Analysis of factors influencing nutritional risk in patients with acute leukemia and its correlation with psychological status

Abstract

BACKGROUND

Chemotherapy is a primary treatment for acute leukemia (AL). Favorable nutritional status enhances tolerance to chemotherapy, alleviates adverse reactions, and reduces the occurrence of adverse events.

AIM

To analyze the factors influencing nutritional risk in patients with AL and examine its correlation with psychological status.

METHODS

Using convenience sampling, patients with AL treated at the Zhengzhou People’s Hospital between January 2022 and January 2025 were enrolled as study participants. Nutritional status was assessed using the Nutrition Risk Screening 2002 (NRS2002). Patients with scores ≥ 3 were classified as the nutritional risk group, and those with scores < 3 as the non-risk group. General and clinical data were collected from patients. Anxiety, depression, and psychological resilience were evaluated using the Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), and Connor-Davidson Resilience Scale, respectively. Pearson correlation analysis assessed the relationships between nutritional risk and psychological variables. Factors influencing nutritional risk in patients were identified using univariate and logistic regression analyses.

RESULTS

Among 174 patients, 81 had NRS2002 scores ≥ 3, resulting in a nutritional risk rate of 46.6%. Educational level, personal monthly income, disease perception, number of chemotherapy cycles, and lowest hemoglobin level were identified as influencing factors through univariate analysis. Except personal monthly income, all these factors were confirmed as independent factors by multivariate logistic regression. Compared with the non-risk group, patients at nutritional risk had notably higher SAS and SDS scores and lower scores across all dimensions of psychological resilience. Nutritional risk was positively associated with anxiety and depression and negatively associated with psychological resilience (self-reliance, optimism, and tenacity). Multivariate logistic regression further showed that SAS and SDS scores were risk factors for nutritional risk, whereas tenacity, self-reliance, and optimism were protective factors.

CONCLUSION

Educational level, personal monthly income, disease perception, number of chemotherapy cycles, and lowest hemoglobin level are key factors influencing nutritional risk in patients with AL. Nutritional risk is significantly correlated with psychological status, and poorer psychological status increases the risk of malnutrition.

Key Words: Acute leukemia; Nutritional risk; Psychological status; Psychological resilience

Core Tip: Malnutrition is common in patients with acute leukemia (AL) due to the combined effects of the disease and anticancer therapy. AL, as a major stressor, increases patients’ physical and psychological burdens, which may impair treatment compliance and disease control. However, there is limited evidence on the relationship between psychological status, psychological resilience, and nutritional risk in patients with AL. Therefore, this study investigates the nutritional status of patients with AL, compares relevant data across nutritional groups, and examines psychological factors influencing nutritional risk in this population and its correlation with psychological status.

INTRODUCTION

Acute leukemia (AL) is a hematological disorder featured by clonal proliferation and abnormal differentiation of hematopoietic stem cells; symptoms mainly including fever, anemia, and bone pain. This condition progresses rapidly and is associated with a poor prognosis[1,2], with chemotherapy being the standard therapy[3]. However, malnutrition affects up to 64.7% of patients with AL due to both disease-related factors and the effects of antitumor therapy[4]. Liver damage is often observed in patients undergoing chemotherapy, and complications such as anemia, coagulation disorders, and infections may occur in patients with poor nutritional status, leading to compromised treatment efficacy[5]. In addition, severe gastrointestinal side effects are generally noted following intensive chemotherapy, further increasing the risk of declined nutritional status[6]. Malnutrition is associated with prolonged hospitalization, reduced survival rates, increased mortality, fatigue, impaired physical function, and poor quality of life[7-10]. Conversely, adequate nutritional status is associated with enhanced tolerance to chemotherapy, alleviated adverse reactions, and reduced incidence of adverse events[11]. Hence, to optimize chemotherapy treatment and improve outcomes in patients with AL, it is important to initiate early nutritional risk screening and assessment coupled with appropriate nutritional interventions.

