Introduction

Malnutrition, in terms of undernutrition, is defined as the deficient intake or uptake of nutrients that can manifest as stunting, wasting, underweight status, or micronutrient deficiencies. Malnutrition is a global health problem whose burden disproportionately affects low- and middle-income countries (LMICs). In 2014, approximately 462 million people were estimated to be underweight worldwide1. Elderly patients are particularly vulnerable to diseases2, cognitive impairment3, frequent hospitalization4, poor oral health5, and age-related changes in body physiology, appetite, and taste6,7,8. Malnutrition leads to decreased muscle and fat mass, jeopardizing muscular, cardiac, and respiratory functions. Furthermore, malnutrition can lead to impaired immunity, poor wound healing, and nutrient malabsorption9.

Although the global burden of chronic kidney disease (CKD) did not change between 1990 and 2017, hemodialysis has become a more common treatment option. The age-standardized incidence of hemodialysis increased by 10.7% during the same period, and more than 2.5 million people received hemodialysis in 201710. Malnutrition is common, particularly in patients undergoing hemodialysis, the prevalence of which ranges between 42.4 and 54.5%11,12,13. The serum albumin concentration and prealbumin concentration have been suggested to be indicators of nutritional status and to be predictors of morbidity and mortality in patients undergoing hemodialysis14,15. Malnutrition is also associated with cognitive decline16 and reduced quality of life17,18.

Pain is also a common complaint in patients undergoing hemodialysis. A systematic review by Brkovic et al. reported that the prevalence of pain among patients undergoing dialysis varies widely across settings but can reach 82% and 92% for acute and chronic pain, respectively19. In response to their pain, patients may have negative thoughts and emotions and adopt negative attitudes toward their treatment. Chronic pain can, for example, interfere with people’s participation in life and work activities20, drain cognitive capacity21, invoke hopelessness, and affect their will to live22,23. In addition, pain often worsens the experience of dialysis24 or dialysis-related procedures, such as cannulation25. Notably, pain is associated with a decrease in the glomerular filtration rate (GFR) and an increase in dialysis mortality in patients26,27.

The relationship between pain and nutritional status is bidirectional and complex. A study reported that hospitalized elderly patients with severe pain are nearly 1.4 times more likely to be at risk of malnutrition than patients without pain28. In another study conducted among elderly patients, chronic nonmalignant pain was significantly associated with poor appetite29. However, the mechanism and directionality of the pain-nutrition relationship remain unclear. In view of the potential association between pain and nutrition, improving nutritional status and dietary habits has been suggested as a treatment modality in pain management30.

Malnutrition and pain in patients undergoing hemodialysis have been extensively studied. In Palestine, multiple studies have indicated a high prevalence of malnutrition and pain in this patient population and revealed that various demographic, socioeconomic, and clinical factors are associated with both11,17,31,32. However, no global or local studies have explored the relationship between malnutrition risk and pain intensity in patients undergoing hemodialysis. This is especially crucial in light of the possible relationship between pain and malnutrition among other populations30. Exploring this relationship might contribute to improving hemodialysis care and establish the foundation for further research. This study aimed to investigate the demographic, socioeconomic, and clinical factors that influence pain intensity, with a focus on the risk of malnutrition.

Methods

Study design and settings

This was a single-center cross-sectional study conducted at An-Najah National University Hospital (NNUH) in Palestine from 2021 to 2022. The hospital has a major dialysis center whose catchment area is the northern governorates of the West Bank.

Population and sampling

A total population of 350 patients received dialysis at NNUH in 2019 according to the annual health records published by the Palestinian Ministry of Health33. The Raosoft online sample size calculator was used to determine the minimum sample size. Using a desired confidence level of 95% and an accepted margin of error of 5%, the minimum sample size was estimated to be 184. A larger sample of the total population was obtained via convenience sampling. Patients who had been undergoing hemodialysis sessions for more than three months before the study and were older than 18 years were included. Patients were excluded if they had cognitive or mental limitations that could interfere with their autonomy and understanding of the study.

