The impact of hand grip strength on quality of life, fatigue, physical performance, and functional capacity in patients with post-mastectomy lymphedema

Abstract

Objectives

Lymphedema, which may develop following mastectomy due to breast cancer, can negatively affect hand grip strength in women. This study aims to determine the effects of lymphedema on hand grip strength and quality of life (QoL) and to identify the factors influencing QoL.

Patients and Methods

This study was designed as a cross-sectional observational study involving 100 patients with post-mastectomy lymphedema. The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC-QLQ-C30) was used to assess QoL, and the Checklist Individual Strength (CIS-20) was utilized to evaluate fatigue. Hand grip strength was measured using a hand dynamometer.

Results

A significant difference was found between the hand grip strength measurements of the affected and non-affected upper extremities (p<0.001). Hand grip strength on the affected side was significantly correlated with both the functional subscales of the EORTC-QLQ-C30 and the CIS-20 (p<0.001).

Conclusion

This study objectively evaluated the impact of lymphedema on hand grip strength based on evidence. Assessing QoL, fatigue, and functional status also provided a comprehensive and objective understanding of the multifaceted effects of lymphedema. This study was not registered as a clinical trial.

Approximately 34% of all cancer diagnoses among specific female populations in the United States are attributed to breast cancer. While breast cancer is the second leading cause of cancer-related mortality among women overall, it disproportionately affects certain populations, ranking as the primary cause of cancer death among Black and Hispanic women.^[1] In Türkiye, approximately one in every four women diagnosed with cancer has breast cancer, making it the most common malignancy among females. Furthermore, breast cancer accounts for 19.6% of all female cancer-related deaths in the country, placing it at the top of the list in terms of mortality.^[2] Lymphedema develops in approximately 20% of breast cancer survivors, typically affecting the ipsilateral upper extremity. The five-year cumulative incidence has been reported as 11.9%.^[3] Lymphedema is defined as the pathological accumulation of protein-rich interstitial fluid due to an acquired or congenital disruption in lymphatic circulation.^[4] This condition frequently arises following surgical interventions that involve removal of lymph nodes from the axillary region. The degree of lymph node removal and the surgical method employed are critical factors influencing the risk of developing lymphedema. In addition to surgery, adjuvant treatments such as postoperative chemotherapy (CT) and radiotherapy (RT) have also been identified as contributing factors.^{[5, 6, 7]}

Long-term follow-up studies have revealed that breast cancer survivors tend to report lower scores reflecting quality of life (QoL) scores—especially those related to aspects like as physical functioning (PF), global health status (GHS), and role functioning (RF)—when compared to age-matched healthy control groups.^[8] Symptoms such as swelling, pain, altered skin appearance, and increased susceptibility to skin infections can negatively impact patients’ overall QoL.^[9] Studies have also reported markedly reduced QoL and sleep quality across different stages of lymphedema in specific patient populations post-mastectomy.^[9] From a functional standpoint, comparisons between lymphedema patients and healthy controls have demonstrated impaired hand function, particularly in grip strength.^[10] The current research investigates the difference in grip strength between the affected and non-affected upper extremities in individuals with post-mastectomy lymphedema.

This study represents a comprehensive investigation that evaluates multiple factors potentially influencing QoL. Notwithstanding the primary objective of assessing the impact of multiple factors on QoL, secondary analyses were carried out to examine variables that may influence hand grip strength. Unlike previous studies that primarily focused on limited correlations, the present research also employed regression analyses with a wide range of clinical variables, thereby enhancing both the originality and the clinical applicability of the findings.

PATIENTS AND METHODS

The current study utilized a cross-sectional observational approach and included 100 patients who had been diagnosed with post-mastectomy lymphedema and referred to the physical medicine and rehabilitation outpatient clinic within the past three years. Individuals were invited for reassessment, and data collection was performed between February and March 2025. In our study, which compared grip strength measurements between the affected and non-affected sides as part of the primary hypothesis, assuming a type I error (α) of 0.05, a power (1-β) of 0.80, and an effect size (Cohen’s d) of 0.3, the minimum required sample size for a two-tailed paired-sample t-test was calculated to be 90 participants.

