Background: The etiology of lymphedema is multifactorial, and definition criteria of lymphedema, its limitation, and follow-up must be considered in studies related to risk factors. The aim of this study is to evaluate risk factors related to arm lymphedema in a cohort study with a long follow-up. Patients and Methods: The study was performed in 622 breast cancer patients. The main endpoint reported was the presence of clinical lymphedema reported in medical records. Univariate and multivariate regression analyses were performed to identify factors related to lymphedema. Results: 66.4% of the patients were submitted to mastectomy, 88.4% to level III axillary lymphadenectomy, 34.9% to radiotherapy in the supraclavicular fossa, and 4.3% to axillary radiotherapy. The mean follow-up was 96.7 months. 45 patients (7.2%) developed lymphedema, of which 82.2% had developed lymphedema at 60 months. Univariate regression analysis showed that supraclavicular radiotherapy, adjuvant/palliative chemotherapy, ≥ 15 lymph nodes dissected, and axillary surgery increase the lymphedema rate by 1.87, 2.28, 2.03, and 6.17, respectively. Adjusted multivariate regression analysis showed that the combination of axillary dissection and number of lymph nodes dissected was the main factor related to lymphedema (p = 0.017). Conclusion: In the pre-sentinel era, axillary dissection and the number of lymph nodes resected are related to 10-year lymphedema.