Print version ISSN 0034-7094
Rev. Bras. Anestesiol. vol.57 no.6 Campinas Nov./Dec. 2007
Oxygen concentrators: evolution of inspired concentration of oxygen and repercussions in an anesthetized patient with CO2 absorber system. Pilot study*
Fábricas concentradoras de oxígeno: evolución de la fracción inspirada de oxígeno y repercusiones en el paciente anestesiado en sistema con absorvedor de CO2. Estudio piloto
Jorge R. Moll, TSAI; Ana V. S. Moll, TSAII; Armin Guttman, TSAIII; Ivo P. Torres Filho, M.D., Ph.DIV; Manoel C. S. Almeida RibeiroV; Lígia Andrade da Silva Telles Mathias, TSAVI
em Medicina pela Faculdade de Ciências Médicas da Santa Casa de
São Paulo; Assessor Chefe de Projetos Estratégicos da Secretaria
de Estado de Saúde e Defesa Civil-RJ; MBA Gestão Integrada para
Excelência na FGV; Project Manager Professional, pelo Project Management
Institute; Co-responsável do CET do Serviço de Anestesiologia
do Hospital Geral de Bonsucesso
IIIntensivista do Hospital dos Servidores do Estado-RJ; Co-responsável do CET do Serviço de Anestesiologia do Hospital Geral de Bonsucesso
IIIResponsável pelo CET do Serviço de Anestesiologia do Hospital Geral de Bonsucesso
IVDoutor em Medicina; Professor-Adjunto dos Departamentos de Anestesiologia, Medicina de Emergência e Fisiologia da Virginia Commonwealth University-VCU Reanimation Engineering Shock Center (VCURES), EUA
VDoutor em Medicina; Professor-Assistente do Departamento de Medicina Social da Faculdade de Ciências Médicas da Santa Casa de São Paulo
VIDiretora do Serviço e Disciplina de Anestesiologia da Irmandade Santa Casa de Misericórdia de São Paulo e Faculdade de Ciências Médicas da Santa Casa de São Paulo; Responsável pelo CET-SBA, ISCMSP
OBJECTIVES: Resolution 1355/92 of the Conselho Federal de Medicina approved
minimal standards for the installation and operation of oxygen concentrators
(PSA) and recommended University Hospitals to undertake a prospective analysis
in order to improve the system. It motivated this pilot study whose objective
was to determine the clinical viability of using PSA oxygen by analyzing the
variation in oxygen concentration in the fresh gas flow (FGF) outlet and in
the inspired concentration of oxygen.
METHODS: An observational study with 30 patients, ASA I, undergoing upper abdominal surgeries using a CO2 absorber system and fresh gas flow (FGF) O293 at 500 mL.min-1. Weight, age, type and duration of the surgery, inspired and expired fraction of CO2 (FiCO2, PETCO2); inspired fraction of O2 (FiO2); and O2 concentration (O2ent) in the FGF. The following parameters were measured after intubation and every 10 minutes until the end of the procedure: PETCO2, FiO2, and O2. Results underwent statistical analysis and p < 0.05 was considered significant.
RESULTS: The inspired fraction of carbon dioxide was equal to zero in all patients and moments of the study, but there was a significant reduction in PETCO2 with time. The variables O2ent and FiO2 had similar tendencies with time (p = 0.1283), but the variable O2ent presented higher means (p < 0.001); evolution of mean O2ent and FiO2 was observed (p < 0.05).
CONCLUSIONS: This study demonstrated that the use of PSA oxygen, within the conditions proposed for the experiment, is safe and induced a progressive increase in O293 in the FGF and of FiO2.
Key Words: EQUIPMENT, Oxygen concentrators; VENTILATION: mechanical controlled, basal flow.