The inclusion criteria were defined as being a female between the ages of 30 and 65, having a history of upper extremity lymphedema following mastectomy for breast cancer, and providing written informed consent. Lymphedema diagnosis was made by comparing circumferential measurements taken from specific anatomical points on both upper limbs. Measurements were obtained at the metacarpophalangeal joints, wrist (at the level of the styloid process), 10 cm distal and proximal to the elbow crease, and approximately 15 cm above the elbow in the upper arm region. A side-to-side difference of 2 cm or more at any of these sites was accepted as diagnostic for lymphedema.^[11]

Exclusion criteria included the presence of neurological, cardiovascular, or musculoskeletal conditions that could pose a risk during exercise testing, significant visual impairment, vestibular disorders, and pregnancy. Demographic and clinical variables such as age, body mass index (BMI), year of lymphedema diagnosis, number and type of CT cycles, history of RT, use of hormone therapy (HT), and cancer stage were recorded for all participants.

QoL was assessed using the Turkish version of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC-QLQ-C30), which has been previously validated for reliability and cultural adaptation. The EORTC-QLQ-C30 is a standardized, validated, cancer-specific QoL instrument that includes multiple subscales. It consists of a GHS/QoL scale, five functional subscales—PF/QoL, RF/QoL, emotional functioning (EF/QoL), cognitive functioning (CF/QoL), and social functioning (SF/QoL)—three symptom subscales (fatigue, pain, nausea/vomiting), and several single-item symptom measures (e.g., dyspnea, appetite loss, sleep disturbance, constipation, diarrhea, and financial difficulties). All assessments are rated on a scale ranging from 0 to 100 following the EORTC scoring manual. In this system, elevated scores on functional domains and GHS represent improved functioning and enhanced QoL, whereas higher values on symptom-related scales correspond to increased symptom severity and poorer health outcomes.^[12]

Fatigue was evaluated using the Checklist Individual Strength (CIS-20), a tool with established reliability and validity in Turkish populations. The CIS-20 consists of 20 items grouped into four key domains: subjective fatigue, concentration, motivation, and activity. Each response item is scored using a 7-point Likert format, with the overall score obtained by summing the responses. Higher scores indicate more severe fatigue symptoms, lower energy or drive, decreased cognitive focus, and reduced engagement in daily activities.^[13] The Charlson comorbidity index was used to evaluate coexisting medical conditions, and its prior validation in diverse clinical populations across Türkiye supports its relevance and applicability. The index includes a weighted scoring system for different chronic diseases. Scores range from 0 to 37—where elevated scores denote more complex comorbidity profiles.^{[14, 15]} Nutritional status was assessed using the mini nutritional assessment-short form (MNA-SF), a reliable and widely accepted tool designed to detect the risk of malnutrition in adults. The MNA-SF includes six questions, yielding a cumulative score between 0 and 14. Scores from 12 to 14 indicate adequate nutrition, while values from 8 to 11 signal potential risk, and 0 to 7 reflect clear evidence of malnutrition.^[16]

The level of pain was measured via the visual analog scale (VAS), which uses a 0 to 10 scale, where 0 indicates no pain and 10 represents the most severe pain imaginable. Hand grip strength was recorded in kilograms using the Jamar hand dynamometer in the seated position, with the elbow at 90°. Performance status (PS) was evaluated using the Karnofsky performance scale (KPS) and the Eastern Cooperative Oncology Group-PS (ECOG-PS), both of which have been validated in Turkish for use in oncology populations. ECOG is scored from 0 to 5. A score of 0 indicates full activity with no limitations, 4 refers to complete functional dependency with the patient unable to perform any self-care and confined to bed or chair, and 5 represents death. Lower scores reflect better functional status. The KPS is scored from 0 to 100 in 10-point increments. A score of 100 reflects full independence and normal daily functioning, while a score of 0 indicates death. Higher scores represent better overall performance and functional capacity.^{[17, 18]} The timed up and go (TUG) test, an easy and validated test first presented by Podsiadlo and Richardson,^[19] was used to evaluate functional mobility. It measures the amount of time needed to get out of a chair, walk three meters, turn around, go back, and sit down. A TUG test duration of 12 seconds or more has been linked to a heightened likelihood of fall risk.^[20] Participants were divided into two groups based on their TUG test results: those with a completion time of 12 seconds or more were considered to have a fall risk, while those with a time under 12 seconds were considered to be without a fall risk. The 6-minute walk test (6MWT) is a practical and non-invasive assessment tool commonly used to evaluate functional exercise capacity. It measures the total distance, in meters, that an individual can walk over a six-minute period on a flat, hard surface. The 6MWT has been widely applied in oncology settings to reflect physical endurance in a simple and clinically relevant manner.^[21]

The study was approved by the Erzincan Binali Yıldırım University Ethics Committee (decision no: 2025-03/02, date: 06.02.2025). Written informed consent was obtained, and the study complied with the Declaration of Helsinki.

