- Email: [email protected]
Incidence and prediction of inadequate preoxygenation before induction of anaesthesia
Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58
Original article
Incidence and prediction of inadequate preoxygenation before induction of anaesthesia Incidence et facteurs pre´dictifs d’e´chec de pre´oxyge´nation avant induction anesthe´sique C. Baillard a,*, F. Depret a, V. Levy b, M. Boubaya b, S. Beloucif a a
EA 3409, department of anaesthesiology and critical care medicine, Avicenne university hospital, Paris-13 university, AP–HP, 125, route de Stalingrad, 93009 Bobigny, France b Clinical research unit, Avicenne university hospital, Paris-13 university, AP–HP, 125, route de Stalingrad, 93009 Bobigny, France
A R T I C L E I N F O
A B S T R A C T
Article history: Received 27 October 2013 Accepted 30 December 2013 Available online 24 February 2014
Background. – Preoxygenation aims to obtain an expired oxygen fraction (FEO2) 90%. Little is known about the incidence and predictors of inadequate preoxygenation in the clinical setting. Patients and methods. – Over a 12-month period, 1050 consecutive preoperative patients were prospectively included. Preoxygenation was performed for 3 minutes with a facial mask using a machine circuit and 12-L/min oxygen flow. Inadequate preoxygenation was defined as an FEO2 < 90%. A logistic regression was performed to identify incidence and independent predictors. Results. – The patient characteristics were: age 51 20 years, 47% male, BMI of 26 5 kg/m2, and ASA score (median [extremes]) of 2 [1–4]. Inadequate preoxygenation was observed in 589 patients (56%). The effective FiO2 delivered was lower in the patients with inadequate preoxygenation than in those with adequate preoxygenation, 95 3% vs. 98 2%, P < 0.001. The difference between the FiO2 and the FEO2 was higher (12 6% vs. 6 3%, P < 0.0001) in patients with inadequate preoxygenation compared with those with adequate preoxygenation. The independent risk factors for inadequate preoxygenation were: firstly, bearded male (odds ratio [OR] of 9.1 [2.7–31.4] P < 0.001); secondly, beardless male (OR 2.4 [1.6–3.4] P < 0.001), thirdly, ASA score of 4 (OR 9.1 [2.6–31.2] P < 0.015); fourthly, ASA score of 2–3 (OR 2.4 [1.6–3.4] P < 0.015); fifthly, lack of teeth (OR 2.4 [1.2–4.5] P < 0.006), and lastly age > 55 years (OR 1.8 [1.2–2.7] P < 0.005). Conclusion. – Inadequate preoxygenation, defined as an FEO2 < 90% despite 3-min tidal volume breathing, was a common occurrence. The predictive factors share an overlap with those previously identified for difficult mask ventilation. ß 2014 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved.
Keywords: Preoxygenation General anesthesia Airway
R E´ S U M E´
Mots cle´s : Pre´oxyge´nation Anesthe´sie ge´ne´rale Voies ae´riennes
Objectifs. – Peu de donne´es sont actuellement disponibles concernant l’incidence et les facteurs pre´dictifs d’un e´chec de pre´oxyge´nation en pratique clinique. Patients et me´thodes. – Au cours d’une pe´riode de 12 mois, 1050 patients conse´cutifs ont e´te´ prospectivement inclus. La pre´oxyge´nation e´tait re´alise´e en volume courant pendant 3 minutes en utilisant le circuit machine avec un de´bit d’oxyge`ne de 12 litres par minutes. Une FEO2 < 90 % a` l’issue de la pre´oxyge´nation de´finissait un e´chec. Une analyse univarie´e puis multivarie´e ont e´te´ re´alise´es pour identifier les facteurs inde´pendants d’e´chec de pre´oxyge´nation. Re´sultats. – Les caracte´ristiques des patients e´taient : aˆge 51 20 ans, 47 % hommes, IMC 26 5 kg/m2, et score ASA : 2 [1–4]. L’incidence globale d’e´chec de pre´oxyge´nation e´tait retrouve´e chez 589 patients (56 %). La FiO2 de´livre´e e´tait plus faible chez les patients avec e´chec de pre´oxyge´nation compare´e aux patients correctement pre´oxyge´ne´s, 95 3 % vs 98 2 %, p < 0,001. Les facteurs de risques inde´pendants d’e´chec de
* Corresponding author. E-mail address: [email protected] (C. Baillard). 0750-7658/$ – see front matter ß 2014 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.annfar.2013.12.018
e56
C. Baillard et al. / Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58
pre´oxyge´nation e´taient : premie`rement, les hommes barbus (odds ratio [OR] of 9,1 [2,7–31,4] p < 0,001) ; deuxie`mement, le sexe masculin (OR 2,4 [1,6–3,4] p < 0,001) ; troisie`mement, un score ASA 4 (OR 9,1 [2,6–31,2] p < 0,015) ; quatrie`mement, un score ASA 2–3 (OR 2,4 [1,6–3,4] p < 0,015) ; cinquie`mement, l’e´dentation (OR 2,4 [1,2–4,5] p < 0,006) et finalement, l’aˆge > 55 ans (OR 1,8 [1,2–2,7] p < 0,005). Conclusion. – L’e´chec de pre´oxyge´nation de´fini par une FEO2 < 90 % a` l’issue de 3 minutes de ventilation spontane´e en volume courant est une situation courante. Les facteurs de risques observe´s sont communs a` ceux pre´alablement identifie´s pour la pre´diction de la ventilation au masque difficile. ß 2014 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Publie´ par Elsevier Masson SAS. Tous droits re´serve´s.
