Oxygen therapy constitutes an effective means to treat and prevent the symptoms of hypoxemia, by delivering concentrated oxygen to the airways . Its clinical relevance led to a marketing authorisation for medicinal oxygen in 1997, which thereby became a medicine. Its many therapeutic indications include the correction of hypoxemias of various aetiologies requiring normo- or hyperbaric oxygen therapy (chronical or acute respiratory failures) .
In hospital wards, oxygen is usually supplied via wall outlets in rooms. Mobile oxygen administration devices are also needed for inpatients during transfers or medical examinations. Today, this function is accomplished by portable cylinders filled with pressurized gaseous medicinal oxygen.
The format of these cylinders (volume and colour) are regulated by the norm NF EN 1089–3 (2011) . Two devices guarantee their safe use. (1) A pressure reducer lowers the pressure to 3.5 bars at the outlet of the cylinder. Low-capacity medicinal oxygen cylinders available on the market are mostly equipped with an integrated pressure-reducing valve; this device has gradually replaced manual-connected valves, for safety reasons . (2) A flow sensor with a panel of pre-set flow rates, ranging from 1 to 15 litres per minute (calibrated outlet ports).
The residual oxygen pressure can usually be read directly from a manometer gauge integrated on the front of the cylinder (Figure 1). The initial fill pressure of portable cylinders is 200 bars ; it diminishes as the cylinder is used, along with the cylinder’s autonomy. The remaining gas volume can be estimated by applying Mariotte’s law, with the formula V=P×n; V is the residual gas volume, P the pressure in bars indicated by the manometer, and n the water volume of the cylinder under air pressure .
This estimation of the remaining gas volume is paramount when moving or transferring patients: it must be high enough to administer the prescribed oxygen flow rate, for a sufficient duration. The nursing staff must thus calculate the autonomy of the oxygen cylinder by dividing the remaining oxygen volume by the administered flow rate, before it can be used. These calculation steps are cumbersome and confusing, although they are made easier by a calculation abacus displayed on the side of the cylinder’s sticker. This abacus is made by the supplier (Table 1).
In our establishment, medicinal oxygen cylinders equipped with a manometer gauge that were sent back for replacement often had a residual pressure close to 50 bar, corresponding to a residual oxygen volume of 250 L (Table 2). Calculating the autonomy is constraining and encourages users to replace oxygen cylinders as soon as the red zone of the manometer is reached (< 50 bar). This early replacement leads to 25 % wastage of the oxygen contained in each cylinder.
To address this issue, some suppliers propose a new type of cylinder with a digital display of the remaining autonomy (Figure 2).
On this type of cylinder, the digital display system is coupled to the pressure reducer and flow sensor.
When the pressure reducer is set in the ‘Open’ position, the screen displays the available volume of oxygen; as soon as an administration flow rate is selected, the autonomy is displayed in hours and minutes. This real-time automatic calculation considers the selected flow rate and residual pressure. Aside from the time advantage, this technology avoids eventual manipulation errors linked to incorrect calculations. By assuring the time of administration for pressures under 50 bar, this type of cylinder could help to optimize medicinal oxygen consumption by hospital departments.
We tested a digital autonomy display model by comparing it to the manometer gauge model. The objective of this study was to estimate the potent advantages of this new device on cylinder turnover rate and full utilization (estimated by their residual pressure at replacement).
Materials and Methods
This prospective, open, controlled, “before-after” type study aimed to compare the use of manometer gauge cylinders (Presence™, Air Liquide Santé France) with digital autonomy display cylinders (Takeo™, Air Liquide Santé France). All other characteristics of these cylinders were identical: B5 format (5 L medicinal oxygen, 200 bar pressure), nominal capacity of 1 m3 (at 15 °C), presence of a pressure reducer and flowmeter, with the following pre-set flow rates: 1; 1.5; 2; 3; 4; 6; 9; 12; 15 litres per minute.
Clinical wards and duration of the study
Two clinical wards were selected for this evaluation because of their significant oxygen consumption: the cardiovascular surgery resuscitation unit (RU) and the emergency department (ED). The latter had an important staff turnover. For the first two-month period, these wards had an initial allocation of cylinders with manometer gauges; this stock was then replaced by digital autonomy display cylinders for the next two months (no wash-out period). At the beginning of each period of the study, the initial stock was 7 cylinders for the RU and 17 for the ED. At the end of the first period, practical training was provided by Air Liquide Santé France to the nursing teams, to present the Takeo™ cylinder to them and its conditions of use.