Statistical analysis

All statistical analyses were conducted using the IBM SPSS Statistics software, version 25.0 (IBM Corp., Armonk, NY, USA). The normality of the data distribution was evaluated using the Shapiro-Wilk test, while the homogeneity of variances was assessed via Levene’s test. Descriptive statistics were reported as mean ± standard deviation and median (minimum-maximum) for continuous variables and as frequency and percentage for categorical data. Comparisons between two independent groups in terms of median values were performed using the Mann-Whitney U test. Differences in hand grip strength between the affected and non-affected sides were analyzed using the Wilcoxon signed-rank test. Correlations between continuous variables were examined with the Spearman rank correlation test. The interpretation of correlation strength followed the classification by Hinkle et al.,^[22] which is widely cited in health and behavioral sciences. Accordingly, r values between 0.00-0.30 were considered negligible, 0.30-0.50 low, 0.50-0.70 moderate, 0.70-0.90 high, and 0.90-1.00 very high.^[22] To identify independent predictors of general health status, univariate linear regression analyses were initially conducted. Based on univariate significance and clinical relevance, the following variables were considered for inclusion in the multivariable linear regression model: ECOG-PS, VAS, MNA-SF, 6MWT, TUG, affected-side grip strength, circumference difference, CIS-20, dominant hand (yes/no), and palliative CT (yes/no). Given the limited sample size (n=100) and to avoid overparameterization and model instability, a stepwise multiple linear regression approach (probability-of-F-to-enter ≤0.05, probability-of-F-to-remove ≥0.10) was applied. All listed variables were entered into the stepwise model. This strategy was chosen to ensure model parsimony while retaining clinically meaningful predictors. Multicollinearity was assessed using variance inflation factor (VIF) and tolerance values. VIF values <5 were considered acceptable. Model assumptions were evaluated through residual analyses, including standardized residuals and the Durbin-Watson statistic. Regression diagnostics demonstrated no evidence of problematic multicollinearity, with all VIF values below 5. Residual analyses indicated that model assumptions were satisfied. The final regression model explained a substantial proportion of variance in the dependent variable and demonstrated acceptable goodness-of-fit. A p-value of less than 0.05 was considered statistically significant.

RESULTS

Hand grip strength measurements of 100 post-mastectomy lymphedema patients with a mean age of 52.29±8.23 years and a mean duration of lymphedema of 4.77±3.38 years were analyzed. Of the patients, lymphedema was present in the dominant upper limb in 51 individuals and in the non-dominant side in 49 individuals. RT, CT, HT, neoadjuvant CT, adjuvant CT, and palliative CT were administered to 94%, 96%, 93%, 76%, 90%, and 33% of the patients, respectively.

The QoL of individuals living with lymphedema was evaluated through the EORTC-QLQ-C30, a validated instrument that captures both functional subscales and symptom subscales. Table 1 presents the median (minimum-maximum) values for both the functional and symptom subscales of the EORTC-QLQ-C30.

TABLE 1. The median scores of EORTC-QLQ-C30 subscales.

Functioning subscales EORTC-QLQ-C30
-	Global health status	Physical functioning	Rolefunctioning	Emotional functioning	Cognitive functioning	Social functioning	-
Mean (SD)	55.37(22.62)	51.12(21.03)	52.75(24.07)	61.33(21.27)	74.83(20.17)	72.50(19.59)	-
Median(Min-max)	50.00(0.0-100.0)	53.33(6.67-100.0)	50.00(0.0-100.0)	66.67(8.33-100.0)	83.33(16.67-100.0)	66.67(16.67-100.0)	-
Symptom Subscales EORTC-QLQ-C30
-	Fatigue	Nausea and vomiting	Pain	Dyspnea	Insomnia	Appetite loss	Constipation
Mean (SD)	58.67(27.81)	9.66(14.25)	62.17(20.90)	12.33(21.53)	41.00(22.65)	33.67(35.61)	6.03 (14.63)
Median(Min-max)	55.56(0-100.0)	0(0-100.0)	66.67(0-100.0)	0(0-100.0)	33.33(0-100.0)	33.33(0-100.0)	0 (0-100.0)

EORTC-QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; SD: Standard deviation; Min: Minimum; max: Maximum.