Data collection: procedures, tools, and definitions of variables

Demographic, socioeconomic, and clinical data

The patients were invited to participate in the study and provided information using an interviewer-administered questionnaire, which was constructed and cross-checked for precision and clarity by professors and researchers with relevant experience. Data collection was performed during the dialysis sessions. The first part of the questionnaire included questions on demographic characteristics, including age, gender, height, weight, and socioeconomic characteristics, including residency (city, village, refugee camp), marital status (married, single), education level (did not receive formal education, primary level, high school-level, university-level), occupation (employed, unemployed), monthly income level (low-income level for an income < 2000 NIS; medium income level for an income ranging between 2000 and 5000 NIS; and high-income level for an income > 5000 NIS), and smoking status as a binary variable (smoker if a participant currently smokes and has smoked 100 cigarettes during a lifetime; and previously smoker and nonsmoker if otherwise)34.

The second part of the questionnaire included questions on clinical data, including the duration of dialysis (≤ 4 years, > 4 years); hours of dialysis per session (< 4 or ≥ 4); frequency of dialysis (2 sessions/week, 3 sessions/week, or ≥ 3 session/week); number of chronic diseases (< 4 or ≥ 4); number of medications used (< 4 or ≥ 4); history of kidney transplant (yes/no); and body mass index (BMI) classified into the following: underweight range, BMI < 18.5 kg/m2; normal weight range, BMI = 18.5–24.9 kg/m2; overweight range, BMI = 25.0–29.9 kg/m2l; and obese and morbid obese range, BMI > 30.0 kg/m235.

Malnutrition-Inflammation Score (MIS)

Malnutrition status was assessed using the 10-component Malnutrition-Inflammation Score (MIS), which relies on collecting clinical data and conducting physical examinations and laboratory tests. The ten components are a medical history of disease; dry weight change; gastrointestinal symptoms; ability to function at work; physical examinations of muscle wasting (examined in temporal muscle, rib, clavicular, scapular, quadriceal, and knee); depleted fat stores (examined below eyes, chest, bicep, and triceps, and biceps); BMI (with an assigned score of 0 if BMI ≥ 20 kg/m2; 1 if BMI = 18–19.9 kg/m2; 2 if BMI = 16–17.99 kg/m2; and 3 if BMI < 16 kg/m2); and serum albumin and total iron binding capacity (TIBC). Each of the ten components is graded on a 4-point intensity scale ranging from 0 (normal) to 3 (severe). The final score was calculated by summing each component score and ranged from 0 to 30, with an MIS < 9 indicating no-to-mild malnutrition, 9–18 indicating moderate malnutrition, and > 18 indicating severe malnutrition36,37. The MIS has been shown to be a valid predictor of morbidity and mortality in patients undergoing dialysis36,38.

Brief Pain Inventory (BPI)

The Brief Pain Inventory (BPI) is a self-administered questionnaire that evaluates pain experience across four domains: pain severity, pain interference with functioning, pain location, and pain relief39,40. This study used the two domains of pain intensity and pain interference with functioning, each of which was reported separately as a continuous variable out of 10. First, the domain of severity of pain captures temporal pain variations by subjectively rating pain on a scale from 0 (as mildest) to 10 (as the most severe) using four elements: pain at its least and worst severity during the last 24 h, its average severity, and its current severity at the time of completing the questionnaire. Second, the pain interference domain with functioning assesses pain interference through seven life activities and is graded from 0 (noninterfering) to 10 (most interfering). The final scores for pain severity and dysfunctional dysfunction are reported as the arithmetic means of the individual item scores of pain severity and interference with functioning, respectively41. The pain intensity domain of the BPI has demonstrated acceptable to excellent internal consistency42,43, reliability43,44,45, and responsiveness to changes43,46 in nonmalignant conditions and has good reliability and validity in measuring pain in Palestine18,47,48,49,50,51. We obtained permission to utilize the officially validated Arabic version of the BPI short form provided by the MD Anderson Cancer Center.