1. Introduction The standard practice of preoxygenating with 100% oxygen before the induction of general anaesthesia is strongly recommended because it delays the onset of arterial hemoglobin desaturation during a potential prolonged apnea [1,2]. The oxygen loading process during preoxygenation involves factors related to alveolar, arterial, tissue, and venous compartments and those that interact with the delivered fraction of inspired oxygen (FiO2) [3]. Because most of the oxygen is stored in the lungs as a function of residual functional capacity (FRC), the primary goal is to maximize the alveolar fraction of oxygen, which depends on the effective FiO2 delivered. When most of the nitrogen in the lung is replaced by oxygen, the expired oxygen (FEO2) increases to slightly greater than 90%, satisfying the recommendation for adequate preoxygenation [2]. The FEO2 measurement is clearly the best evidence ensuring proper FRC loading with oxygen. Various techniques for accomplishing preoxygenation have been evaluated, and the most frequently recommended is 3 min of normal tidal volume breathing (TVB) of 100% oxygen using a standard breathing systems. Eight deep breaths in 60 seconds is considered a suitable alternative technique [1,2]. There is a close relationship between the FEO2 that can be reached following preoxygenation and the subsequent duration of apnea without arterial hemoglobin desaturation [4]. However, there are few data regarding the frequency and predictors of failure to achieve adequate preoxygenation (defined as an FEO2 90%). The aim of this study was to determine in a large, unselected population in a clinical setting the incidence of inadequate preoxygenation (defined as an FEO2 < 90% after a 3-min TVB) and its predictive factors. We hypothesized that certain predictive factors for inadequate preoxygenation could also be factors for difficult mask ventilation. 2. Materials and methods The study design was approved by the Institutional Review Board (CEERB, Paris-North Hospital, Paris-7 university, Assistance Publique des Hoˆpitaux de Paris, France). Because there was no randomization and only routine care was performed, waived informed consent was authorized by the CEERB. Adult patients scheduled for orthopedic, vascular, gynecologic, ENT, or abdominal surgery or for gastrointestinal endoscopy under general anaesthesia were prospectively evaluated for inclusion in this study over a 12-month period. Patients undergoing regional anaesthesia were excluded. Since the aim of this study was to investigate the factors associated with an inadequate preoxygenation, regardless of other known factors such as COPD, trauma or shock patients, we excluded patients requiring oxygen upon arrival to the operating room, or presenting with a SpO2 < 98% while breathing room air. The sample size was determined using a method based on a multiple logistic regression model. Assuming from a pilot study an overall probability of inadequate preoxygenation of 50% and a probability of risk factor occurrence between 0.3% and 0.5%, with an alpha of 0.05, a power of 0.8 and a coefficient of determination
between factors between 0.05 and 0.15, a sample size of 1000 subjects was necessary for an odds ratio of 1.5. The following information was collected during the preoperative visit and after the patient had been educated about the preoxygenation technique: age, sex, ASA score, as well as recognized risk factors for difficult mask ventilation (body mass index [BMI] greater than 26, calculated as weight in kilograms divided by the square of the height in meters), presence of a beard, lack of teeth and history of snoring [5,6]. All the patients were supine (excepted for obese patients: headup position 108) and routinely monitored using electrocardiography, non-invasive blood pressure measurement, pulse oximetry, FEO2 and end-tidal carbon dioxide tension (EtCO2) before general anaesthesia induction. After priming the anaesthesia circuit (Primus Dra¨ger, Lu¨beck, Deutschland) with oxygen, the patients were asked to breath normally (tidal volume breathing) for 3 minutes (using a timer). Preoxygenation was performed with a facial mask, by experienced nurses anaesthetits or anesthesiologists, using a machine circuit with a preset gas flow of 12 L/min O2. The size (3, 4, or 5) of the single-use facemask (Intersurgical Limited, Wokingham, United Kingdom). The capnographic waves was used to achieve the best fit for each patient and an appropriate alveolar gases measurement. During preoxygenation, FiO2, FEO2 and EtCO2 were measured breath-by-breath with a calibrated gas analyzer located in the ventilator with a sample line connected to the filter placed between the Y-piece and the mask. The individual study period ended at the outset of preoxygenation, and the anaesthesia procedure was performed as per usual. The data are summarized as the means and standard deviations for continuous data. Categorical variables are presented as absolute and relative frequencies. Marginal associations between an inadequate preoxygenation (FEO2 < 90%) and the patient’s risk factors for inadequate preoxygenation were assessed using a Chi2 test for categorical variables and a Student’s t-test for continuous variables. The data on sex and the presence of a beard were placed into three categories: women, men without beards and men with beards. All the factors with P < 0.20 in the univariate analysis were included in a multiple logistic regression model with a backward stepwise procedure for variable selection. The multivariate analyze were adjusted for FiO2. All the tests were two-sided, at a 0.05 significance level. The analyses were performed using R statistical software version 2.12.0. 3. Results A total of 1050 patients were included in this study (Table 1). Of these, 299 patients were scheduled for orthopedic surgery, 170 patients for gastrointestinal endoscopy, 156 patients for vascular surgery, 155 patients for abdominal surgery, 111 patients for ENT surgery, 107 patients for gynecologic surgery, and 52 patients for other surgeries. The patient characteristics included age 51 20 years, 47% male, a BMI of 26 5 kg/m2 and a median ASA score (median [extremes]) of 2 [1–4]. The incidence of risk factors for
C. Baillard et al. / Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58 Table 1 Comparison of patients (n = 1050) with adequate or inadequate preoxygenation.
