Effects of Subnormothermic Perfusion Before Transplantation for Liver Grafts from Donation After Cardiac Death: A Simplified Dripping Perfusion Method in Pigs.

BACKGROUND: Liver transplantation from donors after cardiac death (DCD) provides a solution to the donor shortage. However, DCD liver grafts are associated with a high incidence of primary graft nonfunction. We investigated the effectiveness of subnormothermic porcine liver perfusion, before transplantation from DCD, on graft viability. METHODS: Landrace pigs (25-30 kg) were randomly allocated to 3 groups (5 per group): heart-beating (HB) graft, transplanted after a 4-hour period of cold storage (CS); DCD graft, retrieved 20 minutes after apnea-induced cardiac arrest (respiratory withdrawal) and transplanted after a 4-hour period of CS; and subnormothermic ex vivo liver perfusion (SELP) graft, retrieved in the same manner as the DCD graft but perfused with a subnormothermic oxygenated Krebs-Henseleit buffer (21-25°C, 10-15 cm H2O) for 30 minutes in a simplified dripping manner, without a machine perfusion system, after the 4-hour period of CS, and subsequently transplanted. RESULTS: Although all animals in the HB group survived for >7 days, all animals in the DCD group died within 12 hours after transplantation. In the SELP group, 2 recipients survived for >7 days and another 2 recipients were killed on day 5. The survival rate was significantly better for SELP than for DCD grafts (P = .0016). The values of tumor necrosis factor α were not significantly different between the SELP and HB groups. Preserved structure of the parenchyma was observed in the SELP group on histologic examination. CONCLUSIONS: A simplified subnormothermic perfusion before liver transplantation is expected to improve graft viability and survival.

Distribution in time of seizures during presurgical EEG monitoring.

We studied the length of stay needed to record a certain number of seizures in a highly selected group of patients with intractable epilepsy in the final stages of presurgical monitoring. The mean length of stay needed to record one seizure was 2.9 to 3.7 days, depending on the recording technique, 4.5 to 5.5 monitoring days to record three seizures, and 6.1 to 7.6 days to record five seizures. It took 5 days to record at least one seizure in 90% of patients, 7 to 10 days to record three seizures, and 8 to 12 days to record five seizures. One-third of all patients had fewer than five seizures during their stay. Extrapolating from the results in our unit, we computed a need for 91 to 227 presurgical epilepsy monitoring beds to evaluate 2,000 to 5,000 similar epilepsy surgery candidates requiring invasive monitoring in the United States each year. Additional beds would be needed to monitor patients admitted for other purposes.