The associations between CT types and the functional subscales of the EORTC-QLQ-C30 are presented in Table 2. CT type significantly influenced all QoL subdomains. Patients receiving palliative CT had significantly lower scores across all subdomains (all p<0.001). Similarly, neoadjuvant CT was associated with lower scores in GHS/QoL, PF/QoL, RF/QoL, and EF/QoL (all p<0.001), as well as CF/QoL (p=0.001) and SF/QoL (p=0.007). The effects of HT, RT, and fall risk on QoL are presented in Table 3. RT was associated with significant differences, particularly in GHS (p=0.049), PF (p=0.009), RF (p=0.008), EF (p=0.022), and SF (p=0.019). Patients with a high fall risk exhibited significantly reduced scores across all QoL subdomains (all p<0.001).

TABLE 2. The association between EORTC-QLQ-C30 functional subscales and types of chemotherapy.

EORTC-QLQ-C30	CT yes median(Min-max)	CT no median(Min-max)	Adjuvant CT yes median(Min-max)	Adjuvant CT no median(Min-max)	Neoadjuvant CT yes median(Min-max)	Neoadjuvant CT nomedian(Min-max)	Palliative CT yes median(Min-max)	Palliative CT no median(Min-max)
Global health status	50.00(0-91.67)	75.00(50.00-83.33)	50.00(0-91.67)	62.50(33.33-83.33)	50.00(0-91.67)	75.00(33.33-91.97)	25.00(0-91.67)	66.67(25.00-91.67)
p	0.163		0.314		<0.001***		<0.001***
Physical functioning	53.33(6.67-86.67)	73.33(46.67-80.00)	53.33(6.67-86.67)	63.33(46.67-86.67)	46.67(6.67-86.67)	70.00(40.00-86.67)	26.67(6.67-86.67)	60.00(26.67-86.67)
p	0.090		0.091		<0.001***		<0.001***
Role functioning	50.00(0-100)	75.00(66.67-83.33)	50.00(0-100.00)	66.67(33.33-100.00)	50.00(0-100.00)	75.00(33.33-100.00)	33.33(0-100.00)	50.00(16.67-100.00)
p	0.032*		0.131		<0.001***		<0.001***
Emotional functioning	66.67(16.67-100)	79.17(58.33-91.67)	66.67(16.67-100.00)	66.67(41.67-91.67)	58.33(16.67-91.67)	83.33(41.67-100.00)	33.33(16.67-91.67)	66.67(41.67-100.00
p	0.132		0.351		<0.001***		<0.001***
Cognitive functioning	83.33(16.67-100.00)	83.33(66.67-100.00)	83.33(16.67-100.00)	75.00(33.33-100.00)	66.67(16.67-100.00)	83.33(33.33-100.00)	66.67(16.67-100.00)	83.33(16.67-100.00)
p	0.491		0.798		0.001***		<0.001***
Social functioning	66.67(16.67-100.00)	75.00(50.00-100.00)	66.67(16.67-100.00)	66.67(50.00-100.00)	66.67(16.67-100.00)	83.33(50.00-100.00)	66.67(16.67-100.00)	83.33(50.00-100.00)
p	0.885		0.896		0.007**		<0.001***

*p<0.05, **p<0.01, ***p<0.001; bold values indicate statistically significant differences.

EORTC-QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; CT: Chemotherapy; Min: Minimum; max: Maximum.

TABLE 3. The effect of radiotherapy, hormone therapy, hand dominance and presence of fall risk on EORTC-QLQ-C30 functional subscales.

