Lack of significant treatment effect of plasma exchange in the treatment of drug-induced toxic epidermal necrolysis?

Objective: Comparison of outcome in patients with toxic epidermal necrolysi s (TEN) in patients who received plasma exchange (PE) compared with the res ults in two other centres that used almost identical treatment protocols bu t without PE. Design: Retrospective comparative case series with two recently published c ase series serving as controls. Setting: National burns intensive care unit (ICU) and Department of Transfu sion Medicine at Linkoping University Hospital, Sweden. Patients: 8 consecutive patients admitted with TEN who received PE during 1 987-1997. Interventions: Neither prophylactic antibiotics nor cortisone were used. Th e patients were given a median of 5.5 PE treatments (range 1-8). Results: Eight patients with a median (range) age of 45 years (5-89) and wi th a median skin involvement total body surface area (TBSA) of 38% (12-100) were treated. The length of stay in the burns ICU was 15 (13-25) days and the time from onset of the cutaneous signs until complete re-epithelialisat ion was 24 days (13-55) for the seven survivors. Five patients fulfilled th e diagnostic criteria of sepsis. One patient with extensive ischaemic cardi ac disease developed septic shock and died (mortality 12.5%). Two patients developed side effects from PE. Conclusions: To our knowledge, this is the largest series yet presented usi ng PE in the treatment of TEN. Our results, in patients with less cutaneous involvement, similar causative agents, and similar demographic data as in the other two studies (controls), were no different as far as mortality, le ngth of stay, or time to re-epithelialisation were concerned. This finding does not support the use of PE in the treatment of TEN.