Data analysis

The data were analyzed using the 25th version of the Statistical Package for the Social Sciences (IBM-SPSS) software. The mean, range, and standard deviation are reported for normally distributed variables, such as age. Percentages and frequencies of categorical variables, including sociodemographic and clinical characteristics. The frequency, percentage, median, and mean rank were reported for continuous variables with a nonnormal distribution. The Shapiro‒Wilk test was used to assess the normality of the MIS and BPI scores. At the bivariate analysis level, the Mann–Whitney U test was used to analyze associations between pain and other variables in two categories, and the Kruskal–Wallis H test was used for variables in more than two categories. The MIS was defined as an ordinal variable at the bivariate level, as previously explained. In the multivariate analysis, a linear regression model was used to test the significance of the relationship between the BPI score and the variables that were found to be significant at the bivariate level. Based on the nonnormality assumption of variable distributions, the Spearman rank correlation method was used to examine the correlation between the BPI and MIS, both of which were defined as continuous variables. The significance level was determined with a p value less than 0.05.

Ethics approval and consent to participate

The study received approval from the Institutional Review Board (IRB) of An-Najah National University, and the necessary permission documents were issued. Participants were given the freedom to accept or decline the invitation to participate voluntarily. The verbal informed consent of each participant who agreed to participate in the study was obtained, ensuring the confidentiality of their data. The IRB of An-Najah National University specifically approved the use of informed verbal consent due to the nature of the study, where participants participated only in interviews and clinical examinations without any potential harm, as long as their privacy was maintained. The authors affirm that all the methods adhere to the relevant guidelines and regulations.

Results

Demographic and clinical characteristics

Of the 280 patients approached, 23 declined to participate (8.2%), and 27 (9.6%) were excluded based on the exclusion criteria. Of the final sample of 230 patients, 145 (63.0%) were males, and 85 (37.0%) were females. The average age of the participants was 58.3 years (range: 18–85 years, SD ± 14.5), and 126 (84.8%) participants were aged older than 60 years. Most of the participants were married (77.4%), unemployed (84.8%), had a low income (64.3%), and had a primary education (46.1%). Most of the participants resided in cities (47.4%) or villages (42.2%), while a tiny minority lived in refugee camps (10.4%).

For BMI, 78 (33.9%) participants were within the normal range, and another 78 (33.9%) were within the overweight range. Less than one-third (27.0%) of the participants were obese or had morbid obesity, while only 12 (5.2%) participants were underweight. More than half had an MIS within the normal-to-mild malnutrition range (54.8%), while slightly less than half had moderate or severe malnutrition (45.2%). Most of the participants were nonsmokers or previous smokers (76.5%). Most of the participants reported having fewer than three chronic diseases (52.2%) and taking more than four medications per day (80.9%). Most had been receiving hemodialysis three times a week (92.6%), for less than 4 years (68.3%), and for less than four hours per session (91.3%) (Table 1).

Table 1 Association between pain severity score and other sociodemographic and clinical variables.
Full size table

Prevalence of chronic pain and malnutrition

The MIS and pain intensity scores were found to be nonnormally distributed based on the normal and detrended normal QQ plots and the Shapiro‒Wilk test results, with a p value < 0.001. The prevalence of pain, regardless of intensity, was 47.0%. The median BPI was 0.0 (Q1–Q3 = 0.0–4.3), the median pain interference score was 0.0 (Q1–Q3 = 0.0–5.2), and the median MIS was 8.0 (Q1–Q3 = 6.0–11.0).

Factors associated with pain severity and interference

Bivariate analysis revealed that gender (p = 0.038), education level (p = 0.026), number of chronic diseases (p = 0.002) and MIS classification (p < 0.001) were significantly associated with the pain severity score (Table 1). The pain interference score was associated with gender (p = 0.011), marital status (p = 0.021), education level (p = 0.003), number of chronic diseases (p = 0.001), and MIS classification (p < 0.001) (Table 2). According to Spearman’s rank correlation, the MIS score was positively correlated with the pain severity score (ρ = 0.308, p < 0.001) and the pain interference score (ρ = 0.324, p < 0.001). The pain severity and interference scores were significantly positively correlated (ρ = 0.941, p < 0.001).