EORTC-QLQ-C30	RT yes median(Min-max)	RT no median(Min-max)	HT yes median(Min-max)	HT no median(Min-max)	Dominant hand median(Min-max)	Non-dominant hand median(Min-max)	TUG results with fall risk median(Min-max)	TUG results without fall risk median(Min-max)
Global health status	50.00 0 91.67	29.17 8.33 66.67	50.00 0 91.67	66.67 50.00 83.33	50.00 0 91.67	66.67 16.67 91.67	50.00 0 58.33	75.00 33.33 91.67
p	0.049*		0.106		0.053		<0.001***
Physical functioning	53.33 6.67 86.67	23.34 13.33 46.67	46.67 6.67 86.67	66.67 53.33 86.67	46.67 6.67 86.67	53.33 13.33 86.67	40.00 6.67 60.00	66.67 40.00 86.67
p	0.009**		0.060		0.401		<0.001***
Role functioning	50.00 0 100.00	33.33 8.33 50.00	50.00 0 100.00	50.00 50.00 100.00	50.00 0 100.00	50.00 16.67 100.00	33.33 0 50.00	66.67 50.00 100.00
p	0.008**		0.247		0.194		<0.001***
Emotional functioning	66.67 16.67 100.00	33.33 16.67 66.67	66.67 16.67 100.00	66.67 66.67 91.67	58.33 16.67 91.67	66.67 16.67 100.00	50.00 16.67 83.33	75.00 58.33 100.00
p	0.022*		0.193		0.148		<0.001***
Cognitive functioning	83.33 16.67 100.00	66.67 33.33 83.33	83.33 16.67 100.00	66.67 66.67 100.00	83.33 16.67 100.00	83.33 33.33 100.00	66.67 16.67 100.00	83.33 66.67 100.00
p	0.240		0.833		0.081		<0.001***
Social functioning	75.00 16.67 100.00	50.00 16.67 83.33	66.67 16.67 100.00	83.33 83.33 100.00	66.67 16.67 100.00	83.33 33.33 100.00	66.67 16.67 100.00	83.33 50.00 100.00
p	0.019*		0.035*		0.162		<0.001***

*p<0.05, **p<0.01, ***p<0.001; bold values indicate statistically significant differences.

EORTC-QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; RT: Radiotherapy; HT: Hormone therapy; TUG: Timed up and go test; Min: Minimum; max: Maximum.

Patients with dominant extremity involvement had significantly lower grip strength compared to those with non-dominant extremity involvement (mean ± standard deviation: 6.37±2.45 vs. 11.14±5.19, p<0.001). Although patients with involvement of the dominant hand had higher scores across all functional subscales of the EORTC-QLQ-C30 compared to those with non-dominant hand involvement, the differences were not statistically significant (all p>0.05).

Table 4 presents the correlation analyses between the EORTC-QLQ-C30 functional subscales and several clinical and demographic variables. Performance-based measures including the VAS, 6MWT, TUG test, ECOG-PS, and KPS were also highly positively correlated with the functional subscales GHS/QoL, PF/QoL, RF/QoL, and EF/QoL of the EORTC-QLQ-C30.

TABLE 4. Correlations between the functioning subscales of the EORTC-QLQ-C30 and various variables.

Variables	Functioning Subscales EORTC-QLQ-C30
	Global health status		Physical functioning		Rolefunctioning		Emotional functioning		Cognitive functioning		Social functioning
	r	p	r	p	r	p	r	p	r	p	r	p
Age	0.127	0.208	0.150	0.137	0.075	0.460	0.134	0.185	-0.065	0.519	0.131	0.193
BMI	0.589	<0.001	0.503	<0.001	0.532	<0.001	0.620	<0.001	0.376	<0.001	0.484	<0.001
Lymphedema diagnosis year	0.143	0.156	-0.017	0.866	0.080	0.431	0.062	0.538	-0.074	0.465	0.105	0.299
Number of chemotherapy cycles	-0.048	0.634	-0.083	0.411	-0.126	0.213	0.008	0.940	-0.144	0.154	0.057	0.571
CCI result	-0.685	<0.001	-0.690	<0.001	-0.703	<0.001	-0.653	<0.001	-0.621	<0.001	-0.684	<0.001
ECOG-PS	-0.858	<0.001	-0.830	<0.001	-0.828	<0.001	-0.810	<0.001	-0.483	<0.001	-0.660	<0.001
KPS	0.808	<0.001	0.785	<0.001	0.764	<0.001	0.805	<0.001	0.539	<0.001	0.700	<0.001
Stage	-0.753	<0.001	-0.694	<0.001	-0.752	<0.001	-0.713	<0.001	-0.505	<0.001	-0.597	<0.001
VAS	-0.919	<0.001	-0.897	<0.001	-0.872	<0.001	-0.902	<0.001	-0.605	<0.001	-0.794	<0.001
6-minute walk test	0.902	<0.001	0.916	<0.001	0.913	<0.001	0.860	<0.001	0.640	<0.001	0.752	<0.001
TUG test	-0.889	<0.001	-0.903	<0.001	-0.888	<0.001	-0.848	<0.001	-0.586	<0.001	-0.731	<0.001
Affected side hand grip strength	0.710	<0.001	0.619	<0.001	0.619	<0.001	0.680	<0.001	0.505	<0.001	0.625	<0.001
Circumference difference	-0.900	<0.001	-0.803	<0.001	-0.804	<0.001	-0.780	<0.001	-0.622	<0.001	-0.740	<0.001
CIS-20	-0.905	<0.001	-0.882	<0.001	-0.894	<0.001	-0.854	<0.001	-0.640	<0.001	-0.766	<0.001
Subjective fatigue	-0.895	<0.001	-0.889	<0.001	-0.863	<0.001	-0.834	<0.001	-0.633	<0.001	-0.759	<0.001
Concentration	-0.838	<0.001	-0.802	<0.001	-0.831	<0.001	-0.791	<0.001	-0.571	<0.001	-0.723	<0.001
Motivation	-0.882	<0.001	-0.853	<0.001	-0.865	<0.001	-0.830	<0.001	-0.610	<0.001	-0.735	<0.001
Activity	-0.857	<0.001	-0.861	<0.001	-0.875	<0.001	-0.794	<0.001	-0.632	<0.001	-0.707	<0.001

