During the induction of anesthesia, manual mask ventilation in children may result in gastric insufflation (GI) and lead to adverse outcomes, i.e., alveolar hypoventilation, increased risk of regurgitation and pulmonary aspiration, hemodynamic complications, or even gastric rupture from high abdominal pressure.1,2 In adult patients, pressure-controlled mask ventilation may reduce the occurrence of these complications.3 For the same tidal volume, inspiratory pressures are significantly reduced with pressure-controlled ventilation compared with manual ventilation. In adults receiving pressure-controlled ventilation with a mask, the occurrence of GI is 5% and 26% when the inspiratory pressure is set at 20 and 30 cm H2O, respectively.4 It is widely accepted that the pop-off valve of the anesthesia machine should be set at 20 cm H2O to prevent the risk of GI in adults. However, no evaluation or recommendation has been given concerning children. Because GI is frequent during mask ventilation in very young children, we evaluated the upper limit of inspiratory pressures in children that allows proper and safe ventilation and does not result in GI.
METHODS
After institutional research committee approval and informed consent from the parents, 100 consecutive children, aged 1 day to 16 years, ASA physical status I to II, scheduled for general anesthesia were included in this prospective observational study. Exclusion criteria included parental refusal, a predicted difficult intubation, a known or predicted respiratory disease (including premature infants), and any contraindication to facemask ventilation, i.e., full stomach or emergency surgery.
Children were premedicated with 0.4 mg · kg−1 midazolam oral or rectal according to age. After 2 minutes of oxygen administration via the ventilator circuit (Felix; Taema, Antony, France), anesthesia was induced with 6% sevoflurane in an equal mixture of oxygen and nitrous oxide for 2 minutes. After loss of eyelid reflex, an oropharyngeal cannula was inserted, a venous catheter was placed in a peripheral vein, and sufentanil 0.5 μg · kg−1 was injected. After induction, the sevoflurane end-tidal concentration was set at 3.5%, and the children were mask ventilated with pressure-controlled ventilation as soon as respiratory support was required. An experienced (>5000 facemask ventilations) anesthesiologist (FS, MB, YM) fixed the facemask with 2 hands during the ventilation protocol. The initial insufflation pressure was set at 10 cm H2O for 5 breath cycles. In the absence of GI as assessed by epigastric auscultation, the insufflation pressures were increased in steps of 5 cm H2O with a maximum of 25 cm H2O. GI was assessed by a second anesthesiologist (SL) who performed continuous epigastric auscultation during the 5 breath cycles. GI was assessed at each pressure increment using a stethoscope placed in a fixed location (5 cm inferolateral to the xiphisternum).5 The 2 anesthesiologists were blinded to the pulmonary monitor but not from the anesthesia machine because applied inspiratory pressure was not randomized. If GI was detected, the inspiratory pressure was decreased to the lower level to verify the disappearance of GI. The protocol was discontinued if either GI occurred at an inspiratory pressure ≤25 cm H2O or if no GI occurred at P25. At the end of the protocol, the inspiratory pressure was set to obtain a tidal volume of 10 mL · kg−1. If GI occurred, then the tidal volume was reduced. If no GI occurred with a 10 mL · kg−1 tidal volume, the inspiratory pressure was set to the higher pressure providing no GI. The respiratory rate was adapted to the child's age. The inspiratory/expiratory ratio (I/E) was set at 1:2 with no positive end-expiratory pressure. The recorded data were age, weight of the child, and for every insufflation pressure step, the respiratory rate, the expired tidal volume, the minute ventilation, and whether GI occurred. Side effects (laryngospasm, bronchospasm, cough, regurgitation, and movement before intubation) were also recorded.
Statistical Analysis
In a preliminary study of 35 patients, GI was detected in 71% of patients at P20. One hundred patients had to be included in this study to allow a more precise estimation of GI rates with a width of the 95% confidence interval (CI) of 10% around the observed rate.6 Results were expressed as median, interquartile range (IQR) and range (minimum-maximum) or mean (SD) as appropriate. Children were classified into 3 groups according to age: children aged 1 year or younger (age group 1), 1 to 5 years (age group 2), and older than 5 years (age group 3). Results were compared among age groups with Kruskal-Wallis analysis and the Mann-Whitney post hoc test or parametric values with 1-way analysis of variance and the Student-Newman-Keuls post hoc test. A P value <0.05 was considered statistically significant. Occurrence of GI in children among inspiratory pressures (P10, P15, P20, and P25) was assessed by using the χ2 test, and P values <0.05 were considered significant. A multivariate analysis was performed with 2 levels of inspiratory pressure as dependent variables (inspiratory pressure ≤15 cm H2O and ≤25 cm H2O) and age, weight, and tidal volume as independent variables. Odds ratios (ORs) were expressed with their 95% CIs.