Long-term, regular, and multi-drug combination chemotherapy remains the primary treatment strategy for AL across the world[12]. Patients often suffer from adverse reactions or poor drug tolerance during chemotherapy, including gastrointestinal toxicity, myelosuppression, and hepatic and renal impairment, due to compromised immune function[13]. About 40%-50% of patients with AL die from to relapse after chemotherapy, which significantly increases their physical and psychological burdens[14]. As a major stressor, exacerbates these burdens, leukemia contributes to reduced treatment compliance and decreased disease control. The risk of malnutrition is also elevated in patients at high risk of depression[15]. Toyoda[16] demonstrated that in animal models, psychosocial stress interventions enhance psychological resilience and support nutritional recovery. Moreover, many studies across various diseases indicate that effective self-management behaviors improve nutritional status[17]. However, there are few studies on the correlation between psychological status, psychological resilience, and nutritional risk in patients with AL. Therefore, this study investigates the nutritional status in patients with AL, compares demographic and clinical data across nutritional groups, and analyzes the psychological factors to identify determinants of nutritional risk and its association with psychological status.

MATERIALS AND METHODS

Study participants

Convenience sampling was used to select patients with AL who visited the Zhengzhou People’s Hospital between January 2022-January 2025 as study participants. Inclusion criteria were as follows: (1) Age ≥ 18 years; (2) Diagnosis of AL based on clinical examination, bone marrow cytology, and relevant auxiliary tests; (3) Receipt of chemotherapy, absence of severe post-chemotherapy infectious disease, and an expected survival time ≥ 6 months after administration; (4) Ability to independently read or complete required questionnaires with investigator assistance; and (5) Complete clinical data. Exclusion criteria were as follows: (1) Incomplete clinical data; (2) Organ dysfunction or infectious disease; (3) Use of immunosuppressive drugs within the past 3 months; (4) Contraindications to chemotherapy; (5) Other malignancies; (6) History of steroid or immunosuppressive agent use or immune system disorders; (7) Pregnancy or lactation; and (8) History of bone marrow transplantation.

Survey tools

General information survey: Patient data, including sex, age, body mass index (BMI), educational level, personal monthly income, marital status, and occupational status, were collected from electronic medical records and a self-designed questionnaire.

Clinical data survey: Clinical variables included disease type and stage, underlying conditions, number of chemotherapy cycles, lowest hemoglobin level, and white blood cell count.

Nutritional risk assessment: Nutritional status was assessed using the Nutrition Risk Screening 2002 (NRS2002) scale, comprising nutritional status (0-3 points), disease severity (0-3 points), and age (≥ 70 years = 1 point; < 70 years = 0 points). The total score ranges from 0-7 points, with scores ≥ 3 indicating nutritional risk and < 3 indicating normal nutritional status.

Anxiety and depression assessment: Anxiety was assessed using the Self-Rating Anxiety Scale (SAS), a 20-item 4-point Likert scale. Raw scores were summed up and multiplied by 1.25, with the integer value was taken as the standardized score. A cutoff of 50 was used (50-59 indicate mild anxiety, 60-69 indicate moderate anxiety, and ≥ 70 indicate severe anxiety). The Cronbach’s α coefficient was 0.929, indicating high reliability and validity. Depression was assessed using the Zung Self-Rating Depression Scale (SDS), also comprising 20 items, each scored 1-4. The raw score was obtained by summing the scores of all 20 items. The statistical criterion for comorbid depression was defined as an SDS score index (total cumulative score/maximum score of 80) ≥ 50, with scores of < 50, 50-59, 60-69, and ≥ 70 indicating no depression, mild depression, moderate depression, and severe depression, respectively. The Chinese version was considered reliable and valid, as evidenced by the Cronbach’s α coefficient of 0.860.

Psychological resilience assessment: The Connor-Davidson Resilience Scale (CD-RISC) was employed to assess psychological resilience, which consists of 25 items across three dimensions: Tenacity, self-reliance, and optimism. Each item is scored from 0 (not true at all) to 4 (true nearly all of the time), yielding a total score of 0-100. Higher scores indicate greater psychological resilience. The Cronbach’s α coefficients were 0.85, 0.88, and 0.84 for the three dimensions.