p<0.05 indicates statistical significance, and bold values indicate significant results.

EORTC-QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; BMI: Body mass index; CCI: Charlson comorbidity index; ECOG-PS: Eastern Cooperative Oncology Group Performance Status; VAS: Visual analog scale; KPS: Karnofsky performance scale; TUG: Timed up and go test; CIS-20: Checklist Individual Strength.

The median grip strength of the affected side was 7 kg (range, 3-25 kg), while that of the non-affected side was 16.5 kg (range, 10-33 kg). The difference between the two extremities was statistically significant (p<0.001).

Correlation analyses between affected side hand grip strength and the EORTC-QLQ-C30 functional subscales revealed a high positive correlation with GHS/QoL (r=0.710, p<0.001). A moderate positive correlation was found between affected side grip strength and PF/QoL (r=0.619, p<0.001), RF/QoL (r=0.619, p<0.001), EF/QoL (r=0.680, p<0.001), SF/QoL (r=0.625, p<0.001), and CF/QoL (r=0.505, p<0.001).

A moderate negative correlation was identified between hand grip strength and symptom subscales of the EORTC-QLQ-C30, including fatigue (r=-0.653, p<0.001), nausea and vomiting (r=-0.603, p<0.001), pain (r=-0.612, p<0.001), and appetite loss (r=-0.631, p<0.001). A low negative correlation was observed between hand grip strength and dyspnea (r=-0.462, p<0.001). In addition, hand grip strength was found to have a high negative correlation with the total CIS-20 score (r=-0.715, p<0.001). Further analysis of the CIS-20 subscales revealed a high negative association with subjective fatigue (r=-0.719, p<0.001), as well as moderate negative correlations with concentration (r=-0.688, p<0.001), motivation (r=-0.668, p<0.001), and activity (r=-0.571, p<0.001).

A moderate negative correlation was observed between limb circumference difference and grip strength on the affected side (r=-0.648, p<0.001). A very high negative correlation was found between circumference difference and GHS/QoL (r=-0.900, p<0.001). Similarly, a high negative correlation was observed between circumference difference and PF/QoL (r=-0.803, p<0.001).

Univariate linear regression and multivariable regression analyses were employed to investigate the associations between each candidate predictor and GHS. The findings are summarized in Table 5. In univariate analyses, several variables demonstrated statistically significant associations with GHS. Specifically, higher values of ECOG-PS, TUG test, CIS-20, VAS scores, and circumference difference were significantly associated with a decrease in GHS (all p<0.001). Conversely, better performance on functional capacity measures, such as longer distance in the 6MWT, was associated with higher GHS (p<0.001).

TABLE 5. Results of the univariate and the multivariable linear regression analysis for global health status.

