished (SGA-B, score 2–8)) and severely malnourished (SGA-C, score ≥ 9) [26]. The total scores range from 0 to 16, with higher scores indicating worse outcomes.

Weight and height were measured with the patient wearing light clothes and no shoes at discharge, and the body mass index (BMI) was calculated as the weight in kilograms divided by the square of the height in meters. The BMI results were classified into four groups according to the recommendations of the World Health Organization: underweight (< 18.5 kg/m2), normal weight (18.5 ~ 25 kg/m2), overweight (25 ~ 30 kg/m2) and obese (> 30 kg/m2).

Assessment of depressive symptoms

The Hospital Anxiety and Depression Scale (HADS) is a patient self-report questionnaire, and its validity and reliability have been demonstrated in cancer patient populations [24, 27]. The depression subscale of the HADS (HADS-D) is a seven-item questionnaire used to screen for the presence of depressive symptoms. The total scores on the HADS-D range from 0 to 21; 0 and 7 indicate normal, and scores above 8 indicate the presence of depressive symptoms.

Laboratory tests

Routine blood samples were collected from the peripheral veins, and hematologic and biochemical markers were obtained, including erythrocytes, hemoglobin, albumin, leukocytes, absolute neutrophil count, absolute lymphocyte count, platelet count, bilirubin, alanine aminotransferase (ALT), and aspartate aminotransferase (AST). The NLR, PLR, and SII were calculated as follows [54]: NLR = neutrophil count (10⁹/L)/lymphocyte count (10⁹/L), PLR = platelet count (10⁹/L)/lymphocyte count (10⁹/L), SII= (platelet count (10⁹/L) × neutrophil count (10⁹/L))/lymphocyte count (10⁹/L)

Ethical considerations

The study was approved by the Institutional Review Board of our hospital (Approval Reg Number JNU20220310IRB23). All the participants signed informed written consent forms prior to the study.

Statistical analysis

Data were analyzed with the SPSS 22.0 statistical package (SPSS Inc., Chicago, IL, USA), and p < 0.05 was considered statistically significant. Quantitative variable normality was assessed using the Kolmogorov–Smirnov test. Continuous variables are expressed as the means ± standard deviations or medians (quartiles) depending on whether the data were normally distributed, and categorical variables are described as percentages. Spearman’s rank (nonnormally distributed variables (e.g., PG-SGA, HADS-D)) correlation coefficient was used to assess the associations of clinical characteristics with PG-SGA and depression. The Mann–Whitney U test was used to determine differences in qualitative independent variables between groups, with the PG-SGA score or presence of depression as the continuous dependent variable. Furthermore, variables differed significantly in univariate analyses, and sex and age were incorporated into a multiple linear regression model to analyze their effect on depression.

Results

Study sample and patient characteristics

A total of 130 patients with HCC undergoing TACE were included (Fig. 1). This sample consisted of 111 (85.4%) males with a mean age of 63 years. Baseline characteristics are displayed in Table 1. The median number of TACE sessions was 2 (range: 1–5), and the proportion of patients receiving c-TACE was 66.2%. Patients were divided in two groups. In total, 76 (58.5%) patients were placed into the Child–Pugh class A group, and 54 (41.5%) patients were categorized into the Child-Pugh class B group. The median BMI of the patients was 22.0 (range: 19.6–24.5) kg/m². According to the PG-SGA evaluation, 89.2% of the patients had moderate or severe malnutrition (B and C), and the mean score was 8.0 (range: 5.0–11.0). The median HADS-D score was 8 (range: 6.0–11.0), and 58.5% of the patients had depression. The median NLR, PLR, and SII were 4.6 (range: 2.7–8.8), 132.1 (range: 76.0-225.7), and 616.1 (range: 266.1—1189.1), respectively.