Oxygen consumption study
During the two periods, the medicinal oxygen cylinders were replaced at the wards’ request. Each cylinder’s residual pressure was recorded by means of the manometer gauge or the digital autonomy display system. The cylinders were then sorted into two categories depending on whether their residual pressure was above or below 50 bar.
For each ward, the turnover rate was measured for each period (corresponding to the ratio: sent back cylinders/total number of dispensed cylinders).
A cost study was performed, based on the pre-tax unit price (PTUP) communicated by Air Liquide Santé France for each cylinder. This PTUP includes the following costs: 5 L medicinal oxygen, cylinder rental and delivery costs. The annual consumption cost was calculated for each cylinder, based on its PTUP and current annual consumption. The difference between the annual costs for the two cylinder types was then related to their respective turnover rates (estimated by the consumption study). To do so, we assumed that a variation in turnover rates would result in a consequent and proportional variation in annual gas consumption, number of rented cylinders and number of deliveries. Thus, we calculated the economic balance associated with the new device.
The mean cylinder turnover rate and the mean percentages of cylinders replaced with a pressure below vs above or equal to 50 bar were compared by chi-square analysis. The significance threshold was set to 5 %. The statistical analysis was first performed based on the overall headcount, and then separately for each ward.
Oxygen consumption study
During the first phase of the study, 69 cylinders were replaced (21 for RU and 48 for ED). During the second phase, the number of replaced cylinders fell to 48 (15 for RU and 33 for ED).
The overall turnover rate for the Presence™ cylinders was 69 of 74 (93 %) whereas it was 48 of 70 (69 %, p<0.001) for the Takeo™ cylinders. This trend was also true for each of the two wards (Figure 3).
The overall proportion of Presence™ cylinders replaced with a residual pressure below 50 bar was 59 %, whereas the proportion of Takeo™ cylinders was 75 % (p=0.08). This trend was also true for the two wards (Figure 4).
The difference in unit price of Takeo™ vs Presence™ cylinders is due to the extra cost of bottle rental, linked to the integration of the new autonomy display technology (Table 3). The other costs contributing to the PTUP (medicinal oxygen and delivery) are the same for the two types of cylinders.
This study shows that replacing manometer gauge cylinders with digital autonomy display cylinders decreases the turnover rate by 24 %. In parallel, use of the Takeo™ model increased the proportion of cylinders replaced with a residual pressure below 50 bar by 16 %. These results suggest optimized use of the oxygen supplied in the Takeo™ cylinders.
This optimization of consumption was particularly marked in the ED, where the cylinder turnover rate was reduced by 29 %. The proportion of cylinders replaced with a residual pressure of less than 50 bar was also significantly greater for cylinders with the digital autonomy display (+17 %). These data suggest that displaying autonomy in time units enables optimized use of the cylinder by the ED, with a lower residual gas volume at replacement. An autonomy display is less alarming for the staff than a needle moving to the red pressure zone of a manometer gauge. A residual pressure of 50 bar allows the administration of oxygen for two hours at a 2 L/min flow rate, sufficient to transfer a patient treated for a mild case of hypoxemic acute respiratory failure , for example.
There was a lower impact of the new cylinder type on oxygen consumption in the RU, with a smaller decrease in the turnover rate than that of the ED (17 %). The increase in the proportion of cylinders with a residual pressure under 50 bar was also less (+14 %). The patients treated in this unit (cardiovascular surgery) may need higher oxygen therapy flow rates (up to 15 L/min) for their exams and transfers. With such a high flow rate, a 100 bar pressure only guarantees 30 minutes of autonomy. The RU thus replaces oxygen bottles as soon as the residual pressure is less than 100 bar.
The 30 % optimization of consumption with the digital autonomy display claimed by the supplier was confirmed for the ED (–29 %) but not for the RU (–17 %). These results suggest that the digital model is better adapted to hospital departments that require lower oxygen flow rates (such as the ED). It is also possible that the high rate of staff turnover in this ward plays a role in oxygen wastage (problem of training), and consequently in the greater impact of using Takeo™ cylinders in this hospital department.
The economic analysis shows that optimising the consumption of B5 oxygen cylinders at the hospital by 25 % through the use of Takeo™ cylinders, should result in 10 % savings though a higher unit price. The turnover rates observed in the two pilot care units were very different. This trial should be complemented by a study expanded to other wards to confirm the global 25 % decrease in oxygen cylinder consumption.
