Weight gain is an early indicator of injury in ex vivo normothermic limb perfusion (EVNLP)

PURPOSE: There are no established criteria for discontinuing ex vivo normothermic limb perfusion (EVNLP) before irreversible damage occurs. This study evaluates weight gain as an indicator of injury during EVNLP. METHODS: Sixteen Yorkshire pig forelimbs were procured and preserved using EVNLP with a hemoglobin‐based oxygen carrier (HBOC‐201) or static cold storage. EVNLP continued until termination criteria were met: arterial pressure ≥ 115 mm Hg, compartment pressure > 30 mm Hg, or 20% reduction of oxygen saturation. Limb weight, contractility, hemodynamics, perfusate electrolytes, metabolites and gases were recorded. Muscles were biopsied 6‐h, and muscle injury scores (MIS) calculated. Forearm compartment pressures and indocyanine green (ICG) angiography were recorded at endpoint. Outcomes were compared at 2%, 5%, 10%, and 20% limb weight gain. RESULTS: EVNLP lasted 20 ± 3 h. Weight gain was observed after 13 ± 5 h (2%), 15 ± 6 h (5%), 16 ± 6 h (10%), and 19 ± 4 h (20%). Weight correlated positively with MIS (ρ = 0.92, p < 0.0001), potassium (ρ = −1.00, p < 0.0001), pressure (ρ = 0.78, p < 0.0001), and negatively with contractility (ρ = −0.96, p = 0.011). At 5% weight gain, MIS (p < 0.0001), potassium (p = 0.03), and lactate (p < 0.0001) were significantly higher than baseline. Median muscle contractility was 5 [3–5] at 2% weight gain, 4 [1–5] at 5%, 3 [0–4] and 2 [0–2] at 10% and 20%, respectively. At 20% weight gain, contractility was significantly lower than baseline (p = 0.003). Percent weight gain correlated negatively with endpoint ICG hoof fluorescence (r = −0.712, p = 0.047). CONCLUSIONS: Weight gain correlated with microscopic muscle injury and was the earliest evidence of limb dysfunction. Weight gain may serve as a criterion for discontinuation of EVNLP.