Y OBJETIVOS: La Resolución 1355/92, del Conselho Federal de Medicina
aprobó los estándares mínimos para la instalación
y el funcionamiento de las fábricas concentradoras de oxígeno
(P.S.A.) y recomendó que los Hospitales Universitarios hiciesen un análisis
prospectivo, permitiendo el perfeccionamiento del sistema, lo que motivó
la realización de este estudio piloto, cuyo objetivo fue el de estudiar
la viabilidad clínica del uso del oxígeno proveniente de las P.S.A.,
por medio del análisis de la variación de la concentración
de oxígeno en el flujo de gas fresco y de la fracción inspirada
MÉTODO: Estudio de observación, 30 pacientes ASA I, sometidos a cirugías del abdomen superior, utilizando sistema con absorvedor de CO2 y flujo de gases frescos (FGF) de O293 500 mL.min-1. Se evaluaron las variables peso, edad, tipo y tiempo de los procedimientos quirúrgicos; fracción inspirada y expirada del CO2 (FiCO2, PETCO2); fracción inspirada O2 (FiO2) y concentración O2 (O2ent) del FGF. Las variables PETCO2, FiO2 y O2 ent fueron medidas después de la intubación y a cada 10 minuto hasta el final de la anestesia. Los resultados fueron sometidos a tratamiento estadístico, considerándose significativo p < 0,05.
RESULTADOS: La FiCO2 presentó un valor igual a cero en todos los pacientes y momentos estudiados, pero ocurrió una disminución significativa de la PETCO2 en el transcurso del tiempo. Las variables O2ent y FiO2 presentaron tendencias similares a lo largo del tiempo (p = 0,1283), siendo que la variable O2 ent presentó medias superiores (p < 0,001); hubo evolución de las medias de O2 ent y FiO2 (p < 0,05).
CONCLUSIONES: Este estudio verificó que el uso de oxígeno proveniente de P.S.A., dentro de las condiciones propuestas para el experimento, mostró ser seguro e indujo un aumento progresivo de la concentración de O293 ofertada en el FGF y de la FiO2.
Medicinal oxygen can be furnished as a gas (in high pressure cylinders) or a liquid (in cryogenic reservoirs). It can also be produced in oxygen concentrators, known as PSA (Pressure Swing Adsorption), by a process based on selective adsorption of air components through zeolitic molecular sieves, which obtain oxygen by a physical method known as adsorption 1,2.
Difficulties in transportation due to climate conditions, lack of or problems with the roads, and absence of plants of cryogen in some countries, stimulated the use of oxygen concentrators, used in aircrafts, hospitals of the British Royal Navy during the Gulf War, during the war in Bosnia, and in humanitarian missions 3-12.
The use of oxygen concentrators, as a primary source of oxygen, connected to the supply network of hospitals, was standardized in 1992 by the National Standard of Canada (CAN/CSA-1991) 13 and by the International Organization for Standardization (ISO), and, in 1996, by the Associação Brasileira de Normas Técnicas, and the oxygen produced was called oxygen 93 (O293) 14.
Studies with oxygen concentrators in anesthesia using a ventilation system without CO2 absorber that do not allow rebreathing of exhaled gases, demonstrated it is safe to use 6,14,15. This was followed by doubts on whether it was safe to be used in systems with CO2 absorbers. Some publications demonstrated that, in systems with CO2 absorbers and fresh gas flow (FGF) of O293 at 0.5 L.min-1, argon can accumulate in the system 16-20.
In Brazil, Resolution 1355/92 of the Conselho Federal de Medicina (CFM) approved minimal standards for installation and functioning of oxygen concentrators and, on item 5, recommended University Hospitals to undertake prospective analysis in order to improve the system 21. This extremely specific recommendation motivated this pilot study, which objective was to study the viability of the clinical use of PSA oxygen by analyzing the variation in oxygen concentration in the fresh gas flow outlet and inspired fraction of oxygen.
After approval by the Ethics Commission of the Hospital Geral de Bonsucesso and signing of the informed consent, 30 adult patients of both genders, physical status ASA I, between 18 and 60 years of age, scheduled for open upper abdominal surgeries, were enrolled in this observational study.
Exclusion criteria included: smoking, use of medications, presence of associated diseases, and history of heart and/or lung disease.
All patients received pre-anesthetic medication and were taken to the operating room, where a catheter was introduced into a peripheral vein, hydration was instituted as well as monitoring of the heart rate (HR); electrocardiogram (ECG); non-invasive systolic pressure (SBP) and diastolic pressure (DBP); temperature (temp); and peripheral saturation of oxygen (SpO2).