FEO2, % (range) Male sex (n) Age, years (range) Age 55 years (n) Weight (kg) Height (cm) BMI (kg/m2) BMI > 26 (n) BMI > 30 (n) History of snoring Presence of beard Lack of teeth ASA score (n) 1 2 3 4 FiO2, % (range)
Adequate preoxygenation FEO2 90% n = 461 (44%)
Inadequate preoxygenation FEO2 < 90% n = 589 (56%)
P value
93 2 (90–98) 173 (38%) 46 21 (18–98) 159 (35%) 71 16 167 9 25 5 (15–45) 176 (39%) 64 (14%) 53 (12%) 5 (1%) 28 (6%)
83 6 (61–89) 322 (55%) 54 19 (18–99) 311 (53%) 73 16 168 9 26 5 (12–58) 229 (41%) 81 (14%) 87 (15%) 41 (7%) 110 (19%)
< 0.0001 < 0.001 < 0.001 < 0.001 0.04 0.02 0.3 0.7 0.7 0.15 < 0.001 < 0.001 < 0.001
212 (46%) 178 (39%) 69 (15%) 2 (0.4%) 98 2 (92–100)
183 (31%) 259 (44%) 129 (22%) 18 (3%) 95 3 (81–100)
e57
Table 2 Identification of risk factors for inadequate preoxygenation using multivariate analysis (n = 1050). Variables
Odds ratio (95% CI)
P value
FiO2 Sex Female Male without beard Male with beard Age > 55 years Lack of teeth ASA score 1 2 or 3 4
13 (6.7–25.4)
< 0.001
1 2.4 9.1 1.8 2.4
< 0.001 < 0.001 0.005 0.006
(1.6–3.4) (2.7–31.4) (1.2–2.7) (1.2–4.5)
1 2.4 (1.6–3.4) 9.1 (2.6–31.2)
0.015 0.024
CI: confidence interval; FiO2: inspiratory alveolar fraction of oxygen. The data are based on the regression coefficients after multiplication by a correction factor.
model. After selection, the final prediction model included sex and beard, age > 55 years, lack of teeth and ASA score (Table 2). < 0.001
The data represent the mean SD or the number (percentage). FEO2: end-tidal expiratory oxygen concentration; BMI: body mass index; FiO2: inspiratory alveolar fraction of oxygen.
difficult mask ventilation was 470 (44%) patients > 55 years, 405 (38%) patients with BMIs > 26 kg/m2, 140 (13%) patients with a history of snoring, 138 (13%) patients lacking teeth and 46 (4%) patients with beards (Table 1). At the end of a 3-minute preoxygenation, the mean (SD) FEO2 value was 87 7% (ranged from 61 to 98%), and the delivered FiO2 ranged from 81% to 100%. Inadequate preoxygenation was observed in 589 patients (56%) (Table 1). The characteristics of patients of the two groups (with adequate or with inadequate preoxygenation) are listed in Table 1. Several risk factors for inadequate preoxygenation were identified in the univariate analysis (Table 1). The following factors were significantly different between the two groups, regardless of inadequate preoxygenation: age > 55 years, male sex, ASA score, presence of a beard, lack of teeth, and the FiO2 delivered. In contrast, BMI and a history of snoring were not significantly different between the two groups. The difference between FiO2 and FEO2 (O2 gradient) was higher in patients with inadequate preoxygenation than in those with adequate preoxygenation: 12 6% vs. 6 2%, respectively; P < 0.001 (Fig. 1). Variables that were significant in the univariate analysis and the history of snoring (P = 0.12) were included in the multivariate
Fig. 1. Difference between FiO2–FEO2 (O2 gradient) in patients with either adequate or inadequate preoxygenation, 6 2% vs. 12 6%, respectively; P < 0.001.
4. Discussion This study constitutes the first evaluation of the incidence of inadequate preoxygenation (defined as an FEO2 < 90% after a 3-min TVB) in a large, unselected population with baseline SpO2 98%, in usual clinical settings. Overall, inadequate preoxygenation was observed in up to 56% of the patients and we further observed that some of the known predictive factors for difficult mask ventilation also proved predictive of inadequate preoxygenation. Five patient characteristics (sex, age > 55 years, lack of teeth, the presence of a beard and an ASA score > 1) were found to be independent risk factors for inadequate preoxygenation. Preoxygenation allows to prolong, in the interest of safety, the period of possible apnea without hypoxia by increasing the pulmonary oxygen reserves [4]. Pulse oximetry is a weak indicator of the effectiveness of preoxygenation and the best indicator of the completeness of preoxygenation is the FEO2. Obtaining a value 90% ensures that the FRC, the main O2 reserve in the body, is adequately loaded with oxygen. Patients known to be at risk of potential rapid oxyhemoglobin desaturation are those with a reduced FRC (such as obese or pregnant patients) and those with an increased oxygen consumption (e.g., as occurs in sepsis). In our study, a BMI > 26 or > 30 was not found to be a predictive factor for inadequate preoxygenation, in accordance with a previous study by Jense et al. [7]. The reduced FRC exhibited by obese patients is consistent with the shorter time required for effective denitrogenation in these patients due to their increased weight. However, the safety margin for maintaining the SpO2 following apnea is also diminished by the reduced FRC [6]. Since preoxygenation should be performed with 100% oxygen, with a fresh gas flow sufficient to preclude rebreathing, we used a preset gas flow of 12 L/min oxygen, as previously recommended [2]. However, the resulting FiO2 was not 100% in all the patients, and the effective FiO2 delivered was generally lower in the patients with inadequate preoxygenation than in those with adequate preoxygenation. Our findings emphasize that particular attention should be paid to establishing the appropriate delivery of FiO2 prior to beginning preoxygenation. A preset gas flow above 12 L/min would be used in patients at risk for inadequate preoxygenation. The second main technical consideration performed was to ensure a tight fit between the mask and the patient’s face, thus avoiding air leakage, which might result in a suboptimal delivered alveolar oxygen concentration [8]. In our study, the capnographic waves was used to achieve the best fit for each patient. However, an adequate seal can be difficult to achieve in certain patients, such
e58
C. Baillard et al. / Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58
