Unanticipated difficult airway comprises of unexpected difficult bag mask ventilation or unforeseen difficult laryngoscopy. The incidence of difficult laryngoscopy or unanticipated difficult intubation in children was varied from 1.2 %to 4.77% depending on general or specific population and type of surgery. The known risk factors of difficult intubation in children were young age, associated syndrome or congenital abnormality and obstructive sleep apnea. Moreover, the predictors of difficult laryngoscopy by physical examination were associated with short interincisors distance, high frontal plane to chin distance, short thyromental distance and Cormack & Lehane classification 3 or 4. However, little knowledge is known regarding difficult bag mask ventilation in children. The incidence of difficult bag mask ventilation was 6.6% according to the single study. A few study reported the independent risk factor of difficult bag mask ventilation in children which were young age, obesity, use of neuromuscular blocking agent and airway surgery. In addition, the association between difficult bag mask ventilation and difficult intubation are still unknown. To understand more of difficult bag mask ventilation in children and factor-association may reduce incidence of morbidity and mortality by identifying difficult airway, preparing personnel and equipment tool in order to improve clinical outcome in pediatric anesthesia. The objectives of the study were to determine the predictors of difficult bag mask ventilation and the association with unexpected difficult intubation in children who came for elective surgery in tertiary care hospital of southern Thailand.

Study protocol and anesthesia practice Patient's demographic and history of snoring or asthma/hyper-reactive airway were obtained by research assistant at preinduction period. Standard monitoring including non-invasive blood pressure, pulse oximetry, electrocardiography and capnography were applied to all patients before starting general anesthesia. All patients were received general anesthesia with oroendotracheal tube intubation. Induction techniques were either inhalation anesthetic agent by N2O-O2-sevoflurane or intravenous agent with 100% O2. Type of anesthetic circuit used (Circle or Jackson-Rees) was depending on anesthesiologist in charge. Patients head position were in neutral under small head ring to prevent head movement during induction. Maneuvers on bag mask ventilation such as application of CPAP with opening airway maneuvers, oropharyngeal airway or two person ventilations were recorded by research assistant at induction period. When positive pressure ventilation was difficult after the patients had been slept, the CPAP ≥5 cmH2O was applied incorporated with opening airway by head tilt/chin lip or jaw thrust until assisted spontaneous ventilation was successful. BMV was performed during induction with 100% O2. If performing BMV was possible, a non-depolarizing muscle relaxant was administered to facilitate the laryngoscopy procedure. If performing BMV was difficult, deep anesthesia using propofol, or a non-depolarizing muscle relaxant, or succinylcholine was given to facilitate the BMV procedure. The anesthesia personnel who handled the BMV also performed the first laryngoscopy. The laryngoscopists included anesthesia residents (more than 1-year experience), nurse anesthetists (at least 1-year training), and anesthesiologist attendants. Number of intubation attempts, Cormack-Lehane laryngoscopic view, intubation time in second and presence of desaturation (SpO2 < 95%) were recorded by research assistant at intubation period. There were many optioned used to aid successful intubation such as applying BURP maneuver, endotracheal tube with stylet used, change to smaller endotracheal tube size, change laryngoscope type or change to more experience laryngoscopists. Attending anesthesiologists take responsibility over anesthesia resident and nurse anesthetists for first or second failure of intubation attempts.

Outcome of interest Primary outcomes were to determine the predictors of difficult bag mask ventilation. The difficult bag mask ventilation was defined as the occurrence at least 2 following events which were requiring application of CPAP with opening airway maneuvers, requiring oropharyngeal airway or nasopharyngeal airway, requiring 2- person bag mask ventilation, unable to perform bag mask ventilation or presence of desaturation (SpO2 < 95%).

Secondary outcome was to find out the association between difficult bag mask ventilation and unexpected difficult laryngoscopy/intubation.

Difficult laryngoscopy was defined as the Cormack-Lehane laryngoscopic view grade 3 or 4. Difficult intubation was defined as the occurrence of at least two of the following: 1) at least three intubation attempts using a conventional laryngoscope blade, either Miller or Macintosh; 2) intubation time (the summation of all attempts from applying a laryngoscope to the endotracheal tube that passed the vocal cords) >300 seconds; and 3) the occurrence of desaturation (SpO2 <95%)The intubation time was defined as the time starting from applying laryngoscope to presence of positive capnography wave form. In case of fail intubation, the intubation time started from applying laryngoscope to remove laryngoscope from children's mouth. The total intubation time was the combined timing of each intubation attempts

Inclusion Criteria:

• aged up to 9 years
• ASA physical status of 1-3
• came for general anesthesia with oroendotracheal tube intubation
• elective surgery

Exclusion Criteria:

• had difficult airway or having airway anomaly or syndrome related difficult airway
• presence of active pulmonary disease
• had preoperative oxygen saturation at room air < 95%

We recruited children aged up to 9 years, ASA physical status of 1-3, who came for general anesthesia with oroendotracheal tube intubation (OETT) for elective surgery at Songklanajarind Hospital

data was also sent to Anesthesia Department, McGill University, Canada