Table 1

Clinical and demographic characteristics of participants
Variable	Total (n = 130)
Variable	Mean (± SD) or N (%)	Median (IQR)
Gender (%)
Female	19 (14.6)
Male	111 (85.4)
Age (years)	63.0 (± 9.7)
Hypertension (yes, %)	52 (40.0)
Diabetes (yes, %)	27 (20.8)
Child–Pugh class (%)
A	76 (58.5)
B	54 (41.5)
Type of TACE
c-TACE	86 (66.2)
DEB-TACE	44 (33.8)
Possible presence of anxiety (HADS-A ≥ 8)	48 (36.9)
Possible presence of depression (HADS-D ≥ 8)	76 (58.5)
Classification of PG-SGA
SGA-A	14 (10.8)
SGA-B	61 (46.9)
SGA-C	55 (42.3)
Classification of BMI
Underweight (<18.5)	19 (14.6)
Normal weight (≥ 18.5,<25)	82 (63.1)
Overweight (≥ 25,<30)	21 (16.2)
Obesity (≥ 30)	8 (6.2)
Number of TACE		2.0 (1.0–5.0)
Anxiety score (HADS-A)		7.0 (5.0–8.0)
Depression score (HADS-D)		8(6.0–11.0)
Body weight (kg)		22.0 (19.6–24.5)
BMI (kg/m²)		22.0 (19.6–24.5)
PG-SGA score		8.0 (5.0–11.0)
RBC (×10⁹/L)		3.7 (3.4–4.1)
Hemoglobin (g/dL)		117.0 (108.0-130.0)
Albumin (g/L)	33.5 (± 4.1)
WBC (×10⁹/L)		5.5 (4.2–7.9)
Absolute neutrophil count (mm3)		4.3 (2.6–6.4)
Absolute lymphocyte count (mm³)		0.8 (0.6–1.2)
Platelets(x10³/uL)		111.5 (78.3–166.0)
ALT(U/L)		47.0 (25.8–73.5)
AST(U/L)		45.5 (28.0-70.3)
bilirubin		23.3 (15.0-36.1)
NLR		4.6 (2.7–8.8)
PLR		132.1 (76.0-225.7)
SII		616.1 (266.1-1189.1)
Abbreviations: Data are presented as the N (%), Median (interquartile range) or Mean (± standard deviation). TACE, transcatheter arterial chemoembolization; HADS-A, Hospital Anxiety and Depression Scale-Anxiety; HADS-D, Hospital Anxiety and Depression Scale-Depression; PG-SGA, Patient-Generated Subjective Global Assessment; c-TACE, conventional TACE; DEB-TACE, drug-eluting beads TACE; BMI, body mass index; RBC, red blood cells; WBC, white blood cells; ALT, alanine aminotransferase; AST, aspartate aminotransferase; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index.

Associations between nutritional status, depressive symptoms and clinical characteristics.

Correlation analyses and Mann–Whitney U tests were conducted to verify the relationships between nutritional status, depressive symptoms and clinical characteristics (Tables 2, 3). There was a significant correlation between nutritional status and depressive symptoms (Table 2). The PG-SGA score tended to be higher in patients with worse depressive symptoms (r = 0.651, p < 0.001). PG-SGA was correlated with the number of TACE sessions (r=-0.254, p = 0.004). In addition, the PG-SGA score was negatively correlated with the hemoglobin and albumin levels (r=-0.196, p = 0.025; r=-0.231, p = 0.008, respectively) but positively correlated with the bilirubin level (r = 0.120, p = 0.174). Moreover, the PG-SGA score was not correlated with the NLR, PLR, or SII (r = 0.085, p = 0.066; r = 0.162, p = 0.066; r = 0.130, p = 0.140, respectively).

Table 2

Correlation analyses demonstrating the relationship between PG-SGA, HADS-D and clinical characteristics
Variable	PG-SGA		Depression (HADS-D)
Variable	r	p value	r	p value
Age (y)	0.099	0.264	0.030	0.736
Body weight (kg)	-0.138	0.116	-0.085	0.335
BMI (kg/m²)	-0.049	0.579	-0.056	0.529
Number of TACE	-0.254*	0.004*	-0.173*	0.049*
PG-SGA score	—	—	0.651*	<0.001*
HADS-D score	0.651*	<0.001*	—	—
RBC (×10⁹/L)	-0.043	0.628	-0.066	0.454
Hemoglobin (g/dL)	-0.196*	0.025*	-0.090	0.306
Albumin (g/L)	-0.231*	0.008*	-0.202*	0.022*
WBC (×10⁹/L)	0.004	0.967	0.007	0.935
Absolute neutrophil count (mm3)	0.036	0.687	0.049	0.581
Absolute lymphocyte count (mm³)	-0.118	0.182	-0.124	0.158
Platelets (×10³/u L)	0.068	0.440	0.188*	0.032*
ALT(U/L)	0.090	0.310	0.120	0.173
AST(U/L)	0.136	0.123	0.070	0.428
bilirubin	0.120*	0.174*	0.177*	0.044*
NLR	0.085	0.066	0.098	0.267
PLR	0.162	0.066	0.294*	0.001*
SII	0.130	0.140	0.197*	0.024*
Abbreviations: TACE, transcatheter arterial chemoembolization; HADS-D, Hospital Anxiety and Depression Scale-Depression; PG-SGA, Patient-Generated Subjective Global Assessment; BMI, body mass index; RBC, red blood cells; WBC, white blood cells; ALT, alanine aminotransferase; AST, aspartate aminotransferase; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index.
*p<0.05