as edentulous and/or bearded patients, both of which are known risk factors for difficult mask ventilation [5]. In edentulous patients, the air leak results from reduced contact between the cheeks and the mask, and we used the recently proposed alternate mode of facemask application for edentulous patients [9]. As expected, the present study showed that the presence of a beard and a lack of teeth were independent predictive factors for inadequate preoxygenation. In the elderly, the effects of preoxygenation using the 3-min TVB breathing technique have been investigated previously. In a study by Bhatia et al., [10], the FEO2 reached 90% after 3 minutes of preoxygenation in only 19 out of 25 elderly patients, compared with 23 out of 25 young adult (< 60 years) patients. Kang et al., [11] have shown that in the elderly, the rate of increase of FEO2 was slower than that observed in a younger ( 65 years) group, suggesting that the elderly require more than 3 minutes to achieve an FEO2 of 90% during preoxygenation. Presumably, the underlying mechanism involves the functional changes in the lung that occur with aging such as the increase of closing volume, the decrease in minute ventilation which lead to a greater risk of a ventilationperfusion mismatch [12,13]. Another explanation could be the higher prevalence of edentulism among the elderly. Accordingly, our study found that age > 55 years and a lack of teeth were independent predictive factors for inadequate preoxygenation. In the present study, although severe comorbidities were excluded since only patients with a baseline SpO2 98% while breathing room air were investigated, we found that the ASA score was directly related to the risk of inadequate preoxygenation. This relationship is significant because arterial hemoglobin desaturation during apnea would likely be poorly tolerated in this fragile population. Finally, the role of male sex as a predisposing factor for inadequate preoxygenation was previously uninvestigated. That male sex itself is a risk factor for inadequate preoxygenation, with or without the presence of a beard, is noteworthy. Interestingly, male sex has also been established as a predisposing factor for impossible mask ventilation [14]. Although they are derived from a relatively significant number (1050) of patients reflecting usual uncomplicated surgeries, our data have several limitations. Due to the study design, the relationship between inadequate preoxygenation and possible subsequent arterial hemoglobin desaturation during apnea was not evaluated. However, previous investigations have demonstrated that a low FEO2 during the routine induction of general anaesthesia is associated with a limited window of time before an unacceptable desaturation occurs [4]. Additional possible predictive factors remain unstudied, such as those involved in difficult or impossible ventilation, the Mallampati score, jaw protrusion, snoring, changes induced by neck radiation and sleep apnea [5,6,14]. 5. Conclusion We have demonstrated that inadequate preoxygenation is a common event in the clinical setting. The main predictive factors (OR > 9) are the FiO2 delivered, ASA score 4 and man with beard. Particular attention should be paid to set the appropriate FiO2 delivery and higher than 12 L/min gas flow are probably useful in
some patients. The relationship with ASA score 4 is important to consider and attempts should be made to improve the effectiveness of preoxygenation in this high risk population. Several solutions have been previously described to solve the problem of hairy face such as the use of defibrillator pads or the ‘‘lower lip’’ face mask placement which is also useful in the edentulous [9]. Some authors have even suggested to shave the patient before the surgery. We have confirmed our hypothesis that the predictive factors for inadequate preoxygenation share an overlap with those previously identified as predictive of difficult ventilation. We propose that the recognition of this association could have a significant clinical impact because it could lead to the simple and safe enhancement of the overall security of anaesthesia induction. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Financial support was provided solely from institutional and/or departmental sources. Authors’ contributions: Christophe Baillard, Vincent Levy and Sadek Beloucif participated in the design of the study, carried out the study and drafted the manuscript. Franc¸ois Depret participated in the design of the study, inclusion of patients and data analysis. Marouane Boubaya and Vincent Levy participated in the study analysis. All authors read and approved the final manuscript. References [1] Benumof JL. Preoxygenation: best method for both efficacy and efficiency. Anesthesiology 1999;91:603–5. [2] Socie´te´ franc¸aise d’anesthe´sie et de re´animation. Prise en charge des voies ae´riennes en anesthe´sie adulte a` l’exception de l’intubation difficile - Confe´rence de consensus - Recommandations du Jury. Texte long. Ann Fr Anesth Reanim 2003;22:3–17. [3] Tanoubi I, Drolet P, Donati F. Optimizing preoxygenation in adults. Can J Anaesth 2009;56:449–66. [4] Edmark L, Kostova-Aherdan K, Enlund M, Hedenstierna G. Optimal oxygen concentration during induction of general anaesthesia. Anesthesiology 2003; 98:28–33. [5] Langeron O, Masso E, Huraux C, Guggiari M, Bianchi A, Coriat P, et al. Prediction of difficult mask ventilation. Anesthesiology 2000;92:1229–36. [6] Kheterpal S, Han R, Tremper KK, Shanks A, Tait AR, O’Reilly M, et al. Incidence and predictors of difficult and impossible mask ventilation. Anesthesiology 2006;105:885–91. [7] Jense HG, Dubin SA, Silverstein PI, O’Leary-Escolas U. Effect of obesity on safe duration of apnea in anesthetized humans. Anesth Analg 1991;72:89–93. [8] McGowan P, Skinner A. Preoxygenation: the importance of a good face mask seal. Br J Anaesth 1995;75:777–8. [9] Racine SX, Solis A, Hamou NA, Letoumelin P, Hepner DL, Beloucif S, et al. Face mask ventilation in edentulous patients: a comparison of mandibular groove and lower lip placement. Anesthesiology 2010;112:1190–3. [10] Bhatia PK, Bhandari SC, Tulsiani KL, Kumar Y. End-tidal oxygraphy and safe duration of apnoea in young adults and elderly patients. Anaesthesia 1997;52: 175–8. [11] Kang H, Park HJ, Baek SK, Choi J, Park SJ. Effects of preoxygenation with the three minutes tidal volume breathing technique in the elderly. Korean J Anesthesiol 2010;58:369–73. [12] Holland J, Milic-Emili J, Macklem PT, Bates DV. Regional distribution of pulmonary ventilation and perfusion in elderly subjects. J Clin Invest 1968; 47:81–92. [13] Leblanc P, Ruff F, Milic-Emili J. Effects of age and body position on ‘‘airway closure’’ in man. J Appl Physiol 1970;28:448–51. [14] Kheterpal S, Martin L, Shanks AM, Tremper KK. Prediction and outcomes of impossible mask ventilation: a review of 50,000 anesthetics. Anesthesiology 2009;110:891–7.
