Skip to main content
. 2020 Nov 27;185(3):709–740. doi: 10.1007/s10549-020-05988-6

Table 3.

Annualized incidence of breast cancer-related lymphedema, by study type

No intervention (background rate) Monitoring with BIS Monitoring with circumference Comparative rate ratio
Included studies (n) Incidence (%; 95% CI) Included studies (n) Incidence (%; 95% CI) Rate ratioa p-value Included studies (n) Incidence (%; 95% CI) Rate ratiob p-value BIS vs. circumference p-value
Overall pooled estimate 35 4·9 (4·3–5·5) 7 1·5 (0·6–2·4) 0·31  < 0·001 11 7·7 (5·6–9·8) 1·57  < 0·001 0·19  < 0·001
By follow-up duration
 ≤ 2 years 18 10·1 (8·0–12·2) 5 1·9 (0·7–3·2) 0·19  < 0·001 7 9·4 (5·5–13·3) 0·93 0·36 0·20  < 0·001
 > 2 years 17 3·4 (2·8–3·9) 2 0·8 (− 0·6–2·2) 0·24  < 0·001 4 6·9 (4·1–9·8) 2·03  < 0·001 0·12  < 0·001
By study type
 Prospective 19 7·9 (6·1–9·8) 4 2·1 (0·0–4·1) 0·27  < 0·001 6 11·0 (5·8–16·3) 1·39  < 0·001 0·19  < 0·001
 Randomized clinical trials 6 10·1 (6·1–14·0) 1 0·5 (− 0·2–1·2) 0·05  < 0·001 3 7·1 (0·4–13·8) 0·70 0·04 0·07  < 0·001
 Retrospective 10 3·5 (2·7–4·4) 2 1·6 (1·1–2·2) 0·46  < 0·001 2 5·8 (1·5–10·2) 1·66  < 0·001 0·28  < 0·001

aBIS versus no intervention

bCircumference versus no intervention

BIS Bioimpedance spectroscopy