Table 3

The relationships between PG-SGA and HADS-D scores and clinical characteristics
Variable	PG-SGA			Depression (HADS-D)
Variable	Median	Z	p value	Median	Z	p value
Gender		-1.356	0.175		-0.804	0.421
Female	10			9
Male	8			8
Hypertension (yes, %)	8	-0.103	0.918	8	-0.131	0.896
Diabetes (yes, %)	10	-2.383*	0.017*	9	-1.092	0.275
Type of TACE		-0.941	0.347		-0.659	0.51
c-TACE	8			8
DEB-TACE	8			9
Child–Pugh stage (%)		-1.293	0.196		-2.189*	0.029*
A	8			9
B	7			7
Abbreviations: TACE, transcatheter arterial chemoembolization; c-TACE, conventional TACE; DEB-TACE, drug eluting beads TACE.
*p<0.05

Depression was associated with the number of TACE sessions (r=-0.173, p = 0.049), albumin level (r=-0.202, p = 0.022), platelet count (r = 0.188, p = 0.032), bilirubin level (r = 0.177, p = 0.044), PLR (r = 0.294, p = 0.001), and SII (r = 0.197, p = 0.024) (Table 2).

In Table 3, no significant relationship was found between the PG-SGA score and any sociodemographic or clinical variables except for the prevalence of diabetes (Z=-2.382, p = 0.017). With regard to depressive symptoms, only the Child–Pugh classification (Child–Pugh class A and B) was associated with the HADS-D score (Z=-2.189, p = 0.029).

Multiple linear regression with depression as a dependent variable

As shown in Table 4, variables with significant differences in Tables 2 and 3, gender and age were included in the model. The PLR and SII were separately assessed for their impact on depression due to collinearity. In the first step of the model (Model 1), the Child–Pugh stage (Child–Pugh class A and B) and bilirubin level were significantly associated with depression (β=-1.430, p = 0.038, 95% CI: -2.782—-0.079; β = 0.033, p = 0.012, 95% CI: 0.008—0.059). When the PG-SGA was added to this model (Model 2), the explanatory powers of the model increased from 9.5–40.1% (adjusted R²). The results showed that the PG-SGA score and Child–Pugh stage (Child–Pugh class A and B) could significantly predict depression (β = 0.473, p<0.001, 95% CI: 0.355—0.590; β=-1.196, p = 0.033, 95% CI:-2.297—-0.095). In the model including both the PG-SGA score and PLR (Model 3), the PG-SGA and Child–Pugh stage (Child–Pugh class A and B) were still associated with depression (β = 0.469, p<0.001, 95% CI: 0.351—0.586; β=-1.200, p = 0.033, 95% CI:-2.299—-0.100, respectively), whereas the PLR was not significantly associated with depression (β = 0.003, p = 0.230, 95% CI: -0.002—0.007).

