Limb volume reduction and infection outcomes following vascularized lymph node transfer for cancer treatment-related lymphoedema: A systematic review and meta-analysis.

e24040 Background: Cancer treatment-related lymphoedema (CTRL) is a chronic and disabling complication that carries a significant health burden during cancer survivorship. Vascularized lymph node transfer (VLNT) is employed increasingly to microsurgically reconstruct physiological lymphatic flow in affected limbs. We set out to synthesise and assess the efficacy of VLNT for reducing limb volume and infection outcomes in CTRL. Methods: A systematic review was conducted by searching PubMed, Medline and EMBASE databases during early 2020. All full-text English-language articles reporting VLNT as the sole therapeutic intervention for the management of CTRL that provided volumetric limb and/or patient-reported lymphatic quality of life and/or cellulitis-related outcomes were included. All studies were scored for methodological quality (STROBE and CONSORT checklists) and lymphoedema-specific data reporting quality. Using a random-effects model, we estimated the pooled limb volume reduction and 95% CI for upper and lower limbs stratifying according to site of measurement and VLNT donor. We also estimated the pooled reduction in cellulitis episodes and mean improvement in lymphoedema-related quality of life. PROSPERO ID: CRD42020204080. Results: The search strategy identified 277 articles. Thirty-one studies encompassing 581 patients were included with the overall methodological quality of studies assessed to be low. Meta-analysis demonstrated that performing VLNT for CTRL led to limb volume reductions in both upper limb (above elbow circumferential reduction rates (CRR) = 42.7 % (95% CI: 36.5-48.8, I 2 : 94.3 %, p = 0.000, 153 patients, 10 studies); below elbow CRR = 34.1 % (95% CI: 33.0-35.1, I 2 : 0.0 %, p = 0.919, 144 patients, 9 studies)) and lower limb (above knee CRR = 46.8 % (95% CI: 43.2-50.4, I 2 : 92.4 %, p = 0.000 26 patients, 3 studies); below knee CRR = 54.6 % (95% CI: 39.0-70.2, I 2 : 97.6%, p = 0.000, 26 patients, 3 studies)) CTRL. Similarly, extra-abdominal VLNT, abdominal VLNT and autologous breast reconstruction with VLNT reduced CTRL by CRRs of 49.5 % (95% CI: 46.5-52.5, I 2 : 8.3 %, p = 0.365, 7 studies, 108 patients), 39.6 % (95% CI: 37.2-42.0, I 2 : 89.5 %, p = 0.000, 3 studies, 15 patients) and 32.7 % (95% CI: 11.1-54.4, I 2 : 96.9 %, p = 0.000 3 studies, 29 patients). VLNT reduced mean annual cellulitis episodes by 2.1 (95% CI: -2.7- -1.4) episodes per annum and improved lymphoedema-specific quality of life was improved by 4.26 (LYMQOL “overall domain” scale). Conclusions: Performing VLNT for CTRL results in reductions of limb volume and cellulitis episodes for both upper and lower limbs. However, published studies were methodologically heterogeneous and of low quality highlighting the need for standardized outcome reporting and further well-designed randomized controlled trials comparing VLNT against existing therapies.