RESULTS
One hundred children were included. One child was excluded from the study because of difficult mask ventilation due to a large air leakage occurring at 10 cm H2O. He was then intubated without any problem. Demographic data of children are described in Table 1. There was a high correlation between age and weight (r2 = 0.85; P < 0.001; Fig. 1). No child was obese. GI occurred in 78 children: 5 at P10, 16 at P15, 36 at P20, and 21 at P25. The incidence of GI was 95% in age group 1 and 93% in age group 2. It was significantly lower in age group 3 (56%, P = 0.001).
;)
Table 1: Demographic Data
Figure 1: Correlation between age and weight of children.
The incidence of GI increased with inspiratory pressure, i.e., cumulated observed rates of GI at P15 and P20 were 21% (95% CI, 13–29) and 58% (95% CI, 48–68), respectively, for this study. The pressure threshold where GI occurred (PGI) increased with age (Fig. 2): the younger the child, the lower the PGI (P < 0.05, Student-Newman-Keuls). Only age group 1 children experienced GI at 10 cm H2O; moreover, in this group, PGI was ≤15 cm H2O in 50% of cases. PGI was ≥20 cm H2O in 90% of cases in age group 3 and was equal to 15 cm H2O in only 9% of them. In age group 2, PGI was ≥20 cm H2O in 72% of cases. Univariate regression analysis considering age and weight as independent variables showed that at P15, age was an independent risk factor (OR = 0.6 for increase in 1 year [95% CI, 0.4–0.8; P < 0.001]) as well as weight (OR = 0.8 for increase in 1 kg [95% CI, 0.7–0.9; P < 0.001]).
Figure 2: Age of children with or without gastric insufflation (GI) associated with different inspiratory pressure levels. Data are expressed as median (IQR [interquartile range]). White box represents group without GI (no GI). Dark box represents group in which GI occurred at 10 cm H2O inspiratory pressure (P10). Gray boxes represent group with GI occurring at 15, 20, or 25 cm H2O inspiratory pressure, respectively, designated as P15, P20, and P25.*P < 0.05 versus no GI group was considered significant.
Tidal volume increased with inspiratory pressure between 10 and 15 cm H2O. The difference was statistically significant between 10 cm H2O and the 4 levels of inspiratory pressure (P15, P20, P25, and no gastric leak) (P < 0.05, Student-Newman-Keuls). Thereafter, however, increasing inspiratory pressure did not result in significantly higher tidal volumes (Fig. 3). At each inspiratory pressure, tidal volume was lower in the group of children who did not experience GI (Table 2), but the difference was not significant. When children's lungs were ventilated with a tidal volume between 8 and 12 mL · kg−1, GI occurred in 30% of children in age group 1, in 10% of children in age group 2, and in 4% of children in age group 3. In most cases, a tidal volume >7 mL · kg−1 was obtained with an inspiratory pressure ≤15 cm H2O.
Figure 3: Tidal volume applied in children with or without gastric insufflation (GI) associated with different inspiratory pressure levels. Data are expressed as median (IQR [interquartile range]). White box represents group without GI (no GI). Dark box represents group with GI occurring at 10 cm H2O inspiratory pressure (P10). Gray boxes represent group with GI occurring at 15, 20, or 25 cm H2O inspiratory pressure, respectively, designated as P15, P20, and P25.*P < 0.05 versus P10 group was considered significant.
Table 2: Tidal Volume Relative to Different Inspiratory Pressures in Children With and Without Gastric Insufflation
No episode of laryngospasm, bronchospasm, cough, movement, or desaturation, i.e., SpO2 <95%, occurred in any child. No gastric overdistension occurred, and insertion of a nasogastric tube was not necessary in any child. No regurgitation occurred in the pharynx during mask ventilation as can be observed during tracheal intubation.
