Introduction

Lymphedema is the accumulation of protein-rich fluid in tissues due to impaired lymphatic outflow. Primary causes include genetic diseases, whereas secondary causes include mechanical insufficiency resulting from conditions such as tumors, infections, surgery, radiation, or trauma (1, 2).
Breast cancer is the most common cancer in women worldwide, with higher survival rates attributed to advances in screening, detection, and treatment (3, 4). Breast cancer treatments, particularly surgery, are primary causes of upper extremity lymphedema (5) and associated complications such as stiffness, numbness, pain, and nerve injury, resulting in reduced quality of life, emotional distress, and depression in affected women (6, 7). Lymphedema incidence varies from 5% to 60% worldwide (8). Its incidence rises from 13% within the first 2 years to 41% in 10 years (9).
Treatment and non-treatment factors are both considered risk factors for lymphedema. Compared to sentinel lymph node biopsy (SLNB), axillary lymph node dissection (ALND) carries a significantly higher risk, and the risk rises with the number of nodes removed: 6% for <10 nodes and 27% for more than 10 nodes (10). Radiotherapy, particularly with ALND, increases lymphedema incidence (11, 12). Chemotherapy, especially taxane-based regimens, may increase lymphedema risk due to fluid retention, as observed by Kilbreath et al., who linked taxane therapy to arm swelling at 6 and 12 months (13). The impact of neoadjuvant chemotherapy on lymphedema risk remains uncertain but may reduce risk by decreasing positive lymph nodes (14).
Well-established non-treatment-related factors including younger age, aggressive carcinoma, high body mass index (BMI), and infections such as cellulitis, edema, and weight changes, all enhance the risk of upper limb lymphedema, ranging from 4.7% to 41.2% (5, 15, 16). Disipio et al. in a meta-analysis found that a higher BMI was identified as a significant independent risk factor for the development of breast cancer related lymphedema (BCRL) (Relative risk: 5.5) (17). Cellulitis infections resulted in an approximate 3% rise in arm volume alterations (16).
Considering the global rise in the incidence of breast cancer and its related post-operative complications, it is of public health interest to understand in our setup the incidence of secondary lymphedema and its associated risk factors. We therefore carried out a study to estimate the proportion of patients with lymphedema following breast surgery in one of the largest referral hospitals in the East African region with a multidisciplinary breast clinic. We as well aimed to determine its risk factors and their associations.

​

Materials and Methods

This was a retrospective cross-sectional study that analyzed data from patients operated on between January 2014 and December 2019 to calculate lymphedema prevalence. The retrospective nature of the study was undertaken because the facility utilizes the health information management system that facilitates retrieval of patient records.
The study was carried out in the Medical Records department of a referral hospital in our country. This is the largest referral hospital with a multidisciplinary breast clinic. Before attaining access to patient records, we secured ethical approval and institutional authority.
The study included patients who had histologically confirmed breast cancer and patients who had breast cancer surgery at our facility between January 2014 and December 2019. Patient records were selected via convenience sampling, with the exception of incomplete data, primary lymphedema, and non-breast cancer-related lymphedema instances. Data were taken from eligible patient records.
The study examined independent variables such as age, breast cancer pathology, axillary recurrence, BMI, tumor location, chemotherapy, radiotherapy, ALND, SLNB, and the number of lymph nodes removed, with lymphedema as the dependent variable. The main diagnostic criteria for lymphedema at our facility were done clinically (history and physical examination including sequential circumferential arm measurement). However, in selected cases, imaging was used. The lymphedema stages are classified as follows: stage 0 indicates a normal extremity clinically but with abnormal lymph transport; stage 1 is early edema, which improves with limb elevation; stage 2 represents pitting edema that does not resolve with elevation; and stage 3 describes fibroadipose deposition and skin changes (18).
We used the Cochrane formula and calculated the sample size as follows:

Formula
 
The expected proportion was 50% (the expected proportion is between 5% and 60%) (8). This proportion was applied only for prevalence calculation to determine our sample size.
Therefore,
Calculation 

hence, n=384.
A convenience sampling technique was used to select records of patients who met the selection criteria.
Data collection involved a data collection sheet, patient medical records, and institutional ethical approval (P538/09/2020). We applied for a waiver of consent that was granted by the ethical committee. The principal investigator and two trained undergraduate assistants, after 2 weeks of training, gathered demographic, diagnostic, and clinical data including post-operative notes, ensuring confidentiality through serial numbers, secure storage, and password protection. Incomplete records were excluded. Data analysis was performed using SPSS version 24 (IBM Corp., Armonk, NY, USA), employing analysis of variance, T-tests, logistic regression, chi-squared, and Fisher’s exact tests, with p-values < 0.05 considered significant. Results were presented using tables, text, and figures.