Original article
Incidence and prediction of inadequate preoxygenation before induction of anaesthesia Incidence et facteurs pre´dictifs d’e´chec de pre´oxyge´nation avant induction anesthe´sique C. Baillard a,*, F. Depret a, V. Levy b, M. Boubaya b, S. Beloucif a a
EA 3409, department of anaesthesiology and critical care medicine, Avicenne university hospital, Paris-13 university, AP–HP, 125, route de Stalingrad, 93009 Bobigny, France b Clinical research unit, Avicenne university hospital, Paris-13 university, AP–HP, 125, route de Stalingrad, 93009 Bobigny, France
A R T I C L E I N F O
A B S T R A C T
Article history: Received 27 October 2013 Accepted 30 December 2013 Available online 24 February 2014
Background. – Preoxygenation aims to obtain an expired oxygen fraction (FEO2) 90%. Little is known about the incidence and predictors of inadequate preoxygenation in the clinical setting. Patients and methods. – Over a 12-month period, 1050 consecutive preoperative patients were prospectively included. Preoxygenation was performed for 3 minutes with a facial mask using a machine circuit and 12-L/min oxygen flow. Inadequate preoxygenation was defined as an FEO2 < 90%. A logistic regression was performed to identify incidence and independent predictors. Results. – The patient characteristics were: age 51 20 years, 47% male, BMI of 26 5 kg/m2, and ASA score (median [extremes]) of 2 [1–4]. Inadequate preoxygenation was observed in 589 patients (56%). The effective FiO2 delivered was lower in the patients with inadequate preoxygenation than in those with adequate preoxygenation, 95 3% vs. 98 2%, P < 0.001. The difference between the FiO2 and the FEO2 was higher (12 6% vs. 6 3%, P < 0.0001) in patients with inadequate preoxygenation compared with those with adequate preoxygenation. The independent risk factors for inadequate preoxygenation were: firstly, bearded male (odds ratio [OR] of 9.1 [2.7–31.4] P < 0.001); secondly, beardless male (OR 2.4 [1.6–3.4] P < 0.001), thirdly, ASA score of 4 (OR 9.1 [2.6–31.2] P < 0.015); fourthly, ASA score of 2–3 (OR 2.4 [1.6–3.4] P < 0.015); fifthly, lack of teeth (OR 2.4 [1.2–4.5] P < 0.006), and lastly age > 55 years (OR 1.8 [1.2–2.7] P < 0.005). Conclusion. – Inadequate preoxygenation, defined as an FEO2 < 90% despite 3-min tidal volume breathing, was a common occurrence. The predictive factors share an overlap with those previously identified for difficult mask ventilation. ß 2014 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved.
Keywords: Preoxygenation General anesthesia Airway
R E´ S U M E´
Mots cle´s : Pre´oxyge´nation Anesthe´sie ge´ne´rale Voies ae´riennes
Objectifs. – Peu de donne´es sont actuellement disponibles concernant l’incidence et les facteurs pre´dictifs d’un e´chec de pre´oxyge´nation en pratique clinique. Patients et me´thodes. – Au cours d’une pe´riode de 12 mois, 1050 patients conse´cutifs ont e´te´ prospectivement inclus. La pre´oxyge´nation e´tait re´alise´e en volume courant pendant 3 minutes en utilisant le circuit machine avec un de´bit d’oxyge`ne de 12 litres par minutes. Une FEO2 < 90 % a` l’issue de la pre´oxyge´nation de´finissait un e´chec. Une analyse univarie´e puis multivarie´e ont e´te´ re´alise´es pour identifier les facteurs inde´pendants d’e´chec de pre´oxyge´nation. Re´sultats. – Les caracte´ristiques des patients e´taient : aˆge 51 20 ans, 47 % hommes, IMC 26 5 kg/m2, et score ASA : 2 [1–4]. L’incidence globale d’e´chec de pre´oxyge´nation e´tait retrouve´e chez 589 patients (56 %). La FiO2 de´livre´e e´tait plus faible chez les patients avec e´chec de pre´oxyge´nation compare´e aux patients correctement pre´oxyge´ne´s, 95 3 % vs 98 2 %, p < 0,001. Les facteurs de risques inde´pendants d’e´chec de
* Corresponding author. E-mail address: [email protected] (C. Baillard). 0750-7658/$ – see front matter ß 2014 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.annfar.2013.12.018
e56
C. Baillard et al. / Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58
pre´oxyge´nation e´taient : premie`rement, les hommes barbus (odds ratio [OR] of 9,1 [2,7–31,4] p < 0,001) ; deuxie`mement, le sexe masculin (OR 2,4 [1,6–3,4] p < 0,001) ; troisie`mement, un score ASA 4 (OR 9,1 [2,6–31,2] p < 0,015) ; quatrie`mement, un score ASA 2–3 (OR 2,4 [1,6–3,4] p < 0,015) ; cinquie`mement, l’e´dentation (OR 2,4 [1,2–4,5] p < 0,006) et finalement, l’aˆge > 55 ans (OR 1,8 [1,2–2,7] p < 0,005). Conclusion. – L’e´chec de pre´oxyge´nation de´fini par une FEO2 < 90 % a` l’issue de 3 minutes de ventilation spontane´e en volume courant est une situation courante. Les facteurs de risques observe´s sont communs a` ceux pre´alablement identifie´s pour la pre´diction de la ventilation au masque difficile. ß 2014 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Publie´ par Elsevier Masson SAS. Tous droits re´serve´s.