DISCUSSION
GI occurred during facemask ventilation mainly in children younger than 1 year. In adults, it is recommended not to use inspiratory pressures >20 cm H2O to prevent GI.7 In children, no previous study determined the pressure threshold where GI can occur. We now show that the pressure threshold where GI can occur depends on the child's age. It was low in infants and increased with age. In children younger than 1 year, PGI was ≤15 cm H2O and in some cases ≤10 cm H2O. In children older than 1 year, the threshold was ≥15 cm H2O. This is not surprising because the esophageal sphincter is not mature in children younger than 1 year and esophageal sphincter tone increases with age.8 Our results are in accordance with those published by Wahlen et al.9 who found that the pressure threshold for GI with a malpositioned laryngeal mask airway is 17 cm H2O in 3- to 11-year-old children.
GI was detected with epigastric auscultation in this study. Brimacombe et al.5 described that this reliable technique in adults can detect insufflation ≥4 mL air after 1 breath with 95% CI, and the incidence of false negatives was 0% after 4 breaths. Whereas previous comparative studies described prolonged acceptable equilibration time before measuring GI during manual ventilation,3,10,11 we chose a short equilibration time before listening to the epigastric area, as recommended by Brimacombe et al.5 However, the technique we chose was qualitative, thus preventing the risk of gastric overventilation and regurgitation. This technique did not allow a precise evaluation of gastric inflation volume. Measurement of the gastric inflation volume required the use of either a nasogastric tube as described in a group of 10 premature infants11 or a microphone integrated into the head of a stethoscope and taped to the skin in the epigastric area.7 These methods have limitations, and although epigastric auscultation does not quantify gastric inflation volume, it can detect the inspiratory pressure threshold.
When the inspiratory pressure was >15 cm H2O, GI occurred in >58% of cases, which is a higher percentage than in adults. For example, GI occurred in 5% of adult patients ventilated by mask with an inspiratory pressure of 20 cm H2O.4 These results are in agreement with another study in which GI occurred in 30% of patients ventilated by mask with an inspiratory pressure of 20 cm H2O during fiberscope intubation.12 Pressure-controlled ventilation is recommended for decreasing inspiratory pressure during facemask ventilation or ventilation with a laryngeal mask airway.3 Inspiratory pressure was lower for the same tidal volume with pressure-controlled ventilation compared with volume-controlled ventilation. Similar results were reported in children whose lungs were ventilated with a laryngeal mask airway.13,14 In these crossover studies, no GI occurred when children's lungs were ventilated with pressure-controlled ventilation, whereas GI occurred with volume-controlled ventilation in some children.
In children who experienced GI, tidal volumes were higher than in the others at each inspiratory pressure. However, these differences were not significant. This could have been caused by a greater air leakage around the facemask in children who did not experience GI, resulting in a decrease in expired tidal volume and thus preventing GI. Large oropharyngeal leakage was shown to prevent GI in patients whose lungs were ventilated with a laryngeal mask airway.4 Similar results were found in mask-ventilated adults: decreasing the tidal volume from 800 to 500 mL to decrease the peak flow rate led to fewer occurrences of GI.15 In another study, GI occurred during mask ventilation when the tidal volume was >1000 mL.7
This study showed that an inspiratory pressure >15 cm H2O resulted in more occurrences of GI, whereas tidal volume did not change significantly. Moreover, in 77% of cases, an inspiratory pressure ≤15 cm H2O was sufficient to obtain a tidal volume >7 mL · kg−1. This inspiratory pressure could be recommended in children as a standard limit. However, this recommendation holds only in our study conditions, i.e., children's lungs ventilated with pressure-controlled ventilation and with an I/E ratio set at 1:2.
To improve mask ventilation and limit air leakage, we handled the facemask with 2 hands, used an oropharyngeal airway in every child, and applied mandibular advancement if necessary. In adults, this maneuver improved mask ventilation and decreased airway pressure.16 Increasing I/E to 1:1.5 or 1:1 during pressure-controlled ventilation increases tidal volume without changing inspiratory pressure. In this case, a proper tidal volume, i.e., between 7 and 10 mL · kg−1 would have been obtained with a lower pressure, limiting the risk of GI. However, we set the I/E ratio at 1:2 to prevent an intrinsic positive end-expiratory pressure effect. In children younger than 1 year in whom the pulmonary time constant is lower, setting the I/E ratio at 1:1 may make it possible to use a lower pressure to obtain an adequate tidal volume and consequently to decrease the risk of GI.