​

Results

A total of 364 participants from our initially estimated sample size of 384 were included in the study. The mean age of respondents was 50.5±13.8 years and ranged from 12 to 96 years. The majority of the participants were between 41 and 50 years (Figure 1).
Invasive carcinoma constituted a majority of the diagnoses, 343 (94.2%), compared to non-invasive carcinoma, 21 (5.7%). The mean BMI of participants was 22.1±2.1 and ranged from 17 to 35 kg/m2.

 

Figure 1.

Age vs frequency distribution table.

Figure 2.

Tumor location.

Table 1.

Distribution by interventions
ALND, axillary lymph node dissection; SLND, sentinel lymph node biopsy.

Most patients had breast cancer located in the upper outer quadrant (61%) followed by the lower outer quadrant (18%) (Figure 2).
Among the patients in the study, 265 patients (72.8%) did not receive neoadjuvant chemotherapy (Table 1).
At surgery, most patients had mastectomy done, with 353 patients (97%), while ALND was the main axillary management procedure done (97.3%) (Table 1).
The mean number of nodes removed was 8.4±4.1 nodes, ranging from no nodes to 25 nodes.
After surgery, 19 patients (5.2%) developed axillary recurrence. Upper limb lymphedema developed in 26 patients (7.1%), of whom 22 patients (84%) had stage 1 lymphedema (Table 2).

​

Table 2.

Frequency distribution of complications

​

Correlations/bivariate analysis
Clinically, 96% of patients with lymphedema had invasive cancer, translating to about 7% of all patients with invasive cancer. Of the patients with non-invasive cancer, only 1 out of 14 developed lymphedema, representing 6.7%. This correlated to a likelihood ratio of 0.08 with a chi-squared coefficient of 0.08. However, with a p-value of 0.928, this relationship was not statistically significant (Table 3).

 

Table 3.

Bivariate analysis of patient/disease related and pre-operative management factors
BMI, body mass index.

Sixty-one percent of all tumors were located in the upper outer quadrant. Of these, 7.3% developed lymphedema. With a p-value ranging between 0.56 and 0.99 (upper outer quadrant, 0.90; upper inner quadrant, 0.56; lower outer quadrant, 0.90; lower inner quadrant, 0.90), and a chi-squared coefficient of 1.675, with a p-value of 0.795, this relationship was deemed not statistically significant (Table 3).
Even though the patients who had neoadjuvant chemotherapy were a smaller proportion compared with those who did not receive it (27.2% vs 22.7%), the proportion of patients who developed lymphedema in each category was similar (6.1% vs 7.5%, respectively). This correlated to a likelihood ratio of 0.248 with a chi-squared coefficient of 0.24. However, with a p-value of 0.63, this relationship was not statistically significant (Table 3).
The mean BMI of participants was 22.1±2.1 and ranged from 17 to 35 kg/m2. When analyzed against lymphedema development, the likelihood ratio was 18.6, and the chi-square was 25.3. This was statistically significant, as the p-value was 0.021 (Table 3).
At surgery, most patients had mastectomy done with 353 patients (97%). None of the patients who underwent WLE developed lymphedema. Of the 352 patients who underwent mastectomy, 7.4% (26 patients) developed lymphedema. The calculated p-value of 1 meant that the relationship was not statistically significant (Table 4).

 

Table 4.

Surgery-related factors bivariate analysis
ALND, axillary lymph node dissection; CI, confidence interval; OR, odds ratio; SLNB, sentinel lymph node biopsy.

Descriptive statistics showed that only 2.7% of patients underwent SLNB, compared with the majority (97.3%) who underwent ALND. There was no lymphedema development in the patients who had undergone SLNB. However, the patients who developed lymphedema (26 patients) had all undergone ALND. The p-value, however, was 1; hence, this relationship was not statistically significant (Table 4).
Just slightly less than half (47.5%) of total patients underwent post-operative radiotherapy. Of these, 17% developed lymphedema. The odds ratio (OR) was 2.2 [confidence interval (CI) 1.0–5.1], with a chi-square of 3.4. The p-value, however, was 0.06; hence, this relationship was not statistically significant (Table 5).

 

Table 5.

Post-operative management
CI, confidence interval; OR, odds ratio.