-	Univariate				Multivariable
	B	β	95% CI	p	B	β	95% CI	p
ECOG-PS (1 unit increase)	-23.347	-0.869	[-26.014, -20.681]	<0.001	-	-	-	-
VAS (1 unit increase)	-10.671	-0.877	[-11.845, -9.498]	<0.001	-	-	-	-
MNA-SF (1 unit increase)	-25.311	-0.768	[-29.542, -21.081]	<0.001	-5.076	-0.154	[-7.527, -2.625]	<0.001
6MWT (1 unit increase)	0.147	0.889	[0.132, 0.162]	<0.001	-	-	-	-
TUG test (1 unit increase)	-2.191	-0.898	[-2.406, -1.976]	<0.001	-0.726	-0.298	[-0.974, -0.478]	<0.001
ASGS (1 unit increase)	2.762	0.570	[1.965, 3.559]	<0.001	-	-	-	-
Circ. diff. (1 unit increase)	-26.543	-0.900	[-29.124, -23.961]	<0.001	-10.406	0.353	[-13.527, -7.648]	<0.001
CIS-20 (1 unit increase)	-0.741	-0.886	[-0.819, -0.664]	<0.001	-0.221	-0.265	[-0.304, -0.138]	<0.001
Dominant hand (yes)	-8.618	-0.191	[-17.477, -0.241]	0.056	-	-	-	-
Palliative chemotherapy (yes)	-25.950	-0.542	[-34.013, -17.887]	<0.001	-	-	-	-

Model statistics: R²=0.937, adjusted R²=0.935, Durbin-Watson=1.68.

B: Unstandardized beta; β: Standardized beta; CI: Confidence interval; ECOG-PS: Eastern Cooperative Oncology Group Performance Status; VAS: Visual analog scale, MNA-SF: Mini nutritional assessment-short form, 6MWT: 6-minute walk test; TUG: Timed up and go test; ASGS: Affected side grip strength, Circ. diff.: Circumference difference; CIS-20: Checklist Individual Strength.

In the context of MNA-SF scoring, a one-unit increase reflects a transition from normal nutritional status to risk of malnutrition, and subsequently to malnutrition. MNA-SF scores were significantly associated with general health status, with worsening nutritional status corresponding to lower general health scores (p<0.001). In univariate analyses, dominant hand (p=0.056) showed a non-significant association, whereas palliative CT was significantly associated with GHS (p<0.001).

In the final multivariable model, circumference difference, TUG test, MNA-SF, and CIS-20 remained independently associated with GHS. Greater circumference difference (B=-10.406, p<0.001) and longer TUG test times (B=-0.726, p<0.001) were associated with lower GHS. Poorer nutritional status (MNA-SF) was also an independent predictor of worse GHS (B=-5.076, p<0.001). Additionally, higher CIS-20 scores were associated with reduced GHS (B=-0.221, p<0.001). The model explained a substantial proportion of the variance in GHS (R²=0.937, adjusted R²=0.935; p<0.001).

DISCUSSION

Hand grip strength was significantly lower in the affected limb post-mastectomy, showing a strong positive correlation with EORTC-QLQ-C30 functional subscales, highlighting its impact on QoL in lymphedema patients. According to a previous study comparing patients with post-mastectomy lymphedema to healthy controls, hand function and grip strength were found to be significantly reduced in the lymphedema group.^[9] However, comprehensive studies specifically examining the impact of hand grip strength on QoL remain limited.

In a study by John et al.^[23] using regression analysis, both increased hand grip strength and higher BMI were found to significantly reduce upper extremity disability. Consistent with this, our findings also revealed a positive correlation between BMI and the functional subscales of QoL. On the other hand, age—one of the primary demographic factors—was not associated with QoL in our sample. This may be attributed to the exclusion of patients over the age of 65 from the study.

In a 10-year cohort study by Gao et al.^[8] health-related QoL (HRQoL) temporarily declined during RT but showed improvement by the sixth week and nearly returned to baseline at follow-up. In our study, patients without RT had lower HRQoL scores in all functional subscales except CF/QoL.

Adequate hand grip strength is a key determinant of upper limb functionality in performing daily activities. In a recent study by Shinde et al.,^[24] grip strength in the affected arm was also found to be significantly lower, consistent with the findings of our study. Unlike studies focused solely on hand function, our research aimed to explore the broader impact of hand grip strength on overall QoL. We found that improvements in hand grip strength were positively associated with better outcomes across all functional subscales of QoL, highlighting its potential influence beyond localized hand function.

