OBJECTIVE: Lymphedema is classified as primary or secondary according to the underlying cause. Primary lymphedema is hereditary and is considered a consequence of an inherited abnormality of the lymphatic system. Secondary lymphedema, however, is a consequence of lymphatic failure resulting from trauma, parasitic infection, or iatrogenic obstruction. Primary lymphedema is divided into three broad groups, namely, lymphedema congenita, lymphedema praecox, and lymphedema tarda. With the exception of lymphedema tarda, it is thought that age-related deterioration in lymphatic pump function is caused by oxidative stress. The aim of this study was to evaluate and to classify indocyanine green (ICG) lymphography findings in patients with lower limb lymphedema to ascertain whether there is a pattern to age-related deterioration. METHODS: There were 56 patients (104 edematous lower limbs) who had undergone ICG lymphography and for whom the lower extremity lymphedema (LEL) index had been calculated enrolled in this study. Specific inclusion criteria were used to exclude other causes of edema. ICG lymphography images were recorded in the plateau phase (12-18 hours after injection), when no further changes of images would be expected. The LEL index was calculated by summation of the squares of the circumference for five areas in each lower extremity divided by the body mass index. RESULTS: The clinical lymphedema pattern was determined as bilateral in 48 patients and unilateral in 8 patients. Patients with bilateral lymphedema were significantly older than those with unilateral lymphedema (76.40 ± 8.03 years vs 53.13 ± 14.12 years; P < .01). The ICG lymphography pattern was categorized as linear, low enhancement (LE), distal dermal backflow (DB), or extended DB in bilateral lymphedema. ICG lymphography showed the DB pattern on both the thigh and lower leg regions in all eight legs with unilateral lymphedema. There were also significant between-group differences in the LEL index (linear vs distal DB, P < .05; linear vs extended DB, P < .01; linear vs unilateral, P < .01; LE vs extended DB, P < .01; LE vs unilateral, P < .01; distal DB vs extended DB, P < .05; and distal DB vs unilateral, P < .01). CONCLUSIONS: In this study, unilateral lymphedema, with its younger age at onset, severity, and unilateral dominance, corresponded to lymphedema tarda. In contrast, bilateral lymphedema corresponded to senile lymphedema, which is distinct from primary lymphedema in general and lymphedema tarda in particular. Age-related deterioration in lymphatic pump function rather than iatrogenic obstruction or genetic abnormality is likely to account for the characteristic older age at onset of lymphedema and its progression from the distal region.