Background: As an essential hospital facility, the central oxygen supply system should be designed with features allowing backup and/or redundancy in the event of system failure. As part of an organized institutional review of The Cleveland Clinic Foundation hospital inpatient central oxygen supply system, we undertook a survey of all hospitals in two Ohio cities to determine the characteristics of hospital central supply systems.
Methods: The questionnaire was developed and completed by structured telephone interview during calls placed to managers of facilities engineering departments in 35 hospitals in the greater Cleveland and Columbus, Ohio, metropolitan areas. To encourage candid responses to the telephone interview, respondents were assured that institutional names would not be presented in published reports. The questionnaire addressed the type of primary and reserve oxygen sources in the hospital, whether a backup system exists, and if so, in what configuration. The questionnaire also addressed whether any unplanned interruption or other problem (such as contamination of the piped-in oxygen supply) had ever occurred in the facility.
Results: Of the 35 eligible hospitals, responses were available from 32 (91.4%). The mean number of beds in the hospitals responding was 397 +/- 251 (standard deviation), and the original construction dates of the responding hospitals ranged from 1887 to 1982. All 32 responding institutions reported a reserve system, described as a liquid reservoir in 72% (23/32), manifolded cylinders in 16% (5/32), and both in 13% (4/32). Twenty-six (81%) of those responding reported having the reserve supply liquid or manifolded gas cylinders at the same location as the primary liquid vessel. The supply lines of these contiguous primary and reserve containers were reported to join proximal to entering the hospital structure, so at each of these 26 hospitals the primary and reserve systems depend on a single length of pipe. Only 4 (13%) of the hospitals have manifolded cylinders in addition to the primary and reserve liquid supplies. These manifolds are in different locations from the primary and reserve, have physically separate feed lines, and represent the only true examples of redundant piped-in oxygen supplies recorded during the survey. Of the 32 hospitals surveyed, 5 (16%) reported having experienced mishaps with the bulk liquid supply.
Conclusions: (1) Not surprisingly, most of the hospitals in these two urban areas use bulk liquid oxygen systems (with primary and reserve liquid reservoirs) as the main central supply source, with some providing manifolded cylinders as backup. (2) Mishaps regarding the main supply line from the bulk oxygen reservoir were reported by 16% (5/32) of responding institutions. (3) In this context, the fact that most main and reserve tanks were contiguous and fed through a single line to the hospital facility suggests ongoing risk for interruption of an oxygen supply by line mishaps (e.g., street repair). (4) Contingency planning to lessen the risk of an interrupted supply should involve back-up systems with physically separated feed lines, as well as tanks of manifolded cylinders along the course of the main hospital oxygen circuit line.