Table 2 Association between the pain interference score and other sociodemographic and clinical variables.
Full size table

Multiple linear regression analysis

According to the linear regression model, only the association with the MIS classification (p < 0.001) remained significant for the pain severity score (Table 3). The results of the linear regression models for the pain interference score demonstrated a different pattern than that of the pain severity score, and marital status (p = 0.045), the number of chronic diseases (p = 0.012), and the classification of MIS (p < 0.001) retained significance (Table 3).

Table 3 Multivariate linear regression analysis of pain severity and interference with functioning scores.
Full size table

Discussion

Hemodialysis, a renal replacement therapy, is an effective solution for treating renal failure. However, this solution comes at physical and psychological costs. Pain and malnutrition, particularly during dialysis, are common and are associated with increased morbidity and mortality. Given that little research has investigated the interaction between pain and nutritional status, this study focused on exploring the correlation between pain and the risk of malnutrition, in addition to the possible factors influencing pain in patients undergoing hemodialysis. The study showed that pain severity was associated with malnutrition, and pain interference with function was associated with malnutrition, marital status, and the number of comorbidities. The study also found a high prevalence of moderate to severe malnutrition.

The reported prevalence of pain in this study was 47.0%. Two meta-analyses of studies conducted among patients undergoing hemodialysis estimated a consistent and high prevalence of pain (60% and 60.5%, respectively), which is slightly greater than that reported in the present study52,53. Musculoskeletal, peripheral neuropathic, and dialysis-related causes are commonly cited by dialysis patients as perceived causes of pain49,54. Due to the high prevalence of pain and its association with mortality and deterioration of kidney function, pain management is key for improving health outcomes during hemodialysis. The responsibility of nephrologists to provide comprehensive dialysis care involves the ability to detect and treat pain while referring to pain specialists in advanced cases only. Therefore, fellowship training programs must integrate pain management into the curriculum, especially since nephrologists receive inadequate training in pain assessment and management55.

The prevalence of moderate to severe malnutrition in this study was 45.2%. This finding is in line with the findings of local studies that used the MIS tool, which ranged between 55.9 and 66%17,32. Globally, studies have used various tools or classifications to measure the risk of malnutrition but have consistently shown that the prevalence of malnutrition exceeds 50% in most cases36,56,57,58,59. Malnutrition in patients undergoing hemodialysis can be attributed to iatrogenic factors, such as dialysis-induced nutrient loss and inflammation60,61, and/or noniatrogenic factors, such as loss of appetite62, taste alteration63, and insulin resistance61. Furthermore, common comorbid conditions, such as heart failure, may contribute to the worsening of malnutrition in patients undergoing hemodialysis64. Given the impact of malnutrition on mortality, psychological state, and cognitive function, optimizing the nutritional status of patients undergoing hemodialysis is the key to improving health outcomes. Therefore, nutritional assessment and management, preferably by a nutritionist, should be an integral part of hemodialysis care.

In addition, malnutrition was significantly correlated with pain in the present study. While no study has investigated the relationship in this patient population, research conducted among other populations has shown similar associations, including among elderly patients28,65, patients visiting general practice66, and cancer patients living with cancer67,68. Another study conducted among a large sample of the public showed that patients with malnutrition are almost 1.5 times more likely to experience chronic pain69. However, the directionality and pathophysiology underlying this relationship are unclear. Malnutrition can cause pain through the effects of inflammation, oxidative stress, and the gut-brain axis30,70. Pain, on the other hand, can cause depression and affect the cognitive state, leading to loss of appetite and contributing to malnutrition29,71,72. The correlation found in the current study is particularly relevant to dialysis patients, as they are a population that has a high prevalence of malnutrition and pain, both of which are associated with considerable morbidity and mortality. Therefore, interventions and training programs should integrate simultaneous management of pain and malnutrition, as these interventions can be mutually beneficial, efficient, and effective.

