CORRESPONDENCE major rural, intermediate rural, small rural) from 1993 to 1994 were reviewed. Furthermore, the successful discharge rate for selected patients was remarkably low. Of patients selected by emergency physicians forthe proposed EDOU,only 32.2% (range 66.7% to 8.3%)were dis- charged from the hospital within 3 days. Most studies demonstrate that 60% to 90% of observed patients are successfully dis- charged from the EDOU within less than 24 hours.3-6This may be why adequate nursing FTEs were not realized for more beds. If the cohort of identified patients had an 80% suc- cessful discharge rate, displaced staff would have justified 4.2 FTEsor 2.5 beds. Identification of hypothetical, appropri- ate EDOUcases by emergency physicians who were not experienced with EDOUsmay have limited patient selection. At our institu- tion, it took 6 months from the opening of our ED0U for our emergency physicians to maxi- mally use the EDOU,despite physician train- ing in patient selection. Because of these shortcomings, the data do not support the conclusions of the article. Michael A Ross, MD Emergency Center Observation Unit William Beaumont Hospital Royal Oak, MI Robert J Zalenski, MD Department of EmergencyMedicine Wayne State University School of Medicine Detroit, MI 47/8,/99688 1. Sinclair D, Green R: Emergency department observation unit: Can it befunded through reduced inpatient admission? Ann Emerg Med 1998;32:670-675. 2. Decoster C, Peterson S, Kasian P: Alternatives to Acute Care. Manitoba, Canada: Manitoba Centrefor Health Policy and Evaluation (University of Manitoba), July 1996:37-40. 3. McDermott M, Murphy D, Zalenski R, et al: A compari- son between emergency diagnostic and treatment unit and inpatient care in the management of acute asthma. Arch Intern Med 1997;157:2055-2062. 4. Bril]manJ, Dunbar L, Graft L, et al: Management of observation units. Ann Emerg Med 1995:823-830. 5. Graff L, DelaraJ, Ross M, et al: Impact on the care of the emergency department chest pain patient from the chest pain evaluation registry (CHEPER)study. Am J Cardiol 1997:563-568. 6. Gibler WB, Runyon J, Levy R, et al: A rapid diagnostic and treatment centerfor patients with chest pain in the emergency department. Ann Emerg Med 1995;25:1-96. In reply. We thank Drs Rossand Zalenski for their thoughtful letter concerning our article on the funding of EDOUs.As they point out, our study was a theoretical one and the actual practice of utilization of such a unit may be very different, depending on the practice set- ting. We made the assumption that 2 of the 4 beds would be filled with patients who are presently "observed" in the main emergency department, and that 1 bed would be filled with an "avoided admission" patient. We know that patients presently "observed" in EDs do not receive optimum nursing care, and if an observation unit was constructed, extra staff would have to be added to the nursing complement. Five FTEs would be required and the potential bed reductions would only equal 1.68 FTEs.Furthermore, because the effect of an observation unit has a scattered effect on the institution, it is doubtful that single beds on units could be closed and yield staff savings. It is correct to point out that the potential number of patients is very small. We specifi- cally excluded patients with chest pain, because we believe that this patient group requires a specialized dedicated unit (chest pain evaluation unit) with cardiac monitoring capacity. We do support the assumption that the discharge rate from an EDOU would be higher than from inpatient beds. The key finding in the study was that the effect of such a unit is quite diffuse in its effect on inpatient bed utilization, and thus savings from bed closures will not justify the fund- ing foran EDOU. Douglas E Sinclair, MD, FRCPC Robert Green, MD Department of EmergencyMedicine Queen Elizabeth II Health Sciences Center Halifax, Nova Scotia, Canada 47/8/99687 Pressure Immobilization for Neurotoxic Snake Bites Tothe Editor. The description of successful treatment of neurotoxic envenomation by a king cobra ( Ophiophagus hannah) (article #94149) raises several important points. 1The authors overlooked the potentially critical issue of first aid. In Australia the pressure immobilization technique2 has proved both experimentally3 and clinically4.5 useful in controlling the movement of elapid snake venom from the bite site into the systemic circulation. This can provide the emergency physician, faced with a potentially life- threatening situation, with vital time to identify the snake, obtain appropriate and sufficient monovalent antivenom, and undertake investigations into the severity of the evenornation. By contrast, the usefulness of this tech- nique for evenomations associated with major local tissue necrosis, such as those caused by many Asian cobras and vipers, is controversial .6This is related to the hypothe- sis that local tissue damage may be intensi- fied bythe immobilization of venom cytotox- ins and proteases. Local effects are not usually problematic with Australian elapid envenomations. Experimentally, this tech- nique is effective in immobilizing the venom of the Indian cobra (Naja naja)and the Eastern diamond back rattlesnake (Crotalus adamanteus) 7 without exacerbation of local tissue necrosis. At present, however, aside from Australian elapids and sea snakes, this first-aid technique is only recommended after bites by dangerous neurotoxic species such as kraits (Bungarus), cobras which cause little local envenoming (eg, Naja philippinensis), and the king cobra. 8 Its use in the published case 1might have obviated the need for intubation and venti- lation. Therefore, herpetologists dealing with these snakes and their treating physi- cians should be aware of this technique and its potential value. The early application of pressure immobilization of first aid may prove lifesaving particularly when antivenom may not be readily available. In addition, it might reduce the total amount of antivenom required and thence the risk of early and delayed antivenom reactions-- and total cost. Ineffective and sometimes dangerous first-aid procedures such as arterial tourniquets, incision of wounds, hypothermia, electric shock, and the use of aspiration devices should be actively dis- couraged. However, further research is 294 ANNALS OF EMERGENCY MEDICINE 34:2 AUGUST 1999