This study has some limitations. Only 1 child experienced difficult mask ventilation, and respiratory complications did not occur during mask ventilation, i.e., laryngospasm or bronchospasm. Moreover, obese children17 and/or children presenting facial dimorphic features or a cleft palate were not included.
In conclusion, we showed that the PGI in children is age dependent. This threshold is low in infants, in some cases ≤10 cm H2O, and increases with age. In most cases, an inspiratory pressure ≤15 cm H2O is sufficient to provide adequate and safe mask ventilation. Increasing inspiratory pressure above these values is not recommended because it results in more occurrences of GI without increasing the tidal volume.
ACKNOWLEDGMENTS
The authors thank Ray Cooke, PhD (DLC, Université Victor Segalen Bordeaux 2, Bordeaux, France), for proofreading the manuscript.
REFERENCES
1. Moynihan RJ, Brock-Utne JG, Archer JH, Feld LH, Kreitzman TR. The effect of cricoid pressure on preventing gastric insufflation in infants and children. Anesthesiology 1993;78:652–6
2. Ricard J, Boboyono JM, Postel JP, Collet LM, Canarelli JP. Gastric rupture in children. Apropos of a case. Ann Pediatr (Paris) 1989;36:479–81
3. von Goedecke A, Voelckel WG, Wenzel V, Hormann C, Wagner-Berger HG, Dorges V, Lindner KH, Keller C. Mechanical versus manual ventilation via a face mask during the induction of anesthesia: a prospective, randomized, crossover study. Anesth Analg 2004;98:260–3
4. Ho-Tai LM, Devitt JH, Noel AG, O'Donnell MP. Gas leak and gastric insufflation during controlled ventilation: face mask versus laryngeal mask airway. Can J Anaesth 1998;45:206–11
5. Brimacombe J, Keller C, Kurian S, Myles J. Reliability of epigastric auscultation to detect gastric insufflation. Br J Anaesth 2002;88:127–9
6. Triboulet-Beau C, Bordes M, Breton Y, Roux G, Semjen F, Cros AM. Pression d'insufflation maximale lors de la ventilation au masque facial en pédiatrie. Ann Franc Anesth Réan 2003; 22:R337
7. Weiler N, Heinrichs W, Dick W. Assessment of pulmonary mechanics and gastric inflation pressure during mask ventilation. Prehosp Disaster Med 1995;10:101–5
8. Spitz L, McLeod E. Gastroesophageal reflux. Semin Pediatr Surg 2003;12:237–40
9. Wahlen BM, Heinrichs W, Latorre F. Gastric insufflation pressure, air leakage and respiratory mechanics in the use of the laryngeal mask airway (LMA) in children. Paediatr Anaesth 2004;14:313–7
10. Zmora E, Merritt TA. Control of peak inspiratory pressure during manual ventilation.A controlled study. Am J Dis Child 1982;136:46–8
11. Admani M, Yeh TF, Jain R, Mora A, Pildes RS. Prevention of gastric inflation during mask ventilation in newborn infants. Crit Care Med 1985;13:592–3
12. Aoyama K, Yasunaga E, Takenaka I, Kadoya T, Sata T, Shigematsu A. Positive pressure ventilation during fibreoptic intubation: comparison of the laryngeal mask airway, intubating laryngeal mask and endoscopy mask techniques. Br J Anaesth 2002;88:246–54
13. Keidan I, Berkenstadt H, Segal E, Perel A. Pressure versus volume-controlled ventilation with a laryngeal mask airway in paediatric patients. Paediatr Anaesth 2001;11:691–4
14. Bordes M, Semjen F, Degryse C, Bourgain JL, Cros AM. Pressure-controlled ventilation is superior to volume-controlled ventilation with a laryngeal mask airway in children. Acta Anaesthesiol Scand 2007;51:82–5
15. Wenzel V, Idris AH, Banner MJ, Kubilis PS, Williams JL Jr. Influence of tidal volume on the distribution of gas between the lungs and stomach in the nonintubated patient receiving positive-pressure ventilation. Crit Care Med 1998;26:364–8
16. Isono S. One hand, two hands, or no hands for maximizing airway maneuvers? Anesthesiology 2008;109:576–7
17. Jolicoeur P, Pontier J, Pernin MO, Sempe M. A lifetime asymptotic growth curve for human height. Biometrics 1988;44:995–1003