Table 4

Hierarchical multiple regression estimates of the impact of PG-SGA and inflammation (PLR) on depression
Variable	Model 1 ^#				Model 2 (Model 1 ་PG-SGA)^##					Model 3 (Model 2 + PLR) ^###					Model 4 (Model 3 + PG-SGA×PLR) ^####
Variable	β	p	95%CI	β		p		95%CI		β	p		95%CI		β	p	95%CI
Gender	1.289	0.150	-0.471—3.05	0.497		0.498		-0.949—1.942		0.514	0.482		-0.93—1.957		0.514	0.484	-0.934—1.962
Year	0.001	0.974	-0.065—0.067	-0.017		0.525		-0.071—0.037		-0.023	0.404		-0.078—0.032		-0.021	0.452	-0.076—0.034
Child–Pugh stage (%)	-1.430	0.038*	-2.782—-0.079	-1.196		0.033*		-2.297—-0.095		-1.200	0.033*		-2.299—-0.100		-1.193	0.034*	-2.296—-0.090
Number of TACE	-0.008	0.935	-0.207—0.191	0.057		0.488		-0.105—0.220		0.048	0.565		-0.116—0.211		0.052	0.533	-0.112—0.216
Albumin (g/L)	-0.100	0.205	-0.255—0.055	-0.008		0.907		-0.136—0.121		0.005	0.941		-0.125—0.135		0.006	0.927	-0.124—0.136
Platelets (×10³/u L)	0.006	0.214	-0.003—0.015	0.004		0.317		-0.004—0.011		0.001	0.753		-0.007—0.010		0.001	0.789	-0.007—0.010
Bilirubin	0.033	0.012*	0.008—0.059	0.017		0.128		-0.005—0.038		0.016	0.138		-0.005—0.038		0.016	0.136	-0.005—0.038
PG-SGA					0.473		<0.001*		0.355—0.590	0.469		<0.001*		0.351—0.586	0.508	<0.001*	0.320—0.695
PLR										0.003		0.230		-0.002—0.007	0.005	0.315	-0.005—0.015
PG-SGA×PLR															0.000	0.596	-0.001—0.001
F	2.929				11.789					10.681					9.583
Adjusted R²	0.095				0.401					0.403					0.400
F change	2.929				63.336					1.457					0.283
Significant F change	0.007				0.000					0.230					0.596
Abbreviations:
^#Model 1 includes variables that were associated with depression in Table 2 and Table 3, gender, and sex
^##Identical to Model 1, including PG-SGA.
^### Identical to Model 1, including PG-SGA and PLR.
^#### Identical to Model 1, including PG-SGA, PLR, PG-SGA×PLR
TACE, transcatheter arterial chemoembolization; HADS-D, Hospital Anxiety and Depression Scale-Depression; PG-SGA, Patient-Generated Subjective Global Assessment; PLR, platelet-to-lymphocyte ratio
*p<0.05

With regard to depressive symptoms, the size of the effect for nutritional status remained basically unchanged when PLR was included from Model 2 (Model 2: 0.473 (95% CI: 0.355—0.59) to Model 3: 0.469 (95% CI: 0.351—0.586) and increased the model fit (adjusted R²: 0.401–0.403). The addition of the PG-SGA × PLR interaction did not contribute significantly to this model (β = 0.000, p = 0.596, 95% CI: -0.001—0.001).

A similar model was generated to evaluate the contribution of SII instead of the PLR (Table 5). The model indicated that depression was associated with the PG-SGA score (β = 0.472, p<0.001, 95% CI: 0.353—0.591) and with Child–Pugh stage (Child–Pugh class A and B) (β=-1.200, p = 0.034, 95% CI: -2.308—-0.091) but not with SII (β = 0.000, p = 0.926, 95% CI: -0.001—0.001). There was no significant PG-SGA × SII interaction in this model (β = 0.000, p = 0.513, 95% CI: 0.000–0.000).

Table 5

Hierarchical multiple regression estimates of the impact of PG-SGA and inflammation (SII) on depression
Variable	Model 1 ^#			Model 2 (Model 1 ་PG-SGA)^##			Model 3 (Model 2 + SII) ^###			Model 4 (Model 3 + PG-SGA×SII) ^####
Variable	β	p	95%CI	β	p	95%CI	β	p	95%CI	β	p	95%CI
Gender	1.289	0.150	-0.471—3.050	0.497	0.498	-0.949—1.942	0.498	0.499	-0.955—1.950	0.501	0.497	-0.955—1.957
Year	0.001	0.974	-0.065—0.067	-0.017	0.525	-0.071—0.037	-0.018	0.523	-0.072—0.037	-0.015	0.589	-0.070—0.040
Child–Pugh stage (%)	-1.430	0.038*	-2.782—-0.079	-1.196	0.033*	-2.297—-0.095	-1.200	0.034*	-2.308—-0.091	-1.184	0.037*	-2.296—-0.071
Number of TACE	-0.008	0.935	-0.207—0.191	0.057	0.488	-0.105—0.220	0.057	0.492	-0.107—0.220	0.059	0.478	-0.105—0.223
Albumin (g/L)	-0.100	0.205	-0.255—0.055	-0.008	0.907	-0.136—0.121	-0.007	0.913	-0.136—0.122	-0.004	0.953	-0.134—0.126
Platelets (×10³/u L)	0.006	0.214	-0.003—0.015	0.004	0.317	-0.004—0.011	0.004	0.394	-0.005—0.012	0.003	0.444	-0.005—0.012
Bilirubin	0.033	0.012*	0.008—0.059	0.017	0.128	-0.005—0.038	0.017	0.133	-0.005—0.038	0.017	0.118	-0.004—0.039
PG-SGA				0.473	<0.001*	0.355—0.590	0.472	<0.001*	0.353—0.591	0.511	<0.001*	0.344—0.677
SII							<0.001	0.929	-0.001—0.001	0.000	0.528	-0.001—0.002
PG-SGA×SII										0.000	0.513	0.000—0.000
F	2.929			11.789			10.394			9.353
Adjusted R²	0.095			0.401			0.396			0.393
F change	2.929			63.336			0.008			0.430
Significant F change	0.007			0.000			0.926			0.513
Abbreviations:
^#Model 1 includes variables that were associated with depression in Table 2 and Table 3, gender, and sex
^##Identical to Model 1, including PG-SGA.
^### Identical to Model 1, including PG-SGA and SII.
^#### Identical to Model 1, including PG-SGA, SII, PG-SGA×SII
TACE, transcatheter arterial chemoembolization; HADS-D, Hospital Anxiety and Depression Scale-Depression; PG-SGA, Patient-Generated Subjective Global Assessment; SII, systemic immune-inflammation index.
*p<0.05