According to our regression analysis, pain interference with functioning remained associated with the number of comorbidities. This finding is consistent with those of multiple local and global studies conducted among patients who were undergoing dialysis or diagnosed with CKD49,50,73,74,75. The presence of comorbid conditions may cause pain through various mechanisms. Diabetes, atherosclerosis, and bone diseases are common comorbid conditions that can cause pain in patients undergoing dialysis76,77,78. For example, diabetes can cause peripheral neuropathic pain78, which is especially relevant in patients diagnosed with CKD, as coexisting uremia can exacerbate neuropathy79,80. Furthermore, diabetes is associated with high levels of calcium and phosphate products, which can cause musculoskeletal pain76. These findings indicate that comorbid conditions may have synergistic effects on the experience of pain among patients on HD through interrelated mechanisms. The associations of pain with comorbidities and malnutrition further emphasize the need for a comprehensive patient-centered approach to the treatment of CKD in patients undergoing hemodialysis whereby a multidisciplinary team devises management plans customized to the individual patient. These plans should include pain management, nutritional optimization, and integrative treatment of comorbid conditions to reduce the mortality and morbidity associated with hemodialysis.

In addition, the associations of pain with other demographic and socioeconomic factors, including gender and education level, disappeared in the multivariate analysis, except for marital status, which was significantly associated with the pain interference score. These factors represent an interrelated web of environmental factors that are contextual, complex, and variable, especially in different settings. For example, pain in patients undergoing hemodialysis was associated with female gender in a Palestinian study49, while it was associated with male gender in a similar study conducted in Taiwan73. This adds to the complexity that arises from the influence of gender roles on pain expression and the possible biological and psychological factors that affect pain intensity in males and females81.

Additionally, this study revealed that most patients received dialysis for less than four hours per session (91.3%). The duration of the dialysis session is reflective of dialysis adequacy82,83,84. Flythe et al. conducted an observational study that included 10,571 patients and reported that those who underwent dialysis for less than four hours had 26% higher mortality than those who underwent dialysis for more than four hours85. Similar to the present study, previous local studies have revealed that the vast majority of patients undergoing dialysis in Palestine received less than 4-h sessions17,18,86. This might be attributed to the high costs associated with dialysis in Palestine, which was estimated at an average of 16,085 USD per patient per year87. The high cost is fully covered by the government, whose healthcare system faces constant financial crises with resultant shortages in human and physical resources, which explains the reasons behind the short duration of dialysis sessions in pursuit of maintaining services at a lower quality88,89.

This study was the first to specifically investigate the correlation between pain and malnutrition in patients undergoing hemodialysis, especially since these two conditions are common and crucial in determining the health outcomes of hemodialysis patients. This study has several limitations, however. First, the lack of temporal assessment inherent in the cross-sectional design precludes the establishment of a causal relationship between pain and malnutrition, which is especially important since the directionality of this relationship remains unclear in the literature. Second, while MIS has been found to be effective, its classification into mild, moderate, and severe has not been successful. The MIS score, a continuous variable that was measured without error, was redefined as a categorical variable whose categories might correspond to inaccurate thresholding of the original continuous values. This could have led to over- or underestimation of the proportion of patients with moderate to severe malnutrition.

Conclusions

Pain in patients undergoing hemodialysis is associated with mortality and deterioration of kidney function. Malnutrition can also lead to poor health outcomes and increase the odds of mortality in the same patient population. Given that research exploring the pain–nutrition nexus is scarce, this study aimed to explore the potential correlation between pain, on the one hand, and malnutrition and other sociodemographic and clinical factors, on the other, in patients undergoing hemodialysis. The study showed that malnutrition was associated with pain severity. Moreover, pain interference was associated with malnutrition, the number of comorbidities, and marital status. In agreement with most local and global studies, the prevalence of pain and malnutrition was high. To achieve better health outcomes, management plans should adopt an individualized approach that includes pain management, nutritional evaluation, education, and support, with attention given to comorbidities. Furthermore, pain management should be included in nephrology fellowship training to improve nephrologists’ skills in addressing and responding to pain symptomatology.