Survey method and quality control

A pre-survey was conducted with validated scales. The questionnaire was revised based on identified issues to improve data accuracy. The researchers from the hospital’s hematology department received standardized training on psychological scale administration, including simulated survey scenarios and guidance on addressing patient queries to ensure consistency. After the training was completed, the formal survey commenced. Face-to-face assessments were performed following standardized instructions, and questionnaires were checked on-site for completeness and accuracy, with prompt correction of missing or inconsistent responses. For each questionnaire, approximately 30-40 minutes was required. The questionnaire could be completed over 1-3 days during hospitalization. Completed questionnaires were collected on-site and independently reviewed by two researchers for missing data or logical inconsistencies.

Statistical analysis

The database was established by the Microsoft Excel software, with double data entry and cross-checking. Statistical analyses were performed using SPSS 25.0. Continuous variables are expressed as mean ± SD; with comparisons between groups conducted using t-tests or one-way ANOVA as appropriate. Categorical variables are expressed as percentages and analyzed using the χ² test. Variables identified in univariate analysis were entered into multivariate logistic regression to determine factors influencing anxiety and depression in patients. Pearson’s correlation coefficient was used for correlation analyses. A two-tailed α level of 0.05 was applied, with P < 0.05 considered statistically significant.

RESULTS

General demographic data

A total of 188 questionnaires were distributed, of which 179 were returned. After excluding five invalid responses, 174 questionnaires were included in the analysis, yielding a response rate of 92.6%. Based on the NRS2022 scores, 81 of the 174 patients were classified as at nutritional risk and 93 as not at risk. Comparisons between the two groups showed significant differences in educational level, personal monthly income, and disease perception (P < 0.05), whereas no significant differences were observed in sex, age, BMI, marital status, or occupational status (P > 0.05; Table 1).

Table 1 Impact of general data on nutritional risk among patients.

	Risk group (n = 81)	Non-risk group (n = 93)	χ2/t	P value
NRS2002	3.59 ± 0.61	1.18 ± 0.39	31.55	< 0.0001
Gender			0.333	0.564
Male	41	43
Female	40	50
Age (years)	48.09 ± 6.65	49.41 ± 6.47	1.328	0.186
BMI (kg/m2)	22.37 ± 2.94	23.24 ± 3.30	1.821	0.070
Educational level			10.622	0.005
Junior high school or below	29	32
Senior high school/technical secondary school	38	26
College and above	14	35
Personal monthly income (RMB)			4.278	0.039
≤ 3000	51	44
> 3000	30	49
Marital status			1.547	0.461
Unmarried	22	28
Married	43	41
Divorced/widowed	16	24
Occupational status			2.549	0.280
Unemployed	27	34
Employed	30	41
Retired	24	18
Disease perception			9.304	0.002
Good	30	56
Poor	51	37

NRS2002: Nutrition Risk Screening 2002; BMI: Body mass index.

Clinical data of patients

Significant between-group differences were observed in number of chemotherapy cycles and lowest hemoglobin levels (P < 0.05). No significant differences were found in disease type and stage, underlying conditions, white blood cell count, or neutrophil count (P > 0.05; Table 2).

Table 2 Impact of clinical data on nutritional risk in patients.

	Risk group (n = 81)	Non-risk group (n = 93)	χ2/t	P value
Disease type			0.504	0.478
Leukemia lymphocytic	37	47
Myelocytic leukaemia	44	45
Disease stage			2.867	0.090
Initial treatment	47	42
Consolidation	34	51
Diabetes mellitus			0.584	0.445
Yes	21	29
No	60	64
Hypertension			0.140	0.709
Yes	24	30
No	57	63
Number of chemotherapy cycles (n)			6.745	0.009
≤ 3	26	48
> 3	55	45
Lowest hemoglobin level (g/ L)			12.872	0.002
< 60	37	19
60-90	29	52
> 90	15	22
White blood cell count (× 109/L)	13.20 ± 4.63	14.30 ± 5.43	1.429	0.155
Neutrophil count (× 109/L)	2.64 ± 0.77	2.87 ± 0.93	1.772	0.078

Multivariate analysis of nutritional risk in patients

Nutritional risk (present = 1, absent = 0) was used as the dependent variable. Five independent variables statistically significant in the univariate analysis (educational level, personal monthly income, disease perception, number of chemotherapy cycles, and lowest hemoglobin level) were entered into a logistic regression model. Multivariate analysis identified educational level, disease perception, number of chemotherapy cycles, and lowest hemoglobin level as independent factors associated with nutritional risk in patients with AL (P < 0.05; Table 3).