Meta-analyses have detected that breast cancer survivors tend to walk shorter distances in functional tests compared to healthy individuals.^[21] In our study, which included post-mastectomy patients with lymphedema, longer walking distances measured by the 6MWT were significantly associated with higher scores in GHS/QoL, PF/QoL, RF/QoL, EF/QoL, and CF/QoL. Nevertheless, further comprehensive analyses are needed in breast cancer populations without lymphedema to better understand the specific impact of walking capacity on QoL.

The TUG test, which is commonly utilized to assess lower limb functionality, also provides insight into fall risk.^{[25, 26]} Beyond assessing functional mobility, the test serves as a practical tool for identifying potential fall risk.^{[19, 20]} Falls were reported in 23.58% of breast cancer patients during the period of treatment.^[27] While patient-reported fall history was not evaluated, 54% of participants demonstrated TUG durations above 12 seconds, reflecting an increased fall risk.

Increased pain levels have been reported in the literature to be associated with reductions across all functional subscales of the EORTC-QLQ-C30.^[28] In our study, correlational analyses revealed a very high negative correlation between VAS and key QoL subscales including GHS/QoL, EF/QoL. Moreover, high negative correlations were found for PF/QoL, SF/QoL and RF/QoL, while a moderate negative correlation was observed for CF/QoL. These findings underscore the profound impact of pain on overall QoL, with higher pain scores consistently associated with significantly lower functioning across multiple dimensions.

Lacourt et al.^[29] identified a reduction in mitochondrial oxygen consumption following neoadjuvant CT in patients with breast cancer, and this physiological change was linked to elevated levels of fatigue, as measured by the CIS-20. In another prospective study evaluating fatigue over time, the prevalence of fatigue was reported to be 68.75% shortly after treatment completion, while 41.2% of patients continued to experience persistent fatigue five years later.^[30] While fatigue is a well-documented and prevalent symptom among patients with post-mastectomy lymphedema in breast cancer, few studies have directly examined its association with hand grip strength in the affected extremities. Our study uniquely fills this gap by assessing the correlation between clinically measured hand grip strength and CIS-20, revealing a high negative correlation.

For the majority of elderly patients with solid tumors, adjuvant CT and radiation did not negatively impact their QoL.^[31] In the review study published by Ho et al.^[32] HRQoL was reported to be lower among patients with comorbidities, those undergoing CT, individuals with limited social support, and those with unfulfilled requirements. In our study, although adjuvant CT did not significantly impact QoL in functional subscales, the presence of neoadjuvant and palliative CT was associated with a marked deterioration across all functional QoL domains.

One of the most notable contributions of our study is the demonstration of the independent effects of multiple clinical variables on QoL through regression analysis. Beyond hand grip strength, factors such as year of lymphedema diagnosis, CT, RT, HT, performance-based measures (VAS, 6MWT, TUG, ECOG-PS, KPS), fatigue assessments, nutritional status, and comorbid conditions were shown to have significant associations with QoL. Unlike earlier research that often examined a narrower set of parameters or relied mainly on correlation analyses, our inclusion of a broader clinical spectrum and the use of regression methods enhance the originality of this work and provide more robust insights for clinical practice in patients with post-mastectomy lymphedema. In the multivariable regression model for GHS/QoL, limb circumference difference, TUG test, MNA-SF, and CIS-20 score remained significant predictors. These findings underscore the multifactorial nature of QoL in patients with breast cancer-related lymphedema. While physical impairments such as increased limb circumference difference and decreased walking capacity directly influence functional status, nutritional condition and fatigue—reflected in the MNA-SF and CIS-20 scores—highlight the systemic and psychosocial dimensions of the disease. These results support the importance of a holistic evaluation and multidisciplinary care in addressing QoL in this population.

Our findings revealed that patients with involvement of the dominant upper extremity exhibited lower grip strength compared to those with non-dominant limb involvement. This highlights the potential functional vulnerability associated with impairment of the dominant side. While GHS scores appeared higher in the dominant-side group, the difference did not reach statistical significance. Nevertheless, future studies may consider adjusting for hand dominance in statistical analyses to further clarify its impact.

Although this study provides a broad evaluation of QoL, functional mobility, grip strength, and performance in breast cancer survivors with post-mastectomy lymphedema, larger population-based studies comparing similar age groups or patients without lymphedema are still needed to better interpret the results.