1. Introduction The standard practice of preoxygenating with 100% oxygen before the induction of general anaesthesia is strongly recommended because it delays the onset of arterial hemoglobin desaturation during a potential prolonged apnea [1,2]. The oxygen loading process during preoxygenation involves factors related to alveolar, arterial, tissue, and venous compartments and those that interact with the delivered fraction of inspired oxygen (FiO2) [3]. Because most of the oxygen is stored in the lungs as a function of residual functional capacity (FRC), the primary goal is to maximize the alveolar fraction of oxygen, which depends on the effective FiO2 delivered. When most of the nitrogen in the lung is replaced by oxygen, the expired oxygen (FEO2) increases to slightly greater than 90%, satisfying the recommendation for adequate preoxygenation [2]. The FEO2 measurement is clearly the best evidence ensuring proper FRC loading with oxygen. Various techniques for accomplishing preoxygenation have been evaluated, and the most frequently recommended is 3 min of normal tidal volume breathing (TVB) of 100% oxygen using a standard breathing systems. Eight deep breaths in 60 seconds is considered a suitable alternative technique [1,2]. There is a close relationship between the FEO2 that can be reached following preoxygenation and the subsequent duration of apnea without arterial hemoglobin desaturation [4]. However, there are few data regarding the frequency and predictors of failure to achieve adequate preoxygenation (defined as an FEO2 90%). The aim of this study was to determine in a large, unselected population in a clinical setting the incidence of inadequate preoxygenation (defined as an FEO2 < 90% after a 3-min TVB) and its predictive factors. We hypothesized that certain predictive factors for inadequate preoxygenation could also be factors for difficult mask ventilation. 2. Materials and methods The study design was approved by the Institutional Review Board (CEERB, Paris-North Hospital, Paris-7 university, Assistance Publique des Hoˆpitaux de Paris, France). Because there was no randomization and only routine care was performed, waived informed consent was authorized by the CEERB. Adult patients scheduled for orthopedic, vascular, gynecologic, ENT, or abdominal surgery or for gastrointestinal endoscopy under general anaesthesia were prospectively evaluated for inclusion in this study over a 12-month period. Patients undergoing regional anaesthesia were excluded. Since the aim of this study was to investigate the factors associated with an inadequate preoxygenation, regardless of other known factors such as COPD, trauma or shock patients, we excluded patients requiring oxygen upon arrival to the operating room, or presenting with a SpO2 < 98% while breathing room air. The sample size was determined using a method based on a multiple logistic regression model. Assuming from a pilot study an overall probability of inadequate preoxygenation of 50% and a probability of risk factor occurrence between 0.3% and 0.5%, with an alpha of 0.05, a power of 0.8 and a coefficient of determination
between factors between 0.05 and 0.15, a sample size of 1000 subjects was necessary for an odds ratio of 1.5. The following information was collected during the preoperative visit and after the patient had been educated about the preoxygenation technique: age, sex, ASA score, as well as recognized risk factors for difficult mask ventilation (body mass index [BMI] greater than 26, calculated as weight in kilograms divided by the square of the height in meters), presence of a beard, lack of teeth and history of snoring [5,6]. All the patients were supine (excepted for obese patients: headup position 108) and routinely monitored using electrocardiography, non-invasive blood pressure measurement, pulse oximetry, FEO2 and end-tidal carbon dioxide tension (EtCO2) before general anaesthesia induction. After priming the anaesthesia circuit (Primus Dra¨ger, Lu¨beck, Deutschland) with oxygen, the patients were asked to breath normally (tidal volume breathing) for 3 minutes (using a timer). Preoxygenation was performed with a facial mask, by experienced nurses anaesthetits or anesthesiologists, using a machine circuit with a preset gas flow of 12 L/min O2. The size (3, 4, or 5) of the single-use facemask (Intersurgical Limited, Wokingham, United Kingdom). The capnographic waves was used to achieve the best fit for each patient and an appropriate alveolar gases measurement. During preoxygenation, FiO2, FEO2 and EtCO2 were measured breath-by-breath with a calibrated gas analyzer located in the ventilator with a sample line connected to the filter placed between the Y-piece and the mask. The individual study period ended at the outset of preoxygenation, and the anaesthesia procedure was performed as per usual. The data are summarized as the means and standard deviations for continuous data. Categorical variables are presented as absolute and relative frequencies. Marginal associations between an inadequate preoxygenation (FEO2 < 90%) and the patient’s risk factors for inadequate preoxygenation were assessed using a Chi2 test for categorical variables and a Student’s t-test for continuous variables. The data on sex and the presence of a beard were placed into three categories: women, men without beards and men with beards. All the factors with P < 0.20 in the univariate analysis were included in a multiple logistic regression model with a backward stepwise procedure for variable selection. The multivariate analyze were adjusted for FiO2. All the tests were two-sided, at a 0.05 significance level. The analyses were performed using R statistical software version 2.12.0. 3. Results A total of 1050 patients were included in this study (Table 1). Of these, 299 patients were scheduled for orthopedic surgery, 170 patients for gastrointestinal endoscopy, 156 patients for vascular surgery, 155 patients for abdominal surgery, 111 patients for ENT surgery, 107 patients for gynecologic surgery, and 52 patients for other surgeries. The patient characteristics included age 51 20 years, 47% male, a BMI of 26 5 kg/m2 and a median ASA score (median [extremes]) of 2 [1–4]. The incidence of risk factors for
C. Baillard et al. / Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58 Table 1 Comparison of patients (n = 1050) with adequate or inadequate preoxygenation.