Table 3 Multivariate analysis of factors influencing nutritional risk in acute leukemia patients.

Variable	β	SE	Wald	P value	HR	95%CI
Constant	-0.435	0.568	0.587	0.443	0.647	-
Educational level (0: Junior high school or below)	-	-	7.328	0.026	-	-
1: Senior high school/technical secondary school	0.382	0.404	0.894	0.345	1.464	0.664-3.230
2: College and above	-0.922	0.466	3.912	0.048	0.398	0.159-0.992
Personal monthly income (0: > 3000 yuan, 1: ≤ 3000 yuan)	0.142	0.365	0.150	0.698	1.152	0.563-2.357
Disease perception (0: Poor, 1: Good)	-0.710	0352	4.078	0.043	0.492	0.247-0.979
Number of chemotherapy cycles (0: ≤ 3 cycles, 1: > 3 cycles)	0.775	0.346	5.003	0.025	2.170	1.101-4.278
Lowest hemoglobin level (0: > 90 g/L)	-	-	11.741	0.003	-	-
1: 60-90 g/L	-0.297	0.447	0.440	0.507	0.743	0.309-1.786
2: < 60 g/L	1.061	0.473	5.026	0.025	2.890	1.143-7.307

HR: Hazard ratio; CI: Confidence interval.

Anxiety and depression scores of all patients

The mean SAS and SDS scores for all patients were 50.8 ± 8.08 and 52.59 ± 10.09, respectively. Using the thresholds of ≥ 50 for anxiety and ≥ 53 for depression, 102 patients exhibited anxiety and 89 exhibited depression. Stratified analysis showed that patients at nutritional risk had significantly higher SAS and SDS scores than those without risk (P < 0.05; Table 4).

Table 4 Anxiety and depression scores in acute leukemia patients.

	Total score (n = 174)		Score	t	P value
SAS score	50.8 ± 8.08	Risk group (n = 81)	52.64 ± 6.92	2.754	0.007
		Non-risk group (n = 93)	49.68 ± 8.61
SDS score	52.59 ± 10.09	Risk group (n = 81)	56.15 ± 8.62	5.610	< 0.0001
		Non-risk group (n = 93)	50.36 ± 9.15

SAS: Self-Rating Anxiety Scale; SDS: Self-Rating Depression Scale.

Psychological resilience scores of all patients

The mean total psychological resilience score was 69.59 ± 11.39. Scores for tenacity, self-reliance, and optimism were 37.17 ± 8.15 points, 21.25 ± 5.45 points, and 11.16 ± 2.87 points, respectively. The non-risk group demonstrated significantly higher scores across all dimensions and total resilience compared with the risk group (P < 0.05; Table 5).

Table 5 Patient psychological resilience scores.

Item	Total score (n = 174)	Risk group (n = 81)	Non-risk group (n = 93)	t	P value
Total score of psychological resilience	69.59 ± 11.39	63.32 ± 8.95	75.04 ± 10.48	7.872	< 0.0001
Tenacity dimension	37.17 ± 8.15	33.81 ± 7.57	40.10 ± 7.51	5.482	< 0.0001
Self-reliance dimension	21.25 ± 5.45	19.38 ± 4.81	22.88 ± 5.47	4.451	< 0.0001
Optimism dimension	11.16 ± 2.87	10.12 ± 2.91	12.06 ± 2.53	4.707	< 0.0001

Correlation between nutritional risk score and psychological status

Nutritional risk correlated positively with anxiety and depression (SAS and SDS) and negatively with psychological resilience (self-reliance, optimism, and tenacity) (P < 0.05; Table 6).

Table 6 Correlation between patient Nutrition Risk Screening 2002 scores and psychological status.