Oxygenation started with a face-mask for three minutes with a CO2 absorber system and PSA O2 (O293) at 2.0 L.min-1 as fresh gas flow throughout the experiment. This was followed by anesthetic induction with 2.0 µg.kg-1 of fentanyl and endotracheal intubation.
Patients were maintained on controlled mechanical ventilation with a CO2 absorber system and a volumetric ventilator set for 10 breaths per minute, I/E ratio of 1:1.5, and tidal volume of 10 mL.kg-1.
Maintenance of anesthesia was accomplished with halothane and O293 using fresh gas flow (FGF) of 500 mL.min-1. A T-connector was placed between the 800S expiratory valve and the connection with the tracheal tube, with the lateral opening connected to the anesthesia monitor. This set up allowed monitoring of the inspired and expired fractions of the inhalational agent, carbon dioxide (FiCO2 and PETCO2), and oxygen (FiO2 and FeO2). FiO2 and FeO2 were measured by the paramagnetic method. Another T-connector with a lateral outlet was placed at the frontal gas exit of the anesthesia device, and connected to the monitor for paramagnetic analysis of oxygen concentration (O2ent).
After the end of the surgery and after they awaken from anesthesia, patients were extubated and transferred to the post anesthesia care unit (PACU).
The parameters investigated included: 1) inspired fraction of oxygen and oxygen concentration in the fresh gas flow outlet determined immediately after tracheal intubation (int) and every ten minutes until the end of anesthesia; 2) weight, height, age, type and duration of the surgery; 3) SpO2 on the arrival of the patient to the operating room, during initial oxygenation, at induction, immediately after tracheal intubation, and every ten minutes until the end of anesthesia; 4) PETCO2 determined immediately after intubation, and every ten minutes until the end of anesthesia.
Results are presented in descriptive tables containing means, standard deviation, and minimal and maximal values. Analysis of Variance for repeated measures was used for SpO2 and PETCO2.
Analysis of multivariate variance for repeated measures was used for the inspired fraction of oxygen and oxygen concentration in the FGF, and analysis of measurements along time.
A difference was considered statistically significant when p < 0.05. The tests used are part of the statistical package Sigma Stat for Windows, version 2.03, SPPS Inc.
The duration of the surgeries varied from 40 to 180 minutes, with a mean of 99 ± 34 minutes with decreasing number of patients whose surgeries lasted more than 50 minutes, with 50% of the surgeries lasting 110 minutes or less. Eighty per cent of the total was considered to be the number of patients necessary for statistical purposes; therefore, six patients were excluded from the study. After that, the mean duration of the surgeries was 80 minutes.
The final population was composed of 19 females (79.2%) and five males (20.8%), ages 47.4 ± 10.3 years, weighing 62.7 ± 5.8 kg, and height of 165.8 ± 9.4 cm (mean ± SD). The surgeries performed included 22 cholecystectomies and two antrectomies.
Analysis of Variance for repeated measures of SpO2 at the different moments did not show statistically significant differences (SpO2 p = 0.127).
FiCO2 was equal to zero at every moment of the study.
Table I shows the maximal, minimal, and mean values, along with standard deviations, of the expired fraction of carbon dioxide (%).
Analysis of Variance for repeated measures of the expired fraction of carbon dioxide at the different moments showed a statistically significant difference (p = 0.002).
Table II show the minimum, maximum, means, and standard deviation of the inspired fraction of oxygen, oxygen concentration in the FGF, and the differences between the mean values of oxygen concentration in the FGF and inspired fraction of oxygen during the study.
Multivariate Analysis of Variance for repeated measures showed:
- A lack of interaction among the variables studied and time: both variables demonstrated similar tendencies with time (p = 0.1189), but O2ent had higher means (p < 0.001);
- Evolution of the mean levels of O2ent and FiO2 (p < 0.05);
- Comparisons (linear and quadratic effects) revealed that mean O2ent and FiO2 demonstrated a quadratic effect (linear effect, p < 0.001; quadratic effect, p < 0.05).