Finally, an environmental impact study performed by Air Liquide Santé France shows a 200-gram decrease in the amount of carbon emitted during the production and transport of Takeo™ cylinders, relative to Presence™ cylinders. This estimation is based on a 30 % decrease in the turnover rate for Takeo™ cylinders. This figure must be eventually adjusted based on the actual level of optimisation.
The implementation of B5 medicinal oxygen cylinders with digital autonomy display instead of manometer gauge cylinders appears to be a useful approach for optimising the use of this medication by hospital departments and limiting its wastage. The ergonomics of the evaluated device avoids the need to calculate the autonomy, which is cumbersome and time-consuming for the care staff; it results in more efficient use of the amount of gas available in each cylinder. Finally, this change in practice would reduce associated expenditures.
We warmly thank:
The cardiovascular surgery resuscitation unit and the emergency department for participating in this study;
Mr François Berge for the technical implementation of this work;
Ms Marie-Laure Lucas for the data analysis;
Drs Hélène Blasco and Adeline Bourdareau for their scientific proofreading.
1. AARC. AARC Clinical Practice Guideline: Oxygen therapy for adults in the acute care facility - 2002 revision & update [Internet]. 2002 [cited 2015 Feb 22]. http://www.rcjournal.com/cpgs/pdf/06.02.717.pdf
2. Résumé des Caractéristiques du Produit - Oxygène médicinal Air Liquide Santé France 200 bar, gaz pour inhalation, en bouteille [Internet]. 2012 [cited 2015 Feb 22]. http://www.airliquidesante.fr/file/otherelement/pj/f3/92/7c/31/amm-rcp-oxygene-gc-06-07-7582762510900882941.pdf
3. AFNOR. NF EN 1089–3 - Bouteilles à gaz transportables - Identification de la bouteille à gaz (GPL exclu) - Partie 3 : code couleur [Internet]. 2011 [cited 2015 Feb 22]. http://www.boutique.afnor.org/norme/nf-en-1089-3/bouteilles-a-gaz-transportables-identification-de-la-bouteille-a-gaz-gpl-exclu-partie-3-code-couleur/article/633129/fa165917
4. Giraud C. Les bouteilles d’oxygène à détendeur intégré. Ann Fr Anesth Réanim 2003;22:301–11. Google Scholar
About the article
Elsa Bodier-Montagutelliis a PharmD candidate. She studied pharmacy in University Paris-Sud 11 (Châtenay-Malabry, France) and is currently resident at Tours University Hospital (France) since november 2013. During her residency, she spent one year in the field of hospital procurement.
Antonin Maréchal is currently a pharmacist in the chemotherapy preparation unit at the University Hospital of Reims. He obtained his PharmD degree in 2015 at the University of Tours, where he was a former resident of Tours University Hospital.
Richard Porcher has been working in the area of medicinal fluids since 1991. Hewas previously an installer, then technician at Air Liquide Santé before integrating Tours University Hospital in 2003, asmedicinal fluids technician. He also supervises the technical workshop. His job consists in managing the distribution of medicinal gas cylinders to care units (in collaboration with the pharmacy), the maintenance of technical and medicinal fluids networks as well as external companies’ works within the hospital.
Francis Remérand is Head of the department of anaesthesiology and critical care at the Pharmacy Department of the University Medical Center of Tours (France) since 2014. He performed an internship at the Lariboisiere University (Paris VII, France), where he completed his MD in 2001. He worked as an assistant professor at the Pitié Salpétrière University (Paris VI), and as a physician in the Tours University Hospital, where he completed his PhD in 2011. His special interests and research projects include lung ultrasound, pleural drainage, hemodynamic monitoring, quality control, post operative pain, regional anesthesia, and daycase surgery.
Renaud Respaud is a lecturer at the University François Rabelais of Tours in Pharmaceutical Sciences and Pharmacist in the sector of quality control and clinical research at University Hospital of Tours. Former resident at Assistance Publique – Hôpitaux de Paris, he obtained is PharmD degree in 2008 and completed his PhD in analytical sciences in 2011. His work and researches are focused on formulation and analytical development of biotherapeutics to treat pulmonary diseases through inhalation.
Published Online: 2016-06-08
Published in Print: 2016-06-01
Conflict of interest statement: Air Liquide Santé France funded Antonin Maréchal’s business trip to the Rencontres Convergences Santé Hôpital in September 2014, for the presentation of this study. Other than this, authors state no conflict of interest. All authors have read the journal’s publication ethics and publication malpractice statement available at the journal’s website and hereby confirm that they comply with all its parts applicable to the present scientific work.