Characteristic	NRS2002
	r	P value
SAS	0.170	0.025
SDS	0.372	< 0.0001
Tenacity dimension	-0.430	< 0.0001
Self-reliance dimension	-0.281	0.0002
Optimism dimension	-0.315	< 0.0001
Total score of psychological resilience	-0.521	< 0.0001

NRS2002: Nutrition Risk Screening 2002; SAS: Self-Rating Anxiety Scale; SDS: Self-Rating Depression Scale.

Impact of psychological status on nutritional risk in patients with AL

Logistic regression analysis, using nutritional risk (present = 1, absent = 0) as the dependent variable and psychological status scores as independent variables, showed that SAS and SDS scores were risk factors, whereas tenacity, self-reliance, and optimism were protective factors (P < 0.05; Table 7).

Table 7 Multivariate analysis of psychological status on nutritional risk occurrence in patients.

Variable	β	SE	Wald	P value	HR	95%CI
Constant	0.498	2.264	0.048	0.826	1.646	-
SAS	0.068	0.028	6.153	0.013	1.071	1.014-1.130
SDS	0.113	0.025	19.697	0.000	1.119	1.065-1.176
Tenacity dimension	-0.106	0.028	14.592	0.000	0.900	0.852-0.950
Self-reliance dimension	-0.135	0.042	10.530	0.001	0.874	0.806-0.948
Optimism dimension	-0.303	0.080	14.304	0.000	0.738	0.634-0.864

SAS: Self-Rating Anxiety Scale; SDS: Self-Rating Depression Scale; HR: Hazard ratio; CI: Confidence interval.

DISCUSSION

AL, characterized by abnormal hematopoietic stem cell function, is associated with genetic factors, environmental exposures, electronic device usage, and ionizing radiation and represents a common malignant hematological disorder[18,19]. Compared with other cancer patients, those with AL often have compromised immunity and are more susceptible to metabolic disorders. Poor nutritional status may in turn aggravate metabolic disturbances, weaken immune function, and adversely affect patient prognosis[20]. Therefore, assessing patients’ nutritional status during chemotherapy is essential.

In this study, 81 of 174 patients (46.6%) were identified as at nutritional risk based on NRS2002 scores. Comparative and multivariate analyses revealed that educational level, disease perception, number of chemotherapy cycles, and lowest hemoglobin level were independent factors influencing nutritional status in patients with AL patients. Higher educational level (college or above) and better disease perception were protective factors against nutritional risk. Given the prolonged course of AL, variable chemotherapy efficacy, and associated adverse reactions, patients with AL require long-term monitoring and follow-up[21]. Additionally, patients with higher educational levels may have greater access to health information and better understanding of nutritional management, which may support better adherence to treatment and dietary recommendations[22]. More than three chemotherapy cycles and a lowest hemoglobin level < 60 g/L were identified as risk factors for nutritional risk. Increased chemotherapy cycles may lead to cumulative toxicity, including gastrointestinal adverse effects and heightened susceptibility to infection, which can reduce nutrient intake and absorption. In addition, low hemoglobin levels are correlated with anemia and poor nutritional status[23]. These findings support the need for nutritional support strategies in patients with AL. The development of individualized interventions considering nutritional assessment and clinical status may help regulate dietary intake, enhance immunity, improve nutritional status, and mitigate chemotherapy-related adverse effects. As a result, patients had better treatment tolerance.