Figure 1 shows the differences between the mean oxygen concentration in the FGF and inspired fraction of oxygen with time and their standard deviation. The confidence interval of the differences between mean oxygen concentration in the FGF and inspired fraction of oxygen increased with time due to a reduction in the number of observations, which increased the imprecision of the differences observed.
Patients did not develop any side effects during the study. All patients awakened in the surgery room and were transferred to the post anesthesia care unit extubated.
Questions regarding the use of oxygen from PSA in a system with CO2 absorber and Resolution 1355/92 of the CFM 21 recommending University Hospitals to undertake prospective studies stimulated health professionals to develop research studies on the final product and the consequences of its use in anesthesia 22, and they also motivated this study.
Another factor that led to this study was the virtual lack of publications on the subject in the national and very few in the international medical literature. Besides, most of the studies found in the international literature used laboratorial models, and prospective studies with patients under anesthesia had a reduced number of subjects, five patients per study.
In the present study, all measurements of the inspired fraction of carbon dioxide were equal to zero, which confirms that there was no rebreathing of CO2, attesting to the efficacy of the system used. Amaral et al. 24 stated that values of partial pressure of CO2 above zero in the inspired mixture indicated inhalation of this gas and that it is often due to failure of the system to absorb CO2 (soda lime) or defective unidirectional valves.
Although there had been significant reduction in the expired fraction of carbon dioxide during the study period, there were no clinical implications and, even with the ventilation regimen proposed, all patients awakened from anesthesia and were extubated in the operating room.
Analysis of Variance of the inspired fraction of oxygen and entrance oxygen demonstrated that the evolution of mean O2ent and FiO2 during the study had similar tendencies, but O2ent showed higher means. At the beginning of the study, probably due to pollution of the system with atmospheric air, inspired fractions of oxygen were lower than at the end of the experiment (100 min). Parker 17 eliminated this contamination initiating measurements after a period of 20 minutes, when the oxygen fraction in the gas sample had stabilized, and only then FGF was reduced to 500 mL.min-1.
Three experimental studies 17-19 demonstrated that the use of PSA oxygen in CO2 absorber systems could lead to an increase in argon concentration in the system. However, in those studies the authors included two studies with five patients in which O293 was used with a fresh gas flow of 0.5 L.min-1 and concluded that O293 was safe to be used since nitrous oxide was not added to the system.
Rathgeber et al. 20 studied the efficacy and possible disadvantages of using PSA oxygen, in an experimental lung model, using fresh gas flow of 0.5 to 1.0 L.min-1, and concluded that O293 did not cause additional problems, suggesting that O293 would be a viable alternative in anesthesia.
Friesen et al. 5 surveyed 53 hospitals in Canada that used PSA oxygen in their intensive care units, emergency rooms, and operating rooms. They observed that most places used fresh gas flow of 2.0 L.min-1. Adverse events were not reported.
The present study demonstrated that the use of PSA oxygen, within the conditions proposed, induced a progressive increase of O293 in the FGF and in FiO2, with statistically significant differences between them that might have been caused by argon accumulation. However, we did not observe any adverse events during the study period of 80 minutes.
These results reinforce the studies of Friessen et al. 5, who considered that it is important to monitor oxygen concentration in the terminal, and of Parker 18, who stated that: "Since it is possible to monitor oxygen concentration in the ventilation system during anesthesia with low flow or closed systems independent from the supply source, the use of oxygen from oxygen concentrators cannot prevent the use of low flow anesthetic techniques."
Thus, this pilot study, by confirming the results of Parker 17-19, led to the conclusion that oxygen from oxygen concentrators is safe to be used in the ventilation system during low flow anesthesia of 500 mL.min-1, in the conditions of the present study, with continuous monitoring of oxygen concentration.