Approximately 5.2% (18) of patients had axillary recurrence. Of these, four (21.1%) developed lymphedema. This correlated to an OR of 3.91, with a chi-square of 5.1. The p-value was 0.02; hence, this factor was found to be statistically significant (Table 5).
There was no significant difference in mean age of participants who developed lymphedema (51.7±16.2 years) compared to those who did not (50.4±13.7 years). With a chi-squared value of 6.79, there was no significant difference between the two groups in relation to age (Table 6).

​

Table 6.

Cross tabulation of age vs. lymphedema

​

Multivariate analysis
Variables with p < 0.20 on univariate analysis (post-operative radiotherapy, adjuvant chemotherapy, and axillary recurrence) were considered candidates for multivariable logistic regression analysis. After adjustment for age, BMI, neoadjuvant chemotherapy, post-operative radiotherapy, adjuvant chemotherapy, axillary recurrence, and number of lymph nodes dissected, only BMI and axillary recurrence remained statistically significant predictors. Patients with higher BMI had significantly increased odds of developing lymphedema [adjusted OR (AOR) 1.40, 95% CI 1.07–1.83, p = 0.015]. For every one-unit increase in BMI, the odds of lymphedema increased by approximately 40%. Similarly, patients who developed axillary recurrence had nearly sixfold higher odds of developing lymphedema compared with those without recurrence (AOR 5.86, 95% CI 1.41–24.36, p = 0.015) (Table 7).

​

Table 7.

Multivariate analysis
BMI, body mass index.

​

Discussion

From our study, the reported incidence of lymphedema was found to be 7.1%. This finding is lower than that found in Ghana by Sekyere et al., which reported an incidence of 9.9% (19). Our findings are, however, higher than the point prevalence of 6.8% reported by Ren et al. in a study that assessed the burden of lymphedema in breast cancer (20). The high incidence discrepancy could be attributed to the longer follow-up period and their study focusing solely on patients with ALND.
A BMI of 30 kg/m2 or more was found to be a risk factor for the development of lymphedema in our study. This is similar to studies carried out by Leray et al. that found a BMI of 30 kg/m2 or more as a strong independent risk factor for the development of severe lymphedema (21). Similarly, Gillespie et al. found that patients with high BMI were 3.6 times more likely to develop lymphedema (12). Conversely to the above findings, BMI was not found to be statistically significant in a study by Sekyere et al. (19). Other non-treatment-related factors including age, tumor location, and histology did not show a significant correlation with lymphedema in this current study.
Axillary recurrence was found to be significantly associated with lymphedema in our study. Though not well established as a risk factor, this finding could be due to tumor infiltration into the lymphatic vessels or as a result of fibrosis from radiation or lymph node dissection.
ALND and regional lymph node radiation are strongly associated with an increased risk of lymphedema (11, 12). In this study, all lymphedema cases occurred in patients who underwent ALND, highlighting SLNB as a safer alternative for axillary staging in early-stage or clinically node-negative breast cancer. Despite its clinical significance, bivariate analysis did not show a statistically significant correlation between ALND or SLNB and lymphedema development (p = 1). Additionally, there was no statistically significant difference between lymphedema occurrence and the type of breast surgery (mastectomy vs. wide local excision) or between chemotherapy (adjuvant or neoadjuvant) and lymphedema (p = 1). Adjuvant radiotherapy also did not show a statistically significant correlation with lymphedema in this study (p = 0.06). The current study found no significant relationship between chemotherapy and lymphedema. This finding is similar to other studies (11). Despite the minimal number of SLNBs performed in our study, it is still a viable surgical alternative in our setup. The greater prevalence of ALND is likely due to the advanced stages, bulky nodal disease, and late presentation of breast cancer patients encountered in our region.
This study examines the risk factors and incidence of lymphedema subsequent to breast cancer surgery. This will assist clinicians in advising patients regarding the potential for post-operative lymphedema, especially in the context of high-risk surgeries. Our study had significant limitations. The study design increased the chances of selection bias and did not allow for the evaluation of the relative risk of the various risk factors. The use of convenience sampling may have introduced selection bias, limiting the representativeness of the sample and reducing the generalizability of the findings to the broader population of breast cancer survivors.
Furthermore, the study’s breadth was constrained due to its single-centered, retrospective cross-sectional design. The inability to account for confounding variables limited the outcomes of the investigation. The inability to assess long-term follow-up and time to lymphedema occurrence limited our data and results. Our emphasis was limited to in-hospital complications, making it impossible to evaluate long-term results. Incomplete or missing patient records lead to limited data gathering. We therefore advocate performing a multi-center prospective cohort study in this area.