FEO2, % (range) Male sex (n) Age, years (range) Age 55 years (n) Weight (kg) Height (cm) BMI (kg/m2) BMI > 26 (n) BMI > 30 (n) History of snoring Presence of beard Lack of teeth ASA score (n) 1 2 3 4 FiO2, % (range)
Adequate preoxygenation FEO2 90% n = 461 (44%)
Inadequate preoxygenation FEO2 < 90% n = 589 (56%)
P value
93 2 (90–98) 173 (38%) 46 21 (18–98) 159 (35%) 71 16 167 9 25 5 (15–45) 176 (39%) 64 (14%) 53 (12%) 5 (1%) 28 (6%)
83 6 (61–89) 322 (55%) 54 19 (18–99) 311 (53%) 73 16 168 9 26 5 (12–58) 229 (41%) 81 (14%) 87 (15%) 41 (7%) 110 (19%)
< 0.0001 < 0.001 < 0.001 < 0.001 0.04 0.02 0.3 0.7 0.7 0.15 < 0.001 < 0.001 < 0.001
212 (46%) 178 (39%) 69 (15%) 2 (0.4%) 98 2 (92–100)
183 (31%) 259 (44%) 129 (22%) 18 (3%) 95 3 (81–100)
e57
Table 2 Identification of risk factors for inadequate preoxygenation using multivariate analysis (n = 1050). Variables
Odds ratio (95% CI)
P value
FiO2 Sex Female Male without beard Male with beard Age > 55 years Lack of teeth ASA score 1 2 or 3 4
13 (6.7–25.4)
< 0.001
1 2.4 9.1 1.8 2.4
< 0.001 < 0.001 0.005 0.006
(1.6–3.4) (2.7–31.4) (1.2–2.7) (1.2–4.5)
1 2.4 (1.6–3.4) 9.1 (2.6–31.2)
0.015 0.024
CI: confidence interval; FiO2: inspiratory alveolar fraction of oxygen. The data are based on the regression coefficients after multiplication by a correction factor.
model. After selection, the final prediction model included sex and beard, age > 55 years, lack of teeth and ASA score (Table 2). < 0.001
The data represent the mean SD or the number (percentage). FEO2: end-tidal expiratory oxygen concentration; BMI: body mass index; FiO2: inspiratory alveolar fraction of oxygen.
difficult mask ventilation was 470 (44%) patients > 55 years, 405 (38%) patients with BMIs > 26 kg/m2, 140 (13%) patients with a history of snoring, 138 (13%) patients lacking teeth and 46 (4%) patients with beards (Table 1). At the end of a 3-minute preoxygenation, the mean (SD) FEO2 value was 87 7% (ranged from 61 to 98%), and the delivered FiO2 ranged from 81% to 100%. Inadequate preoxygenation was observed in 589 patients (56%) (Table 1). The characteristics of patients of the two groups (with adequate or with inadequate preoxygenation) are listed in Table 1. Several risk factors for inadequate preoxygenation were identified in the univariate analysis (Table 1). The following factors were significantly different between the two groups, regardless of inadequate preoxygenation: age > 55 years, male sex, ASA score, presence of a beard, lack of teeth, and the FiO2 delivered. In contrast, BMI and a history of snoring were not significantly different between the two groups. The difference between FiO2 and FEO2 (O2 gradient) was higher in patients with inadequate preoxygenation than in those with adequate preoxygenation: 12 6% vs. 6 2%, respectively; P < 0.001 (Fig. 1). Variables that were significant in the univariate analysis and the history of snoring (P = 0.12) were included in the multivariate
Fig. 1. Difference between FiO2–FEO2 (O2 gradient) in patients with either adequate or inadequate preoxygenation, 6 2% vs. 12 6%, respectively; P < 0.001.
4. Discussion This study constitutes the first evaluation of the incidence of inadequate preoxygenation (defined as an FEO2 < 90% after a 3-min TVB) in a large, unselected population with baseline SpO2 98%, in usual clinical settings. Overall, inadequate preoxygenation was observed in up to 56% of the patients and we further observed that some of the known predictive factors for difficult mask ventilation also proved predictive of inadequate preoxygenation. Five patient characteristics (sex, age > 55 years, lack of teeth, the presence of a beard and an ASA score > 1) were found to be independent risk factors for inadequate preoxygenation. Preoxygenation allows to prolong, in the interest of safety, the period of possible apnea without hypoxia by increasing the pulmonary oxygen reserves [4]. Pulse oximetry is a weak indicator of the effectiveness of preoxygenation and the best indicator of the completeness of preoxygenation is the FEO2. Obtaining a value 90% ensures that the FRC, the main O2 reserve in the body, is adequately loaded with oxygen. Patients known to be at risk of potential rapid oxyhemoglobin desaturation are those with a reduced FRC (such as obese or pregnant patients) and those with an increased oxygen consumption (e.g., as occurs in sepsis). In our study, a BMI > 26 or > 30 was not found to be a predictive factor for inadequate preoxygenation, in accordance with a previous study by Jense et al. [7]. The reduced FRC exhibited by obese patients is consistent with the shorter time required for effective denitrogenation in these patients due to their increased weight. However, the safety margin for maintaining the SpO2 following apnea is also diminished by the reduced FRC [6]. Since preoxygenation should be performed with 100% oxygen, with a fresh gas flow sufficient to preclude rebreathing, we used a preset gas flow of 12 L/min oxygen, as previously recommended [2]. However, the resulting FiO2 was not 100% in all the patients, and the effective FiO2 delivered was generally lower in the patients with inadequate preoxygenation than in those with adequate preoxygenation. Our findings emphasize that particular attention should be paid to establishing the appropriate delivery of FiO2 prior to beginning preoxygenation. A preset gas flow above 12 L/min would be used in patients at risk for inadequate preoxygenation. The second main technical consideration performed was to ensure a tight fit between the mask and the patient’s face, thus avoiding air leakage, which might result in a suboptimal delivered alveolar oxygen concentration [8]. In our study, the capnographic waves was used to achieve the best fit for each patient. However, an adequate seal can be difficult to achieve in certain patients, such
e58
C. Baillard et al. / Annales Franc¸aises d’Anesthe´sie et de Re´animation 33 (2014) e55–e58