Discussion

This study revealed a significant correlation between nutritional status and depressive symptoms according to the PG-SGA and HSDS-D subscale scores in patients with HCC undergoing TACE. In addition, the PLR and SII were correlated with depression. Multiple linear regression showed that the PG-SGA was correlated with depression, whereas PLR and SII were not significantly correlated with depressive symptoms. The mechanism underlying the association between nutrition and depression does not seem to be straightforward. The interaction effect between the PG-SGA score and the PLR or SII was not significantly different, suggesting no synergistic effects in predicting depression.

Our study found that 89.5% of the patients were malnourished. Recently, similar proportions of patients with malnutrition as determined by the PG-SGA score was also reported in populations of patients with advanced lung cancer, gastric cancer and esophageal cancer (86.7%, 71.6%, and 83.8%, respectively) [28–30]. Furthermore, the prevalence of depression was 58.5%, which is similar to the results of previous studies conducted in advanced lung cancer or colorectal cancer patients [31, 32]. This suggests that there is a need to examine nutritional status and mental health to support the selection of interventions aimed at reducing the incidences of malnutrition and depression.

As expected, our observation suggested a positive correlation between nutritional status and depression, which is in agreement with the results in previous studies [33, 34]. Chabowski, Mariusz et al. assessed nutritional status using the Mini-Nutritional Assessment questionnaire and identified that better nutritional status was significantly associated with lower levels of depression in patients with lung cancer [33]. More recently, Sánchez-Torralvo, Francisco José et al. reported that depressive symptoms were 6.29 times more common in malnourished patients assessed with Global Leadership Initiative on Malnutrition criteria than in well-nourished patients [34]. Therefore, there may be some interaction between these factors.

In our study, Spearman’s correlation analysis showed that the PG-SGA score was not correlated with the NLR, PLR or SII. This contradicts the finding in the Cai et al. study that showed that the PG-SGA score was positively connected with the NLR and C-reactive protein level in patients with chronic radiation enteritis [16]. A study also showed that the PLR and NLR were negatively associated with nutrition in patients with colorectal cancer [35]. The inconsistency of these findings might be due to the investigation of different types of disease or differences in disease severity, which should be further confirmed.

Current insight into the effects of inflammation on psychological factors suggests that agents such as chemotherapeutic drugs, which can act as a trigger for inflammation, could specifically cause mental disorders [36, 37]. With respect to depressive symptoms, in accordance with the above conclusions, we also found that the PLR and SII were associated with depression. Therefore, a high PLR and high SII could be early indicators of depression in patients undergoing TACE, and the early detection of depression may improve patient quality of life. This will also help researchers determine how the inflammatory response interacts with depressive symptoms, promoting our understanding of the etiopathogenesis of depression.

Previously, evidence has shown that a higher NLR was positively associated with increased severity of depression symptoms and worse nutritional status in patients with cancer [38, 39]. However, our study found that there was no association of the NLR with depression or nutrition, which may be due to the fact that immune-related genes affect those relationships [40]. Interestingly, the NLR, an inflammatory marker, has recently been studied as a prognostic factor for survival in HCC patients who have undergone TACE [41, 42].