​

Conclusion

The incidence of breast cancer-related lymphedema remains within the global ranges. BMI and axillary recurrence were associated with increased risk for the development of lymphedema.

​

Author contributions

JMO led in Conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, and in writing, reviewing & editing of the original draft. All other authors equally contributed.

​

References

1. Kayıran O, De La Cruz C, Tane K, et al. Lymphedema: from diagnosis to treatment. Turk J Surg. 2017; 33(2): 51-7.

2. Grada AA, Phillips TJ. Lymphedema: pathophysiology and clinical manifestations. J Am Acad Dermatol. 2017; 77(6): 1009-20.

3. Nounou MI, ElAmrawy F, Ahmed N, et al. Breast cancer: conventional diagnosis and treatment modalities and recent patents and technologies. Breast Cancer (Auckl). 2015; 9(Suppl 2): 17-34.

4. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020; 70(1): 7-30.

5. Zhang X, Tang B, Zou D, et al. Discussion of relationships among changes of pathological indicators, postoperative lymphedema of the upper limb, and prognosis of patients with breast cancer. Biosci Rep. 2019; 39(4): BSR20190231.

6. Chrischilles EA, Riley D, Letuchy E, et al. Upper extremity disability and quality of life after breast cancer treatment in the Greater Plains Collaborative clinical research network. Breast Cancer Res Treat. 2019; 175(3): 675-89.

7. Mbaabu K, Wasike R. Upper arm disability after axillary surgery for early breast cancer. Ann Afr Surg. 2020; 17(3): 99-102.

8. Poage E, Singer M, Armer J, et al. Demystifying lymphedema: development of the lymphedema putting evidence into practice card. Clin J Oncol Nurs. 2008; 12(6): 951-64.

9. Ribeiro Pereira ACP, Koifman RJ, Bergmann A. Incidence and risk factors of lymphedema after breast cancer treatment: 10 years of follow-up. Breast. 2017; 36: 67-73.

10. Kim M, Kim SW, Lee SU, et al. A model to estimate the risk of breast cancer-related lymphedema: combinations of treatment-related factors of the number of dissected axillary nodes, adjuvant chemotherapy, and radiation therapy. Int J Radiat Oncol Biol Phys. 2013; 86(3): 498-503.

11. Singh AP, Singh A, Arya R, et al. Prevalence of lymphedema and associated risk factors following multimodal breast cancer treatment. J Med Sci Health. 2023; 9(3): 310-5.

12. Gillespie TC, Sayegh HE, Brunelle CL, et al. Breast cancer-related lymphedema: risk factors, precautionary measures, and treatments. Gland Surg. 2018; 7(4): 379-403.

13. Kilbreath SL, Refshauge KM, Beith JM, et al. Risk factors for lymphoedema in women with breast cancer: a large prospective cohort. Breast. 2016; 28: 29-36.

14. Specht MC, Miller CL, Skolny MN, et al. Residual lymph node disease after neoadjuvant chemotherapy predicts an increased risk of lymphedema in node-positive breast cancer patients. Ann Surg Oncol. 2013; 20(9): 2835-41.

15. McLaughlin SA, Brunelle CL, Taghian A. Breast cancer–related lymphedema: risk factors, screening, management, and the impact of locoregional treatment. J Clin Oncol. 2020; 38(20): 2341-50.

16. Ferguson CM, Swaroop MN, Horick N, et al. Impact of ipsilateral blood draws, injections, blood pressure measurements, and air travel on the risk of lymphedema for patients treated for breast cancer. J Clin Oncol. 2016; 34(7): 691-8.

17. DiSipio T, Rye S, Newman B, et al. Incidence of unilateral arm lymphoedema after breast cancer: a systematic review and meta-analysis. Lancet Oncol. 2013; 14(6): 500-15.

18. Greene AK, Goss JA. Diagnosis and staging of lymphedema. Semin Plast Surg. 2018; 32(1): 12-6.

19. Sekyere MO. Incidence and risk factors of arm lymphedema following breast cancer treatment. J Glob Oncol. 2018; 4(Suppl 2): 213s.

20. Ren Y, Kebede MA, Ogunleye AA, et al. Burden of lymphedema in long-term breast cancer survivors by race and age. Cancer. 2022; 128(23): 4119-28.

21. Leray H, Malloizel-Delaunay J, Lusque A, et al. Body mass index as a major risk factor for severe breast cancer-related lymphedema. Lymphat Res Biol. 2020; 18(6): 510-6.