01. Penny M Physical and chemical properties of molecular Sieves. The pressure absorption cycle. The pressure absorption cycle. Health Serv State, 1987;61:44-49. [ Links ]
02. Henke I New technology for oxygen supply. Hospital Engineering Connection 1991; 13:11-25. [ Links ]
05. Friesen RM Oxygen concentrators and the practice of anaesthesia. Can J Anaesth 1992;39:R80-R84. [ Links ]
06. Carter JA, Simpson PJ The "Permox" oxygen concentrators. Their performance and potential application. Anaesthesia 1985;40:560-565. [ Links ]
07. Bull PT, Merrill SB, Moody RA et al. Anaesthesia during the Falklands campaign. The experience of the Royal Navy. Anaesthesia 1983;38:770-775. [ Links ]
08. Harding RM The role of molecular Sieves in modern fighter aircraft. Health Serv State, 1987;61:55-56. [ Links ]
09. Acheson DSC Confllict in Bosnia 1992-3. Brit Med J, 1999; 319:1639-1642. [ Links ]
10. Nordmann GR, Woolley T Unusual critical incident: chemical gas alert. Anaesthesia, 2003:58:926. [ Links ]
11. Hodges SC, Hodges AM A protocol for safe anaesthesia for cleft lip and palate surgery in developing countries. Anaesthesia 2000;55:436-441. [ Links ]
12. Shrestha BM, Singh BB, Gautam MP et al. The oxygen concentrator is a suitable alternative to oxygen cylinders in Nepal. Can J Anaesth 2002;49:8-12. [ Links ]
13. Canadian Standards Association. CAN/CSA Z305.6-92: Medical Oxygen Concentrator Central Supply System: for use with nonflammable medical gas piping systems. Toronto, 1992. [ Links ]
14. Associação Brasileira de Normas Técnicas NBR 13587: Estabelecimento assistencial de saúde Concentrador de oxigênio para uso em sistema centralizado de oxigênio medicinal. 1996. [ Links ]
15. Suzuki T, Asai S A simple anesthetic system with oxygen concentrator for war or disaster situation. Masui, 2003;52:662-665. [ Links ]
16. Fenton P The Malawi anaesthetic apparatus for developing countries. Anaesthesia, 1989;44:498-503. [ Links ]
17. Parker CJR The anesthetist's experience of PSA oxygen: the problem of argon accumulation in low flow anaesthesia. Health Serv State, 1987;61:65-70. [ Links ]
18. Parker CJR Concentrator oxygen and flow-low anaesthesia: 1. Eur Acad Anesthesiol 1997;1:55-56. [ Links ]
19. Parker CJR, Snowdon SL Predict and measured oxygen concentrations in the circle system using low fresh gas flows with oxygen supplied by an oxygen concentrator. Brit J Anaesth, 1988;61:397-402. [ Links ]
20. Rathgeber JZ, Kietzmann D, Kraus EJ et al. Efficiency, argon accumulation, and accuracy of anaesthetic gas monitoring during mechanical ventilation in a circle system using an oxygen concentrator. Anaesthesist, 1995;44:643-650. [ Links ]
21. Conselho Federal de Medicina (Brasil) Resolução CFM n° 1.355/92. Estabelece parâmetro mínimo de segurança para concentração de oxigênio utilizado em hospitais. Diário Oficial [da] República Federativa do Brasil, Brasília, DF, p. 12648, 11 set. 1992. Seção 1. [ Links ]
22. Moll JR Usinas concentradoras de oxigênio: estudo da variação da fração inspirada de oxigênio e suas implicações no paciente anestesiado. Rev Bras Anestesiol, 1995;45(Supl 19):238. [ Links ]
23. Conselho Federal de Medicina (Brasil) Resolução CFM n° 1.363/93. Normatiza as condições mínimas de segurança em Anestesiologia. Diário Oficial [da] República Federativa do Brasil, Brasília, DF, p. 3439, 22 mar.1993. [ Links ]
24. Amaral JLG, Ferreira ACP, Ferez D et al. Monitorização da respiração: oximetria e capnografia. Rev Bras Anestesiol, 1992; 42:51-58. [ Links ]
Dra. Lígia Andrade da Silva Telles Mathias
Alameda Campinas, 139/41
01404-000 São Paulo, SP
Submitted em 30
de setembro de 2006
Accepted para publicação em 21 de agosto de 2007
* Received from Hospital Geral de Bonsucesso MS-RJ. Apresentado como Dissertação de Mestrado na Faculdade de Ciências Médicas da Santa Casa de São Paulo.