as edentulous and/or bearded patients, both of which are known risk factors for difficult mask ventilation [5]. In edentulous patients, the air leak results from reduced contact between the cheeks and the mask, and we used the recently proposed alternate mode of facemask application for edentulous patients [9]. As expected, the present study showed that the presence of a beard and a lack of teeth were independent predictive factors for inadequate preoxygenation. In the elderly, the effects of preoxygenation using the 3-min TVB breathing technique have been investigated previously. In a study by Bhatia et al., [10], the FEO2 reached 90% after 3 minutes of preoxygenation in only 19 out of 25 elderly patients, compared with 23 out of 25 young adult (< 60 years) patients. Kang et al., [11] have shown that in the elderly, the rate of increase of FEO2 was slower than that observed in a younger ( 65 years) group, suggesting that the elderly require more than 3 minutes to achieve an FEO2 of 90% during preoxygenation. Presumably, the underlying mechanism involves the functional changes in the lung that occur with aging such as the increase of closing volume, the decrease in minute ventilation which lead to a greater risk of a ventilationperfusion mismatch [12,13]. Another explanation could be the higher prevalence of edentulism among the elderly. Accordingly, our study found that age > 55 years and a lack of teeth were independent predictive factors for inadequate preoxygenation. In the present study, although severe comorbidities were excluded since only patients with a baseline SpO2 98% while breathing room air were investigated, we found that the ASA score was directly related to the risk of inadequate preoxygenation. This relationship is significant because arterial hemoglobin desaturation during apnea would likely be poorly tolerated in this fragile population. Finally, the role of male sex as a predisposing factor for inadequate preoxygenation was previously uninvestigated. That male sex itself is a risk factor for inadequate preoxygenation, with or without the presence of a beard, is noteworthy. Interestingly, male sex has also been established as a predisposing factor for impossible mask ventilation [14]. Although they are derived from a relatively significant number (1050) of patients reflecting usual uncomplicated surgeries, our data have several limitations. Due to the study design, the relationship between inadequate preoxygenation and possible subsequent arterial hemoglobin desaturation during apnea was not evaluated. However, previous investigations have demonstrated that a low FEO2 during the routine induction of general anaesthesia is associated with a limited window of time before an unacceptable desaturation occurs [4]. Additional possible predictive factors remain unstudied, such as those involved in difficult or impossible ventilation, the Mallampati score, jaw protrusion, snoring, changes induced by neck radiation and sleep apnea [5,6,14]. 5. Conclusion We have demonstrated that inadequate preoxygenation is a common event in the clinical setting. The main predictive factors (OR > 9) are the FiO2 delivered, ASA score 4 and man with beard. Particular attention should be paid to set the appropriate FiO2 delivery and higher than 12 L/min gas flow are probably useful in
some patients. The relationship with ASA score 4 is important to consider and attempts should be made to improve the effectiveness of preoxygenation in this high risk population. Several solutions have been previously described to solve the problem of hairy face such as the use of defibrillator pads or the ‘‘lower lip’’ face mask placement which is also useful in the edentulous [9]. Some authors have even suggested to shave the patient before the surgery. We have confirmed our hypothesis that the predictive factors for inadequate preoxygenation share an overlap with those previously identified as predictive of difficult ventilation. We propose that the recognition of this association could have a significant clinical impact because it could lead to the simple and safe enhancement of the overall security of anaesthesia induction. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Financial support was provided solely from institutional and/or departmental sources. Authors’ contributions: Christophe Baillard, Vincent Levy and Sadek Beloucif participated in the design of the study, carried out the study and drafted the manuscript. Franc¸ois Depret participated in the design of the study, inclusion of patients and data analysis. Marouane Boubaya and Vincent Levy participated in the study analysis. All authors read and approved the final manuscript. References [1] Benumof JL. Preoxygenation: best method for both efficacy and efficiency. Anesthesiology 1999;91:603–5. [2] Socie´te´ franc¸aise d’anesthe´sie et de re´animation. Prise en charge des voies ae´riennes en anesthe´sie adulte a` l’exception de l’intubation difficile - Confe´rence de consensus - Recommandations du Jury. Texte long. Ann Fr Anesth Reanim 2003;22:3–17. [3] Tanoubi I, Drolet P, Donati F. Optimizing preoxygenation in adults. Can J Anaesth 2009;56:449–66. [4] Edmark L, Kostova-Aherdan K, Enlund M, Hedenstierna G. Optimal oxygen concentration during induction of general anaesthesia. Anesthesiology 2003; 98:28–33. [5] Langeron O, Masso E, Huraux C, Guggiari M, Bianchi A, Coriat P, et al. Prediction of difficult mask ventilation. Anesthesiology 2000;92:1229–36. [6] Kheterpal S, Han R, Tremper KK, Shanks A, Tait AR, O’Reilly M, et al. Incidence and predictors of difficult and impossible mask ventilation. Anesthesiology 2006;105:885–91. [7] Jense HG, Dubin SA, Silverstein PI, O’Leary-Escolas U. Effect of obesity on safe duration of apnea in anesthetized humans. Anesth Analg 1991;72:89–93. [8] McGowan P, Skinner A. Preoxygenation: the importance of a good face mask seal. Br J Anaesth 1995;75:777–8. [9] Racine SX, Solis A, Hamou NA, Letoumelin P, Hepner DL, Beloucif S, et al. Face mask ventilation in edentulous patients: a comparison of mandibular groove and lower lip placement. Anesthesiology 2010;112:1190–3. [10] Bhatia PK, Bhandari SC, Tulsiani KL, Kumar Y. End-tidal oxygraphy and safe duration of apnoea in young adults and elderly patients. Anaesthesia 1997;52: 175–8. [11] Kang H, Park HJ, Baek SK, Choi J, Park SJ. Effects of preoxygenation with the three minutes tidal volume breathing technique in the elderly. Korean J Anesthesiol 2010;58:369–73. [12] Holland J, Milic-Emili J, Macklem PT, Bates DV. Regional distribution of pulmonary ventilation and perfusion in elderly subjects. J Clin Invest 1968; 47:81–92. [13] Leblanc P, Ruff F, Milic-Emili J. Effects of age and body position on ‘‘airway closure’’ in man. J Appl Physiol 1970;28:448–51. [14] Kheterpal S, Martin L, Shanks AM, Tremper KK. Prediction and outcomes of impossible mask ventilation: a review of 50,000 anesthetics. Anesthesiology 2009;110:891–7.