Nutritional status is associated with depressive symptoms because the microbiome and its metabolites affect brain activity and cognitive functions via the microbiota-gut-brain axis [43]. There are several information exchange networks, including the endocrine, central nervous, and immune systems, that connect the gut and brain. Meanwhile, the dysregulation of gamma-aminobutyric acid, serotonin, and dopamine levels was found to be connected with the pathogenesis of depression. Therefore, the interaction between nutritional status and depressive disorders could be the result of gut microbial-mediated neurotransmitter or neurotransmitter precursors entering the blood circulation and affecting the neuronal activity and cognitive functions of the brain. Some neurotransmitter precursors can pass through the blood–brain barrier and participate in the synthesis cycle of various neurotransmitters in the brain. Some neurotransmitters, such as glutamate and serotonin, can transmit sensory signals to the brain via the vagal pathway, all of which may lead to alterations in brain function and behavior [43, 44]. Evidence has shown that the abnormal expression and function of serotonin is related to the pathogenesis of anxiety-depressive disorder [45].

Limitations

There were some limitations of our study. First, the present study has a cross-sectional design, and the causality between nutritional status and depression should be interpreted carefully. Second, missing data, such as tumor size, tumor number and internal validation, may have biased the nutritional status diagnosis.

Conclusion

The present study indicates high prevalence of poor nutritional status and depressive symptoms in patients after TACE. In addition, the results support the presence of an interaction between nutritional status and depression, but it remains uncertain whether this relationship is associated with inflammation. The findings also help health care providers re-evaluate patient nutritional status and mental symptoms and further provide remote intervention based on networks.

Abbreviations

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BMI

body mass index

c-TACE

conventional TACE

DEB-TACE

drug-eluting bead TACE

HADS

Hospital Anxiety and Depression Scale

HADS-D

Hospital Anxiety Depression Scale depression subscale

HCC

hepatocellular carcinoma

NLR

neutrophil-to-lymphocyte ratio

PG-SGA

Patient-Generated Subjective Global Assessment

PLR

platelet-to-lymphocyte ratio

SII

systemic immune-inflammation index

TACE

transcatheter arterial chemoembolization

Declarations

Acknowledgments:

We would like to thank all patients who participated in this study. Authors are also grateful for the grants received from the Project of Wuxi Institute of Translational Medicine (LCYJ202222).

Funding

This study was supported by the Project of Wuxi Institute of Translational Medicine (LCYJ202222).

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Hongyan Duan, Jie Zhang, Peng Wang, Jiayan Guo, Jianfeng Zhang and Jianwei Jiang. The first draft of the manuscript was written by Hongyan Duan and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the local Clinical Research Ethics Committee (Approval Reg Number JNU20220310IRB23).

Consent to participate

Informed consent was obtained from all individual participants included in the study

Consent to publish

Not applicable.

Code availability

Not applicable.