Subsequently, patients were subjected to psychological state assessment. Compared with the non-risk group, patients at nutritional risk exhibited notably higher SAS and SDS scores and lower psychological resilience scores across all dimensions and in total. AL typically presents with sudden and persistent high fever and rapid progression (positively correlated with severity). Once a diagnosis is confirmed, psychological distress such as anxiety, depression, fear, or even despair may be triggered, thereby affecting treatment compliance and future expectations. In addition, psychological burden is increased by factors such as complex treatment processes, high risk, and poor prognosis[24]. On the other hand, patients’ physical functions and quality of life can be impaired by chemotherapy-related toxicities, which further exacerbates their psychological burden. Body image disturbance, financial burden, poor prognosis, disease recurrence, and prolonged treatment duration are key contributors to anxiety and depression in patients with AL[25]. Psychological resilience, defined as the capacity to cope with and recover from adversity, was significantly lower in patients in the nutritional risk group, as evidenced by reduced CD-RISC scores across all dimensions. This suggests that these patients are more likely to adopt negative coping strategies, have limited ability to utilize resources to solve problems, and are prone to negative emotions, which subsequently negatively affect their nutritional status. Additionally, nutritional risk in patients with AL correlated positively with anxiety and depression (SAS and SDS) and negatively with psychological resilience (self-reliance, optimism, and tenacity). Multivariate logistic regression analysis further identified SAS and SDS scores as risk factors for nutritional risk, whereas tenacity, self-reliance, and optimism were protective factors. Negative emotions such as anxiety and depression may be associated with reduced appetite and inadequate nutrient and energy intake; they may also impair esophageal motility, increasing the risk of dyspepsia[26]. Moreover, anxiety and depression can result in long-term hyperactivity of the hypothalamic-pituitary-adrenal axis, leading to dysregulated cortisol secretion. Stress-induced catecholamine release may cause gastrointestinal motility inhibition and insulin sensitivity change, so that glucose utilization is impaired. Simultaneously, psychoneuroimmunological evidence suggests that depressive states are associated with increase in pro-inflammatory cytokine levels (e.g., interleukin-6 and tumor necrosis factor-α), which may arise from both psychological stress and depressive states. These inflammatory responses can promote muscle protein breakdown, reduce albumin synthesis, and disrupt iron metabolism, thereby increasing the risk of malnutrition[27]. By contrast, psychological resilience may achieve improved mental health and potentially better nutritional outcomes via mitigating negative emotions and enhancing stress tolerance.

Although psychological state was identified as a risk factor for nutritional risk, causal interference is precluded due to the retrospective nature of this study. Malnutrition itself may aggravate psychological distress. For example, the synthesis of neurotransmitters such as serotonin (5-HT) and dopamine can be directly impaired by deficiencies in nutrients such as tryptophan, B vitamins, and essential fatty acids, leading to depression and anxiety. Additionally, there was a close relationship between hemoglobin levels and nutritional risk; anemia-induced cerebral hypoxia and metabolic disturbances often manifest as irritability, low mood, and reduced psychological resilience. In the context of AL, psychological and nutritional factors may interact bidirectionally. Prospective cohort studies incorporating lagged analyses are required in the future to clarify the temporal relationships between these variables.

However, there are still some limitations in this study. First, due to the nature of the study (single-center, convenience-sample retrospective analysis) and the participants enrolled (all of whom were hospitalized patients receiving chemotherapy), selection bias may be introduced. Critically ill patients unable to continue inpatient treatment or those facing extreme financial hardship may have been excluded, limiting participant representativeness. In addition, the results obtained in this study may have been influenced by center-specific factors, including diagnostic and treatment practices, nursing models, dietary habits, and regional socioeconomic characteristics. Second, as a cross-sectional study, the findings reflect correlations at a single time point without longitudinal follow-up. Third, subgroup analyses by different molecular subtypes of AL or specific chemotherapy regimens were not conducted due to the relatively small sample size, so that potential differences in disease progression, drug metabolism, and stress responses may not be noticed, thereby limiting generalizability. Finally, although educational level was correlated with nutritional risk, potential mediating variables could not be quantitatively assessed. Therefore, large-scale, multicenter prospective cohort studies using methods such as structural equation modeling and cross-lagged analyses are warranted to elucidate causal pathways.

CONCLUSION

To sum up, the nutritional risk is high in patients with AL, influenced by factors, including educational level, personal monthly income, disease perception, number of chemotherapy cycles, and lowest hemoglobin level. Additionally, there is a significant association between nutritional risk and psychological status of patients, with poorer psychological status linked to higher nutritional risk. Therefore, it is important to integrate psychological assessment into nutritional management. Targeted interventions such as simplified educational tools for patients with limited education, education for caregivers, early psychological interventions (e.g., cognitive behavioral therapy or mindfulness), and chemotherapy stage-specific nutritional support should be initiated to improve treatment compliance, promote physical and mental well-being, and reduce nutritional risk.