References

Siegel RL, Miller KD, Jemal A (2019) Cancer statistics, 2019. CA Cancer J Clin 69:7–34.
European Association for the Study of the Liver. Electronic address: [email protected]. and European Association for the Study of the Liver (2018) EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol 69:182–236.
Forner A, Reig M, Bruix J (2018) Hepatocellular carcinoma. Lancet 391:1301–1314.
Ahmed S, de Souza NN, Qiao W, Kasai M, Keem LJ, Shelat VG (2016) Quality of Life in Hepatocellular Carcinoma Patients Treated with Transarterial Chemoembolization. HPB Surg 2016:6120143.
Gupta D, Vashi PG, Lammersfeld CA, Braun DP (2011) Role of nutritional status in predicting the length of stay in cancer: a systematic review of the epidemiological literature. Ann Nutr Metab 59:96–106.
Guo ZQ, Yu JM, Li W, Fu ZM, Lin Y, Shi YY et al (2020) Survey and analysis of the nutritional status in hospitalized patients with malignant gastric tumors and its influence on the quality of life. Support Care Cancer 28:373–380.
Reece L, Dragicevich H, Lewis C, Rothwell C, Fisher OM, Carey S et al (2019) Preoperative Nutrition Status and Postoperative Outcomes in Patients Undergoing Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy. Ann Surg Oncol 26:2622–2630.
Kris-Etherton PM, Petersen KS, Hibbeln JR, Hurley D, Kolick V, Peoples S et al (2021) Nutrition and behavioral health disorders: depression and anxiety. Nutr Rev 79:247–260.
Firth J, Gangwisch JE, Borisini A, Wootton RE, Mayer EA (2020) Food and mood: how do diet and nutrition affect mental wellbeing? Bmj 369:m2382.
Klimova B, Novotny M, Valis M (2020) The Impact of Nutrition and Intestinal Microbiome on Elderly Depression-A Systematic Review. Nutrients 12: 710.
Imtiaz F, Shafique K, Mirza SS, Ayoob Z, Vart P, Rao S (2012) Neutrophil lymphocyte ratio as a measure of systemic inflammation in prevalent chronic diseases in Asian population. Int Arch Med 5:2.
Chen H, Luan X, Zhao K, Qiu H, Liu Y, Tu X et al (2018) The association between neutrophil-to-lymphocyte ratio and post-stroke depression. Clin Chim Acta 486:298–302.
Wang L, Wang C, Wang J, Huang X, Cheng Y (2017) A novel systemic immune-inflammation index predicts survival and quality of life of patients after curative resection for esophageal squamous cell carcinoma. J Cancer Res Clin Oncol 143:2077–2086.
Hu B, Yang XR, Xu Y, Sun YF, Sun C, Guo W et al (2014) Systemic immune-inflammation index predicts prognosis of patients after curative resection for hepatocellular carcinoma. Clin Cancer Res 20:6212–6222.
Tan CS, Read JA, Phan VH, Beale PJ, Peat JK, Clarke SJ (2015) The relationship between nutritional status, inflammatory markers and survival in patients with advanced cancer: a prospective cohort study. Support Care Cancer 23:385–391.
Cai Z, Cai D, Yao D, Chen Y, Wang J, Li Y (2016) Associations between body composition and nutritional assessments and biochemical markers in patients with chronic radiation enteritis: a case-control study. Nutr J 15:57.
Siqueira JM, Soares JDP, Borges TC, Gomes TLN, Pimentel GD (2021) High neutrophil to lymphocytes ratio is associated with nutritional risk in hospitalised, unselected cancer patients: a cross-sectional study. Sci Rep 11:17120.
Sato Y, Gonda K, Harada M, Tanisaka Y, Arai S, Mashimo Y et al (2017) Increased neutrophil-to-lymphocyte ratio is a novel marker for nutrition, inflammation and chemotherapy outcome in patients with locally advanced and metastatic esophageal squamous cell carcinoma. Biomed Rep 7:79–84.
Ortiz-Reyes LA, Chang Y, Quraishi SA, Yu L, Kaafarani H, de Moya M et al (2019) Early Enteral Nutrition Adequacy Mitigates the Neutrophil-Lymphocyte Ratio Improving Clinical Outcomes in Critically Ill Surgical Patients. Nutr Clin Pract 34:148–155.
Amitai M, Kaffman S, Kroizer E, Lebow M, Magen I, Benaroya-Milshtein N et al (2022) Neutrophil to-lymphocyte and platelet-to-lymphocyte ratios as biomarkers for suicidal behavior in children and adolescents with depression or anxiety treated with selective serotonin reuptake inhibitors. Brain Behav Immun 104:31–38
Hu J, Wang L, Fan K, Ren W, Wang Q, Ruan Y et al (2021) The Association Between Systemic Inflammatory Markers and Post-Stroke Depression: A Prospective Stroke Cohort. Clin Interv Aging 16:1231–1239.
Huang G, Chen H, Wang Q, Hong X, Hu P, Xiao M et al (2019) High platelet-to-lymphocyte ratio are associated with post-stroke depression. J Affect Disord 246:105–111.
Bauer J, Capra S, Ferguson M (2002) Use of the scored Patient-Generated Subjective Global Assessment (PG-SGA) as a nutrition assessment tool in patients with cancer. Eur J Clin Nutr 56:779–785.
Zigmond AS, Snaith RP (1983) The hospital anxiety and depression scale. Acta Psychiatr Scand 67:361–370.
Detsky AS, McLaughlin JR, Baker JP, Johnston N, Whittaker S, Mendelson RA et al (1987) What is subjective global assessment of nutritional status? JPEN J Parenter Enteral Nutr 11:8–13.
Pham NV, Cox-Reijven PL, Greve JW, Soeters PB (2006) Application of subjective global assessment as a screening tool for malnutrition in surgical patients in Vietnam. Clin Nutr 25:102–108.
Moorey S, Greer S, Watson M, Gorman C, Rowden L, Tunmore R et al (1991) The factor structure and factor stability of the hospital anxiety and depression scale in patients with cancer. Br J Psychiatry 158:255–259.
Lin T, Yang J, Hong X, Yang Z, Ge T, Wang M (2020) Nutritional status in patients with advanced lung cancer undergoing chemotherapy: a prospective observational study. Nutr Cancer 72:1225–1230.
Nikniaz Z, Somi MH, Naghashi S (2022) Malnutrition and Weight Loss as Prognostic Factors in the Survival of Patients with Gastric Cancer. Nutr Cancer 4:1–6.
Pan P, Tao G, Sun X (2015) Subjective global assessment and prealbumin levels of esophageal cancer patients undergoing concurrent chemoradiotherapy. Nutr Hosp 31:2167–2173.
Mercadante S, Valle A, Cartoni C, Pizzuto M (2021) Insomnia in patients with advanced lung cancer admitted to palliative care services. Int J Clin Pract 75:e14521.
Aminisani N, Nikbakht HA, Shojaie L, Jafari E, Shamshirgaran M (2021) Gender Differences in Psychological Distress in Patients with Colorectal Cancer and Its Correlates in the Northeast of Iran. J Gastrointest Cancer 53:245–252
Chabowski M, Polański J, Jankowska-Polańska B, Janczak D, Rosińczuk J (2018) Is nutritional status associated with the level of anxiety, depression and pain in patients with lung cancer? J Thorac Dis 10:2303–2310.
Sánchez-Torralvo FJ, Contreras-Bolívar V, Ruiz-Vico M, Abuín-Fernández J, González-Almendros I, Barrios M et al (2022) Relationship between malnutrition and the presence of symptoms of anxiety and depression in hospitalized cancer patients. Support Care Cancer 30:1607–1613.
Bai X, Feng L (2020) Correlation between Prognostic Nutritional Index, Glasgow Prognostic Score, Systemic Inflammatory Response, and TNM Staging in Colorectal Cancer Patients. Nutr Cancer 72:1170–1177.
Miller AH, Raison CL (2016) The role of inflammation in depression: from evolutionary imperative to modern treatment target. Nat Rev Immunol 16:22–34.
Wang CI, Chu PM, Chen YL, Lin YH, Chen CY (2021) Chemotherapeutic Drug-Regulated Cytokines Might Influence Therapeutic Efficacy in HCC. Int J Mol Sci 22: 13627
Diaz-Martinez J, Campa A, Delgado-Enciso I, Hain D, George F, Huffman F et al (2019) The relationship of blood neutrophil-to-lymphocyte ratio with nutrition markers and health outcomes in hemodialysis patients. Int Urol Nephrol 51:1239–1247.
McFarland DC (2019) New lung cancer treatments (immunotherapy and targeted therapies) and their associations with depression and other psychological side effects as compared to chemotherapy. Gen Hosp Psychiatry 60:148–155.
Sunakawa Y, Yang D, Cao S, Zhang W, Moran M, Astrow SH et al (2018) Immune-related Genes to Dominate Neutrophil-lymphocyte Ratio (NLR) Associated With Survival of Cetuximab Treatment in Metastatic Colorectal Cancer. Clin Colorectal Cancer 17:e741-e749.
Jeong SW (2020) Neutrophil-to-lymphocyte ratio as a prognostic biomarker in hepatocellular carcinoma after transarterial chemoembolization. Ann Transl Med 8:1124.
Wang H, Lin C, Fan W, Zhang J, Zhang Y, Yao W et al (2020) Dynamic Changes in the Neutrophil-to-Lymphocyte Ratio Predict the Prognosis of Patients with Hepatocellular Carcinoma Undergoing Transarterial Chemoembolization. Cancer Manag Res 12:3433–3444.
Chen Y, Xu J, Chen Y (2021) Regulation of Neurotransmitters by the Gut Microbiota and Effects on Cognition in Neurological Disorders. Nutrients 13:2099.
Xia G, Han Y, Meng F, He Y, Srisai D, Farias M et al (2021) Reciprocal control of obesity and anxiety-depressive disorder via a GABA and serotonin neural circuit. Mol Psychiatry 26:2837–2853.
Helton SG, Lohoff FW (2015) Serotonin pathway polymorphisms and the treatment of major depressive disorder and anxiety disorders. Pharmacogenomics 16:541–553.